Copyright and Collective Authorship

COPYRIGHT AND COLLECTIVE
AUTHORSHIP
DANIELA SIMONE
WORCESTER COLLEGE
UNIVERSITY OF OXFORD
SUPERVISOR: PROF GRAEME DINWOODIE
DPHIL
Trinity Term 2014

ABSTRACT
COPYRIGHT AND COLLECTIVE AUTHORSHIP
DANIELA SIMONE, WORCESTER COLLEGE
DPHIL, TRINITY TERM 2014
Many scholars have suggested that current copyright law is ill-equipped to the
challenges of determining the authorship of collaborative work. This thesis analyses
four case studies of large scale collaboration (Wikipedia, Indigenous art, scientific
collaborations and film) in order to consider how best to determine the authorship of the
creative works that they produce for the purposes of copyright law. Current scholarship
and much of the case law has tended to favour a restrictive approach to the grant of
joint authorship status, in order to minimise the number of potential authors of a work.
This is motivated by instrumental/pragmatic concerns related to the ease of exploiting a
copyright work. As joint authors are often joint first owners of copyright, proponents of
this approach fear that a minor contributor might cause hold-up problems by refusing to
consent to licence or assign their copyright interest. This thesis argues that an
instrumental/pragmatic approach to the application of the joint authorship test is
undesirable, because it distances the test both from the creativity reality of collective
authorship and from copyright’s notion of the author. In addition, the
instrumental/pragmatic approach relies upon assumptions about creators, the creative
process and the exploitation of creative works which are not borne out in the case
studies.
Building on the insights from the four case studies, the thesis argues that the best
approach to applying the joint authorship test to works of collective authorship is one
that is inclusive (of all those who have made a more than de minimis contribution of
creative choices to the protected expression) and contextual (in that it takes the context
of creativity into account). In coming to this conclusion the thesis also offers broader
lessons about the nature of authorship and the ongoing relevance of copyright law
standards for the regulation of collaborative creativity.
ACKNOWLEDGMENTS
This thesis would not have been possible without the support of many people. My thesis
has benefited enormously from the formidable expertise and great enthusiasm of my
supervisor, Graeme B. Dinwoodie. I am sincerely grateful to Professor Dinwoodie for
being so generous with his time, for attentive comments on countless drafts, for patient
level-headed responses to moments of imagined crisis and for challenging my thinking
in productive and provocative ways. This project may not have seen completion were it
not for his constant encouragement, especially when my focus was diverted by difficult
personal circumstances.
I am grateful to my examiners, Mireille van Eechoud and Dev Gangjee, for making the
viva voce examination a thought-provoking and enjoyable experience. Their comments
will serve as an invaluable resource for the future development of this work.
I was fortunate to have the opportunity to present my work in progress at a number of

conferences. I am grateful for helpful feedback from participants at graduate
conferences at Queen Mary (2011, 2013), King’s College London (2011, 2013) and
Oxford (2009). My thesis has also benefited from the comments of participants at the
Intellectual Property Scholars Conference at DePaul University (2011) and the
‘Authorship Dynamics and the Dynamic Work’ HERA Workshop held at the
University of Cambridge (2012).
I owe a particular debt of gratitude to a number of intellectual property scholars whose

generous comments and observations on various parts of this thesis have helped in its
development, in particular, Tanya Aplin, Lionel Bently, Laura Biron, Christopher
Buccafusco, Robert Burrell, Elena Cooper, Giuseppina D’Agostino, Séverine Dusollier,
Michael Fraser, Robert Gomulkiewicz, Jonathan Griffiths, Michael Handler, Emily
Hudson, Marta Iljadica, Phillip Johnson, Barbara Lauriat, Justine Pila, Uma
Suthersanen, David Vaver, Kimberlee Weatherall. It would be amiss not to thank the
members of the Oxford Intellectual Property Research Centre, the fantastic group of IP
doctoral students at the University of Oxford (Naomi, Alison, Quentin, Jaani, Graham,
Tom, Karen, Poorna, Henry) and members of the Oxford Law Faculty for many
conversations on fine points of copyright law as well as broader philosophical points.
Particular thanks to Liz Fisher whose energy and great empathy brought the Course in
Legal Research Methodology alive.
Whilst at the University of Oxford I have benefited from the support and assistance of
many administrators and librarians at the Faculty of Law, Worcester College and the
International Office. In particular, I would like to thank Geraldine Malloy, Cath Fraser,
Elizabeth Smith, Sandy Meredith, Caroline Norris, Marianne Biese, Ellen Moilanen,
and Jenny Hassan.
Worcester College provided me with a very supportive atmosphere in which to study. I
am grateful to Donal Nolan for his invaluable advice and to Cathryn Costello for words
of wisdom, which set me upon this path towards the DPhil. I will always remember
Worcester College MCR as a refuge in stressful times, a community of supportive
peers, and the starting place of many great friendships.
I would like to express my gratitude to the many extraordinary teachers who have
encouraged and inspired me over the years, particularly, Deirdre Coleman, Elaine
Moore, Mary Crock, Peter Gerangelos, Hilary Astor, and Patricia Loughlan.
A very big thank you to my dear friends Tania, Nahal, Sam, Megan, Natasa, Bipana and
Erie for their kindness, understanding, support, and for sacrificing so much time at such
short notice to help proof-read parts of this thesis. I would also like to thank good
friends near and far for their encouragement and support, especially: Amanda, Jen,
Rani, Alicia, Chrissy, Fish, Leo, Sarah, Annie, Vasiliki, Maria, Laura, Dan, Chloe,
Clément, May, Rita and Tim. I am very grateful to Cristina, Luciano, Alessandro and
Andrea’s many wonderful friends from Genoa and beyond for their encouragement and
hospitality.
Most importantly, I am grateful to all of my family. My parents’ selfless love created

the many opportunities with which my life has been blessed. They have been a source
of strength and constant support both morally and materially. As my first teachers, the
life lessons that they taught are always with me. My brothers, Mike and Nick (their
partners Tonia and Steph), and my sister, Cath, have loved, supported, and encouraged
me at every difficult turn. My grandparents’ kindness, courage and hard work have
been a great inspiration to me. Thanks to Juliet, Peter, Chris and Sr Teresita for their
prayers and thoughtful cards. Finally, words cannot adequately express my gratitude for
Andrea, who has been there at every step of this journey, for his love and unwavering
support, for his patient counsel, for rigorously challenging my ideas, and for always
believing in me.
Table of Contents
Table of Abbreviations ...................................................................................................... i

Table of Cases ................................................................................................................. iii
Table of Statutes .............................................................................................................. ix
Chapter 1: Introduction .................................................................................................... 1

1.1 Copyright Law and Collective Authorship ............................................................. 1
1.2 Method and Structure of the Thesis ........................................................................ 4
1.3 An Inclusive, Contextual Approach to the Joint Authorship Test .......................... 8
1.3.1 The Joint Authorship Test Should Be Inclusive ............................................ 10
1.3.2 The Joint Authorship Test Should Be Applied Contextually ........................ 11
1.4 Broader Lessons for Copyright Law ..................................................................... 13
Chapter 2: Authorship and Joint Authorship ................................................................. 16

2.1 The Concept of Authorship in the CDPA ............................................................. 17
2.2 The Joint Authorship Test..................................................................................... 29
2.2.1 Contribution Not Distinct .............................................................................. 30
2.2.2 Collaboration or Common Design ................................................................. 31
2.2.3 A ‘Significant’ Contribution of the ‘Right Kind’ .......................................... 33
2.2.4 Questions of Law, Questions of Fact ............................................................. 38
2.3 Three Critiques of the Application of the Joint Authorship Test ......................... 40
2.3.1 The Factual Specificity of the Joint Authorship Test .................................... 41
2.3.2 The Instrumental or Pragmatic Approach ...................................................... 42
2.3.3 A Preoccupation with Aesthetic Neutrality ................................................... 48
2.4 Copyright Scholarship: Theories of Authorship ................................................... 51
2.5 Conclusion ............................................................................................................ 62
Chapter 3: Wikipedia ...................................................................................................... 65

3.1 Authorship Dynamics: Promoting Sharing ........................................................... 66
3.2 Copyright Subsistence on Wikipedia .................................................................... 72
3.2.1 Is Wikipedia (or Parts Thereof) an Original Literary Work? ........................ 72
3.2.2 Are Wikipedia Contributors Copyright Authors? .......................................... 82
3.3 Copyleft Licences and the Ambivalent Role of Copyright Law .......................... 87
3.4 Insights for Copyright Law ................................................................................... 94
Chapter 4: Australian Indigenous Art ........................................................................... 100
4.1 Indigenous Art .................................................................................................... 101
4.1.1 Authorship Dynamics: Building Cultural Identity....................................... 102
4.1.2 Background to the Issue of Protecting Indigenous Cultural Expressions .... 105
4.2 Protecting Indigenous Art with Copyright.......................................................... 110
4.3 Other Solutions for the Protection of Indigenous Cultural Expressions ............. 128
4.3.1 Protocols and Codes of Conduct .................................................................. 129
4.3.2 Collective/Certification Trade Marks .......................................................... 130
4.3.3 Contract ........................................................................................................ 132
4.3.4 Sui Generis Legislation ................................................................................ 133
4.4 Insights for Copyright Law ................................................................................. 134
Chapter 5: Scientific Collaborations ............................................................................. 139

5.1 Authorship Dynamics: Constructing Authority .................................................. 140
5.2 Regulating Authorship with Private Ordering .................................................... 150
5.2.1 Biomedical Science Collaborations: An Authorship Crisis ......................... 150
5.2.2 Particle Physics Collaborations: The Bureaucratisation of Authorship ....... 158
5.3 The Application of Copyright Law ..................................................................... 162
Chapter 6: Film ............................................................................................................. 177

6.1 Authorship Dynamics: The Pragmatic Value of Authorship .............................. 179
6.2 The Subsistence of Copyright ............................................................................. 190
6.2.1 A Brief Historical Note ................................................................................ 191
6.2.2 The Complexity of Film Copyright ............................................................. 193
6.2.3 Film as a First Fixation ................................................................................ 198
6.2.4 Film as a Dramatic Work ............................................................................. 203
6.2.5 The Pitfalls of Pragmatic Reasoning ........................................................... 211
6.3 Private Ordering .................................................................................................. 214
Chapter 7: Characteristics of Collective Authorship and the Role of Copyright Law . 226
7.1 The Nature of Collective Authorship.................................................................. 228
7.2 The Different Meanings of Authorship for each Collective Authorship Group . 234
7.2.1 Authorship Has a Different Meaning in Each Collective Authorship Group
.............................................................................................................................. 235
7.2.2 Authorship Signifies Responsibility for the Work ...................................... 238
7.2.3 Authorship Signals Belonging to a Particular Community ......................... 240
7.2.4 The Attribution of Authorship Can Be Affected by Power Dynamics ........ 242
7.3 The Gap between Copyright Law and Creative Realities ................................... 243
7.4 Private Ordering: Bridging the Gap between Copyright Law and Creative Reality
.................................................................................................................................. 247
7.4.1 Successful Examples of Private Ordering ................................................... 248
7.4.2 Less Successful Examples of Private Ordering ........................................... 251
7.4.3 The Benefits and Limitations of Relying Upon Private Ordering ............... 253
7.5 The Role of Copyright Law and its Concepts ..................................................... 265
7.6 Summary ............................................................................................................. 273
Chapter 8: An Inclusive, Contextual Approach to the Joint Authorship Test .............. 275
8.1 The Relevance of Social Norms ......................................................................... 277
8.2 The Dangers of Deferring to Social Norms ........................................................ 285
8.3 A Framework for Considering Social Norms ..................................................... 289
8.4 Revisiting the Critiques of the Joint Authorship Test......................................... 295
8.4.1 The Factual Specificity of the Joint Authorship Test .................................. 296
8.4.2 The Preoccupation with Aesthetic Neutrality .............................................. 296
8.4.3 The Instrumental/Pragmatic Approach ........................................................ 297
8.5 An Inclusive and Contextual Approach to the Joint Authorship Test ................ 302
8.6 Conclusion .......................................................................................................... 310
Bibliography ................................................................................................................. 314

Table of Abbreviations
1911 Act – Copyright Act 1911
1956 Act – Copyright Act 1956
Bently and Sherman – L Bently and B Sherman, Intellectual Property Law (OUP 2009)
Berne – Berne Convention for the Protection of Literary and Artistic Works of 9
September 1886
Copinger et al – KM Garnett, G Davies, G Harbottle, WA Copinger and EP Skone

James, Copinger and Skone James on Copyright (16th edn, Sweet & Maxwell 2011)
Cornish et al – W Cornish, D Llewelyn and T Aplin, Intellectual Property: Patents,

Copyright, Trade Marks and Allied Rights (8th edn, Sweet & Maxwell 2013)
CDPA – Copyright Designs and Patents Act 1988
Database Directive – Directive 96/9/EC of the European Parliament and of the Council
of 11 March 1996 on the Legal Protection of Databases
Duration Directive – Directive 2006/116/EC of the European Parliament and of the

Council of 12 December 2006 on the term of protection of copyright and related rights
(codified version) (repealing and replacing Council Directive 93/98/EEC of 29 October
1993 harmonizing the term of protection of copyright and certain related rights)
Gregory Report – ‘Report of the Board of Trade Copyright Committee’ (Her Majesty’s
Stationery Company October 1952)
InfoSoc Directive – Directive 2001/29/EC of the European Parliament and of the

Council of 22 May 2001on the harmonization of certain aspects of copyright and
related rights in the information society
IPQ – Intellectual Property Quarterly
JIPLP – Journal of Intellectual Property Law and Practice
i
Laddie et al – H Laddie, P Prescott and M Vitoria, The Modern Law of Copyright and
Designs (4th edn, LexisNexis 2011)
Myer Report – Commonwealth Department of Communications, Information

Technology and the Arts (Australia), ‘Report of the Contemporary Visual Arts and
Crafts Inquiry’, June 2002
Rental and Related Rights Directive – Directive 2006/115/EC of the European

Parliament and of the Council of 12 December 2006 on rental right and lending right
and on certain rights related to copyright in the field of intellectual property (codified
version) (replacing Council Directive 92/100/EEC of 19 November 1992 on rental and
lending right and on certain rights relating to copyright in the field of intellectual
property)
Senate Report – Australian Senate Standing Committee on Environment,

Communications, Information Technology and the Arts, ‘Indigenous Art – Securing the
Future: Australia’s Indigenous Visual Arts and Craft Sector’, June 2007
ii
Table of Cases
United Kingdom
Abraham Moon v Thornber [2012] EWPCC 37, [2013] FSR 17 23, 24

Adventure Film Productions v Tully [1993] EMLR 376 (Ch) 199
Anya v Wu [2004] EWCA Civ 755 162, 171, 262
Baigent v Random House Group [2006] EWHC 719, [2006] EMLR 16 74
Baigent v Random House Group [2007] EWCA Civ 247, [2008] EMLR 7 74
Bamgboye v Reed [2002] EWHC 2922, [2004] 5 EMLR 61 42, 149, 173
Beckingham v Hodgens [2002] EWHC 2143 (Ch), [2002] EMLR 45
30-1, 35, 50, 85, 173
Beckingham v Hodgens [2003] EWCA Civ 143, [2003] EMLR 18
32, 42, 84-5, 206-7, 305
Beechwood House Publishing t/a Binley’s v Guardian Products [2010] EWPCC 1200
81
Beggars Banquet Records v Carlton Television [1993] EMLR 349 (Ch) 200
Biotrading and Financing Oy v Biohit [1996] FSR 393 (Ch) 75, 78
Biotrading and Financing Oy v Biohit [1998] FSR 109 (CA) 75
Blacklock v Peterson [1915] 2 Ch 376 22
Brighton v Jones [2004] EWHC 1157, [2005] FSR 288 25, 36-7, 41, 44-5, 74, 162, 175
Brown v Mcasso [2005] FSR 846 (EWPCC) 33, 35, 50
Brown v Mcasso [2005] EWCA Civ 1546, [2006] FSR 480 33, 50
Brutus v Cozens [1973] AC 854 (HL) 39
Cala Homes v Alfred McAlpine Homes [1995] EWHC 7, [1995] FSR 818
20, 30, 37-8, 56-7, 74, 83, 124, 174, 239, 272, 309
Century Communications v Mayfair Entertainment [1993] EMLR 335 200
Chapman v Smith (1754) 2 Ves Sen 506, 28 ER 324 (Ch) 293
Chappell v Redwood Music [1980] 2 All ER 817, [1981] RPC 337 (HL) 30, 75, 207
Coffey v Warner/Chappell Music [2005] EWHC 449 (Ch), [2005] FSR 34 79, 198
Creation Records v News Group Newspapers [1997] EMLR 444 (Ch) 194-5
Cummins v Bond (1926) 1 Ch 167 20
Donoghue v Allied Newspapers [1938] 1 Ch 106 (Ch) 25, 37-8, 83, 162, 174
iii
Dramatico Entertainment v British Sky Broadcasting [2012] EWHC 268, [2012] RPC
27 204
Elanco Products v Mandops (Agrochemical Specialists) [1980] RPC 213 (CA) 22, 75
Exxon v Exxon Insurance [1982] Ch 119 (CA) 23, 50, 73
Fisher v Brooker and Onward Music [2006] EWHC 3239, [2007] FSR (12) 255 34
Fisher v Brooker [2009] UKHL 41 [2009] 1 WLR 1764 42, 149, 204
Football Association Premier League v QC Leisure (No 2) [2008] EWHC 1411 (Ch)
199
Francis Day and Hunter v 20th Century Fox [1940] AC 112 (PC) 73
Fylde Microsystems v Key Radio Systems [1998] FSR 449 (Ch)
25, 34, 37, 83, 117, 162, 174, 209
George Hensher v Restawile Upholstery [1976] AC 64 (HL) 48-9
Godfrey v Lees [1995] EMLR 307 (Ch) 29, 33, 82, 149
Green v Broadcasting Corp New Zealand [1989] 2 All ER 1056, [1989] RPC 700 (PC)
75, 203
Hadley v Kemp [1999] EMLR 589 (Ch) 25, 29, 35-7, 43-45, 82-3, 173-5, 208, 231, 280
Hollinrake v Truswell [1894] 3 Ch 420 (CA) 49, 73
Interlego v Tyco Industries [1989] AC 217 (PC) 25, 73, 75-6, 124, 280
IPC Media v Highbury-Leisure Publishing [2004] EWHC 2985, [2005] FSR 20
75-6, 79, 198
Kelly v Cinema Houses Ltd [1928-35] Macq Cop Cas 362 193
Kenrick v Lawrence (1890) 25 QBD 99 117
King Features Syndicate v O and M Kleeman [1941] AC 417 (HL) 193
LA Gear v Hi-Tec Sports [1992] FSR 121 (CA) 74
Ladbroke v William Hill [1964] 1 WLR 273, [1964] 1 All ER 465 (HL) 22, 23, 72
Lauri v Renad [1892] 3 Ch 402 42
Levy v Rutley (1871) LR 6 CP 523 31, 44, 84, 163
Lucasfilm v Ainsworth [2008] EWHC 1878, [2009] FSR 103
25, 49, 193-7, 205, 208, 268
Lucasfilm v Ainsworth [2009] EWCA Civ 1328, [2010] Ch 503
194-7, 205, 208, 268-9, 304
Lucasfilm v Ainsworth [2011] UKSC 39, [2012] 1 AC 208
195-7, 205, 208, 254 268, 304
MacMillan v Cooper (1924) 40 TLR 186 (PC) 75-6
MacMillan Publishers v Thomas Reed Publications [1993] FSR 455 (Ch) 77, 86
iv
Mail Newspapers v Express Newspapers [1987] FSR 90 (Ch) 42, 302
Merchandising Corporation of America v Harpbond [1983] FSR 32 (CA) 111, 194
Millar v Taylor (1769) 4 Burr 2303, 98 ER 201 (Ct of KB) 292
Mirage Studios v Counter-Feat Clothing [1991] FSR 145 (Ch) 193
Moyna v Secretary of State for Work and Pensions [2003] UKHL 44, [2003] 1 WLR
1929 39
Newspaper Licensing Agency v Meltwater [2011] EWCA 890 23, 73
Noah v Shuba [1991] FSR 14 (Ch) 165, 171, 262
Norowzian v Arks (No 1) [1998] FSR 394 (Ch) 49, 198-9, 204
Norowzian v Arks (No 2) [2000] FSR 363 (CA) 178, 197, 203-5, 209, 223
Nova Productions Limited v Mazooma Games [2006] EWHC 24 203
Nova Productions Limited v Mazooma Games [2007] EWCA Civ 219; [2007] RPC 25
203
O’Neill v Paramount [1983] CAT 235 193
Powell v Head (1879) 12 Ch D 686 42
R v Higgs [2008] EWCA 1324, [2009] 1 WLR 73 199
Redwood Music v Chappell (1982) RPC 109 (QB) 30, 75, 207
Rexnold v Ancon [1983] FSR 245 (HC) 75
Robin Ray v Classic FM [1998] FSR 622 (Ch) 20, 29, 74, 82-3, 173, 174
Samuelson v Producers Distributing [1932] 1 Ch 201, (1932) 48 RPC 580 39, 260
SAS Institute v World Programming [2013] EWCA Civ 1482, [2014] RPC 8 23
Sawkins v Hyperion Records [2005] EWCA 565, [2005] 1 WLR 3281
25, 48, 76, 122, 124, 127, 254, 280
Shelley Films v Rex Features [1994] EMLR 134 194
Slater v Wimmer [2012] EWPCC 7 200-1, 221, 223-4, 268, 307
Spelling Goldberg Productions v BPC Publishing [1981] RPC 283 (CA) 198-9
Springfield v Thame (1903) 89 LT 242 (Ch) 31, 44
Stuart v Barrett [1994] EMLR 448 (Ch) 42, 45, 83-4
Sweeney v Macmillan Publishers [2002] RPC 35 (Ch) 74
Tate v Fullbrook [1908] 1 KB 821 (CA) 208-9
Tate v Thomas [1921] 1 Ch 503 209
Taylor v Maguire [2013] EWHC 3804, [2014] ECDR 4 23-4
Taylor v Rive Droite Music [2004] EWHC 1605, [2004] All ER 88 74
Taylor v Rive Droite Music [2005] EWCA 1300, [2005] All ER 72 74
v
Temple Island Collections v New English Teas [2012] EWPCC 1, [2012] All ER 49
23-4
The Reject Shop v Manners [1995] FSR 870 (DC) 73
Ultra Marketing v Universal Components [2004] EWHC 468, [2004] All ER 229 75
University of London Press v University Tutorial Press [1916] 2 Ch 601 (Ch) 21-3, 72
Walter v Lane [1900] AC 539 (HL) 20-1, 25, 56
Waterlow Directories v Reed Information Services [1992] FSR 409 (Ch) 22
Wiseman v George Weidenfeld & Nicolson Ltd [1985] FSR 525 (Ch) 39, 163, 260
European Union
C-203/02 British Horseracing Board v William Hill [2004] ECR I-10415; [2005] RPC
260 81-2
C-444/02 Fixtures Marketing Ltd v Organismos Prognostikon Agonon Podosfairou AE
[2004] ECR I-10549 80-1
C-545/07 Apis-Hirstovich EOOD v Ladorka AD [2009] ECR I-1627 80
C-5/08 Infopaq v Danske Dagblades Forening [2009] ECR I-6569
18, 23, 72-3, 76, 79, 81, 197, 255
C-403/08 & C-429/08 Football Association Premier League v QC Leisure [2012] 1
CMLR 29, [2012] ECDR 8 23, 25, 81, 204, 255, 307
C-393/09 Bezpecnostn´ı softwarováasociace v Ministerstvo Kultury [2011] ECDR 3,
[2011] FSR 18 23-5, 81, 255, 307
C-145/10 Painer v Standard Verlags [2012] ECDR 6 23-5, 81, 209
C-277/10 Luksan v van der Let [2013] ECDR 5 19
C-604/10 Football Dataco v Yahoo! UK [2012] ECDR 10 23-4, 81
Australia
Aboriginal Sacred Sites Protection Authority v Maurice; Re the Warumbingu Land

Claim [1986] FCA 90, (1986) 10 FCR 104 (Fed. Ct. of Australia) 128
Aristocrat Leisure Industries v Pacific Gaming [2000] FCA 1273 (Fed. Ct. of Australia)
127
Australian Competition and Consumer Commission v Australian Dreamtime Creations
[2009] FCA 1545 (Fed. Ct. of Australia) 128
BP Refinery (Westernport) Pty Ltd v Hastings Shire Council (1977) 180 CLR 266 (PC)
118
vi
Bulun Bulun v Nejlam Investments (unreported, Federal Court of Australia, Darwin
1989) 104, 112
Bulun Bulun v R & T Textiles (1998) 86 FCR 244 (Fed. Ct. of Australia)
44, 100-1, 111-2, 115-128, 135-8, 239, 272, 308-9
Burge v Swarbrick [2007] HCA 17 (High Court of Australia) 48
Foster v Mountford [1976] 29 FLR 233 (NT SC) 128
Kalamazoo v Compact Business Systems (1983) 5 IPR 213 (Qld SC) 126
Milpurrurru v Indofurn Pty Ltd (1994) 54 FCR 240 (Fed. Ct. of Australia)
104, 110, 113-15, 122, 128, 135, 231
Neowarra v Western Australia [2003] FCA 1402 (Fed. Ct. of Australia) 118
Prior v Lansdowne Press [1977] RPC 511, [1977] FLR 59 (Vic SC) 42, 163
Sega Enterprises v Galaxy Electronics (1996) 35 IPR 161, (1997) 37 IPR 462 (Fed. Ct.
of Australia) 126-7
Stevens v Kabushiki Kaisha Sony Computer Entertainment [2005] HCA 54, 224 CLR
193 (High Court of Australia) 127
Western Australia v Ward [2002] HCA 28, (2002) 213 CLR 1 (High Court of Australia)
118
Yumbulul v Reserve Bank of Australia [1991] FCA 332, 21 IPR 481 (Fed. Ct. of
Australia) 110, 112-13, 116, 133, 136, 258, 299
United States
Aalmuhammed v Lee 202 F3d 1227 (9th Cir, 2000) 43, 58, 191, 211-13, 221, 224-5
Bridgeman Art Library v Corel 36 F Supp 2d 191 (SDNY, 1999) 122
Childress v Taylor 945 F2d 500 (US CA 2nd Cir, 1991) 32, 36, 58
Chou v University of Chicago 254 F3d 1347 (Fed Cir, 2001) 156, 165
Erickson v Trinity Theatre 13 F3d 1061 (7th Cir, 1994) 36, 58
Gaiman v McFarlane 360 F3d 644 (7th Cir, 2004) 36, 58
th
Garcia v Google (Unreported, Docket Number: 12-57302) (9 Cir, 2014) 213-14
Jacobsen v Katzer 535 F 3d 1373 (US CA Fed Cir, 2008) 89
Johnson v Schmitz 119 FSupp 2d 90 (D Conn, 2000) 156, 165
Thomson v Larson 147 F3d 195 (2d Cir, 1998) 36
Weissmann v Freeman 684 FSupp 1248 (SDNY, 1988) 164-7, 169, 262
Weissmann v Freeman 868 F2d 1313 (2nd Cir, 1989) 164-7, 169, 262
vii
Other Jurisdictions
Carlos RL v Javier AA and Maria Cruz DA [2002] ECDR 23 (Madrid Court of Appeal
12th Section, Spain) 164
Case II CSK 527/10, 22 June 2010 (Polish Supreme Court) 308
Elisha Qimron v Hershel Shanks [1993] 7 EIPR D-157 (Israeli SC) 76
Fabrikant c. Swamy (2011) QCCS 1385 (Superior Court of Quebec, Canada) 165
Heptulla v Orient Longman [1989] 1 FSR 598 (Indian High Court) 38, 174
Neudorf (Darryl) v Nettwerk Productions [2000] RPC 935 (British Columbia SC,
Canada) 32
viii
Table of Statutes
UK
Copyright Act 1911 191
Copyright Act 1956 26, 30, 192, 203, 207
Copyright Designs and Patents Act 1988 2, 5,
12, 15-19, 24, 26-9, 39, 42, 46, 49, 50, 53, 55-6, 62, 66, 72, 80, 82-3, 85, 90, 92, 96,
100, 162, 177, 179, 184, 192, 194-5, 198-9, 202-3, 205, 210, 234, 240, 244, 254-5, 260,
264, 271, 281, 284, 287, 297, 306-7, 310
Dramatic Copyright Act 1833 191
Fine Arts Copyrights Act 1862 191
Europe
92/100/EEC Rental and Related Rights Directive 1992 26, 198
93/98/EEC Duration Directive 1993 26, 184, 198
96/9/EC Database Directive 1996 24, 80
2001/29/EC Information Society Directive 2001 23
Australia
Copyright Amendment (Indigenous Communal Moral Rights) Bill (Cth) (Draft)
109, 129
Copyright Act 1968 (Cth) 100, 114, 117
Native Title Act 1993 (Cth) 116-17
Resale Royalty Right for Visual Artists Act 2009 (Cth) 129
Trade Marks Act 1995 (Cth) 130
Other Jurisdictions
Berne Convention 27, 205
United States Code, Title 17 (Copyright Act 1976) (United States) 27
Pacific Model Law (Model for the Protection of Traditional Knowledge and
Expressions of Culture 2002) 134
ix
Chapter 1: Introduction
This thesis analyses four cases of large scale collaboration (Wikipedia, Indigenous art,
scientific collaborations and film) in order to consider how to best determine the
authorship of the creative works they produce, for the purposes of copyright law.
Building upon insights from these case studies, I argue that the joint authorship test
should be applied in an inclusive and contextual way. In coming to this conclusion, this
thesis also offers broader lessons about the nature of authorship and the ongoing
relevance of copyright law standards for the regulation of collaborative creativity.
1.1 Copyright Law and Collective Authorship
This thesis is written at a moment when copyright law appears to be suffering from a
crisis of legitimacy 1 . In recent decades, the successful lobbying of entertainment
conglomerates and the internationalisation of copyright law has led to the expansion of
copyright protection. This has resulted in a copyright regime which is often accused of
being geared more towards protecting the corporations involved in producing and
distributing creative works, than it is towards rewarding and incentivising authors. At
the same time, non-compliance with copyright law is becoming increasingly
widespread, and in some quarters, normalised (viz. the anti-copyright law platform of
the Pirate Party). The Creative Commons and the Free Software movements, which cast
themselves as an ‘ethical alternative’ to copyright, are also gaining popularity. As
1
See: S Dusollier, ‘Open Source and Copyleft: Authorship Reconsidered?’ (2003) 26 Columbia J of L
and Arts 281: ‘The institution of copyright is in ill repute these days’; N Elkin-Koren, ‘Tailoring
Copyright to Social Production’ (2011) 12(1) Theoretical Inquiries in L 309, 310: ‘… a regime in crisis’.
1
copyright law is frequently accused of being out of touch with modern creative realities,
non-compliance may appear unsurprising 2 . Indeed, psychologists have demonstrated
that people are more likely to obey laws they consider to be legitimate and fair3.
In light of this legitimacy crisis, a search for the best way to apply the joint
authorship test ought to begin with the reality of creativity4. As Jane Ginsburg argued
just over a decade ago, refocusing on the author and the act of creating may help restore
a proper perspective on copyright law5. In this spirit, this thesis focuses on the act of
creating a work of collective authorship and, in particular, analyses the dynamics of
creativity in four cases of large scale collaboration. Throughout the thesis, a ‘work of
collective authorship’ refers to any work that is created by many contributors6.
Large scale collaboration is becoming increasingly widespread and economically
significant. This is due, in large part, to advances in digital and communications
technology. In their best-selling book, Wikinomics, Don Tapscott and Anthony
2
T Tyler, ‘Compliance with Intellectual Property Law: A Psychological Perspective’ (1996) 29
International L and Politics 219, 227 arguing that people are more likely to obey a law that reflects public
morality.
3
For example, the important work of T Tyler, Why People Obey the Law (Princeton UP 2006) and ibid.
4
RR Kwall, The Soul of Creativity: Forging a Moral Rights Law for the US (Stanford UP 2009) 5 draws
upon Tyler’s work to argue that laws governing authors’ rights are likely to be ignored if they fail to
embrace widely shared norms regarding authorship.
5
J Ginsburg, ‘The Concept of Authorship in Comparative Copyright Law’ (2003) 52 De Paul L Rev
1063, 1071.
6
This term is not intended to be confined to the meaning of ‘collective work’ in s178 of the Copyright
Designs and Patents Act 1988 (‘CDPA’), although most works of collective authorship are likely to fall
within this definition. ‘Collective’ is preferred to ‘collaborative’ to avoid confusion when considering
which contributors might be considered to be joint authors of such work (given that ‘collaboration’ is a
requirement for joint authorship). The term ‘group authorship’ has been avoided, as it might seem to
imply some cohesion between contributors, which is unnecessarily under-inclusive.
2
Williams, make the bold claim that ‘mass collaboration changes everything’ 7. Tapscott
and Williams identify a fundamental shift in the way that work and innovation are
conducted, which they foresee will ultimately transform the current economic system8.
In this context, they argue that businesses must ‘collaborate or perish’9. Whether or not
one shares Tapscott and Williams’ view, it is clear that large scale collaboration is an
increasingly important feature of the cultural and economic landscape. Yet, for a
number of years intellectual property scholars have expressed doubts that copyright law
is adapted to the challenge of regulating collaborative works10. If copyright law is to
retain its credibility as the primary tool for regulating creativity it must come to terms
with this challenge.
Although most scholars agree that current copyright law is not well-equipped to
meet the challenges of determining the authorship of collaborative works, they offer
different explanations for this. Some suggest that the ‘Romantic author construct’, a
literary trope which conceives of authors as ‘solitary geniuses’, has influenced
copyright law leaving it ill-adapted to collaborative creativity11. Others offer a more
fundamental critique of copyright law, suggesting that it lacks the conceptual tools to
7
D Tapscott and A Williams, Wikinomics: How Mass Collaboration Changes Everything (Atlantic
Books, London 2008).
8
ibid.
9
ibid 31-33.
10
For example, RC Dreyfuss, ‘Collaborative Research: Conflicts on Authorship, Ownership and
Accountability’ (2000) 53 Vanderbilt LRev 1161; A Stokes, ‘Authorship, Collaboration and Copyright a
View from the United Kingdom’ (2002) Entertainment LRev 121.
11
Much of this scholarship is collected in M Woodmansee and P Jaszi (eds), The Construction of
Authorship: Textual Appropriation in Law and Literature (Duke UP, 1994). See also: M Rose, Authors
and Owners: The Invention of Copyright (Harvard UP 1993); D Saunders Authorship and Copyright
(Routledge 1992); J Boyle, Shamans, Software and Spleens: Law and the Construction of the Information
Society (Harvard UP 1996). Others suggest that there are better explanations of the current state of
copyright law: L Bently, ‘R v Author: From Death Penalty to Community Service’ (2008) 32(1)
Columbia J of L & the Arts 1.
3
deal with the forms of creativity that flourish in the modern digital world (many of
which are highly collaborative)12. This thesis does not ask why copyright might be ill-
adapted to collaborative creativity. Instead, it investigates the underlying assumption
that copyright law lacks the appropriate conceptual tools to deal with collective
authorship. I argue that copyright does contain the appropriate tools to determine the
authorship of works of collective authorship, and I conclude that, in applying the joint
authorship test, judges should make better use of their conceptual tool box.
1.2 Method and Structure of the Thesis
In his recent report for the UK government on the reform of copyright law, Ian
Hargreaves argues for the importance of evidence-based policy making13. This is in line
with an increased interest in economic analysis and empirical investigations among
intellectual property scholars. Although valuable, these frames of reference do not
provide a complete picture of the dynamics of creativity. The thesis engages with a
broad range of scholarship from the Social Sciences and the Humanities, which provide
an equally important perspective on the reality of creativity and the role of copyright
law within creative communities. I embrace an interdisciplinary approach that draws
from a range of literature that concerns itself with understanding the dynamics of
creativity and the role of concepts such as ‘authorship’ in the relevant domains. In
12
JP Barlow, ‘The Economy of Ideas: Selling Wine Without Bottles on the Global Net’
<https://projects.eff.org/~barlow/EconomyOfIdeas.html>. All websites referred to in this thesis were last
accessed on 30 September 2014 unless otherwise indicated.
13
I Hargreaves, ‘Digital Opportunity: A Review of Intellectual Property and Growth for HMG’ (2011)
available at: <http://www.ipo.gov.uk/ipreview>.
4
taking this approach, my research might be located amongst the growing reaction to the
abstract approach of previous copyright scholarship14.
The four case studies considered in this thesis have been chosen because they
provide a typical picture of collective authorship. They concern the creation of different
types of copyright works (literary, artistic, dramatic, film) in very different economic
sectors. They are fairly representative of the range of collaborative practices that
currently exist, including a very new form of creativity (Wikipedia) and one that is
ancient in origin (Australian Indigenous art), as well as hi-tech (Science, Film) and
amateur (Wikipedia) examples. In each case, authorship is driven by different
motivations, from largely commercial motivations (Film), to religious and spiritual
motivations (Indigenous art), and even as a recreational pursuit (Wikipedia). They also
provide examples of a range of different ways in which issues of authorship might be
self-regulated.
In order to investigate how best to determine the authorship of works of collective
authorship, I begin by looking at current copyright doctrine. In Chapter 2 I consider the
concept of authorship in the Copyright Designs and Patents Act 1988 (‘CDPA’)15 and
the application of the joint authorship test in the case law. Despite its central role as an
organising concept in the CDPA, authorship is relatively under-theorised16. Although
the contours of the concept of authorship are uncertain, I identify a stable core
14
See, J Cohen, ‘Creativity and Culture in Copyright Theory’ (2007) 40 University of California Davis
LRev 1151.
15
Unless otherwise indicated throughout this thesis statutory provisions refer to sections of the CDPA.
16
Much of the theoretical scholarship focuses upon the philosophical underpinnings of copyright law,
assigning ‘authorship’ an instrumental role according to the scholars’ preferred view. Ginsburg (n5)
provides a notable exception.
5
requirement of a more than de minimis contribution of creative choices evident in the
protected expression17. To the author’s knowledge, there are no UK cases in which a
court has applied the joint authorship test to determine the authorship of a work of
collective authorship18. The existing case law concerns works where only a few people
might claim to be authors. This case law offers limited guidance because most of these
cases are said to be specific to their facts; and the rationale for determining which
contributors count as copyright authors is often unclear 19 . Chapter 2 offers three
critiques of the application of the joint authorship test, which have led to uncertainty in
the case law. These are: (i) the factual specificity of the joint authorship test; (ii) the
instrumental/pragmatic approach to the implementation of the test; and (iii) a judicial
preoccupation with aesthetic neutrality20.
Then, I look outward at the realities of collective authorship. I consider the
regulation of the attribution of authorship and the social incidents of authorship
(benefits, responsibilities, etc) in four case studies of collective authorship:
(i) Wikipedia (Chapter 3);
(ii) Australian Indigenous art (Chapter 4);
(iii) Scientific collaborations (Chapter 5); and
(iv) Film (Chapter 6).
17
2.1 below.
18
With the exception of some cases on film copyright where the joint authorship test has not been applied
because the principal director and the producer are deemed to be the joint authors of a film.
19
L Zemer, ‘Contribution and Collaborations in Joint Authorship: Too Many Misconceptions’ (2006)
1(4) JIPLP 283.
20
2.3.
6
Each case study has been approached with similar questions in mind and the
Chapters follow a common structure in order to facilitate a comparison of what each
case study reveals for copyright law 21 . The first part of each Chapter contains an
analysis of the dynamics of creativity and the social norms which operate to regulate the
attribution and social incidents of authorship in that particular context. The second part
considers the application of copyright law’s rules of subsistence to the case study,
focusing on identifying any gaps or uncertainties. The third part is an assessment of
private ordering measures that have been adopted to address these gaps. The fourth part
is a comment on the insights for copyright law which might arise from the case study.
In Chapter 7 I build upon the insights gained from the case studies to develop five
broad themes which elucidate the role of copyright law in regulating collective
authorship22. In light of these findings, I suggest that an inclusive, contextual approach
ought to be taken to the application of the joint authorship test. Chapter 8 sets out this
approach in detail and concludes with some other insights for copyright law which arise
from the analysis of collective authorship in the thesis. In particular, the case studies
provide a direct challenge to the incentive theory from copyright law’s heartland, as
they involve works that fall neatly within copyright law’s rules of subsistence.
21
The order has been slightly modified in Chapter 5 (scientific collaborations) in order to aid the clear
presentation of the relevant issues.
22
These are dealt with under five main headings: (i) the nature of collective authorship; (ii) the different
meaning of authorship for each authorial group; (iii) the gap between copyright law and creative reality;
and (iv) the role of copyright law and its concepts.
7
1.3 An Inclusive, Contextual Approach to the Joint Authorship Test
This thesis searches for the best way of applying the joint authorship test in cases of
collective authorship. It is difficult to assess whether or not the current statutory test
provides a suitable mechanism for determining the authorship of works of collective
authorship because: (i) collective authorship has not yet been considered by a court; (ii)
the case law on joint authorship is very factually specific; and (iii) the case law tends to
lack analytical clarity23. Following the interdisciplinary method identified above, this
thesis proceeds on the hypothesis that an examination of the ways in which collective
authorship groups regulate authorship can shed light on how the joint authorship test
ought to be applied so that it can provide a suitable mechanism for determining the
authorship of those works. The research question, thus, is formulated in the following
terms:
What does an examination of examples of works of collective authorship reveal about
the best way of applying copyright law’s joint authorship test so that it can provide a
suitable mechanism for determining the authorship of those works?
For the purpose of this thesis suitable is taken to mean:
 in a way that serves copyright law’s purposes to incentivise and reward
creativity24; and
23
Chapter 2, p22-28.
24
Although these are most commonly cited by commentators, there are a number of other possible
purposes of copyright law. For example, encouraging the distribution of creative works, promoting
individual flourishing or fostering the achievement of a just and attractive culture. See W Fisher
‘Theories of Intellectual Property’ in S Munzer (ed), New Essays in the Legal and Political Theory of
Property (2001) <http://www.tfisher.org/publications.htm> for a helpful overview of the many different
views on the theoretical underpinnings of copyright law. In Chapter 2 I argue that the concept of
authorship might be affected by one’s view of copyright law’s purpose and I offer a definition of the
8
 in a way that is credible to creators and the creative community concerned
(because of the importance of the congruency between law and social norms,
both in enhancing the law’s perceived legitimacy and in promoting compliance).
There are at least two possible approaches to the application of the joint
authorship test to collective authorship: a restrictive approach and an inclusive
approach25. Current scholarship and much of the case law on joint authorship tends to
favour a restrictive approach. Proponents of this approach are persuaded by pragmatic
and instrumental reasons to concentrate the authorship of a work of collective
authorship in the hands of few creators (who will, then, usually be the first owners of
the copyright interest subsisting in the work). This position flows from concerns about
the ability of joint owners to effectively exploit their copyright interest, which
proponents of this approach fear may be impeded where many joint owners must agree
on terms to licence or assign copyright26. I refer to this as the instrumental/pragmatic
approach.
Proponents of a minority view are persuaded that a more inclusive approach to
who constitutes a joint author would facilitate collaboration by incentivising and
minimum core of authorship, which operates as a ‘mid-level principle’ of the sort discussed by R Merges,
Justifying Intellectual Property (Harvard UP, 2011).
25
In the US context, this debate has been explicitly played out between two eminent copyright scholars:
Melville Nimmer and Paul Goldstein, see: Zemer (n19) 288; M LaFrance, ‘Authorship, Dominance, and
the Captive Collaborator: Preserving the Rights of Joint Authors’ (2001) 50 Emory LJ 193, 196-197,
259-261; and PS Fox, ‘Preserving the Collaborative Spirit of American Theatre: The Need for a ‘Joint
Authorship Default Rule’ in Light of the Rent Decision’s Unanswered Question’ (2001) 19 Cardozo Arts
& Entertainment LJ 497, 507-509. The restrictive view is currently favoured by most US courts, although
not without criticism from the academy, see: J Dougherty, ‘Not A Spike Lee Joint? Issues in the
Authorship of Motion Pictures Under US Copyright Law’ (2001) 49 UCLA LRev 225.
26
On this view, the more owners there are the greater the possibility of hold-ups occurring.
9
rewarding a broader range of creators27. Yet, few scholars have explicitly set out the
parameters of such an approach28. This thesis proposes that the best approach to the
application of the joint authorship test to works of collective authorship is an inclusive,
contextual approach. This approach reflects the concept of authorship at the heart of
UK copyright law and is also adapted to the realities of collective authorship. I briefly
outline what I mean by an inclusive, contextual approach below.
1.3.1 The Joint Authorship Test Should Be Inclusive
The view that judges should to take a restrictive approach to who counts as an author of
a collaborative work results from pragmatic or instrumental concerns related to the
difficulty of exploiting copyright where ownership is shared among a number of
individuals. Yet, in copyright law authorship is conceptually distinct from ownership.
Indeed, contributors to works of collective authorship often care more about the
noneconomic consequences of authorship (such as, attribution) than they do about
remuneration and control. Further, an examination of the use of private ordering by
collective authorship groups reveals that the fears that motivate the
instrumental/pragmatic approach are generally not well-founded. Collective authorship
groups tend to be able to manage ownership issues very well by relying on social norms
or other private ordering mechanisms. Even where this is not the case, insights from the
case studies support a broad, rather than a restrictive approach to the application of the
joint authorship test, because this would give authors a valuable bargaining chip in
negotiations with other more powerful players.
27
For example, NB Nimmer and D Nimmer, Nimmer on Copyright (Matthew Bender 1998) 6-23; and
Zemer (n19) 288.
28
Nimmer, ibid is a notable exception.
10
An inclusive approach better reflects the reality of creativity in large
collaborations. Collective authorship has a number of special characteristics: (i) there is
a division of labour; (ii) responsibility for the creative or intellectual content of the
work is shared or diffused among many contributors; and (iii) contributors rely on
social norms to regulate the creative process (and, often, also to determine their rights
and responsibilities to one another). When these three characteristics are considered, it
is apparent that the search for one or two controlling minds who might be identified as
the authors of a work of collective authorship misses the point of how large groups
work together to create. In order to achieve an inclusive approach to the joint authorship
test that more accurately reflects the reality of collective authorship, more emphasis
should be given to the ‘collaboration’ limb and less to whether a particular contribution
is of an ‘authorship’ type. Also, the legal requirement for ‘authorship’ should not be
more demanding in the case of a joint work, than it is in the case of a work of individual
authorship.
1.3.2 The Joint Authorship Test Should Be Applied Contextually
One of the greatest strengths of the joint authorship test is its inbuilt flexibility, which
allows it to adapt to different creative contexts. The case studies reveal that the
dynamics of creativity might vary considerably between creative groups, making this
character of the test an essential feature if copyright law is to remain in touch with the
creative realities of authorship.
Related to this, I argue that the joint authorship test ought to be applied in a way
that is sensitive to the context in which creativity occurs. In Chapter 2 I argue that some
aspects of the joint authorship test are best regarded as questions of fact, ie the
11
requirements for ‘collaboration’; a contribution that is ‘not distinct’; and a contribution
that is ‘significant’29. The social norms which operate to regulate creativity in collective
authorship groups provide an important source of information that can offer answers to
these questions. There are many advantages to incorporating social norms in the
application of the joint authorship test, the most important of which is enhancing the
perceived legitimacy of copyright law decision-making, which is likely to promote
compliance30.
In the CDPA, authorship is, however, ultimately and importantly, a legal
question. In the joint authorship test, the requirement for authorship is expressed as a
requirement that the contribution be of the ‘right kind’. This thesis argues that the same
concept of authorship ought to apply in cases of individual authorship and joint
authorship 31 . For reasons that are set out more fully in Chapter 2, I argue that a
requirement for some creative or intellectual input, which is evident in the expression,
is hard-wired into copyright law’s concept of authorship. This requirement provides an
important filter when it comes to the incorporation of social norms in copyright
decisions.
I argue that the principle that judges ought not to judge the aesthetic merits of a
copyright work is not violated by reference to aesthetic criteria in coming to decisions
on the subsistence of copyright in a work32. Indeed, it is probably impossible not to
29
2.2.4.
30
8.1.
31
p29.
32
2.3.3 and 8.4.2.
12
have reference to such criteria. It would be preferable if such criteria were explicitly
acknowledged in judicial reasoning rather than obscured behind a cloak of aesthetic
neutrality. The social norms that govern creativity within a particular authorship group
provide an important and independent source of information about the aesthetic
significance of a particular contribution to a work.
Yet, social norms might be affected by three potential defects: a lack of certainty;
the influence of power imbalances; and the failure to account for the interests of the
public and other third parties33. Therefore, I suggest a framework that requires proof
that the norm exists in a stable form and that it does not enshrine values that might be
repugnant to the law’s overall purposes. This framework might be used in order to
avoid the incorporation of undesirable social norms in judicial decision-making34.
1.4 Broader Lessons for Copyright Law
The case studies also provide an opportunity to consider the influence of non-legal
forms of regulation on the ways in which authorship is understood and attributed. Thus,
the thesis forms part of an emerging body of literature that explicitly considers the
complicated relationship between copyright law norms and the social norms which
operate to regulate creativity in different creative contexts 35 . The case studies
demonstrate that the noneconomic incidents of authorship are frequently just as
important, if not more important, to creators than the economic incidents. In
33
8.2.
34
8.3.
35
See LJ Murray, S Tina Piper and K Robertson, Putting Intellectual Property in Its Place: Rights
Discourses, Creative Labor, and the Everyday (OUP 2014).
13
considering the noneconomic benefits of authorship and the role of social norms in
regulating creativity, this research crosses the terrain of another significant body of
scholarship, which calls into question the ‘incentive’ story of copyright protection by
exploring noneconomic motivations for authorship36. Much of this literature focuses on
creativity that falls within Intellectual Property’s ‘negative spaces’ 37 . It is open to
debate whether such activities ought to be regulated by copyright law in the first place.
In contrast to this literature, the case studies considered here fall squarely within
copyright law’s domain as they concern the creation of copyright works. As such, they
provide an opportunity to re-examine copyright law’s concept of authorship on its own
terms. The case studies seriously undermine the persuasive power of the ‘incentive
story’ by demonstrating that authorship is a complex phenomenon that has important
intrinsic and social dimensions.
Much of the current copyright debate fails to appreciate the importance of the
noneconomic incidents of authorship in motivating creativity. The limitations of the
incentive justification suggest that it can only provide a shaky foundation for the
copyright expansionism that has occurred in its name. There are important intrinsic and
social dimensions to creativity, which might require a nuanced approach to regulation.
36
Elkin-Koren (n1) Tailoring Copyright to Social Production’ (2011) 12 Theoretical Inquiries in Law
309, J Silbey, The Eureka Myth: Creators, Innovators, and Everyday Intellectual Property (Stanford
Press, forthcoming December 2014).
37
E Rosenblatt, ‘A Theory of IP’s Negative Space’ (2011) 34(3) Columbia J of L and the Arts 317
provides an overview of the literature on this point. See also: K Raustiala and C Sprigman, ‘The Piracy
Paradox: Innovation and Intellectual Property in Fashion Design’ (2006) 92 Virginia LRev 1687; D Oliar
and C Sprigman, ‘There’s No Free Laugh (Anymore): The Emergence of Intellectual Property Norms and
the Transformation of Stand-up Comedy’ (2008) 94 Virgina LRev 1787; E Fauchart and EA von
Hippel,‘Norm-Based Intellectual Property Systems: The Case of French Chefs’ available at
<http://papers.ssrn.com/sol3/papers.cfm?abstract_id=881781>; CJ Buccafusco, ‘On the Legal
Consequences of Sauces: Should Thomas Keller’s Recipes Be Per Se Copyrightable?’ (2007) 24 Cardozo
Arts & Entertainment LJ 1121; J Loshin, ‘Secrets Revealed: How Magicians Protect Intellectual Property
without Law’ in C Corcos, Law and Magic: A Collection of Essays (Carolina Academic Press 2008) 123.
14
An economic lens alone will not provide a clear picture of the best way to regulate
collaborative creativity. An appreciation of the realities of creativity suggests the need
to review the way in which instrumental reasoning shapes the development of copyright
law.
This thesis highlights the importance of copyright law as a source of good
authorship standards. These standards are likely to be most valuable when they are
developed in light of the creative realities of authorship. At its best, copyright law can
provide a valuable bulwark against power dynamics in creative communities that can
cause authorship to gravitate to dominant players at the expense of other creators. At its
worst, out of a desire to simplify rights and ensure efficient exploitation, copyright law
might end up bolstering the positions of dominant players (who tend to be orchestrators
and investors) at the expense of real creators. The latter vision would seem to justify the
view that copyright law is out of touch with creative realities. Thus, this thesis urges an
approach that would allow copyright law to reconnect both with creative realities and
also with its own raison d’être: protecting, rewarding and incentivising creators.
Although the CDPA may seem to offer little guidance on the definition of authorship, in
fact, the description of the author as the one who creates a work in section 9 reveals the
heart of this concept.
This thesis offers a first step towards realigning the legal conception of authorship
with creative realities, arguing for a recalibration of the joint authorship test. Before
doing so, I begin by assessing the status quo, looking at the concepts of authorship and
joint authorship in current copyright law.
15
Chapter 2: Authorship and Joint Authorship
This Chapter provides the legal and conceptual background for the analysis of the case
studies that occurs in the following four Chapters. Its purpose is to provide a brief
overview and critique of the provisions on authorship and joint authorship in the CDPA.
Although the contours of copyright law’s concept of authorship are uncertain, it has a
stable core in that it refers to the creator of the protected expression. I argue that the
best view is that authorship requires at least a modicum of creativity or other
intellectual input that is evident in the author’/s’ expression.
One of the most serious weaknesses in the current jurisprudence is a lack of
analytical clarity in the application of the joint authorship test. This makes it difficult to
predict how the test might be applied to a work of collective authorship. I suggest that
this lack of analytical clarity is a result of instrumental/pragmatic reasoning, which
tends to conflate ‘authorship’ with ‘ownership’, and a judicial preoccupation with
aesthetic neutrality. The former suggests there is a more demanding requirement for
authorship in the case of a joint work than for works of individual authorship, which
might make it more difficult for contributors to such works to establish that they are
authors. It is hard to make sense of this from a conceptual point of view, as authorship
implies the same consequences (and has the same role in the CPDA) regardless of
whether a work is individually or jointly authored.
This Chapter begins by outlining copyright law’s concept of authorship (2.1) and
the provisions of the joint authorship test (2.2). Then, I present three critiques of the
16
way in which the joint authorship test is applied in the case law (2.3). These critiques
explain the difficulty of predicting how the joint authorship test might apply to works of
collective authorship. Section 2.4 searches for insights from the scholarly literature on
authorship about how the joint authorship test ought to apply in cases of collective
authorship, revealing a more inclusive alternative to the instrumental/pragmatic
approach. Although authorship is ultimately a legal question that may be influenced by
one’s view of the purpose of copyright law, it is also inevitably linked to creative
practice. Judges are necessarily influenced by pervasive cultural views about creators
and the creative process in the application of copyright law’s rules of subsistence.
Indeed, a great strength of copyright’s concept of authorship is its inbuilt flexibility,
which allows it to adapt to changes in creative practices over time and in different
creative sectors. The Chapter ends by suggesting that the authorship practices of
creative communities might provide some insight on the question of how best to apply
the joint authorship test to works of collective authorship.
2.1 The Concept of Authorship in the CDPA
Authorship is an important organising concept for copyright law. In determining
authorship courts also often take into account factual matters, such as the context within
which creativity occurs. Yet, ultimately for the purposes of the CDPA, determining who
is an author is a question of law. Thus, the concept of the ‘author’ is understood by
reference to matters that lie outside of a dictionary definition of the word, such as the
policy aims of copyright law1. In this section, I explain the two aspects of the concept of
authorship: its factual/causative dimension and its normative dimension. I argue that a
1
L Bently and B Sherman, Intellectual Property Law (OUP 2009) 120 consider that the concept of the
author operates as a ‘legal fiction’.
17
requirement for a modicum of creativity or intellectual contribution is hard-wired into
the provisions of the CDPA on authorship. However, this important requirement has not
been translated into any clear test. This section argues that authorship requires a more
than de minimis contribution of creative or intellectual choices to the protected
expression. Although this view of authorship has not always prevailed in the case law, it
is more likely to do so in the future as a result of the Court of Justice of European
Union’s application of the harmonised standard for originality (the ‘author’s own
intellectual creation’ test)2.
Authorship is the linchpin of the scheme embodied in the CDPA. Thus,
jurisdictional connection with the author of a work is one of the bases for qualification
for copyright protection3. The author is also a point of reference for the duration of
copyright and moral rights protection 4 . The author is usually the first owner of
copyright subsisting in a work5 and is entitled to various moral rights in respect of the
work, namely, the attribution right and the right to object to derogatory treatment6.
Despite the importance of the concept of authorship, it is very under-theorised7.
There is no definition of ‘author’ at the international level, and the concept has only
2
Following Case C-5/08 Infopaq v Danske Dagblades Forening [2009] ECR I-6569 [45]; see text at n25.
3
CDPA s153, s154.
4
CDPA s12, s13A, s13B.
5
CDPA s11.
6
CDPA s77, s80.
7
J Ginsburg, ‘The Concept of Authorship in Comparative Copyright law’ (2003) 52 De Paul LRev 1063,
1066, observes that the concept has not been very well articulated in the case law.
18
been partially harmonised at the European level8, as will be discussed below. The UK
statute is similarly brief. The CDPA simply provides that the author is the ‘creator’ of a
copyright work, but does not define ‘author’ 9 . Below, I argue that some qualities
inherent in the definition of ‘author’ might be inferred from provisions deeming certain
persons authors (ie, that they be ‘taken to be the author’)10. Before doing so, however, I
consider what the description of the author as someone who ‘creates’ a work might
imply about authorship.
The Oxford English Dictionary defines ‘create’ as follows:
…to make, form, set up or bring into existence (something which has not
existed before); to produce (a work of imagination or invention; an
artefact)11.
This definition captures two senses in which authorship might operate in copyright law.
The first sense, which is the predominant understanding, is of creator as
originator/source. Here the focus is on the causal connection between the creator and
what is created. In this sense a creator is a person who produces an artefact or who
causes it come into existence. The second sense is of creator as innovator and focuses
on the act of creation, that is, the production of something that has a quality which
might be described as inventive, imaginative or creative. It is this second sense of
8
Case C-277/10 Luksan v can der Let [2013] ECDR 5. The emergence of a common harmonised
standard of originality also indirectly partly harmonises authorship given the close link between
authorship and originality.
9
CDPA s9(1).
10
p14 below.
11
This is the most relevant definition (‘Of a human agent’) that is given on the electronic version of the
Oxford English Dictionary <www.oed.com>.
19
‘create’ which primarily distinguishes this word from others such as ‘make’ or
‘produce’.
In the case law, authorship has at least two dimensions: a factual, causative
dimension and a normative dimension. The most obvious dimension of authorship is the
factual, causative dimension. This dimension corresponds to the first meaning of
‘create’. According to this view the author is conceived of as the ‘originator’ of the
work. Thus, in Cummins v Bond, a medium was the author of the automatic writing she
produced during a séance because she had written every word of it12. Although this
might seem to imply the need for ‘writing’, in fact, what is required is ‘something
approximating penmanship’, that is, responsibility for the expression of the work 13 .
Expression is understood in abstract terms as the intangible property that copyright
protects (for example, a ‘literary work’) and ought to be distinguished from the fixation
or physical object in which that expression is embodied. So where a clerk takes
dictation of a letter, the author of the resulting literary work is the person who
composed the letter (not the clerk who has merely ‘fixed’ it to paper)14.
The normative dimension of authorship is less clear. The normative dimension of
authorship is essentially the content given to this concept when it is used to police the
boundaries of copyright protection. In order to understand the normative dimension of
authorship it is helpful to consider another closely related concept: originality.
12
(1926) 1 Ch 167, 175. The medium was the originator of the words, which had not been copied. The
supposed ‘spiritual origin’ of the words was irrelevant in this regard.
13
Cala Homes v Alfred McAlpine Homes [1995] EWHC 7; Robin Ray v Classic FM [1998] FSR 622
(Ch).
14
Walter v Lane [1900] AC 539 (HL), 554 (James LJ).
20
Authorship and originality are like two sides of the same coin 15. It is often an ‘original’
contribution that makes a contributor an ‘author’16. In the classic formulation of the
originality test in University of London Press v University Tutorial Press the word
‘original’ is distanced from its ordinary meaning insofar as it might denote novelty or
inventiveness17. Instead, originality is said to mean that the work originates from the
author in that it is not copied. Yet, original often seems to imply ‘that which copyright
protects’. As such, the concept seems to contain more specific requirements, although
these are uncertain and shifting, as the underlying normative justification for copyright
law relied upon is rarely made clear in the cases18.
As the originality requirement sets the threshold for copyright protection, it is
necessarily affected by one’s view of the purpose of copyright law, which shapes one’s
view on the sorts of creative activities that ought to benefit from copyright protection.
The lack of a consensus on the best theoretical justification for copyright protection
may explain why the cases sometimes appear to offer conflicting guidance 19 . The
predominant justification for copyright law in common law jurisdictions assumes that
authors are incentivised to create by the ability to control reproductions of their work in
15
Existing understandings of the originality requirement developed from judicial interpretation of the
notion of ‘authorship’ before there was an express requirement for originality in the Copyright Act. See,
for example, Walter v Lane, ibid.
16
This is particularly the case for derivative works and in the case of works of joint authorship. See 2.2.3
below for a discussion of the emphasis on the ‘authorship’ limb in the application of the joint authorship
test.
17
[1916] 2 Ch 601, 608-9 (Peterson J): ‘...the act does not require that the expression must be in an
original or novel form, but that the work must not be copied from another work – that it should originate
from the author’.
18
Bently and Sherman 94; S Ricketson, ‘The Concept of Originality in Anglo-Australian Copyright Law’
(1991) 9(2) Copyright Reporter 1.
19
Some of the main views are helpfully summarised by W Fisher, ‘Theories of Intellectual Property’ in S
Munzer (ed), New Essays in the Legal and Political Theory of Property (2001)
<http://www.tfisher.org/publications.htm>.
21
order to extract royalties. Yet, the incentive story has limited explanatory power, which
leads some to prefer natural rights or social planning justifications. The uncertainty as
to the normative content of the concept of originality necessarily affects the concept of
authorship, because an original contribution is generally what makes a person an author.
Persistent disagreement about the philosophical underpinnings of copyright law
complicates any attempt to clarify the breadth of the concept of authorship and its role
within the statutory scheme, which seems to reflect more than one rationale for
protection 20 . Equally as important as economic incentives, for example, the label
‘author’ provides valuable noneconomic benefits both in terms of moral rights and
social recognition21.
Originality is a matter of fact and degree. And the philosophical tension between
UK and European notions of authorship plays out in its interpretation. Originality is
often said by the UK courts to involve skill, labour and/or judgment, although a number
of other wordings of the test exist22. In some cases (most involving compilations or
tables), courts appear to have held that mere routine labour or sweat of the brow is
sufficient to confer the requisite originality upon a work23. Yet, these cases appear to be
more influenced by a desire to provide a remedy for unfair competition than the
20
Fisher ibid. On the limits of the incentive story: D Zimmerman, ‘Copyright as Incentives: Did We Just
Imagine That?’ (2011) 12 Theoretical Inquiries in Law 29; N Elkin-Koren, ‘Tailoring Copyright to Social
Production’ (2011) 12(1) Theoretical Inquiries in Law 309.
21
Attribution is an important incident of authorship, as seen in a number of aspects of literary property
law. The right to object to false designation of authorship is venerable and attribution is a precondition
for relying upon a number of the fair dealing defences. W Cornish, D Llewelyn and T Aplin, Intellectual
Property: Patents, Copyright, Trade Marks and Allied Rights (8th edn, Sweet & Maxwell 2013) 505.
22
Bently and Sherman 95, cite others: ‘work, capital, effort, industry, time, knowledge, taste, ingenuity,
experience or investment’.
23
For example, Ladbroke v William Hill [1964] 1 All ER 465 (HL), 478 (Lord Devlin); Waterlow
Directories v Reed Information Services [1992] FSR 409 (Ch); Blacklock v Peterson [1915] 2 Ch 376;
University of London Press (n17); Elanco Products Ltd v Mandops [1979] FSR 46; [1980] RPC 213
(CA).
22
recognition of authorship24. In a recent line of cases, however, the Court of Justice of
the European Union appears to apply a new harmonised standard of originality: the
‘author’s own intellectual creation’ test25. This test focuses on the creator’s ability to
make free and creative choices and stamp the work with his or her ‘personal touch’26.
Although some UK judges are adamant that this new formulation does not alter
the position in the UK27, in a number of cases, UK courts seems to refer to both tests
simultaneously, as if hedging their bets 28 . Most recently, in SAS Institute v World
Programming the UK Court of Appeal seems to suggest that this test may have raised
the standard29. However, the requirement for some creative choices is not far removed
24
Cornish et al 438-439 observe that copyright tends to be denied where some other form of relief against
unfair competition is available, citing Exxon v Exxon Insurance [1982] Ch 119 (CA). Bently and
Sherman 106 note that instead of focusing on whether the work is original, courts in these cases start
from the premise that any labour or effort exerted in the production of the work ought to be protected (as
long as a quantitative threshold is met) citing the maxim: ‘what’s work copying is prima facie worth
protecting’.
25
This test appears to require that: (i) the work is the author’s own original creation; (ii) the creation
reflects his or her personality; and (iii) the author has been able to express his or her creative ability by
making free and creative choices in the process of creation, thus, stamping his or her personal touch on
the work. See: Infopaq (n2) [45]; C-393/09 Bezpecnostn´ı softwarováasociace v Ministerstvo Kultury
[2011] ECDR 3, [2011] FSR 18 [45]; C-403/08 & C-429/08 Football Association Premier League v QC
Leisure [2012] 1 CMLR 29, [2012] ECDR 8 [97]; C-604/10 Football Dataco v Yahoo! UK [2012] ECDR
10 [37]; C-145/10 Painer v Standard Verlags [2012] ECDR 6 [87]. Cf, Newspaper Licensing Agency v.
Meltwater [2011] EWCA 890 [20].
26
See, for example, Painer ibid.
27
NLA v Meltwater (n25) [20] (Chancellor Sir Andrew Morritt): ‘I do not understand the decision of the
European Court of Justice in Infopaq to have qualified the long standing test established by the
authorities [referring to University of London Press (n17) and Ladbroke (n23)]’.
28
Temple Island Collections v New English Teas [2012] EWPCC [27] (Judge Birss QC): stating the test
as ‘skill and labour (or intellectual creation)’. Taylor v Maguire [2013] EWHC 3804 [6]-[7] (District
Judge Clark): ‘an original artistic work is a work in which the author/artist has made an original
contribution in creating it, for example by applying intellectual effort in its creation... it must have been
produced as a result of independent skill and labour by the artist’. Abraham Moon v Thornber [2012]
EWPCC 37 [56] (Judge Birss QC): ‘skill, labour and judgment... it is his own intellectual creation’.
29
SAS Institute v World Programming [2013] EWCA Civ 1482, [2014] RPC 8 [36]-[37] (Lewison LJ):
‘This test may not be quite the same as the traditional test in English law… If the Information Society
Directive has changed the traditional domestic test, it seems to me that it has raised rather than lowered
the hurdle to obtaining copyright protection’. Lewison LJ cites Advocate-General Mengozzi in Football
Dataco (n25), who suggests that the test resembles the continental European approach and is, therefore,
probably higher than the traditional UK approach in that it excludes mere ‘mechanical’ effort.
23
from some of the terms that have been used to express the originality test in the UK,
such as ‘judgment’ or ‘taste’. At the very least, the author’s own intellectual creation
test seems to allow no room for the full-blown ‘sweat of the brow’ doctrine. It is thus a
matter of debate, the extent to which this doctrine remains part of UK copyright law30.
Despite the ongoing disagreements, the continental European view, which is often seen
to require some amount of creativity, might be gaining more traction in the UK with the
emergence of the own intellectual creation standard of originality beyond the three
categories of software, databases and photography31. This can be seen in the distinct
shift in language to focus on the personal vision of the creator rather than pure time,
effort or investment32 (although the standard remains fairly low33).
It is difficult to pin down the normative dimension of authorship when it is treated
as a proxy for what copyright protects or ought to protect34. In the absence of a clear
guiding account of the purpose of copyright law, some courts have preferred to define
authorship in the negative as not a contribution merely of ideas 35; not a mechanical
30
Bently and Sherman 104-106. Routine labour or sweat of the brow has only been sufficient to confer
originality in limited situations, largely in respect of tables or compilations. Thus, sweat of the brow may
not have survived the implementation of the Database Directive, which resulted in the ‘author’s own
intellectual creation’ standard being applied to establish copyright in a databases (the wording of s3(1)(a)
of the CDPA suggests that where a work is both a database and a compilation it should be treated as a
database only).
31
M van Eechoud, ‘Along the Road to Uniformity – Diverse Readings of the Court of Justice’s
Judgments on Copyright Work’ (2012) 1 JIPITEC 60. Cornish et al 440 describing the extension of this
test to other works as a ‘judicial sleight of hand’.
32
See: Taylor v Maguire (n28), Abraham Moon v Thornber (n28), Temple Island Collections (n28). The
language used in these cases approaches the way in which the CJEU describes authorship in cases such as
Painer (n25).
33
Painer (n25), BSA (n25).
34
Ginsburg (n7) 1067 highlighting the problem of consequentialist reasoning.
35
Donoghue v Allied Newspapers [1938] 1 Ch 106 (Ch).
24
contribution to fixation36; not interpretation or performance37; or not proof-reading or
things akin to proof-reading38.
Indeed, the most conservative view of the recent European case law is that it
simply establishes that there can be no authorship where there is no creative freedom at
all 39 . The positive aspects of authorship are less clear. The aesthetic quality of the
contribution is sometimes a relevant factor, as on occasion the sorts of contributions that
count depend upon the type of work 40 . For example, contributions which affect the
playability of a musical work might be authorial41; whilst contributions which do not
affect an artistic work in a visually significant way will not be authorial42. In this way the
categories of work have been used to police the scope of copyright protection43. This
provides an entry point for judges to take into account creative practices, as the CDPA
provides only skeleton definitions of the categories of work44.
36
Such as a clerk taking dictation: Walter v Lane (n14) 554 (James LJ).
37
Hadley v Kemp [1999] EMLR 589 (Ch); Brighton v Jones [2004] EWHC 1157, [2005] FSR 288.
38
Fylde Microsystems v Key Radio Systems [1998] FSR 449 (Ch).
39
Football Association Premier League (n25); BSA (n25).
40
This may be what Ginsburg (n7) 1092 refers to as the ‘constraints of [the author’s] task’.
41
Sawkins v Hyperion Records [2005] EWCA 565, [2005] 3 All ER 636, 648 (Mummery LJ): ‘The work
of Dr Sawkins has sufficient aural and musical significance to attract copyright protection’.
42
Interlego v Tyco Industries [1989] AC 217 (PC); Lucasfilm v Ainsworth [2009] FSR 2.
43
Some suggest that this approach may no longer be permitted following BSA (n25) and Football
Association Premier League (n25), which seems to suggest that the categories of work are no longer
relevant. See van Eechoud (n31) 70-71. Yet, it still seems likely that the category of work will be
pertinent in identifying the sorts of choices that might be relevant in determining whether a work amounts
to an ‘intellectual creation’. Thus, choices related to lighting, setting and pose allow an author to stamp
his or her personal touch upon a photograph: Painer (n25).
44
The presumptions about authorship also oblige the court to take into account attribution as an author on
a work (which is also likely to have been affected by predominant creative practices) see CDPA s104, 105.
25
The different protection offered to entrepreneurial works (compared to authorial
works) may offer further insight into the meaning of authorship 45 . Entrepreneurial
works have always been treated differently on the basis that they are the result of
investment, not authorship 46 . Copyright was generally conferred on the person who
bore the financial risk of creating an entrepreneurial work47. For a film, for example,
this was the producer 48 . In the 1956 Act the first owner of copyright in an
entrepreneurial work was referred to as a ‘maker’, not an ‘author’. European
harmonisation led to reform in this area and the CDPA now provides that certain
nominated people are ‘taken to be’ the author of an entrepreneurial work 49 . This
phrasing echoes the historical hesitation to call those whose main contribution was
bearing the financial risk, ‘authors’. It suggests that here the word ‘author’ is used as a
shorthand way of referring to particular persons who do not have the inherent qualities
of a true copyright author (that is to say, they are not creators)50. This, in turn, suggests
that the use of ‘creates’ in the section 9 definition of authorship (rather than ‘makes’ or
‘produces’) tells us something about authorship. In particular, it indicates the relevance
of the second meaning of ‘create’ – which implies creativity, imagination or innovation.
45
Bently and Sherman 111 contrasting entrepreneurial works with authorial works.
46
The Gregory Report argued for a fixation only form of copyright protection for film on the basis that it
resembled an industrial product more than an authorial work: ‘Report of the Board of Trade Copyright
Committee’ (Her Majesty’s Stationery Company, October 1952).
47
Cornish et al 450.
48
Now the principal director is also considered to be an author, see further Chapter 6.
49
In particular, implementing the Rental and Related Rights Directive and the Duration Directive. KM
Garnett, G Davies, G Harbottle, WA Copinger and EP Skone James, Copinger and Skone James on
Copyright (16th edn, Sweet & Maxwell 2011) 251 [4-47]; G Dworkin, ‘Authorship of Films and the
European Commission Proposals for Harmonising the Term of Copyright’ [1993] EIPR 151.
50
Copinger et al 248, [4–40]: in this context ‘... the expression [author] is not in fact used otherwise than
as a shorthand in the process of identifying the first owner of the copyright’. Although it might be argued
that this is just a drafting technique used for the avoidance of doubt and that it should not impact the
definition of ‘author’ (or if it should, it should favour a broad intention) – the history of the introduction
of these provisions records a hesitation to call these contributors ‘authors’. See Gregory Report (n46).
26
This also goes some way to explaining why the ‘authors’ of entrepreneurial works are
generally not granted moral rights51.
The existence of these deemed authors suggests something about the meaning of
authorship in UK copyright law: the requirement for at least a modicum of creative or
intellectual effort52. This suggests that a true author must be a natural person53. This is
reinforced by the provisions on works made in the course of employment where
although first ownership of copyright is presumed to belong to the employer (often a
company), the author is still held to be the individual employee who created it 54. So, an
author is a natural person who must have contributed more than assuming the financial
risk or controlling the act of creation to be a ‘true’ author (rather than a deemed author).
Seen from this angle, it is hard to escape the conclusion that when the CDPA provides
that an ‘author’ is a person who creates a work, create is meant in both relevant senses
of the word, and hence, what is required is more than a mere causal connection between
creator and work (ie, an author must do something more than just ‘make’ or ‘produce’ a
work). This demonstrates that the requirement for at least a modicum of ‘creativity’ is
hardwired into the concept of authorship in the CDPA. In the third part of this Chapter,
51
With the exception of those who are also creative contributors, eg the principal director of a film and
the performer of a broadcast or sound recording.
52
Bently and Sherman 120 also argue that in the case of entrepreneurial works and computer-generated
works the author is a legal fiction used to allocate rights.
53
S Ricketson, ‘People or Machines? The Berne Convention and the Changing Concept of Authorship’
(1991) 16 Columbia J of L and the Arts 1, 28 arguing that the Berne Convention requires an author to be
a natural person; thus considering that US ‘work for hire style’ principles run counter to the basic
premises of the concept of authorship in the Convention. See also: A Dietz, ‘The Concept of Authorship
Under the Berne Convention’ (1993) 155 Revue Internationale du Droit d’Auteur 3.
54
This might be contrasted with US copyright law that provides that an employer is considered to be the
author as well as the first owner of copyright subsisting in a work for hire, 17 US Code § 201.
27
I explain that one of the reasons that this requirement for creativity is rarely
acknowledged relates to judicial anxiety about aesthetic neutrality55.
Copyright has often been used as a tool to protect investors and entrepreneurs. For
this reason, some suggest that copyright law is primarily concerned with ensuring the
dissemination of creative works (rather their creation)56. Yet, the fact that authors often
need to give away some of their rights to entrepreneurs to ensure that their works are
disseminated, does not undermine the importance of granting them these rights in the
first place57. Indeed, copyright law derives its legal and moral force from the act of
creativity58. This is the reason why authorship is one of the central organising concepts
in copyright law59. In this section, I have argued that the concept of authorship is linked
to the act of creativity and requires a more than de minimis contribution of creative
choices evident in expression of a copyright work. This stable core concept of
authorship remains at the heart of copyright law regardless of the normative
justification or combination of justifications appealed to in an instant case60.
55
The other probable reason is that some judges have been tempted to expand copyright protection in
order to compensate for the lack of a tort of misappropriation or unfair competition. M Spence,
Intellectual Property (OUP 2007) 80-82.
56
LJ Lacey, ‘Of Bread and Roses and Copyright’ [1989] Duke LJ 1532.
57
M Handler, ‘Continuing Problems with Film Copyright’ in F Macmillan (ed), New Directions in
Copyright Law: Volume 6 (Edward Elgar 2007) 173.
58
W Cornish, ‘The Author as Risk-Sharer’ (2002) 26 Columbia J of L & the Arts 1, 12.
59
L Bently, ‘R v Author: From Death Penalty to Community Service’ (2008) 32(1) Columbia J of L &
the Arts 1, 94: ‘…authorship remains at the heart of what most people think, and most legal systems say,
copyright is about’.
60
This might considered a ‘mid-level’ principle that ties together the disparate practices and doctrines of
intellectual property law and that might help bridge the gaps between different views of the theoretical
underpinnings of the law, see: R Merges, Justifying Intellectual Property (Harvard UP, 2011).
28
2.2 The Joint Authorship Test
Section 10(1) of the CDPA defines a ‘work of joint authorship’ as a ‘work produced by
the collaboration of two or more authors in which the contribution of each author is not
distinct from that of the other author or authors’. In addition, case law establishes that in
order to be a joint author a contributor must have contributed a significant part of the
protected skill, labour or judgment, which is sometimes referred to as the authorship
limb of the test61. Therefore, to be a joint author under the CDPA a contributor must
have:
(i) made a contribution that is not distinct;
(ii) in pursuance of some collaboration or common design; and
(iii) he or she must have made a significant contribution of the right kind.
The case law on joint authorship tends to lack analytical clarity as the three limbs
of the test are often considered at once, making it difficult to discern overarching
principles. The uncertainty in the case law is compounded by the fact that the
application of the test is said to be very factually specific62. Despite these difficulties, I
outline each part of the test separately below. I try to draw out common themes from
the case law, filling in gaps where possible with reasoning from first principles.
61
Godfrey v Lees [1995] EMLR 307, 325-8; Ray v Classic FM (n13) 636; Hadley v Kemp (n37).
62
Indeed, some textbooks and commentaries resort to giving long lists of examples to give the reader a
feel for the various circumstances in which joint authorship has and has not been granted to different
contributors to different works, for example, H Laddie, P Prescott and M Vitoria, The Modern Law of
Copyright and Designs (4th edn, LexisNexis 2011) [3.97].
29
2.2.1 Contribution Not Distinct
In order to be a joint author, a contributor’s contribution must not be ‘distinct’. The
requirement is probably not a question of whether it is possible to identify the
contribution. Rather, it is a question of whether the contribution is an integral part of the
work such that without it the work would be different in character 63 . Thus, in
Beckingham v Hodgens (applying the 1956 Act which used ‘not separate’, rather than
‘distinct’), the court held that the violin part of a song was not separate because it was,
heavily dependent upon what is there already. Stripped of the voices and other
instruments, the violin part would sound odd and lose meaning. The final
musical expression – what the audience will hear – is a joint one64.
In other words, a work of joint authorship is not a collection of separate contributions
eg two separate essays in the same book. Distinct contributions may, however,
constitute separate works or a collective work65. This part of the test seems to indicate
that a work of joint authorship involves a combination, merging or mixing of
contributions66.
63
Copinger et al [4-37].
64
At first instance [2002] EWHC 2143 [46].
65
On the distinction between co-authors and joint authors see Copinger et al [4-33]. In Redwood Music v
Chappell (1982) RPC 109 (QB), for example, it was held that the music and lyrics to a song were distinct
contributions to two separate works – a literary work and a musical work. In Cala Homes v Alfred
McAlpine Homes [1995] EWHC 7, [1995] FSR 818, 834 it was accepted that there was no material
difference between the wording of the test in 1956 and 1988 (n5 of the judgment).
66
It might also indicate a bias toward individual ownership, as works are treated separately wherever
possible. On the disaggregation of authorship in this way: P Jaszi, ‘On The Author Effect: Recovering
Collectivity’ in M Woodmansee and P Jaszi (eds) The Construction of Authorship: Textual Appropriation
in Law and Literature (Duke UP 1999) 29, 51-56.
30
2.2.2 Collaboration or Common Design
The requirement for ‘collaboration’ distinguishes a derivative work from a work of joint
authorship67. A derivative work is created where a work is changed by a subsequent
author who is not working together with the original author (for example, a translation).
In order to be joint authors, contributors must have a loose plan to work together or
some sort of preconcerted design to produce the work. For this, joint authors do not
have to be located in the same place and do not need to work simultaneously.
The cases do not provide much clear guidance beyond this. In Levy v Rutley, the
introduction of a new scene and some alterations to the dialogue of a play by a theatre
company proprietor, without the co-operation of the commissioned writer, was not
found to be in furtherance of a common design68. On the other hand, in Springfield v
Thame it was suggested that the sub-editor of a newspaper who touched up a paragraph
composed by another sub-editor might be a joint author of the article 69 . In Levy v
Rutley, ‘mere additions’ to a complete piece were not sufficient to demonstrate
collaboration as they were only intended to make the play more attractive to the
audience. In that case Justice Keating seemed to consider authorial intention to be
relevant as he noted that the requirement would have been satisfied if they had agreed
67
Beckingham v Hodgens (n64) [45] (Floyd QC) citing Levy v Rutley (1871) LR 6 CP 523 distinguishing
‘joint labouring in the furtherance of a common design’ from the ‘subsequent independent alteration of a
finished work’.
68
(1871) LR 6 CP 523. The Court failed to discover ‘any cooperation of the two in the design, or
execution of the piece, or in any improvements either in the plot or general structure’ (529, Keating J);
‘The additions do not disturb the drama composed by Wilks: they were made for the mere purpose of
improving or touching up some of its parts. It would be strange indeed, if not unjust, if the author’s rights
could be thus merged into a joint-authorship with another. There are probably very few instances, at least
in modern times, of a play being put upon the stage without some alteration by the manager’ (530,
Montague Smith J).
69
(1903) 89 LT 242 (Ch). Laddie et al [3.97] consider that collaboration might have been assumed in this
case on the basis that both were employed as sub-editors at the same newspaper.
31
or jointly undertaken to write the play. The Court of Appeal in Beckingham v Hodgens,
however, held that the requirement of a ‘common design’ did not mean the common
intention to be a joint author70. Yet, the best interpretation of Beckingham v Hodgens is
that the Court did not mean that no form of intention is required at all – for surely there
must be some intention to work together for a common design to exist 71 . To hold
otherwise would impermissibly stretch the common language meaning of the word
‘collaboration’. Rather, the Court’s decision should be construed to mean that there is
no need for a subjective intention in a narrow sense: to be a joint author for the
purposes of copyright law.
In jurisdictions that require a specific intention to be a joint author such as the US,
this part of the test tends to take on great importance72. In the UK, instead, cases focus
almost exclusively on the authorship limb which means that decisions on joint
authorship tend to hinge on the characterisation of the nature of a contribution. As a
result, there is an impoverished understanding of collaboration in the case law. There is
virtually no elaboration of the meaning of this term and its impact on the other parts of
the test, ie does the nature of the common design shed any light on whether a
contribution is distinct or whether it is significant? The neglect of the requirement for
70
Beckingham v Hodgens [2003] EWCA Civ 143, [2003] EMLR 18 (Jonathan Parker LJ): there is no
requirement for intention for joint authorship, just a common design to produce the work. The Court
rejected the approaches taken in Canadian and US cases as looking ‘beyond the section into the uncertain
realm of policy’ and introducing undesirable requirements of proof, referring to Childress v Taylor 945
F2d 500 (1991) (US CA 2nd Cir) and Neudorf (Darryl) v Nettwerk Productions Ltd [2000] RPC 935
(British Columbia SC) [52]. Cf L Zemer, The Idea of Authorship in Copyright (Ashgate 2007) 217
arguing that despite this decision UK courts tend to silently embrace a requirement for the intention to co-
author to some extent. See also: Ginsburg (n7).
71
L Zemer, ‘Is Intention to Co-author an Uncertain Realm of Policy?’ (2007) 30(4) Columbia J of L and
the Arts 611, 617: ‘a joint enterprise, by its very nature, is based on some intention by parties to
collaborate’.
72
L Zemer, ‘Contribution and Collaborations in Joint Authorship: Too Many Misconceptions’ (2006)
1(4) JIPLP 283.
32
collaboration in favour of the requirement of authorship does not seem justified by the
wording of section 10, which specifically requires the former and only implies the
latter. This impoverished understanding of collaboration makes the joint authorship test
difficult to apply to works of collective authorship73.
The requirements for collaboration and for contributions which are not distinct,
when considered together, indicate that a work of joint authorship involves a working
together to create something that is more than the sum of its parts. This thesis will argue
that if this view of the joint authorship test is taken seriously, the test would be well-
suited to collective authorship.
2.2.3 A ‘Significant’ Contribution of the ‘Right Kind’
As mentioned above, the case law overwhelmingly focuses on the authorship limb of
the test. In order to be a joint author, a contributor must have made a significant
contribution of the right kind74. Joint authors do not need to contribute equally75. Courts
consider the nature of the work at issue as well as the quantity and quality of the
contribution in question76. Where a qualitative approach is taken it is not clear whether
this is in relation the requirement that a contribution be ‘significant’ or of the ‘right
kind’. It is difficult to identify any over-arching guiding principles from the case law as
73
This problem is particularly acute in relation to a perpetual work in progress like Wikipedia. The
current test for collaboration fails to convincingly resolve the issue of whether Wikipedia is best seen as a
series of derivative works or a work of joint authorship. 3.2.1 below.
74
Although this part of this test tends to be referred to collectively as the authorship limb, the authorship
part really only relates to the requirement that the contribution be of the ‘right kind’.
75
Godfrey v Lees (n61) [1995] EMLR 307, 325 (Blackburne J): ‘It is not necessary that his contribution
to the work is equal in terms of either quantity, quality or originality to that of his collaborators’.
76
Brown v Mcasso [2005] FSR 846 [42] (EWPCC) (affirmed on appeal [2005] EWCA Civ 1546).
33
decisions on joint authorship are often said to be limited to the facts at issue. Yet, this
also provides scope for the court to take predominant creative practices into account. In
this way, the joint authorship test contains a significant amount of flexibility that
enables it to adapt to changing practices and different creative contexts.
There is little judicial guidance on the meaning of ‘significant’. This is
complicated by the fact that this question is often considered together with the
requirement that the contribution be of the right kind. It is unclear whether the
requirement for a significant contribution is qualitative or quantitative. The only judge
to specifically comment on this point is Justice Blackburne in Fisher v Brooker, at first
instance, who suggested that it means ‘more than merely trivial’ (a de minimis
quantitative standard)77; although he also said that the relevant contribution in that case
was ‘on any view substantial’78. Some commentators have preferred this quantitative
approach, because it has the advantage of appearing aesthetically neutral, which accords
with the traditional reluctance of judges to make aesthetic judgments in copyright
decisions79.
77
[2006] EWHC 3239, [2007] FSR 255 [46]. In Fylde Microsystem (n38) [25] Laddie J thought the
relevant question was whether the contribution was ‘big enough’, which also seems to imply a
quantitative standard.
78
ibid [98]. Mr Justice Blackburne’s decision was partially reversed on appeal, the Court of Appeal
considering it to be an ‘extremely unusual case’ ([2008] EWCA Civ 287 [34]); then upheld in the House
of Lords ([2009] UKHL 41). Mr Justice Blackburne’s ruling on joint authorship was not challenged in
either appeal.
79
L Bently, ‘Authorship of Popular Music in UK Copyright Law’ (2009) 12(2) Information
Communication & Society 179, 197 acknowledges that it might not be possible, or even always desirable,
to completely exclude aesthetic value judgments from legal decision-making.
34
Bently has argued that the requirement for a significant contribution has tended to
invite judges to make value judgments about the quality of particular contributions80.
Zemer agrees that many of the cases demonstrate a subjective, qualitative approach to
this criterion81. In Beckingham v Hodgens, for example, it was thought significant that
the violin riff was memorable and catchy82. Similarly, in Hadley v Kemp it was thought
relevant that the saxophone fills were not ‘particularly memorable, tuneful or original...
they were just the sort of thing which any accomplished professional saxophonist would
have provided’ 83 . There was also a quantitative element in that case as the court
considered that the saxophone interlude was only 9% of the duration of the track. The
court in Brown v Mcasso also took a qualitative approach when it considered that
Brown was a joint author, his main contribution being to give the song an ‘authentic rap
feel’84.
Bently and Sherman consider a quantitative approach to be preferable to an
assessment of the aesthetic significance of a contribution85. This might be due to the
number cases in which the determinations on this point appear fairly subjective and
arbitrary 86 ; or where inappropriate weight appears to have been given to expert
evidence that does not take into account the particular features of the relevant genre87.
80
Bently ibid 190.
81
Zemer (n72) 287.
82
(n64) [49].
83
[1999] EMLR 589, 650.
84
[2005] FSR 40 [30].
85
Bently and Sherman 126.
86
Zemer (n72).
35
Vaver goes further, suggesting that the requirement for a significant contribution might
be an unnecessary gloss88.
These concerns might be better addressed by taking a third way. In line with the
importance of collaboration as indicated by the wording of section 10, ‘significant’
might be determined in light of the common design of the contributors. This implies a
qualitative approach and entails reference to aesthetic criteria drawn from the potential
authors’ shared sense of their creative realities (rather than the application of aesthetic
criteria, determined judicially, in isolation from that context)89.
In addition to making a significant contribution, a joint author must make a
contribution of the right kind. This is an entry point in the joint authorship test for much
of the normative content of the concept of authorship discussed above. Although there
is no requirement that a joint author’s contribution must be separately copyrightable,
the jurisprudence on originality is relevant to whether or not a contribution is of the
right kind90. The contribution must be ‘in the nature of authorship’, which means that it
must be a contribution to the creation of the work91. Additional suggestions after the
substance of the creative process is complete will rarely be sufficient 92. The application
87
Bently (n79) 198 discussing the inappropriate weight given to the evidence of classically trained
musicologists in cases involving popular music.
88
D Vaver, Intellectual Property: Copyright, Patents and Trade-Marks (2nd ed, Irwin Law 2011) 120-1.
89
I argue that there is a reliable external standard from which to judge the significance of a contribution:
the common design of the contributors as revealed by the social norms which regulate their collaborative
activities. This is explained in detail in Chapters 7 and 8.
90
In the US, some courts have required a joint author to provide a separately copyrightable contribution,
eg Childress v Taylor (n70); Erickson v Trinity Theatre 13 F3d 1061 (7th Cir, 1994); Thomson v Larson
147 F3d 195 (2d Cir, 1998). Cf. Gaiman v McFarlane 360 F3d 644 (7th Cir, 2004).
91
Hadley v Kemp (n37).
92
ibid; Brighton v Jones (n37).
36
of the authorship limb of the test depends upon the nature of the work in issue. So, for
example, contributions to performance of a musical work were not of the right kind93;
and contributions to the interpretation, and theatrical presentation, of a dramatic work
were not of the right kind94. In the case of a literary work, contributions akin to proof-
reading do not amount to authorship95.
The contribution must also be to the protectable expression, not just the ideas 96.
The idea/expression dichotomy is a notoriously slippery distinction. In Cala Homes the
court held that a person can be a joint author even if he has not put pen to paper.
According to Justice Laddie: ‘… to have regard merely to who pushed the pen is too
narrow a view of authorship’ 97 . Justice Laddie stressed the importance of the
contributions of concepts and emotions, over physical or mechanical contributions98. In
Robin Ray v Classic FM Justice Lightman took a narrower approach, maintaining that
Cala Homes might be an exceptional case and distancing himself from any suggestion
that ‘there is no restriction on the way in which a joint author’s contribution may be
funnelled into the finished work’ 99. In what he sees as restoring the primacy of the
idea/expression dichotomy, Justice Lightman insists that a joint author must ‘participate
93
94
Brighton v Jones (n37).
95
Fylde Microsystems (n38).
96
Donoghue v Allied Newspapers (n35); Brighton v Jones (n37).
97
Cala Homes (n65) 835-6.
98
ibid 835: ‘It is both the words or lines and the skill and effort involved in creating, selecting or
gathering together the detailed concepts, data or emotions which those words or lines have fixed in some
tangible form which is protected’.
99
Robin Ray (n13) 636 (Lightman J): ‘it appears to me the architects in [Cala Homes] were in large part
acting as “scribes” for the director. In practice such a situation is likely to be exceptional’.
37
in the writing’ 100 . This presents an unduly narrow view of authorship that fails to
appreciate the distinction between the expression and the fixation of the work, and does
not sit well with much of the case law101. In requiring responsibility for what appears on
the page, it is probable that Justice Laddie was suggesting responsibility for the
expression (not the fixation). It will be apparent from the case studies in this thesis that
Justice Laddie’s approach in Cala Homes is better adapted to the unique way in which
large groups of authors work together to create a copyright work. Despite this, there is a
real risk that Justice Lightman’s restrictive approach might be taken to the requirement
of ‘responsibility for what appears on the page’ for pragmatic reasons (to reduce the
number of potential joint authors) as I discuss in the third part of this Chapter 102.
2.2.4 Questions of Law, Questions of Fact
In light of the lack of analytical clarity in the case law, it may be useful to suggest a
more defined structure and, in particular, to separate the parts of the joint authorship test
which ought to be treated as questions of law from the parts that ought to be treated as
questions of fact. Although some commentators suggest that the distinction between
questions of law and questions of fact is ultimately a pragmatic one, Endicott carefully
demonstrates that an analytical approach is possible 103 . According to Endicott, a
100
ibid 636 Lightman J stressing ‘[w]hat is essential is a direct responsibility for what actually appears on
the paper’. Similarly at 637 when Lightman J states ‘… the plaintiff…was in no wise a collaborator in its
production’, he seems to be referring to the physical fixation of the work.
101
For example, Heptulla v Orient Longman [1989] 1 FSR 598 (Indian High Court); Donoghue v Allied
Newspapers (n35) 109.
102
2.3. See also M Rimmer, ‘Heretic: Copyright Law and Dramatic Works’ (2002) 2(1) QUT LRev 131,
139, 144.
103
T Endicott, ‘Questions of Law’ (1998) 114 LQR 292 noting that the analytical approach he proposes
incorporates both pragmatic and normative considerations. Endicott argues that it is possible to determine
which questions are questions of law by asking what the point is of treating certain questions as questions
of law.
38
question of law arises where the law requires the question to have a particular answer or
where the appellate tribunal wishes to elaborate on the law104. Adopting this approach,
it seems likely that the requirement that the contribution be of the ‘right kind’ or ‘in the
nature of authorship’ is a question of law, as it requires a particular answer. The law
requires a particular answer to this question because of its normative dimension105. As
an important organising concept in the CDPA, ‘authorship’ is shaped by the purpose of
copyright law and policy concerns that limit its scope106. Indeed, it is recognised that
although contributors might make agreements about who owns the copyright in a work,
they may not make an agreement about who counts as an author of the work 107 .
Although authorship is ultimately a legal question, the joint authorship test also
contains questions of fact.
According to Endicott, questions of fact are those questions that are capable of
decision either way; or to put it another way, the law does not require a particular
answer to these questions. The parts of the joint authorship test that are likely to be
questions of fact include: the requirement for a significant contribution; for
collaboration; and that the contribution not be distinct 108 . The last two questions
104
Endicott ibid. It may not always be immediately obvious which questions are questions of law,
particularly in the latter case (which is likely to entail some pragmatic considerations). Endicott does not
make this point in relation to a particular application of the law/fact distinction; rather, he proposes a
general analytical framework that might be used to distinguish questions of law and questions of fact.
105
The law might also require a particular answer to a question where delegating that answer to
prevailing social norms would conflict with other important policy goals that the law seeks to uphold. For
more on the limits of incorporating social norms in judicial decision-making, see 7.4.3 below.
106
In this way copyright law provides a good source of standards, etc. The importance of retaining a legal
dimension of authorship in the application of the joint authorship test is discussed in Chapters 7 and 8.
107
Copinger et al [4-38]; Samuelson v Producers Distributing [1932] 1 Ch 201, (1932) 48 RPC 580, 586;
Wiseman v George Weidenfeld [1985] FSR 525.
108
The meaning of these parts of the test should be determined in accordance with ordinary language:
Lord Reid in Cozens v Brutus [1973] AC 854, 861; Moyna v Secretary of State for Work and Pensions
[2003] UKHL 44, [2003] 1 WLR 1929.
39
indicate that a work of joint authorship involves a working together to create something
that is more than sum of its parts. These are matters that are capable of decision either
way; that is, copyright law does not require a particular answer to these questions109. If,
as I have suggested, the requirement for a significant contribution is best determined in
light of the particular common design shared by the contributors, it would also be a
question of fact 110 . This is the preferable approach as it leaves issues with a likely
aesthetic dimension to those in the best position to answer such questions (creators)111.
2.3 Three Critiques of the Application of the Joint Authorship Test
It is difficult to tell how the joint authorship test would apply to works of collective
authorship for two reasons. The first reason is that the application of the joint
authorship test is said to depend upon the facts of the case. The second reason is that
there is a lack of analytical clarity in the case law on joint authorship112. I suggest that
this lack of analytical clarity might explained by two preoccupations: a concern to
reduce the number of potential joint owners of copyright; and an anxiety about aesthetic
neutrality. These two reasons can be broken down to three critiques of the application
of the joint authorship test:
(i) the factual specificity of the joint authorship test;
(ii) the instrumental/pragmatic approach; and
109
Whether or not there is collaboration, for example, depends upon the shared understanding of the
contributors.
110
At 8.5 I argue that there are good reasons to consider social norms at this point in the analysis.
111
Endicott (n103) 294 argues that the most common use of the notion of questions of law or questions of
fact is to distribute decision-making power and responsibility.
112
Indeed, some textbooks resort to giving long lists of examples to explain the application of the test,
see text at n62.
40
(iii) a preoccupation with aesthetic neutrality.
This section explores these three critiques in order to foreshadow the insights for
copyright law that emerge from the consideration of the case studies of collective
authorship in the following Chapters.
2.3.1 The Factual Specificity of the Joint Authorship Test
The application of the joint authorship test is frequently said to be a question of degree
that depends upon the facts of the case 113 . This can make it difficult to predict the
outcome of a case, particularly where there is no precedent, as is the case for collective
authorship114. Although the factual specificity of the joint authorship test might result in
considerable uncertainty, ultimately this aspect of the test is a strength 115 . This is
because it provides inbuilt flexibility in that allows the concept of authorship to adapt
over time to new creative forms and techniques 116 . The factual specificity of the
application of the joint authorship test also allows judges to calibrate the provisions of
the joint authorship test to the creative context in which authorship occurs. This
provides a workable method of implementing a copyright scheme that provides a single
test for authorship and joint authorship that is meant to apply in a diverse range of
creative contexts.
113
Brighton v Jones (n37) [31] (Park J): ‘…to a considerable extent the question whether a person is or is
not the author of a work within [s10] is one of fact, assisted of course by the wealth of case law which has
considered the matter’.
114
A notable exception is Bulun Bulun v R & T Textiles (1998) 86 FCR 244 (Fed. Ct. of Australia) which
involved an Indigenous elder who claimed that his tribe had a communal authorship interest in an artwork
produced in accordance with customary law (4.2 below).
115
On the potential benefits of vagueness in the law: M Spence and T Endicott, ‘Vagueness in the Scope
of Copyright’ (2005) 121 LQR 657.
116
See for example, the change in the way that photography has been treated, Bently and Sherman 96.
41
2.3.2 The Instrumental or Pragmatic Approach
The second reason it is difficult to predict how the joint authorship test applies to works
of collective authorship is the lack of analytical clarity in the case law. One reason for
this is the preference for pragmatic or instrumental reasoning which is adopted out of a
concern about one of the potential consequences of finding joint authorship: joint
ownership of copyright117.
Joint authors are usually the first owners of the copyright that subsists in a joint
work118. Joint owners hold their copyright interests as tenants in common 119. There is a
presumption that each joint owner owns equal shares of the copyright interest 120 ;
however, this presumption has been rebutted in a number of cases where copyright was
found to be owned in unequal shares121. Each joint owner’s consent is required to grant
a licence to do any of the protected acts in relation to the work or to assign copyright in
that work, no matter how great or small their share of ownership of the copyright122.
There is no provision in the CDPA that a joint owner must not unreasonably withhold
117
This is not the only example of consequentialist reasoning. A classic example is the ‘principle’ that
what is copying is worth protecting, which seems to be a judicial sleight of hand designed to transfer
complicated considerations of policy to the infringement side of the analysis shifting focus from the
proper scope of the rights of the copyright owner to the conduct of the potential infringer.
118
CDPA s11(1), s10(3).
119
Lauri v Renad [1892] 3 Ch 402. They may hold their copyright interest as joint tenants as in Mail
Newspapers v Express Newspapers [1987] FSR 90, although this may be a special case given that it
concerned a prospective bride and groom who jointly commissioned the taking of wedding
photographs.
120
Prior v Landsdowne Press [1977] FLR 59, [1977] RPC 511; Stuart v Barrett [1994] EMLR 448;
Beckingham v Hodgens (n70).
121
For example, Bamgboye v Reed [2002] EWHC 2922 [42] (ownership divided one third to two thirds);
Fisher v Brooker [2009] UKHL 41 (Fisher’s organ solo entitled him to 40 per cent ownership of the
copyright in the song ‘A Whiter Shade of Pale’).
122
Powell v Head (1879) 12 Ch D 686. One owner can sue in relation to an infringement of copyright
subsisting in the work, but they may have to account to the other owners. One owner may also assign his
or her ownership share.
42
his or her consent. This makes the joint ownership rules appear poorly adapted to
situations involving multiple owners123, because they allow one contributor (who may
have made a relatively small contribution) a seemingly disproportionate power to chill
the exploitation of a copyright work at the expense of other contributors’ interests in
generating royalties and the general public’s interest in the dissemination of creative
works124.
Frequently, judges take a restrictive approach to the joint authorship test out of a
concern to reduce the number of potential joint owners of a work125. In doing so they
appear motivated by consequentialist reasoning: they seek to reduce the possibility of a
situation arising in which one contributor holds up the exploitation of the work.
Simplifying ownership might also mean that assignments and licences of copyright are
easier and cheaper to effect126. Pragmatic, instrumental reasoning of this type is evident
in many of the cases. Justice Park in Hadley v Kemp commented, for example: ‘…[i]t
would be surprising if a slight contribution was enough to make a person a joint author
and thereby make him an equal owner with another or others who had contributed far
more than he had’127. Although this quote seems to rely upon the misassumption that
123
Copyright also provides little guidance as to contributors’ rights with respect to one another and in
relation to the work as a whole, see: Elkin-Koren (n20); RC Dreyfuss, ‘Collaborative Research: Conflicts
on Authorship, Ownership and Accountability’ (2000) 53 Vanderbilt LRev 1161.
124
The position in the US is quite different. One joint owner can licence any use of the work, with only
the requirement to account to the other authors. Interestingly, a restrictive approach to the application of
the joint authorship has also been adopted in this context based on the concern that creators may refrain
from asking for suggestions of others out of fear that those people might claim authorship and gain
control over subsequent uses of the work: see, for example, Aalmuhammed v Lee 202 F3d 1227 (9th Cir,
2000).
125
D Vaver, Copyright Law (Irwin Law, 2000) 76; D Vaver (n88) 121; Rimmer (n102) and L Bently,
‘Copyright and the Death of the Author in Literature and Law’ (1994) 57 MLR 973.
126
Bently ibid.
127
Hadley v Kemp (n37) 643.
43
joint authors are joint owners in equal shares (which is not always the case), Justice
Park repeats this sentiment in Brighton v Jones128.
In order to restrict the number of joint authors, judges use a combination of
strategies. One involves finding there has not been enough of a contribution129. In so
doing, the cases seem to imply a higher standard for joint authorship than individual
authorship. Another strategy is to privilege some contributors over others. Most of the
authorship requirements can be employed strictly to privilege certain contributors over
others; for example, finding that there has not been a contribution of an authorship-type
(eg, contribution to interpretation or performance, not creation) 130 . The fact that
decisions depend on their facts makes it even easier to employ the
instrumental/pragmatic approach. Most frequently, it is unclear whether the
contribution is not ‘significant’ or not of the ‘right kind’131. In some of the older cases
the entire joint authorship test is run together, so that it is difficult to tell whether the
decisive factor was an absence of collaboration or not enough of a contribution of the
right kind132. To get a flavour of the inconsistencies one might compare Brighton v
Jones with Hadley v Kemp. In Brighton v Jones the judge focused on contributions to
the fixation, considering that although some of the concepts and dialogue came from
Brighton, it was Jones that actually wrote the words on her laptop (she was thought to
128
Correcting the point on equal shares at [34].
129
Brighton v Jones (n37) [34].
130
131
In Brighton v Jones (n37), for example, a director did not make enough of the right kind of
contribution to be a joint author despite the judge describing her contribution as ‘valuable and important’.
132
Levy v Rutley (n68); Springfield v Thame (n69).
44
be the sole author)133. In Hadley v Kemp the court took the opposite approach. The
actual process of fixation seems to have been completely disregarded. The musical
works were held to have been already ‘fixed’ in Gary Kemp’s musical consciousness
before he presented it to the band; therefore he was the sole author134.
In adopting the instrumental/pragmatic approach, judges seek to find one person
or a small number of persons who most ‘deserve’ to own the copyright. This leads them
to restrict authorship to those who appear to have done most of the work to the
exclusion of contributors who have made quantitatively less important contributions in
all but the most explicitly egalitarian environments135. This approach is likely to tend to
favour parties who have exercised control over the creative process 136. Where many
contributors have made creative contributions to a work it might be tempting to
consider the most dominant or powerful contributors to be the only authors. Dominant
players in the relevant industry will be the most likely to be able to exercise a great deal
of control over the creative process 137 . These players might also seem to deserve
copyright protection as they may appear best placed to ensure the work’s effective
exploitation. Granting authorship only to those figures with the most control at the
133
(n37) [56].
134
(n37) 639. Cf. D Free, ‘Beckingham v Hodgens: The Session Musician’s Claim to Music Copyright’
(2005) 1(3) Entertainment and Sports LJ 93 <http://www.warwick.ac.uk/go/eslj/issues/volume1/
number3>. The band members were said to be ‘interpreting’, not creating. Free argues that if they were
interpreting something, it was not something created by Gary Kemp.
135
Compare Hadley v Kemp (n37) and Stuart v Barrett (n120) which involved ‘collective jamming’.
136
For example in Hadley v Kemp (n37) 641 Gary Kemp was described as a ‘control freak’ and the band
as ‘not a democracy’; see also Brighton v Jones (n37). RR Kwall, ‘Author-Stories: Narrative’s
Implications for Moral Rights and Copyright's Joint Authorship Doctrine’ (2001) 75 Southern California
LRev 1, 5 warns that the restrictive application of the joint authorship test in the US also tends to
privilege dominant contributors.
137
On the dangers of allowing authorship to gravitate to the most powerful players, see Dreyfuss (n123)
1206, 1209. Also 8.2 below.
45
expense of other creative contributors, however, tends to distance authorship from its
core meaning in copyright law (a contribution of creative choices evident the
expression)138.
In trying to restrict authorship to few contributors, the case law seems to imply
that there is a higher standard of authorship for joint works than for individually
authored works. This does not make much sense on the conceptual level as ‘authorship’
implies the same consequences and has the same role in the CDPA whether a work is
139
individually or jointly authored . Collaboration is becoming an increasingly
important, and in some cases an essential, way of working. Sometimes groups can
create works that individuals cannot. If one of the goals of copyright is to promote
cultural flourishing it seems perverse to encourage individual authorship at the expense
of collaborative endeavours. On the view that copyright’s purpose is to reward and
incentivise authors, there seems little basis to arbitrarily exclude valuable contributors
to a collaborative project in the search of a small number of potential ‘owners’140. In
some cases the work may not have existed at all without the efforts of large number of
people141. Surely issues of ownership ought to be addressed by reconsidering the rules
on joint ownership rather than tinkering with the concept of authorship142.
138
This can be seen in the US where control has become an important part of the joint authorship test, see
FJ Dougherty, ‘Not A Spike Lee Joint? Issues in the Authorship of Motion Pictures Under US Copyright
Law’ (2001) 49 UCLA LRev 225.
139
There may be a stronger argument for different notions of authorship for the other guises which this
concept might take in the CDPA. L Bently and L Biron, ‘Discontinuities Between Legal Conceptions of
Authorship and Social Practices: What, if anything, is to be done?’ forthcoming in M van Eechoud (ed),
The Work of Authorship (Amsterdam UP 2014) 237 suggest the possibility of broader rights of attribution
as a way to address the discontinuities between legal conceptions of authorship and social practices.
140
Kwall (n136) 58.
141
Vaver (n88) 52-53.
142
The instrumental/pragmatic approach is arguably worse than this as it appears to empty the concept of
authorship of most of its meaning – treating it instead, as a placeholder for the separate concept of
46
The instrumental/pragmatic approach, in focusing upon one outcome of a decision
to include a contributor as a joint author (joint ownership), ignores another important
consequence of this decision: the denial of moral rights. Yet, moral rights (particularly
the attribution right) might sometimes be more significant to creators than copyright
ownership. The instrumental/pragmatic approach ignores the realities of collaboration
and the need to incentivise contributors to collaborative work 143 . Furthermore, as
instrumental/pragmatic concerns are accommodated at the expense of a thorough
analytical approach to the joint authorship test, this approach tends to result in legal
uncertainty. This uncertainty has a potentially chilling effect on creativity with a
collaborative element 144 . The impoverished understanding of ‘collaboration’ and the
diminished importance of this part of the test in the UK is also likely to contribute to a
difficulty in conceptualising collaborative creative processes. This leads to the
restriction of authorship to the most obvious/substantial/dominant contributor and a
failure to explicitly consider the important value of other contributions to a work. In this
way the instrumental/pragmatic approach might be seen as an instantiation of the
romantic authorship myth: the assumption that creativity is the product of the mind of a
singular genius figure.
ownership. For suggestions on how the rules of joint ownership might be better adapted to collective
authorship see 8.4.3.
143
Bently (n125) noting that copyright law prefers to minimise the number of authors rather than reflect
the ‘realities’ of collaboration.
144
PS Fox, ‘Preserving the Collaborative Spirit of American Theatre: The Need for a ‘Joint Authorship
Default Rule’ in Light of the Rent Decision’s Unanswered Question’ (2001) 19 Cardozo Arts &
Entertainment LJ 497, 498.
47
2.3.3 A Preoccupation with Aesthetic Neutrality
I have suggested that an unwillingness to explicitly consider relevant aesthetic
considerations also leads to a lack of analytical clarity in joint authorship cases 145. The
application of the joint authorship test has been affected by the ‘supposed terrors for
judicial assessment of matters involving aesthetics’ 146 . The principle that originality
does not require any ‘novelty, usefulness, inventiveness, aesthetic merits, quality or
value’ is a long-standing feature of copyright law147. In this subsection I argue that the
(valid) concern that copyright decisions ought not to depend upon the aesthetic merit of
a work has developed into a general judicial anxiety about making any sort of aesthetic
value judgement. This preoccupation with aesthetic neutrality, however, is misplaced
because aesthetic criteria are inescapably relevant to a determination of authorship148.
The preoccupation with aesthetic neutrality is based upon a legitimacy concern:
that judges are poorly placed to make aesthetic judgments about a work149. As Lord
Reid puts it:
Judges have to be experts in the use of the English language but they are
not experts in art or aesthetics.… we must avoid philosophic or
metaphysical argument about the nature of beauty, not only because there
does not seem to be any consensus about this but also because those who
145
J Cohen, ‘Creativity and Culture in Copyright Theory’ (2007) 40 U of California Davis LRev 1151
discusses copyright scholars’ preoccupation with neutrality and abstraction.
146
Burge v Swarbrick [2007] HCA 17 [63].
147
Sawkins v Hyperion (n41) [31].
148
Recently, Barton Beebe has discussed the role of the concept of aesthetic progress in the US copyright
tradition, ‘Intellectual Property and the Problem of Aesthetic Progress’, Ninth Annual International
Intellectual Property Lecture, Emmanuel College, University of Cambridge, 11 March 2014.
149
Cf. George Hensher v Restawile [1976] AC 64, 97 Kilbrandon LJ did not think that aesthetic matters
were difficult or unseemly for a court to decide (although he thought this unwarranted in that case).
48
are ignorant of philosophy are entitled to have opinions about what is
artistic.150
Despite this concern to maintain neutrality, a number of scholars have convincingly
argued that copyright cases are affected by cultural biases151. This seems to suggest that
aesthetic value-judgments do, in fact, slip into judicial decisions.
It is difficult to make sense of the concept of authorship, especially as it is framed
in the CDPA, without any resort to aesthetic criteria. In determining authorship, judges
often assess a putative author’s contribution in terms of the category of work concerned.
It is difficult to assess the literary, dramatic, musical or artistic nature of a contribution
without making some sort of aesthetic judgment about what makes something literary,
dramatic, etc152. Indeed, such judgments are apparent in the case law. In Norowzian v
Arks the court held that dramatic work must be capable of being ‘performed’ 153; in
Lucasfilm v Ainsworth, a sculpture was held to be something that is intended to be
enjoyed visually, not something that is merely utilitarian or functional 154; and the Court
of Appeal in Hollinrake v Truswell considered that a literary work must be ‘intended to
afford either information and instruction, or pleasure in the form of literary
150
George Hensher ibid 78, also 94 (Simon LJ), 96 (Kilbrandon LJ).
151
A Barron, ‘Copyright Law and Musical Practice: Harmony or Dissonance?’ (2006) 15(1) Social and
Legal Studies 25 discussing copyright’s complicity with the cultural bias against the popular. K Bowrey,
‘The Outer Limits of Copyright Law – Where Law Meets Philosophy and Culture’ (2001) 12 L and
Critique 75 discusses the influence of Western views about creativity in denying the communal
authorship claim of an Indigenous tribe, see 4.2 below.
152
Only very minimal guidance about the meaning of these terms can be found in the CDPA. It is a
matter that has been essentially left to case law to define, see further, J Pila, ‘An Intentional View of the
Copyright Work’ (2008) 71(4) MLR 535.
153
Norowzian v Arks (No 1) [1998] FSR 394 (Ch).
154
Lucasfilm v Ainsworth [2008] EWHC 1878, [2009] FSR 103.
49
enjoyment’155. Earlier in this Chapter, I argued that the requirement for some creativity
is bursting from the seams of the CDPA. In light of this preoccupation with aesthetic
neutrality, it becomes apparent why judges have been hesitant to explicitly embrace
such a requirement in the past.
The requirement of authorship is often seen as a shorthand way of asking: is this
the sort of thing that copyright ought to protect? This inquiry is far from value-neutral.
Indeed, in many cases the application of copyright’s rules of subsistence appears to
have be influenced by public policy concerns156. Of course, the consideration of policy
concerns and the use of aesthetic criteria to guide decision making where these yield
principles which shed light on the meaning of the CDPA is not just helpful, it is
essential. The elision of such matters, however, leads to unhelpfully opaque reasoning
as is evident in some of the joint authorship cases. In assessing the value and
significance of a contribution to a work relative to other contributions judges
necessarily have regard to aesthetic matters. For example, in Beckingham v Hodgens it
was thought relevant that a violin riff was memorable and catchy; and in Brown v
Mcasso changes to the lyrics of a song which gave it an authentic rap feel amounted to
joint authorship. Yet, in the reasoning of these cases it is not clear whether these
considerations relate to whether the contribution was significant or of the right kind.
Instead of striving for neutrality, a better approach would be to avoid subjective
judgments of the aesthetic merit of a work. The aesthetic characteristics of a particular
contribution are likely to be highly relevant to questions of authorship or joint
155
[1894] 3 Ch 420 (CA), 428 (Davey LJ); approved in Exxon (n24) 142-143 (Stephensen LJ).
156
Spence (n55) 80-82.
50
authorship. Instead of being preoccupied with trying to achieve aesthetic neutrality,
judges ought to explicitly engage with questions of how best to determine and apply
aesthetic criteria. This includes considering the relevance of the intention of the author,
the views of experts and social norms which operate in the particular creative context.
An explicit, analytical approach is more likely to enhance the perceived legitimacy of
judicial decisions than the surreptitious consideration of aesthetic criteria. Such an
approach is also likely to result in greater analytical clarity in the case law.
2.4 Copyright Scholarship: Theories of Authorship
The three critiques of the joint authorship test offered in the previous section suggest
that it is difficult to predict how the joint authorship test might apply to works of
collective authorship. This complicates the task of assessing whether the current test
provides a suitable mechanism for determining the authorship of such works. In this
section I consider whether the scholarly literature on authorship in copyright might shed
some light on this question. I begin with the scholarship criticising the influence of the
romantic concept of authorship on copyright law. I note that proponents of this view
have offered few concrete, implementable suggestions on how the authorship of
collaborative works ought to be determined. Then, I situate this thesis in the context of
broader scholarship addressing the concept of authorship in copyright, particularly, the
influential work of Jane Ginsburg. I note an alternative to the pragmatic/instrumental
approach that has been proposed by prominent US copyright scholar, Melville Nimmer.
I also consider recent trends in copyright scholarship which stress the value of taking
into account the context in which creativity occurs. I conclude this section with a
51
comment on the inter-relationship of legal and cultural notions of authorship, which
underlies the method of this thesis157.
The debate on the role of literary theory and the influence of the ‘romantic’
conception of authorship in copyright law has been the subject of many pages of
academic writing. Some commentators have called for a conception of authorship in
copyright law that is more aligned with literary theory158. These commentators consider
the historical origin of copyright law’s conception of authorship as being intertwined
with the romantic literary tradition of the solo genius, which conceives of an author’s
work as an organic emanation from one individual and the embodiment of his or her
personality 159 . This conception might be contrasted with a post-modern vision of
authorship as a process of cultural negotiation between author and audience in which
the work functions as a cultural pastiche 160 . The observation typically made is that
whilst notions of authorship have progressed in literary theory under the influence of
post-modernist and post-structuralist ideas, copyright law’s conception of authorship
has remained frozen, dominated by romantic conceptions of authorship. Many
commentators blame this romantic view of authorship for a copyright system which is
157
See 1.2.
158
For example: M Woodmansee, ‘On the Author Effect: Recovering Collectivity’ in Woodmansee and
Jaszi (n66) 15; M Woodmansee, ‘The Genius and the Copyright: Economic and Legal Conditions of the
Emergence of the “Author”’ (1983-1984) 17 Eighteenth Century Studies 425; D Nimmer, ‘Copyright in
the Dead Sea Scrolls: Authorship and Originality’ (2001) 38(1) Houston LRev 1.
159
Woodmansee and Jaszi (n66); J Boyle, Shamans, Software and Spleens: Law and the Construction of
the Information Society (Harvard UP 1996); D Saunders, Authorship and Copyright (Routledge 1992); M
Rose, Authors and Owners: The Invention of Copyright (Harvard UP 1993); M Rose, ‘The Author as
Proprietor: Donaldson v Beckett and the Geneology of Modern Authorship’ (1988) 23 Representations 51.
Cf. M Lemley, ‘Romantic Authorship and the Rhetoric of Property’ (1997) 75 Texas LRev 873 and A
Adler, ‘Against Moral Rights’ (2009) 97 California LRev 263. See also discussion in Bently (n125).
160
JP Barlow, ‘The Economy of Ideas: Selling Wine Without Bottles on the Global Net’ (1993)
<http://homes.eff.org/~barlow/EconomyOfIdeas.html>.
52
seen to grant rights which are too broad in scope and granted in too narrow
circumstances161.
Aspects of copyright law appear to support this critique. There is some foundation
to the claim, for example, that basic concepts of copyright subsistence are inadequately
equipped to deal with some situations of collective authorship; and specifically, that the
concepts of the ‘work’ and of the ‘author’ are not broad or flexible enough to
encompass certain works or to extend to certain creators perceived to merit copyright
protection162. Many elements of copyright law, however, are at odds with a claim that
copyright law has a romantic view of the author, for example:
 the relatively low standard of the originality test;
 the application of the idea/expression dichotomy which implicitly recognises
that authors are not the source of original knowledge, but rather are compilers
assembling their works from a pre-existing stock of ideas which belong to the
public domain;
 judicial reluctance to assess the aesthetic merits of a work in applying the
provisions of the CDPA; and
161
Woodmansee and Jaszi (n66) 10; A Chander and M Sunder, ‘The Romance of the Public Domain’
(2004) 92 California LRev 1331; P Jaszi, ‘Towards a Theory of Copyright: The Metamorphoses of
Authorship’ (1991) 2 Duke LJ 455 argues that this romantic conception of authorship in copyright law
operates to conceal the influence of economic forces in discussions of literary property; M Rose, Authors
and Owners (n159) 133: ‘The story of copyright…is one of steady expansion’. Cf. Lemley (n159), O
Bracha, ‘The Ideology of Authorship Revisited: Authors, Markets, and Liberal Values in Early American
copyright (2008) 118 Yale LJ; Bently (n59).
162
R Merges, ‘Locke for the Masses: Property Rights and the Products of Collective Creativity’ (2008)
36(4) Hofstra LRev 1179; DJ Halbert, Intellectual Property in the Information Age: The Politics of
Expanding Ownership Rights (Quorum 1999); A Stokes, ‘Authorship, Collaboration and Copyright a
View from the UK’ (2002) Entertainment LRev 121.
53
 a bias in favour of corporations over individuals (for example the rules about the
ownership of copyright subsisting in works created by employees and the ease
of transferring and assigning copyright interests)163.
It follows from these elements that even if copyright law did reflect a romantic notion
of authorship164, such a notion does not completely explain the modern legal rules for
determining the subsistence of copyright 165 . Perhaps this is unsurprising as law and
literary theory are engaged in two very different enterprises166. Literary theory seeks
conceptions of authorship which aid in understanding the meaning of a text. In this
context post-modern ideas about authorship may well be revealing. Copyright law is
usually thought to be largely unconcerned with the meaning of a work, seeking instead
to grant property rights to particular identifiable authors in respect of original works.
Regardless of whether the romantic author model is an adequate explanation of
current copyright law 167 , it is generally agreed that copyright is not well adapted
163
Lemley (n159) 883.
164
ibid 879.
165
Lemley (n159); Adler (n159); Bently (n59).
166
M Price and M Pollack, ‘The Author in Copyright: Notes for the Literary Critic’ in Woodmansee and
Jaszi (n66) 439; Saunders (n159) 233: ‘…to have believed that a mutation in the literary theory of
authorship must produce a corresponding mutation in the law of copyright was perhaps a sign of naivety
or aesthetic arrogance’; R Posner, Law and Literature. A Misunderstood Relationship (Harvard UP 1988)
17: ‘…the problems of literary and of legal interpretation have little in common except the word
“interpretation”’. Posner argues that literature and literary theory are unlikely to help lawyers and judges
to interpret legal texts.
167
Some argue that the romantic notion of authorship may not have existed or might be greatly
exaggerated: A Rahmatian, Copyright and Creativity: The Making of Property Rights in Creative Works
(Edward Elgar 2011).
54
collaborative authorship168. Critiques of the supposed romantic notion of authorship in
copyright law, however, offer few concrete solutions to the problem of determining the
authorship of collaborative works 169 . A post-modern definition of authorship is
essentially a definition of non-authorship, that is, the ‘death of the author’ to make way
for the ‘birth of the reader’170. This can be demonstrated by considering Zemer’s work
on the concept of authorship in copyright law, which might be seen to approach a post-
modern view 171 . Zemer argues that the general public ought to be considered joint
authors of all creative works, because a work is the product of a cultural discourse that
is constituted by the general public. Zemer’s proposal, although well-argued and in
some ways conceptually attractive, would be extremely difficult to implement172. From
a practical point of view, granting everyone authorship rights is almost the same as
granting no one authorship rights. If every person is an author of every work then the
label of author is emptied of any meaning because it cannot distinguish between
readers, writers and the general public as they relate to works of authorship. In addition,
such a conception does not reveal any criteria for determining rights between primary
and derivative authors173. It seems that neither a romantic nor a post-modern view of
168
Stokes (n162) 121, 124: ‘…hopes for a simple, pragmatic, fair and meritocratic solution to the
questions “who owns a collaborative work” and “what is each collaborator’s share of it” are
comprehensively dashed by the actual operation of the CDPA’.
169
Commentators tend to focus on the infringement side of the analysis, suggesting a more liberal
approach to copyright defences.
170
See R Barthes, ‘The Death of the Author’ in S Heath (ed) Image, Music, Text (Fontana 1977); M
Foucault, ‘What is an author?’ in D Bouchard (ed) Language, Counter-Memory, Practice: Selected
Essays and Interviews by Michel Foucault (Cornell UP 1977).
171
Zemer (n63). Although he does not consider his work in these terms: ‘I do not announce the “death of
the author”’ (228); ‘I distance my approach from Foucault and Barthes’ (1450). For an argument in
favour of ‘postmodern law’, see: A Todd, ‘Painting a Moving Train: Adding “Postmodern” to the
Taxonomy of Law’ (2008) 40(1) U of Toledo LRev 105.
172
RR Kwall, ‘The Author as Steward “for Limited Times”: A Review of “The Idea of Authorship in
Copyright” (Lior Zemer, Ashgate Press, 2007)’ (2008) 88(3) Boston U LRev 685.
173
Lemley (n159) 885.
55
authorship is helpful in determining the boundaries and content of the concept of the
‘author’ in copyright law.
The concept of authorship is surprisingly under-theorised given its importance as
an organising concept in the CDPA. Whilst many scholars have written about the
different theoretical justifications of copyright law, few have provided comprehensive
theories of authorship. Although it is difficult to identify a fully articulated, coherent
concept of authorship in the case law174, some scholars have taken some positive steps
in this direction175. This thesis develops a number of the points suggested by one such
scholar: Jane Ginsburg.
Ginsburg has identified six principles ‘in search of an author’ which characterise
many legal systems’ attempts to identify the author of a copyright work 176. Ginsburg
argues that an ‘author’ in copyright law generally refers to the person who
conceptualises and directs the development of a work, rather than a person who follows
those directions to execute it177. This seems in line with Justice Laddie’s flexible view
in Cala Homes that authorship is about more than who pushed the pen. Ginsburg notes
that authorship is a shifting concept that is sensitive to context and that might depend
upon the nature of a work and the number of putative authors. This aspect of authorship
174
Saunders (n159) 213, 235; Ginsburg (n7) 1066.
175
R Versteeg, ‘Defining “Author” for Purposes of Copyright’ (1996) 45 American U LRev 1323 argues
that the heart of the concept of authorship is the communication of original expression. A Drassinower,
‘From Distribution to Dialogue: Remarks on the Concept of Balance in Copyright Law’ (2009) 34(4) J of
Corporation L 991 presents a concept of authorship as a mode of communication or public address. A
similar definition of authorship has been adopted by C Graber and G Taubner, ‘Art and Money:
Constitutional Rights in the Private Sphere?’ (1998) 18 OJLS 61.
176
Ginsburg’s principles point to characteristics that sometimes indicate authorship, such as, sweat of the
brow, highly skilled labour, intent to be a creative author, and investment: (n7) 1071.
177
ibid 1072 noting that Walter v Lane (n14) may not sit perfectly with this.
56
is useful because it provides copyright law with the flexibility to adapt to new creative
practices. Ginsburg argues that much of the confusion about the concept of authorship
arises from eliding the concept of authorship and ownership (as has been observed at
2.3.2). The thesis explicitly considers the gap between authorship and ownership,
reasserting the importance of preserving a distinction between these two concepts.
Ginsburg calls for an emphasis on those principles which focus on the author
themselves and the act of creating 178 . According to Ginsburg an ‘author’ for the
purposes of copyright law is, or should be,
…a human creator who, notwithstanding the constraints of her task,

succeeds in exercising minimal personal autonomy in her fashioning of the
work. Because, and to the extent that, she moulds the work to her
vision…179
A point of commonality between Ginsburg and Justice Laddie’s observations in Cala
Homes is that both suggest that authorship concerns a link between the subjectivity of
the author and his or her creation that is difficult to express in precise terms 180. I have
argued that this might best be expressed as a minimum core requirement for authorship:
a more than de minimis contribution of creative choices that are evident in the
expression.
Although US copyright law is different to UK law in crucial respects, it is useful
to note the opposing views of two eminent US copyright scholars on the proper
178
Ginsburg is not alone in calling for renewed focus on the role of the concept of authorship in copyright
law, see: Bently (n59).
179
ibid 1092.
180
Also, Nimmer (n158) 159: ‘…copyright protection arises only for works that reflect an intent to
produce something personal or subjective’.
57
approach to determining the authorship of collaborative work 181 . Melville Nimmer,
concerned to ensure that collaboration is incentivised and rewarded, proposed that all
those who have made a more than de minimis contribution of an authorship-type,
pursuant to a common design, ought to be recognised as joint authors 182 . Paul
Goldstein, on the other hand, proposed that a contributor must have made an
independently copyrightable contribution to count as a joint author183. US courts, in
recent years, have tended to favour Goldstein’s approach, which has generally resulted
in awarding authorship only to one or two dominant contributors 184. Courts have been
concerned that Nimmer’s approach might result in too many authors, which may
interfere with the efficient exploitation of the work185. Further, there was a concern that
a more inclusive approach might discourage creators from consulting with others in
order to avoid losing sole authorship186. Many scholars criticise the current approach on
the basis that it does not correspond well to the creative reality of many collaborative
endeavours, particularly, because under this test, some highly collaborative works may
seem to have no authors 187 . The debate is significant because it suggests a viable
181
M LaFrance, ‘Authorship, Dominance, and the Captive Collaborator: Preserving the Rights of Joint
Authors’ (2001) 50 Emory LJ 193, 196–197, 259–261; PS Fox, ‘Preserving the Collaborative Spirit of
American Theatre: The Need for a ‘Joint Authorship Default Rule’ in Light of the Rent Decision’s
Unanswered Question’ (2001) 19 Cardozo Arts & Entertainment LJ 497, 507–509; T Huang, ‘Gaiman v
McFarlane: The Right Step in Determining Joint Authorship for Copyrighted Material’ (2005) 20(1)
Berkeley Technology LJ 673.
182
MB Nimmer and D Nimmer, Nimmer on Copyright (1976 edition) 283 (‘more than a word or a line’);
(2011 edition): ‘…copyright’s goal of fostering creativity is best served, particularly in the motion picture
context, by rewarding all parties who labor together to unite idea with form, and that copyright protection
should extend both to the contributor of the skeletal ideas and the contributor who fleshes out the project’.
183
P Goldstein, Copyright: Principles, Law and Practice (Little Brown & Co 1989) 4.2.1.2.
184
This is due to additional requirements that have been added including: mutual intent to be joint authors
and the control test in Aalmuhammed v Lee (n124) discussed at 6.2.5 below.
185
Erickson v Trinity Theatre (n90) 1069-71 (7th Cir, 1994); Childress v Taylor (n70) 506-7.
186
Aalmahummed v Lee (n124).
187
Nimmer (2011 edn) (n182) 6.07[A][3][c] citing, in particular, the opinion of Judge Posner in Gaiman
v McFarlane (n90) 659 that in some cases this would result in ‘peeling the onion until it disappeared’.
58
alternative to the restrictive instrumental/pragmatic approach. This thesis follows the
path that Nimmer has lit, arguing for an inclusive approach to the application of the
joint authorship test.
Many commentators have noted a gap between social practices and copyright
law’s concept of authorship 188 . Dreyfuss offers a new concept of the ‘collaborative
work’ to realign copyright law with social practices in large scientific collaborations189.
Biron and Cooper offer insights from aesthetics, arguing that the concepts of role,
intention and authority might be usefully employed to enhance copyright law’s
flexibility and make it more adaptable to changes in creative practice over time 190 .
Kwall stresses the importance of considering the perspectives of authors (as reflected in
the narrative of creation)191. These theorists are part of a growing body of literature that
stresses the relevance of the context in which creativity occurs192. Some scholars take
this even further, arguing that in some cases social practices might provide an effective
substitute for copyright law193. In Chapter 8, I argue that there are good reasons to have
regard to social practices in the application of copyright (as well as some potential
dangers).
188
Bently and Biron (n139). On the gap between copyright law and practices of attribution see also: C
Fisk, ‘Credit Where It’s Due: The Law and Norms of Attribution’ (2006) 95 Georgetown LJ 49; G
Lastowska, ‘The Trade Mark Function of Authorship’ (2005) 85 Boston U LRev 1171.
189
Dreyfuss (n123).
190
L Biron and E Cooper, ‘Copyright and Multiple Authorship: Insights from the Philosophy of Art’,
HERA Working Paper, 21 June 2013, referring to the relational theories of Arthur Danto, George Dickie
and Jerrold Levinson.
191
Kwall (n136).
192
LJ Murray, S Tina Piper and K Robertson, Putting Intellectual Property in Its Place: Rights
Discourses, Creative Labor, and the Everyday (OUP 2014); Cohen (n145).
193
E Rosenblatt, ‘A Theory of IP’s Negative Space’ (2011) 34(3) Columbia J of L and the Arts 317; see
the literature on intellectual property’s negative spaces referred to on p8 in the text at n37.
59
It might be helpful to make some general remarks about the relationship between
copyright’s notion of authorship and other notions of authorship. There are many
possible dimensions of authorship including: a historical dimension (located
temporally); a social dimension (who judges who is an author); a cultural dimension
(what it means to be an author of a work); a technological dimension (the mechanical
aspects of creation)194; and a legal dimension (who the law considers to be an author for
the purposes of copyright law and related rights). In cultural terms, whether or not a
creator is an author is a socio-historical question based upon technology and politics195.
In this context, what it means to be an author will be different to different interpretative
communities (ie literary critics, authors, copyright lawyers and the public). In cultural
terms, the label ‘author’ is important in ensuring that creators are accountable for what
they create by identifying an originator of a work who is socially and legally
responsible for its content 196 . Thus authorship has a specific meaning which allows
readers to make generalisations about the truth and reliability of a work by reference to
broadly located sources of meaning in respect of the particular author and in respect of
the aesthetic or expressive tradition to which the work belongs197.
194
J Griffin, ‘The Changing Nature of Authorship: Why Copyright Law Must Focus on the Increased
Role of Technology’ (2005) IPQ 135, 137 argues that technology is a central component of authorship.
195
M Ross, ‘Authority and Authenticity: Scribbling Authors and the Genius of Print in Eighteenth
Century England’ (1992) 10 CAELJ 495, 495.
196
Griffin (n194) 193: ‘authorship is used as a method of quality control by readers’; Lastowska (n188); J
Ginsburg, ‘The Author’s Name as a Trade Mark: A Perverse Perspective on the Moral Right of
“Paternity”?’ (2005) 23 Cardozo Arts and Entertainment LJ 379.
197
Shapin considers the significance of trust and its relation to cultural ideas about the location of textual
authority in constituting knowledge: S Shapin, A Social History of Truth: Civility and Science in
Seventeenth-Century England (U of Chicago P 1994); see also J Pila, ‘Authorship and e-Science:
Balancing Epistemological Trust and Skepticism in the Digital Environment’ (2009) 23 Social
Epistemology 1 on the role of authorship in balancing epistemological trust and skepticism in e-science.
60
Scholars agree that legal definition of authorship might be influenced by other
dimensions of authorship (ie what it means to be an author or to create a literary work
in social and cultural terms, etc)198. Coombe argues that the relation between law and
culture is best conceived of as one of ‘mutual rupturing’199. The attribution of the label
of ‘author’ in copyright law has significance for the public at large because Intellectual
Property Law can serve as a valuable source of ‘default rules, policy and public interest
safeguards’200. By defining who is an author, copyright law has an important role to
play in constituting textual authority (who can be held responsible for that work)201. The
wider interpretative community’s attribution of textual authority is necessarily
influenced by whom copyright law considers to be an author. But, because the
relationship between law and culture is not simple or linear, it may be informative to
consider significant reasons for the attribution of authorship that are external to the
legal definition202.
Cultural ideas have a place within legal notions of authorship. Once it is
acknowledged that it is virtually impossible to decide who counts as an author without
reference to aesthetic criteria, it must be recognised that there are better sources of
information about these criteria external to legal discourse. Musicians have a valuable
198
Pila (n152) 557 argues that legal conceptions of authorial works ought to be anchored in non-legal
conceptions of them.
199
R Coombe, ‘Contingent Articulations: A Critical Cultural Studies of Law’ in A Sarat and T Kearns
(eds) Law in the Domains of Culture (U of Michigan P 1998); R Coombe, ‘Fear, Hope and Longing for
the Future of Authorship and a Revitalized Public Domain in Global Regimes of Intellectual Property’
(2003) 52 DePaul LRev 1171.
200
Pila (n197); Dreyfuss (n123); R Coombe, The Cultural Life of Intellectual Properties: Authorship,
Appropriation and the Law (Duke UP 1998) particularly the Introduction.
201
Saunders (n159) 213: ‘in a culture as juridified as ours, the phenomenon of authorship cannot be
defined independently of, or prior to, its legal conditions’.
202
ibid.
61
perspective, and greater knowledge, of what constitutes a musical contribution than
judges or lawyers. The creative context might provide important information that can
help answer questions of fact. It may shed light on what contributors intend to do, how
they see each other, and which aspects of the creation are meaningful to them203. These
perspectives are particularly relevant to the determination of whether a contribution is
distinct, whether it is significant, and whether contributors are collaborating. Perhaps
the most important reason for a link between the legal conception of authorship and
social or cultural understandings of authorship is that this is important for copyright
law’s legitimacy 204 . I will consider how social norms can usefully inform legal
decisions in Chapters 7 and 8.
2.5 Conclusion
Although the contours of the concept of authorship in copyright law are uncertain, I
have argued that authorship has a stable core meaning, which requires a more than de
minimis contribution of creative choices evident in expression of a copyright work. A
number of scholars have suggested that copyright law’s rules of subsistence are poorly
adapted to collaborative activities, because of the influence of a literary trope: the
romantic author construct. An analysis of some of the case law does suggest a tendency
to concentrate authorship in the hands of one or two dominant contributors. I have
argued that this does not necessarily follow from the provisions CDPA; rather, it results
from the way in which those provisions have sometimes been applied. In fact, it is
203
Some see intention as a crucial aspect of the determination of copyright subsistence: Nimmer (n158);
Pila (n152).
204
Bently and Biron (n139); A Barron, ‘Copyright and the Claims of Art’ (2002)4 IPQ 368, 399: ‘… the
relation between art and copyright law matters; and in particular that copyright law is in some important
sense answerable to the claims of art, and amenable to being judged by reference to whether and how it
responds to those claims’. See also 8.1 below.
62
difficult to tell how the joint authorship test would apply to a work of collective
authorship because of the factual specificity of the test and the lack of analytical clarity
in the case law. The lack of case law applying the joint authorship test in situations of
collective authorship and the lack of analytical clarity in the existing case law results in
legal uncertainty that has the potential to chill collaborative creativity.
I presented three critiques of the application of the joint authorship test. Two of
these critiques suggest that the joint authorship test may not be able to determine the
authorship of a work in a way that reflects the reality of collective authorship. The
tendency to apply a restrictive approach to the application of the joint authorship test to
concentrate ownership in the hands of a few, based upon pragmatic reasoning, is
unnecessarily restrictive. This pragmatic/instrumental approach divorces the application
of the joint authorship test from both the realities of collaboration and the core meaning
of authorship in copyright law. Second, a judicial preoccupation with aesthetic
neutrality makes it difficult to take the context of creativity into account, leaving the
joint authorship test lacking adequate tools to differentiate between authors and mere
contributors where many contributors share responsibility for the creative or intellectual
content of a work.
This thesis proceeds on the basis that analysing the ways in which creative
communities regulate authorship themselves in cases of collective authorship can shed
some light on how the joint authorship test ought to be applied so that it can provide a
suitable mechanism for determining the authorship of such works. The best approach to
applying the joint authorship test should retain the strengths of the current law,
particularly, the flexibility of the concept of authorship. It should also uphold copyright
63
law’s purpose to incentivise and reward authors, ie those who make contributions of
creative choices to protected works. The best approach is one which enhances, or at
least does not detract from, the credibility of copyright law as a tool for regulating
authorship.
In the next four Chapters, the case studies reveal that there are good reasons for
UK courts to follow Nimmer’s lead and favour an inclusive approach to the application
of the joint authorship test in cases of collective authorship. They show that the
assumptions upon which the pragmatic/instrumental approach is premised do not
correspond with the creative realities of collective authorship. Then, I suggest an
approach to the application of the joint authorship test that is inclusive (because it
includes all those who would qualify as authors); and contextual (as it takes collective
authorship group social norms into account in answering relevant questions of fact).
64
Chapter 3: Wikipedia
Wikipedia, the popular online encyclopaedia, is one of the most visited websites on the
internet1. It is the site for a unique form of creativity that is interactive, dynamic and
highly iterative. Wikipedia promotes a model of creativity that intentionally blurs the
distinction between reader and writer. It is an example of the increasing use of
applications which facilitate the interactive and collaborative use of the internet, which
is often described as Web 2.0 2 . Web 2.0 technology has transformed processes of
cultural production by providing authors with new sources of inspiration and greatly
facilitating the creation, dissemination and publication of works thereby permitting
more people than ever before to participate in creative processes. A number of
commentators have observed that copyright law provides a significant impediment to
much of the innovation facilitated by Web 2.03. Some claim that copyright law is at a
moment of crisis4, or at least, in need of substantial reform as many of its concepts seem
to conflict with norms and community expectations which have developed in the
context of this new digital environment 5 . As the ‘poster child for the collaborative
1
At the time of writing, Alexa ranked www.wikipedia.org sixth globally in terms of internet traffic:
<http://www.alexa.com/siteinfo/wikipedia.org#>. This thesis primarily refers to the English language
version of Wikipedia, see <http://en.wikipedia.org/wiki/Main_Page>.
2
T O’Reilly, ‘What is Web 2.0?’ <http://www.oreillynet.com/pub/a/oreilly/tim/news/2005/09/30/what-
is-web-20.html>.
3
L Lessig, Remix: Making Art and Commerce Thrive in the Hybrid Economy (Penguin Press 2008); D
Tapscott and A Williams, Wikinomics: How Mass Collaboration Changes Everything (Atlantic Books
2008) 25-27; N Elkin-Koren, ‘Tailoring Copyright to Social Production’ [2011] 12 Theoretical Inquiries
in Law 309.
4
S Corbett, ‘Creative Commons Licences, the Copyright Regime and the Online Community: Is there a
Fatal Disconnect?’ (2011) 74(4) MLR 503.
5
R Merges, ‘The Concept of Property in the Digital Era’ (2008) 45(4) Houston LRev 1239 provides a
summary and critique of some of these arguments. See also Lessig (n3); DJ Halbert, Intellectual Property
in the Information Age: The Politics of Expanding Ownership Rights (Quorum 1999).
65
construction of knowledge and truth that the new, interactive Web facilitates’ 6 ,
Wikipedia makes an interesting case study of collective authorship and the challenges
that it might pose to copyright law’s joint authorship test.
This Chapter begins by considering the dynamics of creativity on Wikipedia:
briefly describing how it works; and suggesting some reasons why it works. I explain
that authorship on Wikipedia is understood as a donation of creative efforts to an
altruistic community project. In the second section, I consider the subsistence of
copyright on Wikipedia, asking whether: (i) Wikipedia as a whole, or individual
Wikipedia pages, would be recognised as original literary works in which copyright
subsists for the purposes of the Copyright, Designs and Patents Act 1988 (CDPA); and
(ii) whether Wikipedia contributors would be recognised as joint authors and thus
copyright owners under the CDPA. I conclude that it is at most unlikely, and at least
uncertain, that Wikipedia contributors would be considered to be joint authors of
individual pages to which they have contributed. The third section examines the role of
copyleft licences in sustaining collective creativity on Wikipedia. I argue that copyleft
licences are more effective as a means of reinforcing Wikipedia’s sharing norms, than
they are as legal instruments. The fourth section concludes with some insights for
copyright law that arise from this case study.
3.1 Authorship Dynamics: Promoting Sharing
This section considers the dynamics of authorship on Wikipedia. I argue that these
dynamics are influenced by three significant features of Wikipedia, namely that: (i) it is
6
W Richardson, Blogs, Wikis, Podcasts, and Other Powerful Web Tools for Classrooms (Corwin Press
2006) 61-62.
66
a perpetual work in progress; (ii) it requires collaboration between people from all over
the world who are likely to have never met; and (iii) it relies upon social norms that
emphasise sharing. On Wikipedia, authorship is a very inclusive notion that is defined
in terms of participation in a non-profit community project with public-minded goals.
Wikipedia calls itself ‘the free encyclopedia that anyone can edit’ 7 . There are
many differences between Wikipedia and traditional paper encyclopaedias like
Encyclopaedia Britannica. Encyclopaedia Britannica commissions experts to write
articles which are presented as an evaluative synthesis of views which approach the
‘truth’ on a subject. The subject of each article is chosen by an editorial board with
almost no input from readers and each edition is published (and sold) only when
complete. Wikipedia, by contrast, is created by an extensive, loose, global network of
collectively organised volunteers and is made freely available to anyone with an
internet connection8. It aims to present a selection of currently prevalent representations
of knowledge, rather than the truth on any subject 9 . Contributors have an almost
unrestrained freedom to determine its content and scope. Contributions can be made by
anyone, regardless of their level of expertise, at any time 10. Wikipedia is a constant
7
<http://www.wikipedia.org>.
8
This is made possible by wiki software which allows web pages to be created, edited and linked to one
another by anyone in real time using a common web browser. See: A Ebersbach, M Glaser and R Heigl,
Wiki: Web Collaboration (Springer 2006) 14-17.
9
A Bruns, Blogs, Wikipedia, Second Life, and Beyond: from Production to Produsage (Peter Lang 2008)
114. The debate on the reliability of Wikipedia is outside the scope of this article. See: A Keen, The Cult
of the Amateur: How Today's Internet is Killing our Culture (Doubleday/Currency 2007); L Sanger,
‘WHO SAYS WHAT WE KNOW: On the New Politics’ (The Edge) <http://www.edge.org/3rd_culture
/sanger07/sanger07_index.html>; cf. Bruns 121.
10
To adopt Eric Raymond’s often-quoted observation on the creation of Linux, Wikipedia is not built like
a cathedral ‘carefully crafted by individual wizards’, but more like ‘a great babbling bazaar of different
agendas and approaches…out of which a coherent and stable system could seemingly emerge only by a
succession of miracles’: ‘The Cathedral and the Bazaar’ <http://www.catb.org/~esr/writings/cathedral-
bazaar/cathedral-bazaar/>.
67
work in progress that grows by the accumulation of a great many small contributions.
Wikipedia works on the assumption that many different users will continually refine or
‘refactor’ each individual page until the page stabilises and comes to represent the voice
of the community11. In this way, it makes more sense to think of Wikipedia as a creative
process than a creative product. Indeed, Wikipedia is best conceived of as a perpetual
work in progress.
Although it may seem that such a process could only result in chaos, Wikipedia
has organised itself over time through the consensus-driven activities of its community
of regular contributors. In the ‘Village Pump’ section of Wikipedia users collaborate to
create policies, style guides, codes of conduct and other self-governing mechanisms to
regulate the community’s activities. Perusal of this behind-the-scenes space reveals that
Wikipedia sees itself as defined by five fundamental principles, or ‘pillars’:
 Wikipedia is an encyclopaedia;
 Wikipedia is written from a neutral point of view;
 Wikipedia is free content that anyone can use, edit, modify and distribute;
 Editors should treat each other with respect and civility; and
 Wikipedia does not have firm rules12.
Wikipedia has developed complex dispute resolution mechanisms for dealing with
editorial and other conflicts.
Wikipedia contributors might undertake a variety of different activities including:
contributing content, editing, correcting spelling and grammar, devising and running
11
Bruns (n9) 321.
12
<http://en.wikipedia.org/wiki/Wikipedia:Five_pillars>.
68
‘bots’ 13 , creating and enforcing community policy, helping to resolve disputes, etc.
Many of these are community sustaining activities rather than purely content creating
activities. There are a variety of levels of access to the functionalities of Wikipedia’s
wiki software which a user might attain over time if they participate as a good
Wikipedia community member. Each Wikipedia page has a history page associated
with it which allows users to see previous versions of the page so that quick reversions
might be made if it is affected by vandalism or spam and so that contributions might be
clearly identified by user name (for a registered user) or IP address (for an ‘anonymous’
user). Every time a contributor edits a Wikipedia page that page is reproduced and
published online. In order to avoid future claims of copyright infringement, when a
contributor adds anything to Wikipedia he or she must agree to allow that contribution
to be reproduced or modified by anyone for free (under certain conditions) pursuant to
copyleft licences14.
Wikipedia works because the barriers to contribution (money, technical expertise,
time, etc) are minimal. A sense of community is often seen as a key factor in ensuring
that there are more positive than negative contributions15. Wikipedia fosters this sense
of community through the activities of its welcoming committee16, the discussion pages
13
These are semi-automated tools or ‘robots’ which carry out mundane or repetitive tasks.
14
When a contributor makes changes to Wikipedia, he or she sees the following message displayed above
the ‘Save Page’ button: ‘By clicking the “Save Page” button, you agree to the Terms of Use, and you
irrevocably agree to release your contribution under the CC-BY-SA 3.0 License and the GFDL. You
agree that a hyperlink or URL is sufficient attribution under the Creative Commons license’ (including
hyperlinks to the licences and Terms of Use).
15
Bruns (n9) 110; S Rafaeli and Y Ariel, ‘Online Motivational Factors: Incentives for Participation and
Contribution in Wikipedia’ in A Barak (ed), Psychological Aspects of Cyberspace: Theory, Research,
Applications (CUP 2008). It is also important to have enough contributors for this to work.
16
<http://en.wikipedia.org/wiki/Wikipedia:Welcoming_committee/Welcome_to_Wikipedia>.
69
linked to each page, the use of consensus-driven policies 17 and the principle of
Wikiquette18 . People contribute to Wikipedia for a variety of non-financial reasons:
because they want to teach the world about something that they are enthusiastic about,
out of a sense of altruism, just for fun, or ‘because there’s nothing better on
television’19 . Some contributors might be motivated by a desire to attain status and
respect within this community. Contributors grant each other ‘barnstars’ for positive
contributions to Wikipedia and its community20. Although contributors to Wikipedia
might have different motivations, they are bound together by the act of contributing
towards the shared goal of the community: providing a free encyclopedia that anyone
can edit21.
Cass Sunstein suggests that the concept of authorship makes no sense in the
context of Wikipedia22. This is because of the unique way in which the wiki software it
relies upon allows people to add, delete, and change the content of any page in real
time. Implicit in Sunstein’s suggestion is that the notion of authorship implies control or
dominion over a work. This notion of authorship appears strongly linked to a particular
17
<http://en.wikipedia.org/wiki/Wikipedia:List_of_policies_and_guidelines>.
18
<http://en.wikipedia.org/wiki/Wikipedia:Etiquette>.
19
Bruns (n9) 112; Lessig (n3) 162.
20
<http://en.wikipedia.org/wiki/Wikipedia:Barnstars>. Barnstars are displayed on a contributor’s
userpage and are available to reward virtually any type of contribution. Some examples include: the
Barnstar of Good Humour (for lightening the mood); Anti-Vandalism Barnstar; the Barnstar of
Diplomacy; the Copyright Clean-up Barnstar. Giving barnstars, when they are earned, is particularly
encouraged by a group of contributors who have formed what is known as the ‘Kindness Campaign’.
21
There is a more ambitious ultimate vision for Wikipedia: as a tool for disseminating the sum of human
knowledge to everyone on the planet.
22
C Sunstein, Infotopia: How Many Minds Produce Knowledge (OUP 2006) 153.
70
view of copyright’s notion of the author as first owner of copyright23. Rather than being
a space of non-authorship, the Wikipedia community creates its own concept of
authorship. In doing so the community taps into some of the values of copyright’s
notion of authorship, whilst discarding other aspects of it. Wikipedians have re-invented
their own concept of authorship which is tailor-made to the project’s needs. In the
context of Wikipedia, authorship is an extremely inclusive and community-orientated
notion built on sharing rather than exclusion. It is a dynamic concept that is defined in
terms of participation in a shared project with altruistic goals. Participation is shaped by
adherence to a set of community-developed policies and norms that are crafted to
reflect the project’s aims24.
A crucial feature of Wikipedia’s sense of community is the notion that Wikipedia
is a sharing economy which operates in the absence of the usual notions of authorship
in the sense of the ownership and control of works 25 . Contributors make their
contributions for free and expect others to modify or even destroy them. Wikipedia is
thought of as owned by no one, to be shared by all 26 . The next section considers
whether Wikipedia is in fact owned by no one in copyright terms; and the following
section considers whether the operation of copyleft licences really does ensure that
content may be shared by all.
23
To some extent, this view appears to conflate authorship and ownership. In Chapter 2, I argued that
these concepts ought to remain conceptually distinct within copyright law.
24
So, for example, the no original research policy reflects the anti-expertise bias of the site; and the
neutral point of view policy reflects the desire to create a factual work of reference.
25
Sunstein (n22) 153.
26
<http://en.wikipedia.org/wiki/Wikipedia:Ownership_of_articles> accessed 30 September 2014, still
substantially reflecting a statement made in the original version of this page on 24 September 2003, that:
‘No one person “owns” the articles in the Wikipedia. They are the common property of all humankind.
The license known as [[GFDL]] guarantees this’.
71
3.2 Copyright Subsistence on Wikipedia
Wikipedia encourages users to download and print articles and even to create ‘books’ of
collections of Wikipedia pages27. Whilst there are many pages of discussion dedicated
to dealing with the problem of identifying and removing copyright infringing material,
the Wikipedia community spends considerably less time considering the extent to
which copyright subsists in Wikipedia itself and which contributors, if any, might own
that copyright 28. Presumably, this is because the act of contributing to Wikipedia is
often seen to imply the relinquishment of the usual incidents of authorship, such as the
right to control certain reproductions of the work, for the greater good of the Wikipedia
project (the creation of a free encyclopaedia that anyone can edit) in the spirit of its
ethic of sharing. Yet, this question merits some consideration.
3.2.1 Is Wikipedia (or Parts Thereof) an Original Literary Work?
Section 1(1)(a) of the CDPA provides that copyright subsists in an ‘original literary,
dramatic, musical or artistic work’. Copyright may subsist in: an individual Wikipedia
page as a literary work; or in Wikipedia as a whole as a literary work which is a
compilation or a database. Individual Wikipedia pages are clearly ‘literary’ in the
relevant sense29. They are also likely to be original as long as they are an ‘intellectual
creation’, the result of some labour, skill or judgment30. Some ‘stub’ articles may not be
27
Facilitated by links on the left-hand side of each Wikipedia page.
28
To the author’s knowledge no court has considered the subsistence of copyright in Wikipedia to date.
Wikipedia’s copyright policy can be found at: <http://en.wikipedia.org/wiki/Wikipedia:Copyrights>.
29
CDPA s3(1), s178.
30
Although this has usually been taken to be a low threshold test: University of London Press v
University Tutorial Press [1916] 2 Ch 601, 608-9 (approved in Ladbroke v William Hill [1964] 1 WLR
273, [1964] 1 All ER 465 (HL)). There remains considerable uncertainty as to the full impact of Case C‐
5/08 Infopaq Intl v Danske Dagblades Forening [2009] ECR I-6569 and subsequent CJEU decisions
72
sufficiently substantial or original for copyright to subsist in them31. It may be difficult
to determine the contours of a literary work on Wikipedia given the interactive and
transient nature of most of its content32. A Wikipedia page could be conceived of either
as a series of different works or as a series of draft versions of one final work. It seems
that a Wikipedia page might more closely resemble the latter case, although a truly final
version may never exist. This view is supported by the fact that Wikipedia presents
itself as a perpetual work in progress: almost every page invites the reader to interact
with it by editing it or commenting on its content on the associated ‘talk’ page.
It is unclear how copyright law would deal with this situation. Although copyright
generally does not subsist in mere copies of a work with minor alterations33, each draft
which imply a new harmonised European standard of originality of ‘the author’s own intellectual
creation’. Some UK jurisprudence suggests that Infopaq has not materially altered the originality
requirement in the UK: Newspaper Licensing Agency v Meltwater [2011] EWCA 890 [20]. See further:
M van Eechoud, ‘Along the Road to Uniformity – Diverse Readings of the Court of Justice’s Judgments
on Copyright Work’ (2012) 1 JIPITEC 60; E Rosati, ‘Originality in a Work, or a Work of Originality: the
Effects of the Infopaq Decision’ (2011) 33(12) EIPR 746.
31
Stub articles usually consist of a short title or a few sentences, which act as an invitation to the reader
to add content. They might be seen as ideas, lacking originality, or not ‘intended to afford either
information and instruction, or pleasure in the form of literary enjoyment’ (Hollinrake v Truswell [1894]
3 Ch 420 (CA) (Davey LJ, 428) approved in Exxon v Exxon [1982] Ch 119 (CA) (Stephensen LJ, 142-
143)). Although titles and short works have often been held either not to be a ‘work’ or to be
insufficiently original, there is authority to suggest that in some cases they might be extensive and
important enough to be considered original literary works: Francis Day and Hunter v 20th Century Fox
[1940] AC 112. In NLA v Meltwater (n30) [19]-[22] the Court of Appeal affirmed the first instance
decision that newspaper headlines might be original literary works. In Infopaq ([45]) the CJEU
considered that an 11 word phrase might be sufficiently original in cases where the author expresses his
or her creativity in an original manner through the choice, sequence and combination of words so as to
achieve a result which is an ‘intellectual creation’.
32
M Chon, ‘New Wine Bursting From Old Bottles: Collaborative Internet Art, Joint Works, and
Entrepreneurship’ (1996) 75 Oregon LJ 257, 263.
33
Interlego AG v Tyco Industries [1989] AC 217 (PC) holding that ‘a well-executed tracing’ is not
sufficiently original, although the decision may have been affected by the court’s perception that
Interlego were attempting to obtain a perpetual monopoly by making minor alterations to their Lego
bricks (255-6). Similarly, an enlarged facsimile copy of a drawing will not suffice: The Reject Shop v
Manners [1995] FSR 870.
73
version of a final work is likely to create a new copyright work 34. In LA Gear v Hi-Tec
Sports, for example, Lord Justice Nourse said that,
If, in the course of producing a finished drawing, the author produces one or
more preliminary versions, the finished product does not cease to be his
original work simply because he adapts it with minor variations, or even if
he simply copies it, from an earlier version. Each drawing having been
made by him, each is his original work35.
Bently and Sherman suggest that these comments only apply in cases where draft
versions have the same author36. Where the same author creates a number of drafts,
courts appear to adopt a more relaxed approach that takes into account the iterative
nature of the creative process, tending to consider each draft to be a potentially
independent copyright work37. In taking this approach, judges have sometimes thought
the point not to be of great consequence since, as a practical matter, even if a later draft
is not an independent work in which copyright subsists (because it lacks originality), it
might still be used as evidence of an earlier draft which is an original work38. Yet,
determining which of a series of drafts constitute an original work becomes important
34
Brighton v Jones [2004] EWHC 1157 [32] (adopting Justice Lightman’s comments in Robin Ray v
Classic FM [1998] FSR 622, 638 that where an earlier work is used as a basis for a later work, copyright
in the earlier work is not somehow ‘subsumed’ into copyright in the later work).
35
[1992] FSR 121 (CA). Approved in Taylor v Rive Droite Music [2004] EWHC 1605, [2004] All ER 88
[247-8] Mr Justice Lewison noted that copyright is capable of subsisting in unfinished versions, but that a
new composition will not be created by simply reworking a previous version of a work unless the
reworking is so extensive as to produce a recognisably new work or a new arrangement of the previous
work (136, Staughton LJ and Sir Michael Kerr agreeing).
36
Bently and Sherman 103.
37
For example, Sweeney v Macmillan Publishers [2002] RPC 35 (HC) [34] (Loyd J): ‘In the case of a
creative process as complex as that of Ulysses, it may be unclear, now, in what form the author’s text, as
a whole, stood at any given moment of time’.
38
Cala Homes (South) v Alfred McAlpine Homes East [1995] FSR 818 (Laddie J); Sweeney v Macmillan
(n37) (although the court may have been influenced by the fact that the previous versions were
unpublished).
74
where the author does not own copyright in each draft39 or where a party appears to be
attempting to create a perpetual monopoly40.
Where a series of drafts involves the contributions of different people, one must
turn to the complex case law on derivative works for assistance. It may be difficult to
distinguish between an infringing copy and a derivative work in which copyright
subsists (especially given that a work might be both41). In such cases judges might be
disinclined to find that a new derivative work has been created where there appears to
have been some misappropriation of the labour of previous authors42. Generally some
material change in character or quality must be brought about in the new work for it to
be considered sufficiently ‘original’. It seems that the skill, labour or judgment applied
in creating the derivative work must exhibit some ‘individuality’, which the Privy
Court’s decision in MacMillan v Cooper might lead one to believe is an exacting
standard, implying the addition of a new quality or character which the raw material did
not possess43. The change must also be of the right kind in relation to the type of work
39
Ultra Marketing v Universal Components [2004] EWHC 468 where the author did not own copyright
in the latest and most detailed draft of a drawing; Biotrading and Financing Oy v Biohit [1996] FSR 393,
395 (HC) (upheld on appeal [1998] FSR 109 (CA)) where it seemed likely there was no copyright in the
earliest drawing.
40
Interlego (n33) 255-6 (Lord Oliver of Aylmerton). Biotrading v Biohit 395 (upheld on appeal [1998]
FSR 109 (CA)) where it was suggested that earlier drawings cannot be relied upon as proof of the
originality of almost identical later drawings to avoid the possibility of creating a perpetual copyright
monopoly in an image (referring to Interlego); cf Rexnold v Ancon [1983] FSR 245, 260 (HC) where it
was considered arguable that later drawings might be sufficiently original if there were some differences
between these and earlier drawings. These cases might also be explained by the need to ensure certainty
in the subject matter of copyright: IPC Media v Highbury-Leisure Publishing [2004] EWHC 2985 [7]
referring to Green v Broadcasting Corp NZ [1989] RPC 700.
41
Redwood Music v Chappell (1982) RPC 109 (QB).
42
For example: Elanco Products v Mandops [1980] RPC 213 (CA) 228 (Goff LJ); 231 (Buckley LJ: ‘on
the material before the court there is a distinct possibility that it might be held that they have not done
sufficient independent work to eradicate the vice of copying’).
43
(1924) 40 TLR 186.
75
considered (ie, having a literary, artistic, dramatic or musical character)44. The amount
and type of change required appear to vary between categories of work and even
between sub-categories of work 45 . On the strict view established in MacMillan v
Cooper, it might seem that few newly updated versions of Wikipedia pages would
qualify as independent original literary works. Yet, this would be to disregard case law
in which judges have taken a less stringent view making allowances for a creative
process that involves building upon pre-existing ideas and works 46 or taking into
account the public interest in encouraging the creation of particular types of derivative
works such as high quality restorations47.
For some theorists, the heart of the matter is whether there has been an injection
of the author’s subjectivity in the derivative work48. Although this approach might have
more currency in the UK following the CJEU’s decision in Infopaq, it remains unclear
how it might be used to guide decision-making. In assessing the subsistence of
copyright in recreative works (such as restorations), Ong persuasively argues that the
best approach is to focus on the process by which the derivative work is created, rather
44
In Interlego (n33) 268 (Lord Oliver) as the case concerned an artistic work, the change needed to be
visually significant. In Sawkins v Hyperion Records [2005] EWCA 565, [2005] 1 WLR 3281 changes to
a musical work were sufficient to establish the originality of a derivative work because they impacted
upon its playability.
45
For example, a higher standard might be required of condensements than abridgements: Macmillan v
Cooper (n43).
46
IPC Media v Highbury-Leisure Publishing [2004] EWHC 2985, [2005] FSR 20 [9]; Baigent v Random
House Group [2006] EWHC 719, [2006] EMLR 16; on Appeal [2007] EWCA Civ 247, [2008] EMLR 7
Lloyd LJ [106].
47
Some have suggested that high quality replicas of works of art may be protected by copyright. See
Laddie et al [4.42]; Copinger et al [3-133] and J Ginsburg, ‘The Concept of Authorship in Comparative
Copyright Law’ (2002) 52 DePaul LRev 1063 cited with approval in Sawkins v Hyperion (n44) (Jacob LJ
[83]). Also see: Elisha Qimron v Hershel Shanks [1993] 7 EIPR D-157 (Israeli SC).
48
D Nimmer, ‘Copyright in the Dead Sea Scrolls: Authorship and Originality’ (2001) 38 Houston LRev
1, 159: ‘Copyright protection arises only for works that express an intent to produce something personal
or subjective’; Ginsburg (n47).
76
than any visual change that is evident in the resulting work49. Although one might make
a variety of policy-based arguments in favour of providing copyright protection for
recreative works that might not equally apply in the context of Wikipedia, Ong does not
confine his argument in this way. Rather, he makes the general point that the case law
on derivative works tends to be unduly concerned with the character of the end-product,
arguing that a determination of the existence of a new original work ought to extend
beyond this to take into account the process by which it has been made50. Copyright
law’s very static conception of a work makes it ill-adapted to Wikipedia’s dynamic,
process-orientated model of creativity.
Iterative processes of creativity have been accommodated within copyright’s
notion of the work on occasions where the precise identification of the moment or
moments in which an original copyright work comes into existence is of less practical
importance, for example, when copyright is owned by the same person(s). In such
cases, drafts might be conceptualised as steps along a process which will end with a
final work, although such a work might have been incomplete in the mind of the author
at the time the drafts were made51. If sufficient labour and skill to establish originality
can be shown at some point in that process, then determining the exact point in time at
which there is a work might only become important where a question is raised in
relation to the duration of copyright. In MacMillan Publishers v Thomas Reed
Publications, for example, the Court was willing to consider a sequence of chartlets in
49
B Ong, ‘Originality from Copying: Fitting Recreative Works into the Copyright Universe’ (2010) 2
IPQ 165.
50
Ong, ibid does not make an argument based on the policy reasons that favour allowing copyright
protection for recreative works, rather, he argues that any policy or doctrinal reasons against such
protection are not insurmountable. He focuses on giving weight to the intellectual skill, labour and
judgment involved in creating many such works.
51
Copinger et al [3-06].
77
successive editions of a nautical almanac all together52. The Court was influenced by
the fact that although the chartlets were the result of the efforts of a number of different
people, any copyright subsisting in them was owned by the plaintiff. Despite the
attractive possibility of considering a series of versions together as the relevant original
literary work which appears to arise from this case, as Wikipedia contributors retain any
copyright that they might have in any part of Wikipedia, there is no similar unified
ownership solution that would unite the rights of contributors in a series of versions of a
Wikipedia page the way the court was able to do in that case.
Where a case involves the adjudication of rights between different contributors to
different versions of a work over time, it is usually resolved by determining whether
there has been a ‘material change’ (although it is difficult to predict what sort of change
would suffice) so that the resulting work might be redefined as something new: an
object of authorship as opposed to a copy. In this way, determinations of the contours of
a work seem to slip into an analysis of the sufficiency of any relevant authorial input,
which as the next section considers, is a part of the joint authorship test which also
proves to be problematic in the context of Wikipedia. Whether the best approach to the
case law on derivative works is to focus on the injection of the author’s subjectivity or
the process of their creation, it seems clear that the current case law is a doctrinal maze,
which makes it extremely difficult to understand how a multi-authored dynamic work
like Wikipedia might fit within copyright’s concept of the work.
The collaboration of many different people by small increments over a significant
period of time on Wikipedia is difficult to fit within the confines of copyright law’s idea
52
[1993] FSR 455, also cited in Biotrading v Biohit (n39) 395 (Robert Walker J).
78
of the work. This is exacerbated by the fact that many potential authors are involved at
different points in time. Some argue that the CJEU’s reasoning in Infopaq might result
in a fracturing of the traditional notion of the copyright work, as the Court seemed to
suggest that parts of a work could be sufficiently original to be protected in their own
right (where by the choice, sequence and combination of those words the author
expresses his creativity in an original manner so as to achieve a result which is an
intellectual creation)53. Although fracturing the work in this way might, at first, appear
to better accommodate the creative process on Wikipedia by allowing each contributor
the copyright in his or her specific contribution, this approach seems very artificial and
is still likely to result in significant legal uncertainty as to the contours of a work 54. It
also appears to be at odds with judicial statements about the necessity of objectively
determining the scope of the work, as copyright should not be seen as a ‘legal
millefeuilles’ in which a claimant might try to claim various layers of protection by
asserting that small parts of a work are independent works, thus making it easier to
establish that an infringer has copied a substantial part of the work55. Considering parts
of Wikipedia pages to be independent works will also not do much to accommodate
pages which cannot be so easily divided as they have been heavily edited by large
numbers of people. Copyright’s fundamental conception of a work as having the
character of a fixed object which is (or might be) completed is poorly adapted to
Wikipedia’s creative process which results in works that do not have such neatly
defined contours.
53
Infopaq (n30) [47]; B Sherman, ‘What is a copyright work?’ (2011) 12(1) Theoretical Inquiries in Law
99, 115.
54
This is because, as argued below, the kind of labour which a contributor must contribute to a work to
be considered an author of the work remains uncertain.
55
IPC Media v Highbury-Leisure (n46) (Laddie J, [23]) approved in Coffey v Warner/Chappell Music
[2005] EWHC 449 (Ch) (Blackburne J, [10]).
79
Further complications arise when considering Wikipedia as a whole. For the
purposes of the CDPA, Wikipedia is probably a database rather than a compilation56. A
database is defined as a ‘collection of independent works, data or other materials which
– (a) are arranged in a systematic or methodical way, and (b) are individually accessible
by electronic or other means’57. These criteria are not explained in the CDPA or the
Database Directive that lead to their implementation. Case law and commentary provide
some guidance and suggest that the term ‘database’ ought to be construed broadly58.
Wikipedia is a ‘collection of independent works’ because Wikipedia pages are
‘separable from one another without their informative, literary, artistic, musical or other
value being affected’59. Wikipedia pages are also ‘individually accessible’ via a search
function 60 . Derclaye suggests that the criterion of arrangement in a ‘systematic or
methodical way’ is probably a fairly low standard, which should be read broadly so as
to exclude only haphazard collections61. The existence of some structure to the data so
that it might be organised after the application of a search programme, for example,
56
The wording of s3(1)(a) suggests that where a work is both a database and a compilation it should only
be treated as a database. The sui generis database right is beyond the scope of this article and has been
considered elsewhere, see: J Lipton, ‘Wikipedia and the European Union Database Directive’ (2010) 26
Santa Clara Computer & Technology LJ 631; J Phillips, ‘Authorship, Ownership and Wikiship:
Copyright in the Twenty-First Century’ (2008) 3 (12) JIPLP 788.
57
CDPA s3A(1).
58
E Derclaye, ‘The Court of Justice Interprets the Database Sui Generis Right for the first time’ (2005)
30(3) Eur LR 420; E Derclaye, ‘Database Sui Generis Right: What is a Substantial Investment? A
Tentative Definition’ (2005) 36(1) Intl Rev of Intellectual Property and Competition L 2; E Derclaye,
‘Databases Sui Generis Right: Should We Adopt the Spin Off Theory?’ (2004) 26(9) EIPR 402; E
Derclaye, ‘Do Sections 3 and 3A of the CDPA violate the Database Directive? A Closer Look at the
Definition of a Database in the UK and its Compatibility with European Law’ (2002) 24(10) EIPR 466; E
Derclaye, ‘What is a Database? A Critical Analysis of the Definition of a Database in the European
Database Directive and Suggestions for an International Definition’ (2002) 5(6) JWIP 981. See Case
444/02 Fixtures Marketing v Organismos Prognostikon Agonon Podosfairou [2004] ECR-I10549 [19]-
[21], C-545-07 Apis-Hirstovich v Ladorka [2009] ECR I-1627 [69].
59
Fixtures Marketing (n58) [29].
60
This threshold requirement is probably not high: Derclaye, ‘Do sections 3 and 3A of the CDPA violate
the Database Directive?’ (n58) 469; Laddie et al [30.23].
61
(n60) 468.
80
would be sufficient62. Wikipedia satisfies this requirement because its software indexes
and stores pages in a database. As Wikipedia is constantly being updated, it might be
difficult to determine when, and if, changes made to it over time result in a new
database. In British Horseracing Board v William Hill a database of horse racing details
updated by around 800,000 changes a year was held to be a single database in a
constant state of refinement63. Wikipedia is likely to be considered similarly.
Although Wikipedia is almost certainly a database, it is probably not sufficiently
original as it currently operates. To meet this requirement the database must constitute
the author’s own intellectual creation by reason of the selection and arrangement (not
the creation) of its contents 64 . Reviewing, marking and editing particular Wikipedia
pages to ensure compliance with Wikipedia’s editorial policies would probably be seen
as a contribution to content. If Wikipedia became more controlled and its administrators
took a more active role in the organisation of pages, it might meet this standard and in
that case (assuming the requirements for joint authorship are met) these administrators
62
Fixtures Marketing (n58) [30]: ‘while it is not necessary for the systematic or methodical arrangement
to be physically apparent…that condition implies that the collection should be contained in a fixed base,
of some sort, and include technical means…or other means, such as an index, a table of contents, or a
particular plan or method of classification, to allow the retrieval of any independent material contained
within it’ (in the Opinion of Advocate General Stix-Hackl this is to ‘ensure that only planned collections
of data are covered’[40]).
63
[2001] EWCA 1268, [2001] RPC 31 [71]-[73]. This approach was supported by A-G Stix-Hackl’s
opinion in C203/02 British Horseracing Board v William Hill [2004] ECR I-10415 in which he described
constantly updated databases as dynamic databases providing a rolling right [148]-[151] (the CJEU did
not specifically consider this point). See also Beechwood House Publishing t/a Binley’s v Guardian
Products [2010] EWPCC 12 [63]-[64].
64
It had been assumed that the requirement in s3A(2) that the database constitute ‘the author’s own
intellectual creation’ is a higher standard of originality than for other literary works: see Bently and
Sherman 108. In C-604/10 Football Dataco v Yahoo! UK [2012] ECDR 10 [37] the CJEU seems to link
this standard to its newly harmonised originality standard post-Infopaq citing this case and C-393/09
Bezpečnostní softwarová asociace - Svaz softwarové ochrany v Ministerstvo Kultury [2011] ECDR 3,
[2011] FSR 18 [45]; C-403/08 & C-429/08 Football Association Premier League v QC Leisure [2012] 1
CMLR 29, [2012] ECDR 8 [97]; and C-145/10 Painer v Standard Verlags [2012] ECDR 6 [87].
81
would own any copyright subsisting in Wikipedia as a database 65 . As it stands,
however, relatively little effort is expended by individual contributors and
administrators in this respect. Indeed, the value of Wikipedia and much of the reason
for its success lies in its lack of explicit structure and the ability of contributors to easily
link pages on the fly. This is despite the fact that, as Wikipedia has grown, efforts have
increased to organise and structure it into categories and to identify which parts need to
be improved or are yet to be written.
In summary, although it seems unlikely that copyright subsists in Wikipedia as a
whole as a database at present, it is likely that copyright would subsist in individual
Wikipedia pages as literary works. It is impossible to predict, however, how a court
might construe the contours of such a work. This is likely to result in uncertainty in
relation to which contributors might be considered to be authors of that literary work.
3.2.2 Are Wikipedia Contributors Copyright Authors?
Assuming that a given Wikipedia page is an original literary work, in most
circumstances it would have multiple potential authors. For a Wikipedia contributor to
be a joint author under the CDPA he or she must have:
(i) made a significant contribution of the right kind;
(ii) in pursuance of some common design or collaboration; and
(iii) his or her contribution must not be distinct66.
65
Administrators are contributing Wikipedia members of long-standing who ought not to be confused
with the WikiMedia Foundation that owns and runs the website (but does not control its content). Lipton
(n56) 647 argues that the WikiMedia Foundation is also unlikely to own any database rights subsisting in
Wikipedia following British Horseracing Board (n63).
66
CDPA s10(1); Godfrey v Lees [1995] EMLR 307, 325-8; Ray v Classic (n34) 636; Hadley v Kemp
[1999] EMLR 589.
82
Although most contributions to Wikipedia will satisfy the last criterion, it is by no
means clear that the others would be met. The case law on joint authorship only
provides limited guidance and is often specific to a particular set of facts.
In order to satisfy the first criterion, the contributor must have contributed some
of the skill, labour or judgment that makes the work original in the sense required by
the CDPA. The contribution must be of the right kind, that is, in relation to the
expression of the work and not just the ideas67. A joint author of a literary work must
share some responsibility for what appears on the page, which has been explained as
‘something which approximates to penmanship’68. This can often be a fine point69. In
Fylde Microsystems v Key Radio Systems, a technician who tested and debugged
software was not a joint author of that software, despite the fact that the contributions
made were extensive, technically sophisticated and involved the expenditure of
considerable time and effort. This was because the technician’s work was more akin to
proof-reading than authorship 70 . Following this reasoning, many valuable Wikipedia
contributors would not qualify as joint authors (for instance, minor editing, involvement
in dispute resolution procedures, developing bots, creating and enforcing community
policy). If a restrictive approach, such as the instrumental/pragmatic approach, is taken
to the joint authorship test it is likely that most valuable contributors to Wikipedia
would not be considered joint authors.
67
Donoghue v Allied Newspapers [1938] 1 Ch 106 (Ch).
68
Cala Homes v Alfred McAlpine Homes [1995] EWHC 7, [1995] FSR 818; Ray v Classic FM (n34) 636
(Lightman J).
69
Bently and Sherman 122. Compare Hadley v Kemp (n66) and Stuart v Barrett [1994] EMLR 449.
70
[1998] FSR 449 (Ch) Mr Justice Laddie stated, rather ambiguously, that section 10 ‘does not turn
someone who is not an author into an author’ suggesting, perhaps, that authorship requires more than the
investment of skill, time and effort relied upon by the defendant in that case.
83
In addition to contributing the right kind of skill and labour, the contribution must
also be significant. Whether or not a contribution is significant is a question of fact that
depends upon the circumstances of the case71. In Fisher v Brooker Justice Blackburne
considered this to mean ‘more than merely trivial’72. If this is an accurate formulation
of the standard, although fairly low, it may still exclude the activities of a large number
of contributors who only make very small contributions to individual pages. Bently has
argued that the requirement for a significant contribution has tended to invite judges to
make value judgments about the quality of particular contributions 73. If significance is a
qualitative requirement, it is difficult to predict how it might apply to contributions to
Wikipedia, especially given the lack of guidance from the case law74.
Section 10(1) requires that a work of joint authorship be ‘produced through the
collaboration of two or more authors’, which has been described in the case law as
‘joint labouring in furtherance of a common design’ 75 . The Court of Appeal in
Beckingham v Hodgens held that the requirement of a ‘common design’ did not mean
the common intention to be a joint author 76. In Chapter 2 I argued that the Court’s
decision should be construed to mean that there is no need for a subjective intention in a
71
Stuart v Barrett (n69).
72
[2006] EWHC 3239, [2007] FSR 255 [46] in that case Mr Justice Blackburne considered the relevant
contribution to be ‘on any view substantial’ [98]. Although the case was partially reversed on appeal
([2008] EWCA Civ 287) the Court of Appeal considering it to be an ‘extremely unusual case’ [34], it was
again reversed by the House of Lords ([2009] UKHL 41). Mr Justice Blackburne’s ruling on joint
authorship was not challenged.
73
Communication & Society 179, 190.
74
See 2.2.3 above.
75
Levy v Rutley (1871) LR 6 CP 523 (Court of Common Pleas) 529 (Keating J).
76
[2003] FSR 238 (Ch) [52]. Cf L Zemer, The Idea of Authorship in Copyright (Ashgate 2007) 217
arguing that despite this decision UK courts tend to silently embrace a requirement for the intention to co-
author to some extent. See also Ginsburg (n47).
84
narrow sense: to be a joint author for the purposes of copyright law. On this view, the
requirement of a common design would still require some intention to work together. If
this approach were adopted, one could argue that most Wikipedia contributors satisfy
this requirement. The existence of discussion pages, consensus-driven policy measures
and the Wikipedia community culture tend to support the existence of a common design
amongst contributors. The longer a contributor has participated and the more involved
they have been as a member of the Wikipedia community, the more likely this
argument is to be persuasive.
Indeed, Chon argues that in determining the authorship of digital creations, judges
ought to favour an objective determination with reference to the existing expectations of
the relevant internet subculture 77 . She suggests inferring the intention to be a joint
author from the fact that an ‘individual author voluntarily created his or her work in
response to a call for participation that emphasised interactivity’78. Wikipedia’s sharing
culture might conversely be relied upon as evidence of the absence of an intention to
collaborate in a narrower sense (that is, in the proprietary sense provided for by the
CDPA) because contributors might be seen to evince an intention to donate or
relinquish copyright ownership of their contributions by agreeing to copyleft licences.
However, this is not a persuasive argument, as copyleft licences involve an implicit
recognition that the contributor owns any copyright subsisting in his or her contribution,
although he or she agrees not to enforce some of the incidents of that ownership
provided for by copyright law. In addition, such an argument relies upon the narrow
interpretation of intention that I argue was rejected in Beckingham v Hodgens.
77
Chon (n32) 271.
78
ibid 272.
85
The requirement for collaboration or a common design provides a way of
distinguishing between cases involving successive derivative works (with different
authors) and works which involve different contributions to a common project that
might be considered to be works of joint authorship. Considering Wikipedia in isolation
from its social norms, it might seem that each contributor potentially creates a new
derivative work when making a small change to a Wikipedia page and that it would be
in rare cases that such a contributor would bring about enough of a material change for
that version to be sufficiently original. I have argued that it makes more sense to think
of Wikipedia as a series of drafts of a final work that may never come into existence.
This contention is supported by the existence of social norms (as embodied by
Wikipedia’s consensually developed policies) which establish collaboration amongst
the majority of Wikipedia contributors. Although contributors will vary in the extent to
which they are familiar with, and subscribe to, Wikipedia’s social norms, a
contributor’s response to the general invitation to the public to participate in the free
Wikipedia encyclopaedia project ought to be enough to raise a presumption that the
contributor is collaborating in the relevant sense. One might then consider a Wikipedia
page at any particular point in time and with the aid of Wikipedia’s history pages,
identify which users have contributed to the written expression on the page as it existed
at that moment taking into consideration its previous versions, as the court did in
MacMillan v Thomas Reed. Yet, it would remain a difficult task to determine which, if
any, have made enough of the right kind of contribution to count as joint authors and
thus owners of any copyright that might subsist in that page. In Chapter 8 I present an
inclusive, contextual approach to the application of the joint authorship test, which
provides a framework for undertaking this task. Specifically, I suggest that the common
design of the contributors ought to take on a more significant role in the application of
86
the joint authorship test. In particular, the significance of a contributor’s contribution
ought to be assessed in light of the common design from which the joint work arises.
In summary, on the basis of current case law it seems at most unlikely, and at
least uncertain, that some Wikipedia contributors would be considered to be joint
authors of individual pages to which they have contributed. If this were established, the
joint authors would hold their copyright interests as tenants in common so that although
one author might sue in relation to an infringement, all joint authors would be required
to grant any licence to reproduce a part of it79. This would present obvious logistical
issues for Wikipedia’s creative model in the absence of copyleft licensing.
3.3 Copyleft Licences and the Ambivalent Role of Copyright Law
Copyleft licensing is commonplace on the internet as a response to the breadth of scope
and perceived uncertainty of copyright restrictions on the reproduction of works 80 .
Licences are used as a means for authors to opt-out of the automatic operation of
copyright law protection so as to authorize the copying of their works in advance in
certain circumstances. The origin of copyleft licensing lies with the Free/Libre/Open
Source Software movement, which adopted licences to enhance the development of a
free collaborative exchange by software developers. Creative Commons then expanded
upon this idea by producing a variety of tailor-made licences which are designed to
cover many different works in which copyright might subsist and which are made
79
For a comparative study of the complexities of joint ownership: M Perry and T Margoni, ‘Ownership
in Complex Authorship: A Comparative Study of Joint Works’ (2012) 34(1) EIPR 22.
80
LP Loren, ‘Building a Reliable Semicommons of Creative Works: Enforcement of Creative Commons
Licenses and Limited Abandonment of Copyright’ (2007) 14 George Mason LRev 271, 273.
87
available online for anyone to use81. For many commentators copyleft licences have a
political significance. Gonzalez, for example, notes that the Free Software Foundation’s
General Public Licence reads as a mixture of a legal contract and an ideological
manifesto82 and Lessig presents Creative Commons as the only ethical way possible for
exercising copyright83.
On Wikipedia, copyleft licences police the interaction between copyright law’s
notion of authorship (implying ownership and control) and Wikipedia’s own notion of
collective creativity (the donation of creative efforts to a community project). Most
content on Wikipedia is co-licensed under the Creative Commons Attribution-
Sharealike 3.0 Unported Licence (‘CC-BY-SA’)84 and the GNU Free Documentation
Licence (‘GFDL’)85. These licences allow anyone to copy or modify text on Wikipedia
as long as they:
 credit the authors (by including a hyperlink/URL to the original Wikipedia page
or by providing a list of all authors which may be filtered to exclude very small
or irrelevant contributions);
81
Proposal sketched in L Lessig, The Future of Ideas: The Fate of the Commons in a Connected World
(Vintage Books 2002); see <http://www.creativecommons.org>.
82
AG Gonzalez, ‘Viral Contracts or Unenforceable Documents? Contractual Validity of Copyleft
Licences’ (2004) 26(8) EIPR 331, 333. The preamble to the GPL has been described as reading like a
‘declaration of independence’: M O’Sullivan, ‘The Pluralistic, Evolutionary, Quasi Legal Role of the
GNU General Public Licence in Free/Libre/Open Source Software (FLOSS)’ (2004) 26(8) EIPR 340, 341;
or as the Free Software Movement’s ‘constitution’: R Gomulkiewicz, ‘General Public License 3.0:
Hacking the Free Software Movement’s Constitution’ (2006) 42(4) Houston LRev 1015.
83
L Lessig, Free Culture: How Big Media Uses Technology and the Law to Lock Down Culture and
Control Creativity (Penguin Press, 2004). S Dusollier, ‘The Master’s Tools v the Master’s House:
Creative Commons v Copyright’ (2005) 29 CJLA 271, 287: Creative Commons tends to see the
commons as a ‘principle, a religion or morality’.
84
<http://en.wikipedia.org/wiki/Wikipedia:Text_of_Creative_Commons_Attribution-ShareAlike_3.0_Un
ported _License>.
85
<http://en.wikipedia.org/wiki/Wikipedia:Text_of_the_GNU_Free_Documentation_License.
88
 re-licence any modifications or additions under the CC-BY-SA or later
versions86;
 indicate if modifications or additions have been made to the original; and
 include a licensing notice stating that the work is released under CC-BY-SA
and either a hyperlink/URL to the text of the license, or a copy of the license87.
There is no requirement that reuse be for non-commercial purposes88.
Although copyleft licences are becoming extremely prevalent, there have been
surprisingly few cases considering their validity89. In Jacobsen v Katzer the US Court
of Appeals (Fed Ct) held that copyleft licences are enforceable as contracts90. Although
convincing arguments might be still made either way91, it seems likely that UK courts
would treat copyleft licences as bare licences, not contracts, due to the apparent lack of
consideration 92 . The court in Jacobsen v Katzer located the reciprocity required for
consideration in the advantages associated with the increased distribution facilitated by
copyleft licenses and the conditions imposed on re-use (ie, attribution). These
86
The validity of this provision is unclear: Gonzalez (n82).
87
<http://en.wikipedia.org/wiki/Wikipedia:Copyrights>.
88
It is likely, however, that the successful commercial exploration of reproductions would detrimentally
affect Wikipedia’s ethic of sharing.
89
For a useful outline of most of the cases to date, see Corbett (n4).
90
535 F 3d 1373 (2008); See: R Gomulkiewicz, ‘Conditions and Covenants in License Contracts: Tales
from a Test of the Artistic License’ (2009) 17(3) Texas Intellectual Property LJ 335; J Wacha, ‘Taking
the Case: Is the GPL Enforceable?’ (2005) 21 Santa Clara Computer & High-Technology LJ 451, 481; W
Reynolds and J Moringello, ‘Survey of the Law of Cyberspace: Electronic Contracting Cases 2008-2009’
(2009) 65 The Business Lawyer available at <http://ssrn.com/abstract=1437162>.
91
Loren (n80) 312-3; Gomulkiewicz (n90) 346 criticises the concept of a bare licence as an artefact; M
Henley, ‘Jacobsen v Katzer and Kamind Associates – an English legal perspective’ (2009) 1(1) IFOSS L
Rev 41.
92
P Johnson, ‘“Dedicating” Copyright to the Public Domain’ (2008) 71(4) MLR 587; E Moglen, ‘Free
Software Matters: Enforcing the GPL, I’ <http://emoglen.law.columbia.edu/publications/lu-12.pdf>.
89
arguments, however, may not be persuasive to a UK court given that the conditions of
re-use require little more than the moral rights provisions in the CDPA and given that
increased (free) distribution is not always advantageous per se93. On this view, copyleft
licences operate as a defence to a claim of copyright infringement rather than as a
foundation for an action for breach of contract, so that if a licence condition is not met
the only legal effect would be that the permission granted by the licence would cease to
be effective.
The symbolic association of copyleft licences with the ideals of the free software
and Creative Commons communities gives them a key role in constituting Wikipedia’s
culture of sharing, which may be more significant than their role as legal instruments 94.
In the context of Wikipedia, copyleft licences appear to operate as a type of customary
law which relies primarily on the effect of social pressures for its moral force 95 .
Copyleft licences rely upon property rights granted by copyright law, but seek to
exercise them differently. Despite their intention to create a copyright-free zone for
creativity96, copyleft licences necessarily rely upon the continuing validity of copyright
law as a source of legal standards97. In this way, they might be seen as playing the game
of copyright, not attempting to abolish it98. Reliance on property language and concepts
93
The former argument might not be as strong in case of Wikipedia as would be for copyleft licences
more generally, given the exception to the attribution and integrity rights in relation to literary works
made for inclusion in an encyclopaedia (CDPA s79(6)(b), s 81(4)(b)).
94
O’Sullivan (n84) 345
95
K Bowrey, Law and Internet Cultures (CUP 2005) 95 argues it makes more sense to think of them as
evidencing a relationship of trust, rather than a commercial relationship.
96
M Fox, T Ciro and N Duncan, ‘Creative Commons: an Alternative, Web-based Copyright System’
(2005) 16(6) Ent LR 111, 116.
97
J Boyle, ‘Second Enclosure Movement and the Construction of the Public Domain’ (2003) 66 Law and
Contemporary Problems 33, 64-6.
98
Dusollier (n83) 278.
90
in copyleft licences may undermine the goal of facilitating a sharing or ‘free culture’99
and the share-alike provision seems to make contractual rights more like property
rights, as they resemble rights against the world that run with the work 100 . The
community spirit which motivates many regular contributors to Wikipedia is built upon
a shared understanding (embodied in the copyleft licences) that contributions are
donated. This assumption subtly reinforces copyright law’s proprietary notion of
authorship, because the grant of permission to reuse or modify a contribution is
premised on the assumption that the contribution is valuable and might be ‘owned’ in a
copyright sense.
Copyleft licences might also tend to expand the perception of what can be
protected by copyright law. The use of these licences tends raises a presumption in the
minds of those encountering the works to which they are attached, that copyright
subsists in those works and that it is necessary to observe the conditions of the licence
to reproduce them101. As this Chapter has argued, in the case of Wikipedia, the extent to
which copyright subsists might be uncertain. In addition, users might not be aware that
uses of the work covered by fair dealing defences need not comply with the licence
conditions (as the licence is likely to operate at most as a defence to a claim of
copyright infringement)102. The copyleft licence used by Wikipedia attempts to expand
the rights of contributors in some respects, for example, by providing for strong
attribution rights and requiring creators of derivative works to re-license their works
99
For example references to a ‘Commons Deed’. On the use of property metaphors in copyright: M Rose,
‘Copyright and its Metaphors’ (2002) 50 UCLA LRev 1.
100
101
Corbett (n4) 527.
102
ibid. This point is highlighted in the most recent version of the CC-BY-SA used on Wikipedia.
91
under the same licence103. Yet many, if not most, contributors to Wikipedia who must
be attributed as authors under the licence will not be ‘authors’ who would have this
right according to the CDPA. The attribution of contributors provides a locus for credit
which might be gained within the Wikipedia community; it demonstrates one’s
belonging to a community that presents itself as altruistic to outsiders; and attribution
might have an intrinsic value in itself to the contributor. In this way copyleft licences
provide an alternative notion of authorship which is better adapted to the Wikipedia
community: modifying the incidents and rewards of authorship and broadening the
category of contributors who might be the beneficiaries of this alternative conception of
authorship.
Dusollier has suggested that it may be possible for the widespread use of copyleft
licences, over time, to re-signify the meaning of authorship in the digital context104.
Caution is warranted in relation to the increasingly widespread use of instruments
which institutionalise a process involving individual creators giving up benefits to
which they may be entitled under copyright law with licences which have been
criticised as overbroad and inflexible105. Whilst some argue that copyleft licences are
empowering, giving authors back their long-lost autonomy106, it is difficult to see how
an author is empowered by giving away copyright entitlements especially when this is
done via a standard form which has been devised by a third party. In the case of
103
It seems unlikely that this provision is legally valid, but even if it were, it would be unenforceable
against third parties. Gonzalez (n82).
104
105
Copyleft licences are difficult, if not impossible, to amend. R Gomulkiewicz, ‘Open Source License
Proliferation: Helpful Diversity or Hopeless Confusion?’ (2009) 30 Washington UJ of L and Policy 261.
106
N Elkin-Koren, ‘What Contracts Cannot Do: The Limits of Private Ordering in Facilitating a Creative
Commons’ (2005) 74 Fordham LRev 375, 386.
92
Wikipedia, copyleft licences do not offer much authorial control as a contributor must
agree to the version of the licence the community has adopted. Dusollier argues that
copyleft licensing is more concerned with user rights and free access to culture than it is
with creators’ rights. She criticises the social forces at play in copyleft licensing,
arguing that they aim to make free access to works the politically correct way for a
creator to exercise his or her rights, which she compares to other social constructions
which can operate to undervalue labour 107 . She argues that this narrative of a gift
society (the sharing culture) could render invisible the work done by authors108.
Furthermore, copyleft licences are not a complete substitute for copyright law
because they do not have the benefits of being produced through a legislative process
which involves the balancing of interests and public policy considerations109. Although
there is a tendency to present copyleft licences as the ‘people’s response’ to the
perception that copyright law is out of step with the norms and expectations which have
developed in the new digital environment, these licences are not produced by any sort
of participatory process and do not reflect customs that have emerged over time –
rather, they have been developed to reflect the will of strong norm entrepreneurs110. In
addition, the standardised form of these licences might reinforce a licensing culture
which some corporations have used to enforce strong proprietary control and restrict
107
108
ibid 293. She argues that authors must give free and informed consent for others to use their works,
which includes freedom from ‘social construction’ 289.
109
Although this point might be made about all contracts, it seems particularly relevant here given the
ideological aims of the copyleft movement.
110
F Marrella and S Yoo, ‘Is Open Source Software the New Lex Mercatoria?’(2006-2007) 47 Virginia J
of International L 807.
93
user rights such as has been the case with the use of End-User Licensing Agreements in
the software industry.
Despite the potential for the widespread use of copyleft licences to slowly devalue
the concept of authorship that copyright law aims to protect, the benefit of their use in
the context of Wikipedia probably outweighs any potential detriment. This is because
copyleft licences allow for a space to be carved out from the application of copyright
law in order to facilitate the survival of, and even nourish, the unique means of
collective creation that thrives on Wikipedia.
3.4 Insights for Copyright Law
Corbett argues that copyright law is poorly suited to the new types of creativity enabled
by modern digital technology because it is not aligned with community norms and
expectations 111 . Wikipedia provides a useful lens through which to examine this
contention. Wikipedia does not fit neatly within copyright law’s principles of
subsistence. Indeed, it appears to challenge some of the notions at the heart of copyright
law’s conception of authorship. For instance, must a ‘work’ be an object capable of
being finished? Which sort, and how much, labour, skill or judgment counts as
‘authorship’? How do the rules on derivative works apply in situations where the reader
is invited to interact with a work and become a creator? What does it mean for joint
authors to ‘collaborate’ or to ‘pursue some common design’? In practice, copyleft
licences alleviate uncertainty in the application of copyright law by limiting potential
claims of copyright infringement which might otherwise impede Wikipedia’s day-to-
111
Corbett (n4).
94
day operation. In fact, Wikipedia’s culture of sharing both relies upon, and is partly
constituted by, copyleft licences. Nevertheless, the legal effect and the general
desirability of copyleft licences still remains unclear.
Authorship and ownership are not intertwined on Wikipedia in the same way that
they are in copyright law. Wikipedia attempts to separate creativity from ownership in
the name of a greater good: the Wikipedia project and its ethic of sharing. The
phenomenal success of Wikipedia seems to undermine instrumental justifications for
copyright law as necessary to incentivise creativity 112 . Contributors participate in
Wikipedia willingly and appear to derive some personal, albeit non-monetary, benefit
from their participation. In so doing they are not alone – many other creators
(magicians, chefs, comedians, etc) do not rely upon copyright law’s incentives 113 .
Although Dusollier argues that the prevalence of sharing economies facilitated by
copyleft licences might undervalue the work of authors, it would be stretching her
argument too far to suggest that copyright law ought to regulate all forms of creativity
which might be conceived of as ‘authorship’. Instead, it provides an argument for
ensuring that copyright remains an important source of good authorship standards, that
establish a presumption that authors are entitled to the pecuniary (as well as reputational
benefits) from their creative activities.
112
Elkin-Koren (n3) makes this point more generally in relation to what she terms social production. See
also: D Zimmerman, ‘Copyright as Incentives: Did We Just Imagine That?’ (2011) 12(1) Theoretical
Inquiries in Law 29.
113
See the literature on intellectual property’s negative spaces referred to in Chapter 2, p37.
95
Some suggest that creativity on Wikipedia is not ‘authorship’ and, therefore, that
it might be the sort of creativity that copyright ought not to concern itself with 114 .
Indeed, Wikipedia’s contributors seem to have no desire for copyright law to directly
apply to their activities, agreeing to copyleft licences which limit the restrictions
copyright law might impose upon Wikipedia’s creative model. Yet, from every angle
that you look at it, Wikipedia is exactly the sort of creation that the CDPA seeks to
regulate. It is a ‘literary work’ and is it clearly a product of authorship (it is created by
many people who have made creative or intellectual choices that are evident in the
expression). The problem is not that Wikipedia is not a work of authorship, but rather
that copyright law’s concepts are not currently applied with enough attention to the
context in which creativity occurs. Three aspects of Wikipedia make it ill-served by a
narrow, acontextual approach to the application of the joint authorship test. These are
the facts that: a number of contributors might make very small contributions; those
contributors are likely to have never met; and Wikipedia is a perpetual work in
progress.
The shadow of copyright law still casts itself over the norms which operate to
create and regulate the alternative system of self-regulated creativity on Wikipedia115.
The fact that copyleft licences rely upon copyright law for their effectiveness and as a
source of standards should not be under-estimated. The Wikipedia community taps into
some of the values of copyright’s notion of authorship, whilst discarding others.
Crucially, contributors retain any copyright in their contributions, and in choosing to
forgo many of the benefits that that copyright interest might entail in the name of the
114
Sunstein (n22) 153.
115
R Coombe, The Cultural Life of Intellectual Properties: Authorship, Appropriation and the Law
(Duke UP 1998).
96
Wikipedia project, they help constitute the sharing norms which sustain this loose-knit
network of contributors. Although some contribute to Wikipedia just for fun or because
they find it intellectually stimulating, many contribute out of a sense of altruism,
because they value interaction with the Wikipedia community or because they wish to
build a reputation within that community. In these situations contributors may value the
fact that however great or small, their contributions are acknowledged in the history
pages and whenever any part of the page is reproduced pursuant to the copyleft licence.
If new forms of creativity, like Wikipedia, ought to be considered ‘authorship’, as
I think they should be, the question arises as to what sorts of accommodations ought to
be made for this fluid, interactive and participatory model of creativity. This thesis
argues for a contextual approach to the application of the joint authorship test that
places more emphasis on the collaboration limb of the test. In the context of Wikipedia,
this means appreciating that there is collaboration present amongst contributors who are
united by their response to a call for participation in the Wikipedia project. This is more
likely to be the case, the more integrated contributors are in the Wikipedia community.
This avoids the unreal conceptualisation of Wikipedia as a series of derivative works,
none of which may contain enough of a material alteration to be considered original. A
contextual approach to the joint authorship test also means taking the common design
of contributors into account when determining whether a contributor has made a
significant contribution of the right kind. This means that contributions that are viewed
as important in the context of the common design will not be immediately excluded. In
this light those who enforce community policy might not be so easily dismissed as
having not made enough of a contribution of the right kind. This leads into the second
aspect of my argument, which is that an inclusive approach should be taken to the
97
application of the joint authorship test. In most cases it will be impossible to isolate
only one or two contributors, and consider them to be the sole authors of a Wikipedia
page, particularly in light of the lack of control which contributors have in respect of
their contributions. If a high standard were required for people to be considered joint
authors, this might lead to the view that Wikipedia has no authors, which, in turn,
would lead to a significant mismatch between copyright law and creative realities.
Wikipedia demonstrates that private ordering can be very successfully used to
regulate authorship issues (including issues relating to copyright ownership). Whilst
there are good reasons to be sceptical about copyleft licences, they do have an
important role to play on Wikipedia in providing a space which allows for this unique
type of creativity to flourish 116 . Although the availability of private ordering
mechanisms has been crucial in allowing Wikipedia to implement a notion of
authorship that is tailor-made to the project, there remains an important role for
copyright law. Copyright law provides an important source of default good authorship
standards. These standards make it possible to constitute the sharing norms that
Wikipedia depends upon for its success. As the example of Wikipedia indicates, critics
have been justified in expressing concern that copyright law may not be well-adapted to
Web 2.0 collaborations. Yet, I have suggested that this might not necessarily entail a
crisis for copyright law. Instead, I argue that if copyright law is to retain its relevance
and credibility as a source of standards it must be clarified and brought closer in line to
community norms and expectations117.
116
They also allow Wikipedia to avoid the rules on joint ownership which are not well-adapted to large
group collaboration. Perry and Margoni (n79).
117
Corbett (n4).
98
These observations do not just relate to the modern digital environment. The next
Chapter considers a case study of collective authorship that carries on a tradition which
is ancient in origin, but which still thrives today: Australian Indigenous art. The case
study of Australian Indigenous art provides a number of insights about the limitations
of the current approach to the application of the joint authorship test (particularly in
relation to the application of the idea/expression dichotomy) and reveals some of the
disadvantages of relying upon private ordering to remedy the gap between copyright
law and creative realities in cases of collective authorship.
99
Chapter 4: Australian Indigenous Art
The limitations of the current approach to the joint authorship test are also apparent
outside the context of new forms of digital collaboration. This Chapter presents a case
study of collective authorship with ancient roots: Australian Indigenous art. I focus on
Indigenous art produced in accordance with customary law that Indigenous
communities consider to be a product of communal authorship. Although there is no
UK case law on the application of the joint authorship test to works of collective
authorship, Australian courts have had the occasion to consider, and reject, a claim of
communal authorship in respect of Indigenous art1. In Bulun Bulun v R & T Textiles the
Federal Court of Australia struggled to find a way to protect an Indigenous
community’s interests in art within the bounds of copyright, because of the apparent
difficulty establishing that these interests amount to authorship interests2. Indeed, one
of the most significant challenges in providing effective protection of Indigenous
cultural expressions in general is accommodating them within existing legal regimes for
the protection of cultural property, such as copyright law. In this Chapter I argue that
copyright law’s difficulties in recognising Indigenous community interests in art stems
from the influence of historically-located, culturally specific understandings of
authorship; and a preference for a restrictive approach (finding few joint authors) to the
application of the joint authorship test. In the absence of copyright protection,
1
Although the distance between UK and Australian copyright law has increased over time, there remains
a sufficient degree of similarity in the application of the joint authorship test to provide a helpful
comparison for the purposes of this thesis. The definition of a work of joint authorship is virtually
identical to the CDPA, s10 of the Copyright Act 1968 (Cth) provides that it: ‘means a work that has been
produced by the collaboration of two or more authors and in which the contribution of each author is not
separate from the contribution of the other author or the contributions of the other authors’.
2
(1998) 86 FCR 244.
100
Indigenous communities must rely on contracts and a patchwork of other legal
doctrines, which only provide partial protection. Ultimately, Indigenous cultural
expressions and traditional knowledge would be best protected by a sui generis regime
that might be better adapted to this task. Nevertheless, I argue that the joint authorship
test is probably more flexible than Bulun Bulun might suggest.
In the first section of this Chapter I consider the dynamics of the authorship of
Indigenous art and I provide some background to the issue of the protection of
Indigenous cultural expressions in Australia. The second section considers a number of
cases in which Indigenous people have sought to use copyright law to protect their
interests in art. I analyse some of the conceptual challenges that prevented the court in
Bulun Bulun from recognising the communal authorship of an artwork. The third
section looks at some of the other ways in which Indigenous people have sought to have
their interests in their cultural property protected. In this context, private ordering and
market-based measures have provided limited assistance because of the inequality of
bargaining power that arises from the poor socio-economic position of many
Indigenous artists and their communities. The final section of this Chapter suggests
some insights for copyright law that arise from this case study, in particular, the need
for a more flexible, contextual approach to the application of the joint authorship test.
4.1 Indigenous Art
The first part of this section considers the dynamics of the authorship of Indigenous art.
I outline the meaning and the consequences of authorship in ‘traditional’ Indigenous
101
communities3. It is worth noting at the outset that a different analysis might apply to the
work of many modern urban Indigenous artists not produced in accordance with
Indigenous customary law. Although there is a danger of adopting a definition of
‘traditional’ Indigenous art which is too static, there are important policy reasons to
treat art produced in accordance with Indigenous customary law as a special case. This
sort of art provides an example of collective authorship as it is communal in origin. The
second part of this section provides a brief contextual background to the issue of
protecting Indigenous cultural expressions in Australia in order to shed light on the
social, cultural and political significance of the problem of protecting Indigenous art.
This provides context to the judicial decisions on the subsistence of copyright in
Indigenous art that are considered in the following section.
4.1.1 Authorship Dynamics: Building Cultural Identity
Indigenous art often has a spiritual or religious significance as it depicts and invokes an
ancestral past, or ‘Dreaming’. For Indigenous communities, the Dreamtime is a period
when ancestral spirit beings emerged from the ground to transform the earth and
determine the form of social life4. These ancestral beings are seen to have moved aside,
often merging into land forms and retaining the power to intervene in the life of man.
Certain pre-existing designs are seen as artistic manifestations of ancestral beings or
3
W Wendland, ‘Intellectual Property, Traditional Knowledge and Folklore: WIPO’s Exploratory
Program – Part 1’ (2002) International Review of Intellectual Property and Competition L 485, 491
objects to the use of the word ‘traditional’ as it seems to imply a quality of stasis that may not adequately
account for the dynamic nature of Indigenous culture. Wendland (495) stresses that the important issue is
not necessarily identifying the precise ambit of terms like ‘traditional knowledge’, but rather delineating
the terms upon which they will be protected. It is in this spirit, and for practical reasons, that I use the
word ‘traditional’ here and I adopt the term ‘Indigenous cultural expressions’ when referring to the
debate more generally.
4
D Ellison, ‘Unauthorised Reproduction of Traditional Aboriginal Art’ (1994)17 UNSWLJ 327, 330-1;
See also Mr Justice Ronald Sackville, ‘Legal Protection of Indigenous Culture in Australia’ (2003) 11(2)
Cardozo J of International and Comparative L 711, 714-15.
102
aspects of ritual knowledge associated with the Dreamtime5. According to Indigenous
customary law, pre-existing designs ought not to be changed. Their efficacy in
activating ancestral power, would be impaired if they deviated too widely from socially
accepted norms 6 . Certain persons (depending upon descent, apprenticeship and
initiation) have the right to depict pre-existing designs, which are seen as owned by the
relevant Indigenous community. The artist is accountable to the community, so in this
sense the group has control over the reproduction of pre-existing designs and the
subsequent uses of those reproductions7. Indigenous art is integral to the social fabric
of ‘traditional’ Indigenous communities, because the transmission of Indigenous
cultural heritage is an essential part of it 8 . Art often provides the framework for
ordering the relationships between people, ancestors, ritual knowledge and the land.
Thus, paintings might have a dual aspect of representing relationships between things
and being integral to those relationships9. If objects of cultural expression are removed
from their context, they often cease to have meaning; and if appropriated or
disrespected their sacred nature might be lost10.
Authorship (of art) might have a number of values within Indigenous
communities governed by customary law. It is often an act of religious significance
5
ibid.
6
ibid.
7
ibid.
8
L Behrendt, C Cunneen, T Libesman, Indigenous Legal Relations in Australia (OUP 2009) 210.
9
H Morphy, Ancestral Connections: Art and an Aboriginal System of Knowledge (U of Chicago P, 1991).
10
A Diver, ‘“A Just War”: Protecting Indigenous Cultural Property’ (2004) 6(4) Indigenous L Bulletin 7.
103
expressing and maintaining a relationship to the land 11 . It is primarily a way of
transmitting and preserving culture, but the sale of certain artworks can also provide a
valuable source of income for the artist and his or her community12. The authorship of
art promotes community cohesion and helps socialise young members into a particular
worldview 13 . This is because Indigenous artists must often undergo a period of
initiation and apprenticeship, learning particular artistic techniques and discovering the
meaning of certain designs and motifs. The artist is also a custodian of cultural
knowledge; as such, he or she is responsible to the community for the appropriate use of
the sacred knowledge embodied in the artwork. This can include subsequent uses of the
original and any copies which are made of it, whether or not those copies have been
authorised by the artist14.
In Indigenous communities the values of individual authorship are often
inextricably linked to community values and norms (such as the hierarchy of access to
particular cultural meanings). Although an artwork might actually be painted by one
individual, from an Indigenous point of view the whole community is involved in the
creation of that artwork. The artist is only able to paint the artwork after a period of
apprenticeship, during which elders pass on secret information and teach the artist the
appropriate techniques for depicting the ritual knowledge. Then, the artist is initiated
11
Morphy (n9). Some Indigenous art is not of spiritual significance, for instance, the result of children
playing.
12
See the testimony of Indigenous artist John Bulun Bulun in Bulun Bulun v Nejlam Investments
(unreported, Fed. Ct. of Australia, Darwin 1989). For more see: C Golvan ‘Aboriginal Art and the
Protection of Indigenous Cultural Rights’ (1992) 14(7) EIPR 227, 228.
13
Morphy (n9).
14
This was acknowledged by the Federal Court of Australia in Milpurrurru v Indofurn (1994) 54 FCR
240, 246, where Justice von Doussa made an award of damages for ‘cultural harm’ as compensation to
Indigenous artists for the stigma, humiliation and punishment that they were likely to suffer within their
communities as a result of the unauthorised and offensive reproduction of their artworks.
104
and given the permission to depict certain aspects of the ritual knowledge that belongs
to his or her tribe. As a delegate of the tribe and a custodian of ritual knowledge an
artist may paint an artwork, but it is the community as a whole who have created it in
the relevant sense. Thus, the resulting artwork is seen as owned by the community as a
whole, as the ritual knowledge depicted in it belongs to, and is controlled by, the
community (as represented by elders).
The communal values of authorship for Indigenous communities might be
contrasted with the values of authorship in the Western art market, which tend to relate
to authenticity – a value which attaches strongly to an artist as an individual (albeit as
an Indigenous individual). This authenticity value is more easily accommodated by
copyright law than communal values of authorship, because copyright law defines
authorship in terms of origination, which tends to be linked to the mechanical act of
fixing a work in a material form. The incompatibility of these two notions of authorship
is evident in recent controversies about the authenticity of paintings, which include
substantial unattributed contributions from family members of the named Indigenous
artist ‘author’15.
4.1.2 Background to the Issue of Protecting Indigenous Cultural
Expressions
The apparent gap in legal protection available for Indigenous artists has been a topical
issue in Australia in recent times16. At the same time, the legal protection of traditional
15
E Burns Coleman, Aboriginal Art, Identity and Appropriation (Ashgate 2005) 84-5.
16
The problem of Indigenous art ‘rip-offs’ in the souvenir market was considered in the ‘Stopping the
Rip-Offs’ issue paper <http://www.ag.gov.au/Publications/Pages/StoppingtheripoffsOctober1994.aspx >
105
knowledge and Indigenous cultural expressions are issues that have featured
prominently on the international intellectual property agenda17. In both the national and
the international context, this debate has been significantly affected by historical
factors, such as colonialisation and social justice concerns. On the international level,
however, the protagonists are different: the debate takes place been states, tending to
elide the issue of empowering indigenous groups to lobby their own state for protection
of cultural expressions where that state is reluctant to take these measures. Despite this
difference, international and national debates feed into one another. Some
commentators characterise this dispute as being one of the developed world’s ideas of
intangible property (embodied in current intellectual property law) versus the
developing world’s (or Indigenous) concepts of intangible property18. Indeed, under the
current intellectual property status quo, developed countries seem to benefit more in
economic terms than developing countries 19 . This suggests that current intellectual
property law is a product of dominant social and economic forces which have shaped
modes of (cultural) production 20 . A central question in the debate on Indigenous
(Australian Federal Government, October 1994); T Janke, Our Culture: Our Future A Report on
Australian Indigenous Cultural and Intellectual Property Rights (Michael Frankel & Co and Terri Janke,
1998); and in the media: ‘The 7.30 Report’, ABC, 19 August 2010 <http://www.abc.net.au/7.30/content/
2010/s2988038.htm>. Recently, there has been controversy over a sculpture featuring unauthorised and
offensive depictions of Wandjina spirits that was eventually removed following an action under planning
regulations by the Blue Mountains City Council: R Ayres, ‘The Wandjina case illustrates the lack of
protection for Indigenous Culture’, Art+Law, Issue 3, September 2010; and the costumes of two Russian
ice dancers that performed an ‘Aboriginal dance’ in the Russian National Figure Skating Championships:
B Manton, ‘Russian Ice Dancers Should Re-think their Routine’ Sydney Morning Herald, 21 January
2010 <http://www.smh.com.au/opinion/politics/russian-ice-dancers-should-rethink-their-routine-2010012
1-mnwj.html>.
17
In October 2000, for example, the WIPO General Assembly established an Intergovernmental
Committee on Intellectual Property and Genetic Resources, Traditional Knowledge and Folklore.
18
Note that the characterisation of such interests as intangible property is not entirely uncontroversial.
19
Some argue that a strong intellectual property regime will stimulate creativity for developing countries
in the long term. It is, however, by no means clear that such a regime would promote creativity to a
greater extent than allowing free copying.
20
Behrendt, Cunneen and Libesman (n8) 224; C Craig, Copyright, Communication and Culture:
Towards a Relational Theory of Copyright Law (Edward Elgar 2011) 12-13.
106
cultural expressions is whether ‘cultural rights’ in the broadest sense are rival to
intellectual property rights, or essentially the same thing? Are such rights necessarily
constructed in opposition to traditionally ‘authored’ works or should they be
incorporated into the copyright fold? These questions reverberate throughout this
Chapter, as they do through the Australian case law considered in the next section.
Indigenous art is of great economic and cultural significance in Australia. The
Indigenous art market is valued at around 300 million AUD each year 21 and there
appears to be annual growth in value of the Indigenous art industry of 40 to 50%22.
Indigenous art can create a ‘window of recognition’ between white and black Australia
and the whole community can share a collective sense of pride in it23. The benefits of
Australian Indigenous art are difficult to quantify. They include:
Pride, self esteem, maintenance of culture, transmission of culture,

intergenerational learning, meaningful activity, purposeful life, creative
achievement, recognition from peers; recognition from national and
international art media, provision of much of our nation’s ‘corporate
identity’; provision of ‘Australia’s greatest cultural export’...24
Unfortunately, Indigenous Australians are very rarely the main financial beneficiaries
25
of their art . Furthermore, there are widespread problems of unauthorised
21
T Janke, Beyond Guarding Ground: A Vision for a National Indigenous Cultural Authority (Terri
Janke & Co, 2009) 6. The Australian Senate Standing Committee on Environment, Communications,
Information Technology and the Arts, ‘Indigenous Art – Securing the Future: Australia’s Indigenous
Visual Arts and Craft Sector’ (June 2007) (‘Senate Report’) placed it around $200-300 million, noting the
dearth of available statistics.
22
Senate Report, ibid 13.
23
Golvan (n12) 227.
24
Submission from Ms Christine Godden quoted in the Senate Report (n21) 15.
25
According to the Senate Report, ibid 13-4, this was one probable explanation of evidence of the
massive growth in the value of the industry combined with no commensurate expansion in the number of
artists. This assumption was supported by evidence presented to the inquiry.
107
reproductions and uses of Indigenous art which have caused offence and cultural harm
to Indigenous communities. The formal legal protection available for Indigenous
cultural expressions is unclear and often seen to be inadequate. This issue has been the
subject of a number of inquiries and public reports, most notably, the ‘Stopping the Rip-
Offs: Intellectual Property Protection for Aboriginal and Torres Strait Islander Peoples’
Issues Paper and the Janke and Frankell report, ‘Our Culture: Our Future’26.
Indigenous people seek to have a number of different rights/interests enforced in
respect of their cultural expressions. These include: the right to economic benefits from
commercialization; the right to control disclosure (including the right to keep certain
things secret); the right to control their use and in particular to prevent derogatory,
offensive and fallacious uses; the interest in their preservation and continuation (for
identity/heritage reasons); the right to be acknowledged and attributed as their ‘author’.
Indigenous artists, communities, and their advocates, have sometimes struggled to
enforce these rights/interests within the framework of existing legal regimes which
provide protection to intangible property. This Chapter focuses on copyright law – the
regime which has often seemed best adapted to address one of the most pressing
problems in relation to Indigenous Australian art: unauthorised and/or offensive
reproductions.
The issue of protecting Indigenous art cannot be separated from its social and
historical context. The legacy of colonialisation, Indigenous people’s historic
experience of discrimination and injustice, the Indigenous struggle for self-
26
See n16 above. Other notable reports include: Commonwealth Department of Communications,
Information Technology and the Arts, ‘Report of the Contemporary Visual Arts and Crafts Inquiry’, ,
June 2002 (‘Myer Report’); T Janke, Minding Culture: Case Studies on Intellectual Property and
Traditional Cultural Expressions (WIPO 2003); and Beyond Guarding Ground (n21).
108
determination and a sense of regret or responsibility for past wrongs from much of the
Australian community, all form part of the social and cultural tapestry upon which
judicial decisions and political debate on this subject are interwoven27. In this context,
improving protection for Indigenous cultural expressions is often seen as a matter of
public importance, part of an ongoing project of reconciliation between Indigenous
people and the wider Australian community for the injustices of the past28.
The process of reconciliation with the Indigenous peoples of Australia has been
slow and fraught with difficulties29. The Australian legal system has proven limited in
its ability to provide adequate and appropriate remedies for Indigenous Australians
across a range of areas. The debate on the ‘recognition’ of Indigenous customary law,
for example, always prioritises Australian law over Indigenous law30. The recognition
of native title rights is also subject to a presumption that generally favours non-
Indigenous land interests over Indigenous interests in land31. In addition, there are a
number of practical problems which affect the ability of Indigenous people to take
advantage of the legal protections which may already be available to them. One such
example comes from the Arts Law Centre of Australia’s newsletter:
Arts Law advised one very senior artist in Arnhem Land who believed the
document he signed (with a cross as he neither read nor wrote English)
27
For further see: Behrendt, Cunneen and Libesman (n8).
28
‘Creative Nation’, Commonwealth Cultural Policy Statement, October 1994, 67; C Hawkins, ‘Stopping
the Rip Offs: Protecting Aboriginal and Torres Strait Islander Cultural Expression’ [1995] Aboriginal L
Bulletin 3.
29
One step forward, such as the Australian Government’s apology in 2008, is often accompanied by steps
backwards, such as the failure to enact the Indigenous Communal Moral Rights Bill.
30
C Graber, ‘Aboriginal Self-Determination vs The Proprietisation of Traditional Culture: The Case of
the Sacred Wanjina Sites’ (2009) 13(2) Australian Indigenous LR 18.
31
Behrendt, Cunneen and Libesman (n8) xiv.
109
was a sale document for his painting. In actuality, the document also
purported to assign all his copyright to a third party who produced
photographic reproductions for online sale with no further royalty or
payment.32
Indigenous Australians have sometimes been reluctant to bring legal claims, because of
their historically uneasy relationship with courts and the police, who have not always
been perceived to be instruments of justice in their eyes 33. This is exacerbated by the
poor socio-economic position of many Indigenous artists. Although many of these
issues are outside the scope of this Chapter, they ought to be borne in mind when
assessing the case law and, in particular, when assessing the adequacy of the current
approach to determining the subsistence of copyright in Indigenous art.
4.2 Protecting Indigenous Art with Copyright
There are often a number of problems establishing the subsistence of copyright in
Indigenous cultural property. Obstacles include, for example, the idea/expression
dichotomy which might preclude protection for styles and themes; the requirement of
originality34; the material form requirement35; and the fact that copyright law struggles
32
D Everard, ‘Code of Conduct for Indigenous Art’, Art+Law, June 2009, available at
<http://www.artslaw.com.au/artlaw/archive/2009/09CodeOfConduct.asp>.
33
For example: ‘Report of the Royal Commission into Aboriginal Deaths in Custody’, tabled in the
Australian Federal Parliament on 9 May 1991 considered the underlying social, cultural and legal reasons
for the disproportionately high rates of incarceration of Indigenous people. Chapter headings used in this
report suggest some of the major issues, they include: the Legacy of History, Aboriginal Society Today,
Relations with the Non-Aboriginal Community, The Harmful Use of Alcohol and Other Drugs,
Schooling, Employment, Unemployment and Poverty, Housing and Infrastructure, Land Needs and Self-
Determination.
34
There was thought to be difficulty establishing the subsistence of copyright in Indigenous art that
depicts pre-existing clan-owned motifs. It is unlikely that Indigenous art will lack originality on these
grounds following: Yumbulul v Reserve Bank of Australia [1991] FCA 332, 21 IPR 481 and Milpurrurru
(n14).
110
to give meaning to Indigenous notions of communal ownership and custodianship of
particular cultural forms 36 . Body art, for example, might be too ephemeral to be
protected37. Copyright law is also probably not an effective tool to protect ancient rock
art, because the author or authors are generally unknown and the term of copyright
protection is likely to have expired38.
This section considers a line of cases in which Indigenous artists have sought to
use copyright law to protect their cultural expressions from unauthorised and offensive
reproductions. These cases culminated with the Federal Court of Australia’s decision in
Bulun Bulun v R & T Textiles, which is an important case because it exemplifies the
difficulties of accommodating Indigenous interests in art within copyright’s conceptual

39
framework . Bulun Bulun concerned a claim by an Indigenous elder, as a
representative of his community, that an artwork created in accordance with customary
law was the product of ‘communal’ authorship and, as such, the Court ought to
recognise the community’s interest in any copyright subsisting in the artwork. Although
the Federal Court of Australia expressed its sympathy for this cause, it ultimately felt
too constrained by the limits of existing legal concepts to be able to offer much
effective relief, acknowledging the inadequacies of copyright law’s provisions in this
context.
35
Dances and stories might lack the material form required for copyright subsistence: E Adeney,
‘Unfixed Works, Performers’ Protection, and Beyond: Does the Australian Copyright Act Always
Require Material Form?’ (2009) IPQ 77 comparing UK and Australian law on this point.
36
In addition, as suggested above, there are practical problems bringing claims and an imbalance in
power relations.
37
Merchandising Corporation of America v Harpbond [1983] FSR 32.
38
On Indigenous rock art see: ‘Chapter 5: Rock Art Revisited’ in S Kleinart & M Neale, The Oxford
Companion to Aboriginal Art and Culture (OUP 2000) 103.
39
(1998) 86 FCR 244.
111
Before considering Bulun Bulun in detail it is worth considering some of the cases
that laid the groundwork for that decision. Many early cases did not make it to court. In
1966, for example, the Reserve Bank of Australia was the subject of much public
embarrassment for reproducing David Malangi’s artwork (a bark painting entitled ‘The
Hunter’) on the one dollar note without seeking his permission, assuming it to be the
work of an ‘anonymous and probably long dead artist’40. This case was settled with the
artist being given $1000, a fishing kit, and a silver medallion. In 1988, there was
another widely publicised case involving the unauthorised reproduction of the artworks
of fourteen Indigenous artists on T-shirts which was settled for $150,000, which the
artists decided to share equally (irrespective of the number of infringing reproductions
of each artwork) in accordance with their customary law41. This case provided hope that
copyright law might prove a useful means of protecting Indigenous communities’
interests in their art.
In another action brought against the Reserve Bank of Australia, Yumbulul v
Reserve Bank of Australia (which concerned the reproduction of Terry Yumbulul’s
‘Morning Star Pole’ sculpture on the ten dollar note), Justice French noted the ‘...
difficulties that arise in the interaction of traditional Aboriginal culture and the
Australian legal system relating to the protection of copyright and the commercial
exploitation of artistic works by Aboriginal people’42 and that Australia’s copyright law
may not provide ‘adequate recognition of Aboriginal community claims to regulate the
40
Janke (n16) 8.
41
Bulun Bulun v Nejlam Investments (unreported, Federal Court of Australia, Darwin 1989). For more,
see: C Golvan, ‘Aboriginal Art and Copyright: The Case For Johnny Bulun Bulun’ (1989) 11(10) EIPR
346.
42
Yumbulul (n34) [1].
112
reproduction and use of works which are essentially communal in origin’43. The court
suggests that this is a matter that ought to be dealt with by legislative intervention44. In
this case, the Court did not need to explicitly consider any copyright issues, focusing
instead on the validity and terms of licences that Yumbulul had granted the Aboriginal
Artists Agency in respect of the work. Evidence presented to the effect that Mr
Yumbulul might not have understood the nature of the licences that he signed was not
accepted by the court, although Justice French acknowledged that ‘…[i]t may be that
greater care could have been taken in this case’45.
In Milpurrurru v Indofurn, which was described as a ‘landmark in the protection
of Indigenous culture’ 46 , the Federal Court explicitly recognised that Indigenous
artworks, which depicted pre-existing clan-owned designs, were original works in
which copyright subsists because of the skill and judgment exercised by the artist in the
interpretation of those designs 47 . This was an important finding because some
commentators had previously expressed concern that Indigenous artworks which
featured traditional designs might lack the requisite originality for copyright to subsist
in them (as the form of these designs is strictly controlled by customary law) 48. In his
43
ibid [21].
44
ibid [2].
45
ibid [21].
46
Ms Bronwyn Bancroft, artist and chairperson of the National Indigenous Arts Advocacy Association: T
Janke, ‘Briefs: Copyright – The Carpet Case’ [1995] Aboriginal L Bulletin 13, (1995) 3(72) Aboriginal L
Bulletin 36.
47
See Mr Justice von Doussa’s comments (n14) 248, although the subsistence of copyright in the
artworks was eventually conceded by the carpet importers.
48
Milpurrurru (n14) 247-8: ‘...a problem perceived to exist at one time in relation to the application of
the Copyright Act to Aboriginal artworks based on pre-existing tradition and images...whether works
incorporating them satisfied the requirement of originality... Although the artworks follow traditional
Aboriginal form and are based on Dreaming themes, each artwork is one of intricate detail and
complexity reflecting great skill and originality’. See also Golvan (n12).
113
judgment, Justice von Doussa acknowledged the use of art by Indigenous communities
as a means of recording stories of the Dreaming and for teaching future generations49.
Much evidence was presented of the importance of art in the Indigenous community
and its regulation under customary law. Testimony was also heard on the nature of
communal ownership under Indigenous customary law50.
In this case, the importers of carpets which infringed the copyright in the artworks
were required to pay additional damages which took into account the ‘culturally based
harm’ suffered by the three living artists in the action 51 . This was intended to
compensate the artists for the stigma, humiliation and punishment that they were likely
to suffer within their communities (because they would be held responsible for the
unauthorised and offensive reproductions of sacred images on the carpets even if they
had no knowledge of what had occurred and no control over it)52. It was not an award to
the community to compensate for the devaluing of their cultural artefacts.
Although the judgment does not explicitly uphold Indigenous customary law
provisions in relation to the ownership and appropriate treatment of the artwork,
accommodations of customary law were made within the scope of judicial discretion as
permitted by court rules and procedures53. In accordance with Indigenous customary
49
ibid 245.
50
ibid 245-6.
51
ibid 279 as permitted by Copyright Act 1968 (Cth) s115(4).
52
ibid 246, 277.
53
ibid 243, 272-3.
114
law the names of the deceased artists were not spoken in court54. Although the damages
award was not made to the whole community as compensation for the offensive
treatment of their cultural artefacts, the court did take the step of making a collective
award of damages rather than individual awards so that the amount could be distributed
amongst the artists equally according to custom 55 . This willingness to use judicial
discretion to accommodate customary law might have seemed, at the time, like a first
step towards the recognition of communal ownership of artistic works by Indigenous
people within the Australian legal system.
The Australian case law on copyright ownership and Indigenous art culminates
with the Federal Court of Australia’s decision in Bulun Bulun. This case concerned the
importation and sale of printed clothing fabric which infringed John Bulun Bulun’s
copyright in the artistic work, ‘Magpie Geese and Water Lilies at the Waterhole’ 56.
Bulun Bulun’s artistic work incorporated sacred subject matter derived from the ritual
knowledge of the Ganalbingu people that was only reproduced because of certain rights
granted by Indigenous customary law. The respondents admitted infringing Bulun
Bulun’s copyright and the trial focused upon the second applicant, George Milpurrurru,
who as a representative of the traditional Indigenous owners of the Ganalbingu country
(Arnhem Land, Northern Territory), brought a number of claims in the alternative:
54
ibid 243. The deceased applicants were only referred to by name once for the purpose of ensuring that
they were identified with certainty.
55
ibid 272-3. The damage award was expressed in terms of an aggregate liability.
56
(n2) 246.
115
(i) that the community are equitable owners of the copyright subsisting in the
artwork57;
(ii) that Mr Bulun Bulun held ownership of copyright in the artwork on trust for the
community; or
(iii) that Mr Bulun Bulun’s ownership of copyright in the artwork gave rise to
fiduciary obligations in favour of the community.
At the heart of these claims was the fact that the relevant artistic work incorporated
sacred and important subject matter derived from the ritual knowledge of the
Ganalbingu people that was only reproduced because of certain rights granted by
Indigenous customary law.
Justice von Doussa highlighted the significance of the case:
These proceedings represent another step by Aboriginal people to have

communal title in their traditional ritual knowledge, and in particular in their
artwork, recognised and protected by the Australian legal system. The
inadequacies of statutory remedies under the Copyright Act as a means of
protecting communal ownership have been noted in earlier decisions of this
court: see [Yumbulul and Milpurrurru]58
And later, ‘... [t]he claim raises important and difficult issues regarding the protection
of the interests of Indigenous people in their cultural heritage’59. Indeed, in recognition
of the importance of the issues, the Court took the step of travelling to Arnhem Land to
57
The Federal Minister for Aboriginal and Torres Strait Islander Affairs and the Attorney-General (NT)
intervened to argue that this claim, in effect, involved an assertion of native title land rights over
Ganalbingu country, which the court lacked the jurisdiction to determine in the absence of a proper
application under Native Title Act 1993 (Cth) s74. The interveners were significant because the
respondent company was under administration and without them there would have been no contradictor.
58
Bulun Bulun (n2) 247.
59
ibid 257.
116
hear evidence from Indigenous elders and to inspect the site that was the subject of the
artwork60.
Justice von Doussa explained that the claims were confined to the recognition of
an equitable interest in the copyright subsisting in the artwork, for a number of reasons:
(i) Section 8 of the Copyright Act 1968 (Cth) precludes the continuing application of
any copyright protection, except as provided by the Act. This was seen to remove
the possibility that Indigenous customary law providing communal title to
artwork could be recognised by the common law.
(ii) Section 35(2), which provides that the author is the owner of any copyright
subsisting in a work, was seen to preclude the notion of group authorship (except
insofar as provided for by the joint authorship test).
(iii) The case law on joint authorship was said to establish that it is not enough to
supply the artistic idea for a work; instead, it envisages a contribution of skill and
labour to the production of the work itself. Justice von Doussa clearly implies that
these criteria are not satisfied in this case as he assumes that the community’s
contribution was only of unprotectable ideas61.
(iv) The case was further complicated by the fact that the claim as originally stated
appeared to imply that the community’s interest in the artwork was an incident of
native title to the land, which ultimately had to be abandoned because no
application for native title had been made in the proper form62.
60
C Golvan, ‘Aboriginal Art and Copyright: An Overview and Commentary Concerning Recent
Developments’ (1999) EIPR 549, 552.
61
Bulun Bulun (n2) 257-8 (von Doussa J) citing Kenrick v Lawrence (1890) 25 QBD 99 and Fylde
Microsystems v Key Radio Systems [1998] FSR 449 (Ch).
62
Native Title Act 1993 (Cth) s74. It is unfortunate that the Court did not consider the possibility that
intellectual property rights might arise as an incident of native title rights. Subsequent cases have
117
Justice von Doussa and counsel for the parties seem so convinced of the
impossibility of success of an argument that the community might own a legal interest
in copyright in the artwork that despite a ‘wide ranging search’ for ways to establish
this, the statement of claim was amended to claim only equitable interests63. This left
Justice von Doussa with limited options. There could be no express trust in favour of
the community because the facts did not establish evidence of an intention to create
such a trust and even if they had, the terms of any such trust would be likely to lack
sufficient certainty64. The court considered the possibility that there might have been an
express contract between the artist and the community, but found that there was no
evidence of an ‘intention to create legal relations’ because there was no suggestion of
an ‘express agreement of a contractual nature in which terms were agreed’ 65 . The
conditions in which a contract might be implied also did not appear to be present. This
is unsurprising given that these conditions are directed towards upholding the
reasonable expectations of parties to a transaction (as indeed is the law of contract)66.
The principles for implying a contract focus on factors such as business efficacy, which
seem to be fundamentally at odds with the moral and deeply religious character of the
relationship between an Indigenous artist and his or her community expressed in the
creation of art which embodies ritual knowledge.
foreclosed this possibility, citing Justice von Doussa’s comments that recognising such rights would
fracture a skeletal principle of the common law in: Western Australia v Ward [2002] HCA 28, (2002) 213
CLR 1 [59], [60], [644]; in dissent [580] (Kirby J); Neowarra v Western Australia [2003] FCA 1402
following Ward and rejecting a claim that a native title right existed to use, maintain, protect and prevent
the misuse of cultural knowledge (specifically, sacred painting and ceremonies) of the Wanjina-
Wunggurr community in relation to a native title land claim. See further: Graber (n30).
63
Bulun Bulun (n2) 256-7.
64
ibid 259.
65
ibid 262-3.
66
See: BP Refinery (Westernport) v Hastings Shire Council (1977) 180 CLR 266, 283. Cf. Mr Justice
von Doussa noted that there has been some success adopting the contract approach in Ghana: Bulun
Bulun, ibid 263.
118
Justice von Doussa found it easier to establish the existence of a fiduciary duty,
given the inherent adaptability of this legal category and its purpose to remedy harms
that arise in situations where a special relationship of trust and confidence exists
between parties. Justice von Doussa held that Bulun Bulun owed a fiduciary duty to his
community arising from the nature of the ownership of artistic works amongst the
Ganalbingu people because the grant of permission by representatives of the
community to paint certain designs is ‘predicated on the trust and confidence which
those granting permission have in the artist’67. In holding that a fiduciary duty exists in
these circumstances Justice von Doussa affirmed that the community’s interest in ‘the
protection of [their] ritual knowledge from exploitation which is contrary to their law
… is deserving the protection of the Australian legal system’ 68. The content of this
fiduciary duty, however, was held to be very narrow: merely requiring the artist to act
in the interests of the community in relation to the artwork, so as to preserve the
integrity of their culture and ritual knowledge69. This duty would not prevent the artist
from pursuing his own interest to a certain degree, for example, by selling the artwork
for profit. In this case the duty did not extend further than to oblige Bulun Bulun to sue
infringers of copyright in the work, a duty which he had satisfied 70. The existence of
this fiduciary duty ‘without more’ was held to be insufficient to vest an equitable
interest in the ownership of the copyright in the Ganalbingu people 71. The fiduciary
duty would only give rise to a constructive trust in the most extreme of circumstances –
67
ibid 261-2.
68
ibid 263.
69
ibid 262.
70
ibid 263.
71
ibid.
119
in order to allow the community to bring an action to enforce copyright in the work in
the artist’s absence.
Bulun Bulun was seen by many as a beacon of hope for the protection of
Indigenous cultural expressions and has been greatly commented upon. This is largely
because the Court recognised the significance of Indigenous cultural expressions for
Indigenous communities, and for Australia in general, admitting the desirability of
protecting such expressions from misuse and unauthorised exploitation. Although Bulun
Bulun might be seen as a symbolic judicial gesture of good faith, ultimately Justice von
Doussa considers the matter to be one that ought to be primarily dealt with by the
legislature72. The importance of Bulun Bulun has often been overstated73. The decision
is of little precedential value, given that it is likely to be restricted to its facts and it
largely ignores the social realities of the production and exploitation of Indigenous art74.
The problem of widespread unauthorised and offensive commercial exploitation of
Indigenous cultural expressions is not likely to be ameliorated by the narrow fiduciary
duty created in Bulun Bulun, as Indigenous artists are rarely unwilling to protect their
community’s interest in artwork produced according to customary law75.
72
ibid 247 (von Doussa J) referring to the ‘inadequacy of statutory remedies’.
73
See also E Mackay, ‘Indigenous Traditional Knowledge, Copyright and Art – Shortcomings in
Protection and an Alternative Approach’ (2009) 32(1) UNSWLJ 1, 6.
74
M Rimmer, ‘The Bangarra Dance Theatre: Copyright Law and Indigenous Culture’ (2000) 9(2)
Griffith L Rev 274, 278.
75
MacKay (n73) 6; K Weatherall, ‘Culture, Autonomy and Djulibinyamurr: Individual and Community
in the Construction of Rights to Traditional Designs’ (2001) 64 MLR 215, 221-2 arguing that Bulun
Bulun will only ever have a limited practical effect because it is response-based and prohibitively
expensive.
120
Bulun Bulun demonstrates an apparent lack of flexibility in copyright law’s
concepts of subsistence, which suggests that copyright is not well-adapted to protect an
Indigenous community’s interests in its cultural expressions. There are unlikely to be
more cases testing the limits of copyright protection in this respect. Aside from the
practical problems of access to the court system, there seems little reason for
Indigenous communities to take the financial risk of bringing another copyright claim
for protection of their interests in art as Bulun Bulun appears to categorise Indigenous
communal ownership of artwork and the various rights this ownership entails under
customary law as something that is fundamentally different from, and which cannot be
contained within, copyright law’s notion of joint authorship. In a number of places in
the judgment, Justice von Doussa characterises the Indigenous community’s ideas of
authorship as fundamentally different from copyright law’s concept of authorship. For
example, recognising the connection between the creation of artistic works and land
ownership was seen to risk fracturing a skeletal principle of the legal system (the
separate recognition of real property and intangible property) 76 . Similarly, it was
considered contrary to established legal principle for the common law to recognise
communal title77.
The judgment involves an odd characterisation of Indigenous communal
ownership under customary law, in one part of the judgment as analogous to copyright
(so as to be extinguished by section 8); and in another part, as something fundamentally
different to copyright (so much so, that recognising it would fracture a skeletal part of
the legal system). The issue of sovereignty is a strong undercurrent throughout the
76
Bulun Bulun (n2) 256.
77
ibid 257.
121
decision. Justice von Doussa affirms previous case law which limits the application of
Indigenous customary law to those outside of the relevant Indigenous community. He
holds that Indigenous customary law might only be taken into account as part of the
factual matrix of a case, rather than as an independent source of obligations for third
parties 78 . Yet, Indigenous customary law seems not to have actually had a material
impact on the assessment of the relevant factual matrix of the community’s copyright
claim, given the arguably pre-emptive dismissal of the joint authorship claim.
The failure to consider the application of the joint authorship test might be
explained by what some have criticised as a judicial tendency to focus upon ‘fixation’,
that is, how the material form of a copyright work comes into existence. In the previous
Chapter, I mentioned Burton Ong’s argument that the case law on derivative works
focuses too much on the form of a work rather than the process by which it is created 79.
This focus might tend to result in the denial of copyright protection for socially valuable
works such as art restorations80. In this way, the originality test appears to privilege
contributions which are easily observed in the material form of the copyright work. This
characteristic affects the joint authorship test, because originality provides the yardstick
for ‘authorship’ and therefore contributes to how judges determine whether a
contributor has made a contribution of the right kind. Determining which sorts of
contributions count becomes more difficult the greater the number of contributors, so
the joint authorship test particularly struggles to come to terms with creations involving
78
ibid 262, 248 noting that the court in Milpurrurru (n14) took the consequences of unauthorised
reproductions of artistic works under Indigenous customary law into account when determining the
quantum of damages to award.
79
B Ong, ‘Originality from copyright fitting recreative works into the copyright universe’ (2010) IPQ
165.
80
There are cases which suggest that this might not be the case: Sawkins v Hyperion Records [2005]
EWCA Civ 565, [2005] 1 WLR 3281. Cf Bridgeman Art Library v Corel 36 FSupp 2d 191 (SDNY 1999).
122
the contributions of large numbers of contributors 81 . This is exacerbated by the
tendency to adopt stricter requirements of authorship in some cases in order to reduce
the number of potential joint authors and, therefore, potential joint owners of copyright
subsisting in the work (the instrumental/pragmatic approach).
The application of the joint authorship test is by no means as clear cut as Justice
von Doussa seems to suggest. First, as the application of the test depends upon the facts
in issue, previous cases do not foreclose the possibility of finding joint authorship in
new circumstances. It is precisely this flexibility which I argue might be used to adapt
the test to the wide range of creative forms. Second, the joint authorship test struggles
to precisely calibrate the necessary connection between a contributor and their creation
for ‘authorship’ due to judicial concern that copyright law ought to be seen to be
aesthetically neutral. This tends to preclude the use of creativity, artistic merit, authorial
intent, or any other value-specific criterion for determining authorship. Instead, judges
have often focused on the literary, artistic, etc character of the work, or the amount of
skill, labour or judgment which went towards its fixation. Although it is not sufficient to
be a joint author just to take dictation (as this is pure expression without any form of
creativity); there is also a need to exert some labour, skill or judgment which is
expressed in the form of the work. This focus on ‘objective’ characteristics makes
copyright law more amenable to creative contributions which reflect ‘traditional’ modes
of cultural production, which might seem more self-evidently ‘authorial’ and less
adaptable to contributions which fall outside the romantic author model. Despite this
81
D Simone, ‘Copyright or Copyleft? Wikipedia as a Turning Point for Authorship’ (2014) 25(1) Kings
LJ 102; D Simone, ‘Recalibrating the Joint Authorship Test: Insights from Scientific Collaborations'
(2013) 26(1) Intellectual Property J 111.
123
focus on ‘objective’ characteristics, some cases are influenced by considerations of
public policy82 and some take a more nuanced and contextual approach83.
Cala Homes v Alfred McAlpine Homes, the classic case on joint authorship,
suggests that the joint authorship test might be able to accommodate communal
authorship of the type that occurred in Bulun Bulun. That case involved drawings made
by draftsmen to precise specifications given by the design director of a company that
designed and built homes. The design director was held to be a joint author along with
the draftsmen – the court emphasising that what was required was ‘something
approximating penmanship’. In this case contribution of the concept, the provision of
precise specifications and ongoing control over the creative process was held to be
sufficient to establish joint authorship. In many ways an Indigenous elder (acting on
behalf of the tribe) is comparable to the design director in that case and will often
exercise even greater control. Ritual knowledge provides the source for the ‘concept’;
the apprenticeship of the artist provides him or her with ‘precise specifications’ in
relation to the execution of the artwork; and customary law provides at least as
significant a means of exercising artistic control as a contract between designer and
draftsmen.
Justice von Doussa held that the common law could not recognise Indigenous
customary law provisions on communal ownership, because section 8 of the Copyright
Act provides that copyright law is ‘entirely a creature of statute’. Yet, he also assumed
that the meaning of joint authorship was limited by certain doctrines drawn from
82
Sawkins v Hyperion (n80); Interlego v Tyco Industries [1989] AC 217.
83
Cala Homes v Alfred McAlpine Homes [1995] EWHC 7.
124
precedent which cannot be found in the Act, such as the idea/expression dichotomy84.
As Indigenous communal authorship had not been previously considered in any case
law, these doctrines might have been applied in a different and more accommodating
manner, especially given the inherent flexibility in the joint authorship test which
allows it to be applied in a factually specific way. Indeed, from the Indigenous point of
view the painting was clearly a collaborative endeavour85.
Justice von Doussa suggests that the Indigenous community may have only
contributed unprotectable ‘ideas’. The idea/expression dichotomy is a notoriously
slippery distinction. Bowrey notes that from an Indigenous point of view, ritual
knowledge (recast as an idea) could not be separated from the artwork (conceived of as
the expression of that idea). She argues that by imposing an interpretation that displaces
the contributions of Bulun Bulun’s ancestors and the ongoing role of contemporary
Indigenous elders to the artistic work (as embodied in ritual knowledge) and focusing
only on Bulun Bulun’s direct contributions, the Court effectively constructed an entirely
new work, by a process which obliterates the significance of the original artwork from
an Indigenous point of view86. The conception of authorship employed by the court
seems to import temporal and mechanical divisions into the production of the work
which operate to separate the idea from the expression87; and, which in this case, led the
judge to suggest that the community had not made the right kind of contribution at the
84
K Bowrey, ‘The Outer Limits of Copyright Law – Where Law Meets Philosophy and Culture’ (2001)
12 L and Critique 75, 81.
85
ibid 81.
86
ibid 82.
87
ibid 81.
125
right moment in time, to be considered ‘authors’88. In so doing, Bowrey criticises the
judge’s commitment to established liberal values which privilege certain interests and
certain legal subjects over others89. In this way the judgment demonstrates a discursive
gap between copyright law’s notion of authorship and Indigenous conceptions of
authorship.
Although judges claim to be applying copyright law’s rules of subsistence in an
aesthetically neutral manner, the application of these rules necessarily involves making
value judgments about who is entitled to credit as an author, a decision which has a
hidden political dimension. Bowrey contrasts Justice von Doussa’s approach with the
more liberal approach taken to the subject matter in which copyright might subsist in
cases involving new technologies90. In Sega v Galaxy, for example, a video game was
held to be sufficiently ‘embodied’ to be a ‘cinematographic film’ as the possible
sequences of 3D images existed in the minds of the designers and in various 2D models,
despite the fact that sequences of images did not actually materialise until the game was
played by a user. Considering this arguably creative conception of the material form
required for copyright subsistence, the court’s decision in Sega might seem analogous
to the Ganalbingu community’s claim: their ritual knowledge existed in their minds
alongside an anticipation that the knowledge would be embodied in an artwork created
under authorisation in a way that allowed community elders to significantly influence
the content of the artwork (as executed by the artist). It may be that Sega was
88
ibid describing this as the requirement of ‘materiality’.
89
ibid.
90
Bowrey cites Kalamazoo v Compact Business Systems (1983) 5 IPR 213 (blank accounting forms held
to constitute a ‘literary work’) and Sega Enterprises v Galaxy Electronics (1996) 35 IPR 161, (1997) 37
IPR 462 (video games were a ‘cinematographic film’) attributing the judicial approach in these cases to a
‘legal infatuation with serving the perceived needs of actors in the so-called information economy’. ibid
78.
126
incorrectly decided or incomparable to the facts of Bulun Bulun 91 . Yet, Bowrey’s
analysis still raises interesting questions about copyright law’s notions of authorship
and joint authorship, which appear to privilege certain legal subjects over others, as the
rules for establishing the subsistence of copyright are malleable allowing judges to
expand the scope (and the beneficiaries) of protection where there appear to be good
reasons for doing so92.
I have argued that the joint authorship test is flexible, so it might be able to adapt
to a variety of different creative contexts. In the case of Indigenous art which
incorporates ritual knowledge and has been created with permission pursuant to
Indigenous customary law, the authorial contribution of the community is directly
reflected in the expression of the artwork – by the control over the subject matter, style,
depiction of encoded meanings; and through the initiation and apprenticeship of the
artist. The significance of the community’s role is reflected by the requirement for
artists to consult and obtain approval from elders in relation to the creation and
subsequent uses of the work. Furthermore, there are good policy reasons to recognise
that Indigenous communities’ interest in art made in accordance with customary law
amounts to an authorship interest as this: (a) recognises the social reality for Indigenous
artists; (b) allows Indigenous communities recognition for their contributions; and (c)
gives those communities the ability to better control and manage artworks (together
with the artist), which in a very real sense, are seen as communally authored.
91
In Stevens v Kabushiki Kaisha Sony Computer Entertainment [2005] HCA 54, 224 CLR 193 [86]
(Gleeson CJ, Gummow, Hayne and Heydon JJ) ‘...amicus curiae rightly pointed to difficulties to which
that case gives rise’ although neither side challenged Sega. Sega was distinguished in Aristocrat Leisure
Industries v Pacific Gaming [2000] FCA 1273 [67] but the principle that intellectual property legislation
must be liberally interpreted, especially in the case of video games, was approved.
92
Bowrey (n84) 95: ‘The reinvention of copyright concepts to accommodate new technologies need not
be confined to the cause of investors in new technologies alone’. See: Sawkins v Hyperion (n80).
127
4.3 Other Solutions for the Protection of Indigenous Cultural
Expressions
After Bulun Bulun, most scholars tend to agree that copyright law has reached its
conceptual limits in providing protection for Indigenous communities’ interests in their
art93. Indigenous people have looked to a patchwork of other legal doctrines to fill the
gaps, such as: judicial discretion in relation to the award of damages 94 ; fiduciary
duties 95 ; breach of confidence 96 ; planning legislation 97 ; contract 98 and consumer
protection law 99 . Overall, most scholars now agree that the issue of protecting
Indigenous cultural expressions, with all its complexities, is probably best addressed
with a sui generis legislative scheme that can incorporate a balancing of policy
concerns. In the absence of the legislative will to enact such a scheme, private ordering
mechanisms can provide a flexible ‘bottom up’ solution, although they are most
effective in relation to those already motivated to protect Indigenous cultural
expressions and they tend to be compromised by imbalances of power. Given the social
and cultural disadvantages that many Indigenous communities face, measures which
rely on market forces, such as collective trade marks; or private bargaining power, such
93
Mackay (n73), K Bowrey, ‘Alternative IP?: Indigenous Protocols, Copyleft and New Juridifications of
Customary Practices’ (2006) 6 Macquarie LJ 65, D Burkitt, ‘Copyrighting Culture – The History and
Cultural Specificity of the Western Model of Copyright’ (2001) IPQ 146.
94
Milpurrurru (n14)
95
For example, the fiduciary duty in Bulun Bulun (n2). See also J Gibson, ‘Justice of Precedent, Justice
of Equity: Equitable Protection and Remedies for Indigenous Intellectual Property’ (2001) 6(4)
Australian Indigenous L Reporter 1.
96
Foster v Mountford [1976] 29 FLR 233, Aboriginal Sacred Sites Protection Authority v Maurice; Re
the Warumbingu Land Claim (1986) 10 FCR 104.
97
See text at n16 and Ayres (n16).
98
See, for example: Rimmer (n74).
99
Australian Competition and Consumer Commission v Australian Dreamtime Creations [2009] FCA
1545.
128
as contract, are likely to have limited success. Legislative attempts to improve
protection for Indigenous artists to date have either only addressed a small part of the
problem (eg, resale royalties legislation100); or have fallen off the legislative agenda (eg,
the Communal Moral Rights Bill101).
In this section, I assess four of the best candidates for a non-copyright based
solution to the problem of protecting Indigenous cultural expressions: protocols/codes
of conduct; collective/certification trade marks; contracts; and sui generis legislation.
4.3.1 Protocols and Codes of Conduct
Protocols and codes of conduct might be useful in enhancing the protection of
Indigenous cultural expressions 102 . They encourage good practice in commercial
dealings in reproductions of Indigenous cultural expressions and often provide informal
dispute resolution mechanisms103. Over the previous years a number of such protocols
have been developed by arts organisations104. In October 2009 the Australian Federal
100
Resale Royalty Right for Visual Artists Act 2009 (Cth).
101
An exposure draft of the Copyright Amendment (Indigenous Communal Moral Rights) Bill (Cth) was
circulated in 2003. The introduction of the bill was postponed ostensibly until 2006, however, it has still
not been reintroduced into Parliament.
102
Janke (n21) 26.
103
Informal dispute resolution mechanisms are considered inappropriate in situations involving
significant imbalances of power, see: G Clark and I Davies, ‘Mediation – When is it not an appropriate
dispute resolution process?’ (1992) 3(2) Australian Dispute Resolution J 70, 78-79; R Field, ‘Mediation
and the art of power (im)balancing’ (1996) 12 QUTLJ 264, 266-267; D Eliades, ‘Power in mediation –
some reflections’ (1999) 2(1) ADR Bulletin 4, 6.
104
Melbourne City Council developed a code of conduct for galleries and retailers of Indigenous art:
‘Code of Practice for Galleries and Retailers of Indigenous Art’ <http://www.melboume.vic.gov.au> 23
December 2008. Protocols have also been developed by the Australia Council for the Arts, ‘Publications’
<www.australiacouncil.gov.au>; Screen Australia, ‘Indigenous Filming Protocol’ <http://www.afc.gov.
au/filminginaustralia/indigproto/fiapage9.aspx>; and SBS, ‘The Greater Perspective – Protocol and
Guidelines for the Production of Film and Television on Aboriginal and Torres Strait Islander
Communities’ (2nd edn, 1997).
129
Government, as well as State and Territory governments endorsed the Indigenous
Australian Art Commercial Code of Conduct 105 . Although protocols and codes of
conduct are helpful in raising awareness of the issues, their effectiveness is undermined
by the fact that they are voluntary so there are limited provisions to ensure compliance
(ie, the publication of non-compliant members’ names online). These measures are
likely to be most effective in relation to parties already interested in, and committed to,
compliance. Even where this is not the case, the poor socio-economic position of many
Indigenous artists may make it difficult for them to negotiate the inclusion of such
protocols in a contract for the sale of their art106.
4.3.2 Collective/Certification Trade Marks
Certification or collective trade marks are often employed by groups of traders to
promote their collective interest107. The success of this strategy is ultimately dependent
upon how well and how widely the marks are implemented, regulated and policed108.
The Our Culture: Our Future Report proposed the adoption of a National Indigenous
Authentication Trade Mark 109 . In 1999 the National Indigenous Arts Advocacy
105
<http://www.indigenousartcode.org>. The Code was a key recommendation of the Senate Report (n21)
which revealed widespread endemic patterns of unscrupulous commercial behaviour by dealers within
the Indigenous arts industry.
106
MacKay (n73) 20 argues that one context in which protocols might be effective is where government
or other funding grants are made conditional on compliance with the protocol.
107
Trade Marks Act 1995 (Cth) s162, s169. See: Janke (n21) 23.
108
L Wiseman, ‘The Protection of Indigenous Art and Culture in Australia: The Labels of Authenticity’
(2001) EIPR 14, 25.
109
Janke (n16) 198. S McCausland, ‘Protecting Communal Interests in Indigenous Artworks after the
Bulun Bulun Case’ (1999) 4(22) Indigenous L Bulletin 4 suggests that notices might also be put on
artwork in order to deter unauthorised reproductions.
130
Association (‘NIAAA’) introduced an Indigenous Label of Authenticity110. However,
the Label of Authenticity suffered from a number of practical problems: delays in
implementing the scheme 111 ; overly strict registration requirements 112 ; a failure to
adequately publicise the scheme; and quality control issues in the administration of the
scheme113. The scheme was not widely accepted by the Indigenous community as the
requirement to prove ‘authenticity’ was considered insulting and inappropriate 114. In
2002 the Australia Council suspended funding to NIAAA, which meant that the label of
authenticity scheme was rendered practically defunct 115 . Some suggest that local
trademarks for Indigenous artistic centres might be more successful and help to
empower local communities (if accompanied with the proper administrative
infrastructure) 116 . The usefulness of such marks is limited, however, because they
depend upon the market: consumers must understand the significance of the mark and
be motivated to buy the authorised/genuine products bearing the mark. In the souvenir
market, where problems of unauthorised and offensive reproductions of Indigenous art
are rife, consumers are likely to be more concerned with price, than provenance.
110
<http://www.culture.com.au/exhibition/niaaa/labelqa.htm>.
111
M Rimmer, ‘Australian Icons: Authenticity Marks and Identity Politics’ 2004(3) Indigenous LJ 139
112
75% of the initial applications failed to have sufficient supporting documentation (the applicant’s
Aboriginality had to be attested to in writing with the common seal of two Indigenous organisations):
Rimmer ibid 156.
113
ibid 164.
114
ibid 262. The scheme also failed to adequately distinguish between Indigenous groups
115
ibid 160-161 this was associated with concerns related to poor governance and management of NIAAA
and the fact that it did not include regional representation.
116
ibid 165. A number of regional arts centres successfully use registered trade marks, eg Desart.
131
4.3.3 Contract
Organisations, such as the Arts Law Centre of Australia, have developed model
contracts and licensing agreements and provide free or low cost legal advice to
Indigenous people to help them to develop arrangements which suit their needs 117. The
Bangarra Dance Troupe provides a good example of the use of contract to protect
Indigenous cultural interests118. The company entered into a private agreement with the
Munyarrun Clan recognising Clan copyright ownership of particular traditional dances
and songs and agreeing to pay a fee for the permission to reproduce their themes in its
performances. The agreement includes provisions designed to ensure that the company
takes care to ensure respectful treatment of the cultural property. Yet, the Bangarra
Dance Troupe is probably an exceptional case in terms of its legal literacy119. Further,
insofar as the agreement intends to circumvent and modify the scope of copyright there
are real questions as to its enforceability120. Similar arrangements are unlikely to be
replicated with parties that do not share the Dance Troupe’s motivation to respect
Indigenous culture. Contracts offer the benefit of promoting self-determination,
allowing for more management and negotiation at the local level 121. Yet, contract does
117
As recommended in Janke (n21) 24. For example Arts Law provide a free sample Indigenous Art
Centre and Gallery Consignment Agreement: <http://www.artslaw.com.au/Indigenous/IndigenousArtist
AndArtCentreAgreement.asp>.
118
Rimmer (n74).
119
ibid 274.
120
Rimmer, ibid cites Netanel’s concerns over contracts which seek to circumvent copyright law
provisions in the digital context – specifically: N Netanel, ‘Copyright and Democratic Civil Society’
(1996) 106 Yale LR 283, 305-6, 382-5.
121
Bowrey (n93) 75 argues that private ordering allows one to capture the voice of a community instead
of presuming to speak for it, as legislation might. Cf P Drahos with J Braithwaite, Information
Feudalism: Who Owns the Knowledge Economy? (Routledge 2002); also S Sell, Private Power, Public
Law: The Globalisation of Intellectual Property Rights (CUP 2003).
132
not work well where there is unequal bargaining power (a key issue in this context)122
and contracts cannot be enforced against third parties.
4.3.4 Sui Generis Legislation
MacKay argues that sui generis legislation might meet the ‘urgent need to reshape the
conceptual landscape’123. Sui generis legislation could be specifically tailored to fit the
needs of Indigenous artists and their communities, for example, providing more scope
for non-Western concepts such as communal ownership; reflecting the unimportance of
originality or novelty; and taking into consideration the intergenerational nature of
Indigenous cultural property. It might also extend to a broader range of subject matter
not protected by conventional intellectual property law, providing a more holistic
approach to protecting Indigenous cultural expressions and traditional knowledge 124 .
Sui generis legislation could take into account the role of customary law in Indigenous
communities. Any such legislation should be drafted with a view, not only to preserving
Indigenous culture, but also to ensuring and promoting its continuation. A sui generis
system might have the additional benefit of being empowering and promoting self-
determination for Indigenous people. Caution must be expressed on this point, however,
as this is likely to depend on the degree of Indigenous involvement in the drafting and
implementation of such a scheme.
122
MacKay (n73) 21. This is supported by the Arts Law Centre of Australia’s experiences (p65) and
Yumbulul (p67).
123
MacKay (n73) 10.
124
T Lopez Romero, ‘Sui Generis Systems for the Protection of Traditional Knowledge’ (2005) 6
International Law: Revista Colombiana de Derecho Internacional 301-339.
133
Yet, a sui generis solution is not wholly uncontroversial. Although few would
dispute that Indigenous artists need and deserve special protection. Some have argued
that sui generis regulation might merely serve to further marginalise Indigenous people
and their cultural interests 125 . Such a ‘top down’ approach risks being seen to lack
legitimacy and, thus, may be ineffective in practice126. This risk could be addressed by
ensuring extensive and broad consultation with Indigenous artists and their
communities. In addition, any such system ought to be complementary to the current
legal system, not constructed in opposition to it127. Unfortunately, there has been very
little investigation of the potential details and scope of a sui generis approach to date128.
There is a tension between the function of the authorship of artistic works within
Indigenous communities and the conception of authorship that is protected by copyright
law. For Indigenous communities the process of creating art is essential to the
preservation and continuation of their culture; it is an act that establishes and reinforces
the identity of the author within the social structure of the community. The author is a
custodian of the ritual knowledge which he or she is allowed to depict and is held
responsible to the community for the appropriate and respectful use of that knowledge
125
Bowrey (n93) 88.
126
On the methodological issue of speaking ‘for’ Indigenous people see: K Bowrey and J Anderson, ‘The
Politics of Global Information Sharing: Whose Cultural Agendas Are Being Advanced?’ (2009) 18
Social and Legal Studies 479.
127
MacKay (n73) 26 citing also M Dodson and O Barr, ‘Breaking the Deadlock: Developing and
Indigenous Response to Protecting Indigenous Traditional Knowledge’ (2007) 11(2) Australian
Indigenous LR 19, 23.
128
This is surprising given that such legislation has been successfully adopted elsewhere, for example:
Pacific Model Law (Model for the Protection of Traditional Knowledge and Expressions of Culture
2002).
134
both in the artwork itself, and in relation to subsequent uses of the artwork by others.
The ritual knowledge is seen as owned by the community as a whole, entrusted to
certain elders and initiated artists for safekeeping, preservation and continuation. There
is a mismatch between the ‘identifying’ function of authorship as an indicator of the
origin of the work in a specific sense linked to its ‘material form’ (a Western notion
reflected in copyright law); and the ‘identity’ function of authorship as an indicator of
status within, and belonging to, a particular community (an Indigenous notion). In
Indigenous communities, authorship has a strong ‘responsibility’ value which
corresponds to the idea of cultural custodianship. The cases demonstrate some capacity
to take this value into account within the Australian legal system, for example, in the
damages award for culturally-based harm in Milpurrurru and in the fiduciary duty in
Bulun Bulun. Yet, courts are unwilling to recognise the flipside of Indigenous authorial
responsibility: the controlling influence of community norms (as embodied in
customary law) over the creative process. Courts are unwilling to apply these norms
especially when this involves claims against those outside of the relevant Indigenous
community, as might be inferred from the extremely narrow scope of the fiduciary duty
in Bulun Bulun.
Bulun Bulun was a lost opportunity to formulate the values which copyright law
protects in cultural terms129. A more thorough application of the test for joint authorship
might have revealed a number of areas in which Indigenous customary law could
legitimately have been be taken into account as part of the relevant factual matrix.
Justice von Doussa’s judgment is cast in an aesthetically neutral light, appearing to
apply copyright law’s rules of subsistence in a way that might seem free of value
129
Bowrey (n84) 82.
135
judgments external to those legal principles. Yet, when employing vague concepts like
the idea/expression dichotomy, Justice von Doussa seems to be influenced by cultural
pre-conceptions about who can be called an ‘author’. The lack of analytical clarity
about the way in which the joint authorship test ought to be applied makes it vulnerable
to cultural pre-conceptions, which are particularly likely to affect the assessment of
whether a putative author has made a significant contribution of the right kind. Judicial
insistence on aesthetic neutrality in decision-making hides the inherently political
dimension of copyright law cases in which judges must decide which subjects are
‘authors’ and what counts as an ‘artistic work’.
A critical analysis of Bulun Bulun reveals the perception of Indigenous communal
authorship as something that is fundamentally different from, and even potentially
threatening to, copyright’s own conceptions of authorship and joint authorship 130 .
Indeed, Indigenous cultural expressions are different from traditional ideas of
intellectual property in many ways: they are seen as the object of communal rather than
individual rights; although they might evolve over time, they are generally not
concerned with originality or novelty. Some argue that it does not make sense to
conceive of Indigenous interests in their cultural expressions in terms of ‘property’131.
From this angle Indigenous people appear to be caught in a difficult situation: in trying
to fit their cultural property into Western categories they may gain partial protection,
however, in trying to conform to dominant notions of ownership Indigenous people
may be displacing their own cultural identity132. This argument might be countered by
130
Hence, Justice von Doussa’s resort to principles of equity in Bulun Bulun (n2).
131
Diver (n16) citing Bulun Bulun (n2) and Yumbulul (n34) as authorities for this proposition.
132
ibid.
136
asserting the empowering potential of ‘authorship’ from the point of view of democratic
participation and cultural self-determination. Whether or not Indigenous cultural
expressions are best protected by intellectual property law is a complex question, which
cannot be adequately addressed within the context of this thesis. It might suffice to say
that copyright law alone is not best-equipped to deal with this issue, although it might
do more to protect Indigenous art than is apparent from the court’s decision in Bulun
Bulun. Ultimately the best solution is likely to be a comprehensive sui generis
approach, which deals with all the interests that Indigenous people seek to protect and
which recognises differences within and between Indigenous communities133.
This case study shines a light on the concepts of authorship and joint authorship
in copyright law, thus providing a clearer view of the limitations and glimpses of the
possibilities of these concepts. Whether or not one accepts the influence of the romantic
author construct upon copyright law, it does appear clear that copyright law envisages a
model of creativity which relies upon individuals or at the most very small groups of
creators. Larger-scale creativity is only explicitly dealt with in the context of
employment (where the employer presumptively owns copyright) or of a collective
work (where those who select and arrange contributions are first owners of copyright in
the whole work, and those who contribute distinct parts are the first owners of copyright
in those parts). This may have contributed to Justice von Doussa’s reluctance to admit
the possibility of ‘communal’ authorship in Bulun Bulun. A commitment to aesthetic
neutrality has meant that judges tend to elide many considerations that relate to the
context of creativity – focusing upon the skill, labour or judgment that go into the act of
fixation rather than the skill, labour or judgment which go into the ‘expression’ of the
133
There may still be 200 Indigenous legal systems in operation in Australia: Dodson and Barr (n127) 25.
137
work in a broader sense. This Chapter has tried to show that this apparently
aesthetically neutral approach might mask the influence of conceptions of authorship
which reflect economic, historical, social and cultural hierarchies of power and
influence.
Courts have been willing to take Indigenous customary law into account in the
assessment of damages and in the procedures adopted during hearings. Indigenous
customary law provides a significant untapped resource that might be used to help
answer questions of fact in the application of the joint authorship test. I have tried to
show how, despite the ruling in Bulun Bulun, the joint authorship test might be applied
in a more contextual manner allowing for the recognition of important contributions to
a work which are directly reflected in its expression, although they might have involved
an intellectual or creative, rather than a mechanical or physical, contribution to the
material form. In doing so, I would like to suggest the possibility of re-imagining the
concepts of authorship and joint authorship in copyright in a more contextual way that
might be better adapted to the dynamics of creativity that occur across the wide range of
creative collaborations which exist – whether they rely upon new technology or
customary practices which are ancient in origin. In the next Chapter, I explore this
argument further considering a very different type of collective authorship: scientific
collaborations.
138
Chapter 5: Scientific Collaborations
Collaboration is becoming an increasingly important, and in some areas an
indispensable, feature of modern science. As a result, lists of authors on scientific
journal articles have tended to grow longer and longer. At the same time, practices of
authorial attribution on journal articles have increasingly been the subject of criticism
for the ‘inflation’ or ‘dilution’ of two traditional values of authorship in science: credit
and responsibility. This Chapter focuses on two disciplines in which there is a
particularly strong imperative for scientists to collaborate: biomedical science and
particle physics. In these fields it has become virtually impossible to make significant
advances without the cooperation of a variety of specialists and technicians. These
disciplines provide an interesting point of comparison as each has attempted different
solutions to the problems of allocating authorship in respect of collaborative work. The
benefits and limitations of these solutions might provide some insights for copyright
law.
The Chapter begins with an introduction to the context of scientific authorship1. I
consider the important functions of authorship within science’s reputational economy
(as a locus for credit and responsibility) and how these functions are challenged by the
attribution practices of large collaborations. The second section of this Chapter contrasts
two different scientific communities’ approaches to the problem of regulating
authorship. Particle physics collaborations’ inclusive approach to authorship seems to
1
This Chapter considers authorship by reference to common practices of authorial attribution on scientific
journal articles; for other views of the ‘author’ in science, see: H Rheinberger, ‘“Discourses of
Circumstance”: A Note on the Author in Science’ in M Biagioli and P Galison (eds) Scientific
Authorship: Credit and Intellectual Property in Science (Routledge 2003) 309, 311.
139
have been more successful than biomedical journals’ attempts to isolate ‘dominant’
authorial contributions. The third section briefly considers the subsistence of copyright
in journal articles which report the research of large scientific collaborations and
comments on the little case law on scientific authorship and copyright. The case law
suggests both the value, and the potential dangers, of incorporating social norms in
copyright decisions. The final section suggests insights that this case study offers for
copyright law, arguing that the special characteristics of large scientific collaborations
provide support for an inclusive, contextual approach to the application of the joint
authorship test.
5.1 Authorship Dynamics: Constructing Authority
This section considers the dynamics of authorship in science. I begin by considering
the norms that govern science and the role of authorship in science’s reputation
economy as a locus for credit and responsibility. Authorship has an important role in
the process of establishing the authority of a scientific claim. The traditional notion
of authorship in science was very individualistic, which has meant that the
contributions of technicians and other less powerful players were often elided. I
consider how the collectivisation of science has disrupted this traditional notion of
scientific authorship.
Historians and sociologists of science have done much work to clarify the pre-
vailing concept of authorship in the sciences, examining how it has transformed in
modern times2. Although attribution practices vary between disciplines3, the scientific
2
M Biagioli and P Galison, Scientific Authorship: Credit and Intellectual Property in Science (n1)
provides a set of interdisciplinary perspectives on this question.
140
community shares some common assumptions about the essential features of scientific
authorship. Science aims to produce knowledge about nature and to be useful 4 .
Scientific authorship is best understood as a step in an epistemological process that
establishes the authority or truthfulness of a scientific claim and, by extension, its
author5. This authority is established by adherence to conventions developed by the
scientific community which establish the basis for trusting the claim. Robert Merton6
famously identified the norms which embody the normal expectations of the scientific
community as:
(i) communalism (discoveries should be freely shared with the scientific
community) 7;
(ii) universalism (impersonal and objective criteria should be used to evaluate
claims);
(iii) disinterestedness (scientists should subordinate their own biases and interests to
the advancement of knowledge);
(iv) organized scepticism (a new claim should not be accepted into the canon of
scientific knowledge until it has been subjected to systematic scrutiny and
validation); and
3
M Biagioli, ‘Rights or Rewards? Changing Frameworks of Scientific Authorship’ in Biagioli and
Galison (n1) 260, 274.
4
J Ziman, Of One Mind: The Collectivisation of Science (American Institute of Physics Press 1995) 239.
5
J Ziman, Science in Civil Society (Imprint Academic 2007) 7; E Garfield, ‘Giving Credit Only Where it
is Due: The Problem of Defining Authorship’ (1995) 9(19) The Scientist 13.
6
RK Merton, The Sociology of Science: Theoretical and Empirical Investigations (Chicago UP 1973)
310. These norms are now widely accepted: D Burk, ‘Research Misconduct: Deviance, Due Process, and
the Disestablishment of Science’ (1995) 3 George Mason U Independent LRev 305, 310.
7
The original term used by Merton was ‘communism’, however, he later adopted the term
‘communalism’, which is now commonly used by modern sociologists of science: eg, J Ziman, Real
Science: What it is, and What it Means (CUP 2000) 33. See also K Strandburg, ‘User Innovator
Community Norms: At the Boundary Between Academic and Industry Research’ (2009) 77 Fordham
LRev 101
141
(v) originality (scientists should investigate novel and unanswered questions)8.
These norms might be said to embody the normal expectations of the scientific
community and are often collectively referred to as CUDOS, an allusion to their role in
establishing the criteria for recognition, which is valuable within science’s
reputational/sharing economy9. Works of scientific authorship demonstrate compliance
with these norms and with the scientific method10. This helps establish their claims as
non-fictional and hence authoritative. Both the presentation and evaluation of scientific
claims depend upon trust11.
In the eyes of the scientific community, an author is not viewed as the originator of
a particular claim about nature, but rather, the person who uncovered it – that is, the
person who first hypothesised, tested, identified and verified the claim. The element of
originality or novelty valued by the scientific community relates to the selection of the
problem and/or the methodological approach taken in investigating it. In contrast to
copyright law, science tends to value creativity or effort expended in relation to the ideas
in a work, rather than in relation to their expression. Despite this, considerable attention
is paid to the presentation of scientific claims. Scientific writing is usually cast in
8
Merton’s original list did not include originality. Originality may be less important than the others given
that it is also seen as good practice to retest a claim. The meaning of this term in the context of science is
explained at p89.
9
The acronym alludes to the importance of adhering to these norms for retaining one’s reputation and
attaining the recognition of one’s peers within the scientific community.
10
This term is surprisingly slippery: the exact content of the ‘scientific method’ is not uncontroversial
and some even claim that it does not exist. S Shapin, ‘How to be Antiscientific’ in JA Labinger and H
Collins (eds) The One Culture? A Conversation about Science (Chicago UP 2001) 99.
11
S Shapin, A Social History of Truth: Civility and Science in Seventeenth-Century England (Chicago UP,
1994); J Pila, ‘Authorship and E-science: Balancing Epistemological Trust and Skepticism in the Digital
Age’ (2009) 23 Social Epistemology 1.
142
emotionally-neutral authoritative tones and phrased in the passive voice 12 . The best
examples of scientific writing are said to be impersonal, brief, clear, precise and direct13.
This is because a scientific claim has the greatest chance of being accepted by the
scientific community when it is framed as an objective statement about nature and not as
the scientist’s opinion14.
Sociologists and historians of science locate the primary motivation for au-
thorship in science in the benefits that it brings within the context of the reputation or
gift economy of academic science 15 . This gift economy is a system of exchange
premised on reciprocity, reputation and responsibility16. Publication is beneficial for
scientists in two ways: (i) it is a marker of achievement and thus often linked to funding,
promotion and professional opportunities; and (ii) it allows a scientist to build a
reputation which is beneficial per se and may also lead to tangible benefits17. The flip
side of the credit value of authorship within the scientific community is responsibility.
Responsibility centres on good faith and scientific rigor reflecting the crucial role that
12
C Bazerman, ‘Emerging Perspectives on the many dimensions of scientific discourse’ in JS Martin and
R Veel (eds) Reading Science: Critical Functional Perspectives on Discourses of Science (Routledge
1988) 15.
13
I Valiela, Doing Science: Design, Analysis, and Communication of Scientific Research (OUP 2001)
108, 127.
14
Ziman (n4) 188.
15
Biagioli (n3) and C McSherry, ‘An Uncommon Controversy’ in C McSherry, Who Owns Academic
Work: Battling for Control of Intellectual Property (Harvard UP 2001) 68.
16
C McSherry, ‘Uncommon Controversies: Legal Mediations of Gift and Market Models of Authorship’
in Biagioli and Galison (n1) 225, 225.
17
Biagioli (n3) 266 claims that scientific authorship is more about ‘reward’ than it is about ‘rights’.
143
trust plays within the scientific community18. Therefore, the author is held responsible
for adhering to the requirements of the scientific method and other ethical obligations19.
This gift economy is constructed in direct opposition to the market economy20.
So, for example, to be credible, scientific authors must be seen as not primarily
motivated by monetary concerns21. Information which might need to be kept secret if it
is to have value in the commercial market, might only be valuable in the reputation
market if it is made public22. By situating scientific authorship outside the realm of the
marketplace, an air of neutrality and, thus credibility, is established23. Yet, there is a
tension between market forces and the reputational economy because monetary
concerns inevitably affect scientists who must earn a living, whose studies require
funding and whose inventions might be economically valuable 24 . McSherry sees
authorship as the primary site for this tension between the gift economy and the market
economy models in science25.
18
Valiela (n13) 274; Shapin (n11) Ch1.
19
This responsibility includes not exaggerating or falsifying results; not negligently failing to control the
appropriate variables; and avoiding potential conflicts of issues or other ethical violations. In this sense
responsibility is more about ‘playing by the rules’ than it is about accountability for the truthfulness of a
particular claim about nature.
20
McSherry (n15).
21
Indeed, conflicts of interest are seen as unethical.
22
Ziman (n4) 333.
23
McSherry (n16) 227 citing M Walshok, Knowledge Without Boundaries: What America’s Research
Universities Can Do for the Economy, the Workplace, and the Community (Jossey-Bass 1995) 191.
24
See R Eisenberg, ‘Proprietary Rights and the Norms of Science in Biotechnology Research’ (1987) 97
Yale LJ 177.
25
McSherry (n16) 245 noting that ‘one of the things that helps authorship mediate gift and market
economies is the shared assumption that the two models do not share important assumptions’.
144
The traditional narrative of science is very individualistic26. The history of science
is filled with stories of extraordinary individuals who resemble the romantic author27. In
the early years of science, scientific knowledge seemed to grow by accretion as a result
of the activities of these observers and thinkers. Yet, although an individual generally
stood behind a scientific claim28, science has always been a collective undertaking in
some sense, developing cumulatively as a result of interactive processes such as peer
review29 . Indeed, the idea of the scientist as a lonely seeker after the truth is often
historically inaccurate. Many such scientists relied on artisans, craftsmen and laboratory
technicians30. The extensive contributions of such helpers were often dismissed as mere
routine labour and rendered virtually invisible on surviving documentation of their
work31. One explanation for this is the tendency for scientific authorship to gravitate to
persons with institutional authority. Thus, Shapin argues that Robert Boyle was the
author of his papers, despite the considerable input of technicians such as Denis Papin
who ran experiments and even wrote up results, because Boyle had authority over the
investigations in the sense of taking responsibility for them. He had control over the
26
Ziman (n4) at 133.
27
For example: S Shapin and S Schaffer, Leviathan and The Air-Pump: Hobbes, Boyle, and The
Experimental Life (Princeton UP 1985).
28
Shapin (n11): these individuals were usually ‘gentlemen’ thought to be bound by a code of personal
honour, which, along with their financial independence, was seen to guarantee the trustworthiness of their
claims.
29
Ziman (n4) 98; NS Steinberg, ‘Regulation of Scientific Misconduct in Federally Funded Research’
(2000) 10 South California Interdisciplinary LJ 39, 43; A Warwick, ‘“A very hard nut to crack” or
Making Sense of Maxwell’s Treatise on Electricity and Magnetism in Mid-Victorian Cambridge’ in
Biagioli and Galison (n1) at 133 offers the example of a technically difficult textbook, which, by the
inclusion of many Cambridge mathematicians’ attempts to understand and clarify it for their students in
the second edition, became an increasingly interactive, collective product.
30
MW Jackson, ‘Can Artisans Be Scientific Authors? The Unique Case of Fraunhofer’s Artisanal
Optics and the German Republic of Letters’ in Biagioli and Galison (n1) 113.
31
S Shapin, ‘The Invisible Technician’ (1989) 77 American Scientist 554.
145
workplace and work agenda, as well as the final decision on the content and form of
the work32.
As scientific research is increasingly done by large scale collaborations, science’s
traditional individualistic narrative is being openly challenged33. In modern scientific
collaborations it is difficult to identify a Boyle-like figure with overall authority or
control over research projects, and the work of many expert contributors is less easily
rendered invisible. This has had consequences for science’s concept of authorship and its
role in establishing the authority of scientific claims.
Over the last decades, there has been a rapid shift in focus for science from
autonomous basic research towards more directed commercial research. Ziman as-
sociates this with the ‘collectivization’ of science34. Others have called this the rise of
‘Big Science’35. According to Ziman now ‘scientists have to work in teams, hunt in
packs and share the use of many expensive instruments’36. Large collaborative projects
such as NASA, the CERN collaborations, or the Human Genome Project are becoming
an increasingly important and sometimes indispensable part of modern science. These
collaborations often require extensive contributions from scientists in other disciplines
32
ibid 560.
33
W Vesterman, ‘The Death of the Scientific Author: Multiple Authorship in Scientific Papers’ (2002) 8
Common Knowledge 439; Ziman (n4) 241-2.
34
Ziman (n4) 242 argues that science is being collectivized in two senses: (1) scientists are increasingly
being forced to work together to achieve a ‘collective effect’; (2) science is increasingly under the control
of collectives such as government agencies and large industrial corporations.
35
This is a term that has been attributed to AM Weinburg in ‘Impact of Large-Scale Science on the United
States’ (1961) 134(3473) Science 161. It is usually used ‘...to describe a series of changes in science
which occurred in industrial nations during and after World War II, as scientific progress increasingly
came to rely on large-scale projects usually funded by national governments or groups of governments’:
<http://en.wikipedia.org/wiki/Big_Science>.
36
Ziman (n4) 241.
146
and a variety of supporting personnel (eg, technicians, statisticians, managers, software
developers). The need to acknowledge and reward a diverse range of contributors to
such projects has resulted in lists of authors on journal articles growing longer 37 ,
resembling the long lists found in film credits38. This fragmentation of authorship has
disrupted the reputational economy of science39.
Currently, being listed as an author may imply any number of different con-
nections to a scientific article. A person might be designated as an author because of a
contribution at any point in the process of creation, such as designing a study, carrying
out experiments, providing funding, supervising the project, designing software or
other equipment used, collecting or analysing data, or providing technical assistance.
Interpersonal relationships affect decisions about the attribution of authorship, with
senior scientists more influential than their junior colleagues in deciding authorship
issues40. Controversially, sometimes authorship may be attributed to someone with a
tenuous connection to the project, for example, a senior colleague whose name is
added to increase chances of publication41. A survey of the world’s 20 most prolific
scientists indicated an average of at least 3.9 days between each of their publications
from 1981 to 199042. It seems obvious from these figures that these scientists could not
37
For example: ‘The ATLAS Experiment at the CERN Large Hadron Collider’ published in the Journal
of Instrumentation listed 2926 authors from 169 research institutions. P Davis, ‘An Authorship
Accelerator’ dated 8 December 2008 <http://scholarlykitchen.sspnet.org/2008/12/08/hep-authorship>.
38
T Conley ‘End Credits’, Biagioli and Galison (n1) 360.
39
Biagioli (n3).
40
AG Mainous III, MA Bowen, JS Zoller, ‘The Importance of Interpersonal Relationship Factors in
Decisions Regarding Authorship’ (2002) 34 Family Medicine 462, 467.
41
This contribution might be characterized as putting one’s reputation on the line.
42
Valiela (n13) at 136.
147
have contributed significantly to all the papers on which their names appear43. This has
implications for the credit and responsibility values attached to authorship in science.
Authorship norms vary depending upon the scientific community to which the
author belongs. They are inevitably affected by structural dynamics and power
relations in particular scientific communities44. Journals often have specific guidelines
on attribution practices45. Scientific readers make inferences about a particular author’s
contribution from their position in the list of authors. In many disciplines it is common
to list the person who is responsible for most of the writing and is most directly
involved in the project first, and the supervisor or person who obtained the funding for
the study last. These are typically the two most prestigious positions. The attribution
of authorship and the order of authors vary from discipline to discipline, sometimes
from institution to institution, and even from project to project. The answer to the
question ‘who is the author?’ in respect of a scientific journal article will depend to
some extent upon who is asking. Whilst journal editors are keen to define a locus of
responsibility for published work, researchers may have more pragmatic concerns just
to ‘get the job done’, using the reward of authorship as a cost-free way of getting busy
people to participate46. The importance of the order in which authors are listed is a
notion that is alien to copyright and even to the attribution right. Indeed, copyright law
tends to weigh the contributions of authors equally when it comes to the distribution of
43
It appears to be common for scientists to be listed as authors on papers that they have not even read: J
Birnholtz, ‘When Authorship Isn’t Enough: Lessons from CERN on the Implications of Formal and
Informal Credit Attribution Mechanisms in Collaborative Research’ (2008) 11(1) J of Electronic
Publishing <http://dx.doi.org/10.3998/3336451.0011.105>.
44
T Scott, ‘Changing Authorship System Might Be Counterproductive’ (1997) 315 British Medical J 744.
45
Rheinberger (n1) 310.
46
R Phillips, ‘Researchers’ Objective is to Get the Job Done’ (1997) 315 British Medical J 747.
148
47
royalties amongst joint copyright owners , except in cases where factual
circumstances rebut this inference48.
The correlation between authorship, professional recognition, and financial
benefit is not easily measured. Yet, it is clearly an issue of great practical
significance49. As author lists grow longer, it becomes increasingly difficult to assess
the contributions of each author 50 . This has led to a concern that credit is being
impermissibly inflated and responsibility becoming impermissibly diffuse. Rennie,
Yank and Emanuel conceptualise scientific authorship as a coin with a ‘credit’ value
on one side and a ‘responsibility’ value on the other51.They argue that as the number of
authors increases, in practice, the amount of responsibility for each author seems to
diminish faster than the credit, separating the two sides of the coin from each other52.
Intuitively it seems fair that if you are to share the reputational benefits of scientific
authorship, then you must also accept some of the responsibilities. This idea unravels,
however, in the context of many large scale collaborative projects where there may be
47
Godfrey v Lees [1995] EMLR 307, 329.
48
For example: Bamgboye v Reed [2002] EWHC 2922 [42]; Fisher v Brooker [2009] UKHL 41.
49
For this reason alphabetical order has been criticized as resulting in winners and losers on arbitrary
grounds. Ziman (n4) 372; Valiela (n13) 135 cites a similar solution: ‘Order of authorship was decided
from the outcome of a game of croquet played on the grounds of CCCC College’.
50
Davis (n37).
51
D Rennie, V Yank and L Emanuel, ‘When Authorship Fails: A Proposal to Make Contributors
Accountable’ (1997) 278 JAMA 579, 580.
52
Biagioli (n3) 263-4 agrees that concerns relating to the inflation of credit may be less well founded
than concerns relating to responsibility. M Biagioli, ‘Documents of Documents: Scientists’ Names and
Scientific Claims’ in Annelise Riles (ed), Documents: Artifacts of Modern Knowledge, (U of Michigan
Press 2006) 129, 149. Cf S Seymore, ‘How Does My Work Become Our Work? Dilution of Authorship
in Scientific Papers, and the Need for the Academy to Obey Copyright Law’ (2006) 12 Richmond J of L
& Tech 11; Vesterman (n33).
149
good reasons to give credit to valuable contributors who could not fairly be asked to
bear responsibility for the article as a whole.
5.2 Regulating Authorship with Private Ordering
Mario Biagioli has explored the different ways in which the attribution practices of
biomedical science and particle physics collaborations have challenged the
conventional notion of scientific authorship53. The diffuse nature of responsibility in
large collaborations has been a cause for concern leading to arguments for stricter
control of authorial attribution in the biomedical sciences. Particle physics
collaborations, on the contrary, seeking to credit a range of valuable contributors, have
expanded the notion of authorship. This section revisits Biagioli’s comparison with a
view to determining what insights these experiences might offer copyright law.
5.2.1 Biomedical Science Collaborations: An Authorship Crisis
The challenges of allocating authorship in large scientific collaborations have been felt
particularly acutely in the biomedical sciences, where there is perceived to be a crisis
in authorship. This is due, in part, to the disruption that long author lists cause to the
authority or responsibility value of authorship. This section argues that the strategy
adopted by many biomedical journals of restricting authorship claims has had only
limited success because it is not adapted to the realities of large collaboration.
Although aimed at identifying those who can be held responsible for an article, this
approach does not seem to have prevented authorship gravitating to those with the
53
Biagioli (n3).
150
most institutional authority. The section ends by considering Sean Seymore’s argument
that scientific attribution practices ought to be realigned with copyright law norms.
Intellectual limitations caused by necessary specialisation make it virtually
impossible for one person to have enough knowledge across specialisations to make
significant advances in modern biomedical science 54 . As author lists on articles
produced by large collaborations have grown longer, attribution practices have come to
be associated with a number of high profile incidents of scientific misconduct, which
in turn, lead to fears of a crisis in authorship 55. Scientists and journal editors have
become concerned that in large collaborations the credit value of authorship might
become inflated and the responsibility value might become impermissibly diffuse –
disrupting science’s reputation economy and undermining the trust on which science
depends. Some authorship practices regularly seen to be inappropriate include:
(i) Gift or guest authorship: adding the name of a person who has contributed little
to the publication. This might be to acknowledge earlier influence on the primary
author’s career; to make the paper appear more authoritative; or to increase its
chances of publication. It may even be done to acknowledge someone who has
provided access to facilities, data or funding. In some cases the gift author may
not even be aware of being listed as an author56.
54
RC Dreyfuss, ‘Collaborative Research: Conflicts on Authorship, Ownership and Accountability’
(2000) 53 Vanderbilt LRev 1161, 1171.
55
R Epstein, ‘Academic Fraud Today: Its Social Causes and Institutional Responses’ (2010) 21 Stanford
L & Policy Rev 135; Steinberg (n29).
56
R Bhopal et al, ‘The Vexed Question of Authorship: Views of Researchers in a British Medical
Faculty’ (1997) 314 (7086) British Medical J 1009: A study of 66 staff of a University medical faculty
revealed that although gift authorship was perceived as unethical it was seen as common practice and,
indeed, seemed to be encouraged by academic reward systems. 64 per cent of respondents had exper-
151
(ii) Coercion authorship: where superiors who have had little or no direct in-
volvement in the research ask to be listed as an author – the importance of
maintaining good working relationships may make it difficult to refuse such a
request57.
(iii) Mutual support authorship: where authors agree to put each other’s names on
papers where no direct contribution is made to ‘pad’ their curriculum vitae.
(iv) Ghost authorship: where someone who has made a significant contribution to the
work is not acknowledged as an author. This may be to make the report appear
independent from that person or their organisation 58 , or may involve junior
researchers whose contributions are not acknowledged by their superiors 59 .
Scientific authorship here is under-inclusive in copyright law terms.
Ghost and guest authorship are particularly problematic, since both have been
frequently associated with cases of scientific misconduct and even fraud 60. Fraudulent
authors have often sought prestigious co-authors to lend credibility to their papers. In
the early 1980s, John Darsee infamously falsified a number of studies at Emory and
ienced difficulties with authorship, the most common difficulty was being excluded from authorship
when it was apparently merited.
57
Providing access to equipment, providing materials or being a head of department may be contributions
relied upon by people making such claims. L Claxton, ‘Scientific Authorship Part 2: History, Recurring
Issues, Practices, and Guidelines’ (2005) 589 Mutation Research 31, 35.
58
S Stern and T Lemmens, ‘Legal Remedies for Medical Ghostwriting: Imposing Fraud Liability on
Guest Authors of Ghostwritten Articles’ (2011) 8(8) PLoS Med. e1001070; A Matheson, ‘How Industry
Uses the ICMJW Guidelines to Manipulate Authorship – And How They Should Be Revised’ (2011) 8(8)
PLoS Med 1001072.
59
Claxton (n57) 36.
60
Stern and Lemmens (n58); AJ Fugh-Berman, ‘The Haunting of Medical Journals: How Ghostwriting
Sold “HRT”’ (2010) 7(9) PLoS Med e1000335: ‘the pharmaceutical company Wyeth used ghostwritten
articles to mitigate the perceived risks of breast cancer associated with HT, to defend the unsupported
cardiovascular “benefits” of HT, and to promote off-label, unproven uses of HT’; L McHenry, ‘Of
Sophists and Spin-Doctors: Industry-Sponsored Ghostwriting and the Crisis of Academic Medicine’
(2010) 8(1) Journalology 129.
152
Harvard Universities. Many of the papers concerned included prominent heads of
department as co-authors. These people had not fabricated the data, but were criticized
for allowing their names to appear on work about which they knew too little61. The
controversy revealed divergent attitudes to the social incidents of authorship, namely,
whether the heads of department had acted appropriately and what sanctions, if any,
ought to be imposed on them.
The pharmaceutical industry, in particular, has been criticized for using senior
scientists to add credibility to studies and to mask the commercial function of journal
articles which may be used to help advertise particular products 62. The use of a well-
known scientist as a first author to give the impression that the scientist instigated or
controlled the study and endorses the results is apparently widespread in biomedical
publications63. Even where the content of the study is not suspect, this practice poses a
serious problem for scientists who need to know the true origin of an article in order to
appropriately judge the information in it64. The trust upon which science depends may
accordingly be undermined.
A number of groups have created guidelines to address this crisis in
authorship 65 . The International Committee of Medical Journal Editors (‘ICMJE’)
61
J Smith, ‘Gift Authorship: A Poisoned Chalice?’ (1994) 309 British Medical J 1456.
62
Matheson (n58); S Sismondo, ‘Ghosts in the Machine: Publication Planning in the Medical Sciences’
(2009) 39 Social Studies of Science 171.
63
Matheson, ibid.
64
ibid.
65
Claxton (n57) 38-43. These include: the UK Organization Committee on Publication Ethics (COPE),
the National Institutes of Health, University of Pennsylvania, University of California at San Francisco,
the American Chemical Society, the American Statistical Society, and the Danish Committee on
Scientific Dishonesty. See also: Rennie, Yank and Emanuel (n51) 584.
153
guidelines are the most accepted66. They are quite prescriptive in requiring that each
name listed in the by-line refers to a person who takes full responsibility for the
contents of the entire article and who has made substantial contributions to (i) the
conception and design, or analysis and interpretation of data; (ii) drafting the article or
revising it critically for important intellectual content; and (iii) who has also given
final approval of the version to be published (the ‘triple lock formula’) 67 . This
approach is designed to reduce the number of contributors who might be named as
authors. It attempts to single out particular types of contributions that, in combination,
count as authorial because they ensure that authors are adequately involved in the
instigation, control, and writing up of a study so as to be properly held responsible for
the article68. These guidelines aim to ensure that authorship is granted only to those
who are best placed to ensure compliance with science’s norms and assume authority
for the article’s claims.
To date, the ICMJE guidelines appear to have had only limited success69. This
seems to be because they largely ignore the exigencies of large scientific collaborations,
where authorship is seen as a cost-free way of getting busy people to participate or as
quid pro quo for funding or access to equipment70. They are also insensitive to the fact
66
Garfield (n5).
67
‘Uniform requirements for Manuscripts Submitted to Biomedical Journals’ (1997) 277 JAMA 928.
Participation solely in the acquisition of funding or collecting of data, and general supervision of the
group will not be sufficient to establish authorship.
68
In this way the guidelines seem to echo historical practices which seemed to ‘erase’ the contributions
of technicians as discussed in the previous section, see Shapin (n31).
69
PC Gøtzsche et al, ‘Ghost Authorship in Industry-initiated Randomised Trials’ (2007) 4 PLoS Med
e535; E Wager, ‘Authors, Ghosts, Damned Lies, and Statisticians’ (2007) 4(1) PLoS Med e34. A Hudson
Jones, ‘Can Authorship Policies Help Prevent Scientific Misconduct? What Role for Scientific Societies?’
(2003) 9 Science and Engineering Ethics 243.
70
Claxton (n57) 35; R Horton, ‘The Signature of Responsibility’ (1997) 350 Lancet 5-6.
154
that collaborative work often necessitates a division of labour. It might not be possible
for any one person to have sufficient expertise fully to understand and be fairly held
responsible for every part of a complex collaborative study. The exclusion of
statisticians has been a particular cause for concern given how vital their work is for
understanding the significance of, and interpreting, results71.
In a recent article Matheson shows how pharmaceutical industry publication
planning strategies rely upon selective interpretations of the ICMJE guidelines to boost
the credibility of their publications by exaggerating the apparent contributions of some
academic authors and understating the contributions of employed commercial
writers72. This is because the so-called ‘triple-lock’ formula allows an academic to be
named as an author as long as he or she makes some contribution to the design of a
study or data analysis, makes some revisions to the manuscript, and approves it.
Conversely, even if industry representatives do most of the design, data collection,
analysis, and writing, if final approval is given by an academic then those
representatives are disqualified from authorship. Even a person who has actually
written an article is denied authorship if he or she was not involved in any other
aspects of the study.
Part of the challenge in addressing pharmaceutical company publication strate-
gies is that doing so might open a Pandora’s Box, as these practices mirror other
practices which are widely tolerated and in some quarters considered entirely
71
Wager (n69); RA Parker and NA Bergman, ‘Criteria for Authorship for Statisticians in Medical
Papers’ (1998) 17(20) Statistics in Medicine 2289.
72
Matheson (n58).
155
appropriate73. Laboratory directors, departmental chairs and supervisors often ask to be
listed as authors on publications on the basis of their institutional position rather than
any significant contribution to the article 74 . The attribution of authorship is often
affected by power relations: the problem of the senior professor who insists on being
listed as an author on all of the papers of junior colleagues is apparently notorious75.
Indeed, the supervisor/student relationship is at the heart of a number of legal disputes
on employment matters and breaches of fiduciary duties, which frequently also affect
the allocation of credit for scientific work 76. As a practical matter, the most senior
scientist involved in the research project will often have the final say about who is
listed as an author, and in what order, even if that scientist may not have been directly
involved in the actual writing of the report. In stressing the importance of giving final
approval, the ICMJE guidelines tend to reinforce these power dynamics.
Sean Seymore suggests that realigning scientific practices of attributing au-
thorship with (US) copyright law norms might help address this crisis of authorship77.
Seymore locates problems of the dilution or inflation of authorship primarily with the
expanding number of authors listed in the middle of a by-line who may have
contributed the mere labour of carrying out experiments or collecting data or may be
73
Stern and Lemmens (n58).
74
PC Gøtzsche et al, ‘What Should Be Done To Tackle Ghostwriting in the Medical Literature?’ (2009)
6(2) PLoS Med e1000023.
75
Ziman (n4) 372.
76
Chou v University of Chicago 254 F3d 1347, 1361 (Fed Cir 2001); Johnson v Schmitz 119 FSupp 2d
90, 91 (D Conn 2000); M Astala, ‘Comment, Wronged by a Professor? Breach of Fiduciary Duty as a
Remedy in Intellectual Property Infringement Case’ (2003) 3 Houston Bus & Tax LJ 31.
77
Seymore (n52) [27]: claiming that ‘[s]cientific misconduct might subside if science professors were
taught basic intellectual property law’ (referring to US law). Although the UK joint authorship test
operates differently to the US test, much of his argument still translates in the UK context.
156
gift or coercion authors. He approves of copyright’s joint authorship test because he
sees it as linking authorship to creativity (ie, some intellectual contribution above mere
labour) and focusing on contributions which go to the planning, writing and reviewing
of an article78. Here Seymore may overstate the level of intellectual or creative input
that the requirement for ‘creative spark’ implies, as it is generally accepted that fairly
mundane works reach this standard. Seymore also approves of the fact that the joint
authorship test has been applied with the deliberate aim of reducing the number of
authors. In the UK context, I have called this the instrumental/pragmatic approach
because it does not follow naturally from the wording of the statute, but rather, is an
interpretation adopted for pragmatic or instrumental reasons. Seymore implies that the
ICMJE guidelines are a step towards copyright norms. The way in which these
guidelines have been implemented by pharmaceutical companies, however,
demonstrates that they might be both under-inclusive and over-inclusive from a
copyright point of view.
Problems of scientific misconduct are unlikely to be solved by deferring to
copyright norms. Although a reduction in the number of authors listed might make it
easier to find individuals to hold responsible for the article, copyright law will not
always select the most appropriate contributors from science’s point of view: those who
establish the authority of the article and can take responsibility for its claims. Copyright
authorship primarily relates to the origin of the expression of a work rather than ethical
responsibility to the scientific community for its content. The role of scientific
authorship in constructing authority would be undermined if it were granted to those
who contributed the written expression of an article over those who take responsibility
78
ibid [29]-[32].
157
for the process of making the claims of scientific significance in the article. Further, this
approach does not address the reason for the increase in named authors: the
globalisation of science and the growing need to co-operate across large teams. Many of
the contributors whose work is considered to be very valuable within particular
scientific collaborations would be unlikely to be considered to be authors for copyright
purposes79. Seymore’s appeal to copyright norms, however, highlights the potential role
of copyright law as a source of good authorship standards. In particular, copyright law
might provide a benchmark minimum standard for determining authorship based upon
responsibility for the creation of the expression of a work. This will be considered in
further detail in Chapters 7 and 880.
5.2.2 Particle Physics Collaborations: The Bureaucratisation of Authorship
Particle physics collaborations have been far more successful than their biomedical
science counterparts at dealing with attribution issues. Large group collaboration is
virtually unavoidable in the particle/high energy physics community. CERN, for
example, is run by 20 European member states and includes scientists from 608
institutions and universities. Half of the world’s particle physicists come to CERN for
their research81. Particle physics collaborations tend to adopt a corporate approach to
credit, responsibility and the attribution of authorship. They have a very bureaucratic
structure and produce standardised protocols on authorship that prioritise the coherence
79
This might be because their contributions: are not directly reflected in the expression; are not the
product of activities that copyright considers ‘authorial’; or because they are distinct. A Stokes,
‘Authorship, Collaboration and Copyright a View from the UK’ (2002) Entertainment LRev 121.
80
See, particularly, 7.8 and 8.2.
81
<http://public.web.cern.ch/public/en/About/Global-en.html>.
158
of the collaboration over the recognition of the specific contributions of individuals82.
Indeed, due to the heterogeneity of such collaborations, it is often true that when the
collaboration speaks it says something that no one individual could say from first-hand
knowledge83. Responsibility is managed collectively through the use of internal review
processes.
The Collider Detector at Fermilab (‘CDF’) Collaboration is a typical example. It
has a standard alphabetic author list which appears on all publications. Criteria for being
added and removed from the list are determined by guidelines developed by a
committee. Authorship is conceived of in terms of credit for accumulated labour
dedicated to the common endeavour 84 . A person may, therefore, be listed on a
publication whilst on a leave of absence: for, without all the contributors no research
could have been done85. Prior to publication an article must be preliminarily approved
by a publication committee; it is then posted to an internal webpage and all members of
the collaboration can comment on it. After a revised version is posted, those on the
Standard Author List may withdraw their names if still unsatisfied with the article. This
leads to an interesting result: an article with fewer names appears to be less credible
than an article with more names (in contrast to the biomedical sciences where the worry
is that more authors dilute responsibility) 86 . The Stanford Linear Detector (‘SLD’)
collaboration policies on authorship differed depending on the scope of the audience for
82
P Galison, ‘The Collective Author’ in Biagioli & Galison (n1) 329, 332; Ziman (n4).
83
Galison (n82) 352.
84
Biagioli (n3) 270.
85
Birnholtz (n43).
86
Biagioli (n3) 273.
159
the publication and the knowledge claim made within it87. Conference proceedings, for
example, were to be signed ‘The SLD Collaboration, presented by a particular person’
with a footnote naming all of the SLD authors; whereas the authorship of internal
memoranda was a matter for the writers to decide.
Names such as ‘SLD Collaboration’ or ‘CERN’ operate like brand names that
perform a trust function, acting as a placeholder for the contributions of many
individuals 88 . This is reminiscent of historical practices whereby the individual
contributions of scientists to large group projects were gathered together under the name
of ‘sacralised’ individuals standing for groups89. The largest hydrogen bubble chamber
collaboration at Lawrence Berkeley Laboratory in the 1960s, for example, was referred
to as the ‘Alvarez Group’. Luis Alvarez could be seen to be responsible for the work
because ultimately he made all the decisions about the physics results to be published
and all funding for the group passed by him90. Particle physics collaborations take this
model, but use consensus-based decision-making processes to establish the authority of
the group over the claims that it makes. Although not the name of an individual
scientist, ‘CERN’ conveys relevant information about the source of its publications and
who takes responsibility for them. Internal processes for gaining or losing authorship,
producing articles, and reviewing them before publication are crucial to establishing
this corporate identity and the sense that the collaboration has a mind of its own. These
87
Galison (n82) 333-4.
88
On the trade mark function of authorship: J Ginsburg, ‘The Author’s Name as a Trade Mark: A
Perverse Perspective on the Moral Right of “Paternity”?’ (2005) 23 Cardozo Arts & Ent LJ 379; G
Lastowska, ‘The Trade Mark Function of Authorship’ (2005) 85 Boston U LRev 1171.
89
H Gusterson, ‘The Death of the Authors of Death: Prestige and Creativity among Nuclear Weapons
Scientists’ in Biagioli and Galison (n1) 281, 282 gives the example of the atomic bomb which was
credited to Robert Oppenheimer despite the key roles of a number of other scientists.
90
Galison (n82) 329-330.
160
procedures play an important role in establishing the authority of the authors over the
claims (thus, maintaining trust), as long alphabetical author lists convey little
information on responsibility.
Particle physics collaborations can adopt corporate solutions to the problem of
authorship, because there is a strong community culture: scientists are often located in
the same place or share a piece of equipment. Such a solution is unlikely to work in the
biomedical sciences where the membership of collaborations tends to fluctuate over
time according to the requirements of the specific task undertaken. There is likely to be
little cohesion amongst a group of scientists collaborating on a clinical trial who may
never have met, might be located in different places, and might even be competing for
funding. In biomedical science, authorship is part of a struggle for professional
advancement, whereas in the small cohesive particle physics community it is just a fact
of life91. This is not to suggest that attribution of authorship is never controversial.
Rather, for individuals working in particle physics the focus may be more on getting
noticed in informal settings than having a publication record92.
In adopting this model, particle physics communities have experienced not just a
redefinition of the role of authorship within the reputational economy, but also an
evolution in attribution norms towards a more expansive view of what counts as
authorship and a procedural means for collectively establishing responsibility or
authority. Particle physics collaborations have been more successful at regulating
authorship than biomedical journals, because they have been able to develop a
91
Biagioli (n3) 272.
92
Birnholtz (n43).
161
framework of processes which are better adapted to the different way that large
collaborations author articles.
5.3 The Application of Copyright Law
The experiences of biomedical science and particle physics collaborations might offer a
number of insights for copyright law. Before considering these, it might be useful to
look at the application of copyright law’s rules of subsistence to journal articles which
report the research of large scientific collaborations.
A scientific journal article will almost always be considered an original literary
work for the purposes of the CDPA. It is a more complicated task to determine which
contributors to a large collaborative study will be considered to be ‘authors’. It seems
likely, however, that many of those listed in the by-line will not pass the joint
authorship test. As discussed in Chapter 2, it is difficult to predict how the test will
apply to cases of collective authorship. A few general comments, however, might be
made.
First, very trivial contributions might not qualify for joint authorship as they are
not ‘significant’. In addition, many contributors will not have made contributions of
the ‘right kind’. This would tend to exclude preparatory contributions not reflected in
the expression of the work; the contribution of ‘ideas’93; and activities which are not of
an ‘authorship type’94. Thus, planning, controlling and directing experiments; carrying
93
Anya v Wu [2004] EWCA Civ 755.
94
Donoghue v Allied Newspapers [1938] 1 Ch 106, Brighton v Jones [2004] EWHC 1157, [2005] FSR
288, Fylde Microsystems v Key Radio Systems [1998] FSR 449 (Ch).
162
out experiments; securing funding; providing access to samples or equipment;
designing software or machines used in experiments; proof-reading papers, etc. may
not be sufficient. Such contributions might be relevant where accompanied by more
than a trivial contribution to the expression of the work, although there is little
guidance in the case law on this point. The authorship limb of the joint authorship test
is likely to exclude a large number of contributors who are currently named authors on
many scientific publications. Those who have made no contribution to the journal
article will certainly be excluded (eg, guest/gift authors, and contributors who have left
particle physics collaborations but who must still be attributed according to the
relevant protocol). Conversely, joint authorship might be available to those who have
made authorial contributions but who are not listed as authors (ghost authors). For the
purposes of UK copyright law it is not enough to agree to be an author95.
The instrumental/pragmatic approach to the joint authorship test is impossible
to implement in the case of scientific collaborations. There are two types of candidates
for authorship under the instrumental/pragmatic approach. The first is the person(s)
responsible for the actual writing of the article (most likely a junior scientist). This
requires one to exclude many contributors, such as those who designed and supervised
the study; those who provided the content for different parts of the article; and those
who checked it. The second is the person(s) who supervised the study and had most
control over the research process (most likely a senior scientist). Yet, that person is
unlikely to have made much of a direct contribution to the writing. Both choices seem
wrong in principle as they tend to divorce authorship from the creation of the protected
95
Copinger et al [4.01], [4.38] citing Levy v Rutley (1871) LR 6 CP 523, 531 and Wiseman v George
Weidenfeld [1985] FSR 525, 529 (where the point was accepted on all sides). Cf Prior v Lansdowne
Press [1977] RPC 511 (SC of Victoria).
163
expression. The first requires a myopic focus on the process of fixation; and the second
grants authorship to the most powerful player. In addition, this approach requires one
to ignore the many people who have made contributions which have valuably added to,
and shaped, the expression of the work. The failure of biomedical collaborations to
reduce the number of authors of a scientific article by focusing only on dominant
contributors is a lesson in the ineffectiveness of such a restrictive approach.
To be a joint author, the relevant contribution made must not be distinct. This
may exclude authors of diagrams 96 . The requirement for collaboration is not
particularly demanding, as it does not require proof of a specific intention to be a joint
author 97 . Compliance with the social norms that govern the authorial group might
indicate that a contributor is part of a collaborative endeavour, ie, that they share a
common design. Even if a ‘gift’ or ‘guest’ author satisfied the authorship limb of the
test (unlikely), they may fail the joint authorship test at this point, as they might not be
seen to be collaborating with the other joint authors.
I have suggested that a restrictive approach to the joint authorship test may make
it difficult to apply in a way that seems credible in light of the creative realities of a
scientific collaboration. In addition, the presumption of authorship which arises where
one’s name is on the work (s104(2)) is likely to be misleading when applied in cases of
collective authorship in science, as the lists of authors on articles are often both over-
96
Where such a contribution is distinct, it may constitute a work in which copyright subsists
independently of the rest of the article. In Carlos v Javier and Maria Cruz [2002] ECDR 23 (Madrid CA,
12th Section) it was held that originality will not be denied to a diagram merely because the concepts
represented in it are commonly known, or are regularly described in, textbooks.
97
Weissmann v Freeman 684 FSupp 1248 (SDNY, 1988); 868 F2d 1313 (2nd Cir, 1989) provides an
interesting example of how an intention requirement might be interpreted in the context of scientific
authorship.
164
inclusive and under-inclusive from a copyright law point of view 98 . Although the
presumption may go some way towards indirectly incorporating the authorial norms of
particular scientific communities in the determination of copyright disputes, it provides
no room to accommodate the different hierarchies of authorial importance
communicated to a scientific reader by the order in which authors are listed.
There are few cases on the subsistence and ownership of copyright in scientific
journal articles99. This is likely to be due to the importance of authorship in science’s
reputational economy, which is constructed in opposition to the market economy. The
right to royalties and control ensured by copyright, for example, seems to clash with
scientific norms which encourage open dissemination and economic disinterestedness.
The attribution of authorship lies right at the point of tension between market and
reputation economies, as authorship (in addition to implying copyright ownership)
influences the allocation of credit for scientific work100. On occasion, where the stakes
in the reputational economy are high, scientists have turned to courts to resolve
attribution issues101. False attribution of authorship was at the heart of Weissmann v
98
Even though this presumption might be easily displaced where there has been no contribution to the
expression whatsoever, it may tip the evidentiary scales in favour of named authors in unclear cases.
99
I have not been able to locate any cases concerning the application of the joint authorship test to a
scientific journal article with a large number of authors. Disputes which implicate issues of authorship in
science have tended to concern access to data or samples; defamation (Noah v Shuba [1991] FSR 14
(Ch)); misleading claims/fraud; or employment issues (eg, unfair dismissal following an authorship
dispute). See: Dreyfuss (n54). In the US there are some cases which have found a fiduciary duty exists
between students and supervising academics: Chou (n76); Johnson v Schmitz (n76).
100
McSherry (n15) 245.
101
In Noah v Shuba (n99) 33, for example, an epidemiologist initiated legal proceedings in relation to a
magazine article which reproduced a portion of his work adding to it in a misleading manner, which
seemed to suggest that he approved of certain products. In awarding damages for false attribution of
authorship in respect of the added sentences, the court took into account the ‘… author’s interests in his
reputation and integrity of his work’. A perception of unfair attribution contributed to the dispute and
tragic events in: Fabrikant c. Swamy (2011) QCCS 1385 (Superior Court of Quebec, Canada).
165
Freeman, a high-profile copyright dispute involving scientists in the US 102. This case
demonstrates that judicial characterisation of the context of creativity can be pivotal
when it comes to applying the joint authorship test.
Dr Heidi Weissmann had worked for many years under the supervision of Dr
Leonard Freeman. They had authored many articles together and had devised a
syllabus together, which over the years they had both updated and revised
independently. Weissmann independently published an updated altered version of the
syllabus (‘P-1’), which Freeman then reproduced without her permission under his
own name. A key question was whether P-1 was a work of joint authorship they both
owned or an original derivative work solely owned by Weissmann. The trial and
appeals courts took significantly different views of the context, which influenced
their assessment of each scientist’s contributions to P-1.
Judge Pollack at trial construed P-1 as a ‘stock piece’ in evolution and therefore a
work of joint authorship owned by both scientists. He was influenced by the pre-
existing professional relationship between the two scientists. He emphasised
Freeman’s senior position and stressed that Weissmann’s career had developed under
Freeman’s ‘supervision, guidance and control’ 103 . Judge Pollack observed that as
‘principal investigator’ Freeman was frequently responsible for the experiments that
they had both been involved in, even though Weissmann typically did most of the
102
Trial judgment: Weissmann v Freeman 684 FSupp 1248 (SDNY 1988). Appeal: Weissmann v
Freeman, 868 F2d 1313 (2nd Cir 1989). See also: C McSherry (n15) 68.
103
More than half the judgment is dedicated to background and the relationship between the parties.
166
writing up104. Judge Pollack upheld Freeman’s joint authorship claim because of his
role in providing the authority for their joint work – focusing on his contribution to the
ideas of scientific significance evidenced in the syllabus, rather than on the origin of its
expression105. Judge Pollack even accepted Freeman’s view as an expert opinion on
whether the syllabus was relevantly original, dismissing Weissmann’s arguments on
this point as ‘merely echo[ing] legal language found in the reported cases’106.
The Court of Appeals by a 2:1 majority reversed this decision, finding that P-1
was an original work solely authored by Weissmann, and that Freeman had infringed
Weissmann’s copyright by reproducing it. Whilst Judge Pollack had characterised the
dispute as an ‘uncommon controversy’, Judge Cardamone of the Court of Appeals
considered the case to be ‘the paradigm of the problems that arise when a long
relationship between accomplished professor and brilliant assistant comes to an
end’107. Judge Cardamone acknowledged Weissmann’s efforts to break away from her
apprenticeship with Freeman by developing her own approach. In that context, their
previous collaboration was seen as an example of science’s slow evolutionary
development in which innovations are necessarily based on pre-existing, often joint
works 108 . In Judge Cardamone’s assessment, Freeman had not made a relevant
contribution to P-1 and there was no intention for joint-authorship. Further, in
determining that the defence of fair use was not available, Judge Cardamone
104
Weissmann v Freeman (trial, n102) 1253-4: noting Freeman’s name ‘lent authority’ to Weissmann’s
work.
105
ibid 1256.
106
ibid 1257.
107
Weissmann v Freeman (appeal, n102) 1315.
108
ibid 1319.
167
considered that in attributing authorship to himself Freeman stood to gain something
valuable within the scientific community, which is ‘ill-measured in dollars’ but which
might often influence professional advancement 109.
In an attempt to reach a decision that took into account the peculiar context of
scientific authorship, the trial judge was unduly influenced by the power dynamics
which have often given senior scientists control over the attribution of authorship 110.
Judge Cardamone’s reasoning in the appeal decision is preferable because it
incorporates a more nuanced understanding of scientific authorship, whilst remaining
attentive to copyright law principles. For copyright, authorship is a question of law
and not a matter of agreement amongst the parties 111. The heart of copyright’s concept
of authorship is the idea that authorship is about more than who ‘signs’ a work – it
relates to the origin of the expression 112 . In this way copyright might provide a
bulwark against power imbalances within authorial communities, a particular issue in
science where authorship is associated with authority and tends to gravitate upwards.
This is not to say that judges ought not to take into account social norms in
determining authorship, but merely that it is inappropriate to defer to such norms113.
109
ibid 1324.
110
(trial, n102) 1252, Judge Pollack even discussed Freeman’s conduct first, ‘in the order of seniority’.
111
Stokes (n79) 124.
112
See, Ginsburg (n88) discussing the dispute between Alexandre Dumas and his ghostwriter, August
Maquet, as dramatized in the play Signé Dumas by Cyril Gely and Eric Rouquette.
113
J Rothman, ‘The Questionable Use of Custom in Intellectual Property’ (2007) 93 Virginia LRev 1899,
1899 warns of the dangers of deferring to custom.
168
Weissmann v Freeman provides a warning of the dangers of uncritically adopting
community norms governing the attribution of authorship114. The previous two sections
have shown how values of authorship in modern scientific collaborations can be
complex, varying from group to group, evolving over time, and even being contested
within particular groups 115 . This case demonstrates how easily such norms are
misconstrued. Yet, even where properly identified, social norms might enshrine power
dynamics that risk distorting copyright’s core notion of the author as the originator of
expression protected as a copyright work. Weissmann v Freeman also provides an
example of the value of retaining a legal notion of authorship in copyright law that is
independent of community norms. It would have been difficult for Weissmann to obtain
any redress within the scientific community. Indeed, during the dispute she was
dismissed and Freeman was promoted116.
Although the reputation economy of science tends to operate in opposition to the
market economy, scientists still depend upon copyright. Publications are necessary in
order to obtain jobs, funding, and promotions, and journals rely on the copyright
subsisting in articles to be able to market their publications 117 . The attribution of
authorship on articles also establishes a locus for credit and responsibility for scientific
claims. In this context, copyright might have an important norm-setting role in
114
ibid.
115
Biagioli (n3).
116
R Kaufman, ‘After 5 Years, Heated Controversy Persists in Science Copyright Case’, The Scientist, 14
September 1992 at: <http://www.the-scientist.com/?articles.view/articleNo/12483/title/After-5-Years-
Heated-Controversy-Persists-In-Science-Copyright-Case/> accessed 26 August 2013.
117
Scientists also benefit from broad circulation of their work to their peers in reputable journals.
Birnholtz (n43).
169
establishing a legally enforceable standard for authorship which, in most jurisdictions,
also entails a moral right of attribution118.
This section considers the insights this case study of scientific collaborations might
offer copyright law. First, such collaborations demonstrate the importance of the
noneconomic benefits of authorship. Second, they provide examples of imperfect
solutions that copyright ought not to adopt wholesale because: (i) the ICMJE
guidelines are under-inclusive; (ii) particle physics approaches are over-inclusive; and
(iii) social norms ought to be approached with caution. Copyright law has an important
role to play as a source of good authorship standards, although it may not be able to
solve the authorship ‘crisis’ in the biomedical sciences. In the previous section I
suggested that it might be difficult to apply the joint authorship test to collective
authorship in science, particularly if the instrumental/pragmatic approach is taken. In
this section I argue that there is enough flexibility within the joint authorship test to
take into account the special characteristics of large collaborations as demonstrated in
this Chapter without compromising copyright’s core notion of authorship.
The example of scientific authorship demonstrates the importance of the
noneconomic consequences of authorship to some creative communities. This raises
broader questions about common incentive-based justifications for copyright which
others have considered elsewhere119.The intrinsic value of the label of ‘author’ is worth
118
Pila (n11); Dreyfuss (n54).
119
The literature on IP’s negative spaces helpfully summarised by E Rosenblatt, ‘A theory of IP’s
Negative Space’ (2011) 34(3) Columbia J of L and the Arts 317. See also p8 text at n37.
170
more within the scientific community than the possibility of recouping royalties.
Attribution of authorship is something that scientists might negotiate prior to
collaborative work and might even go to court to protect. The importance of the
attribution of authorship, independent of economic rights such as copyright, reinforces
the need to keep the concepts of authorship and ownership separate. The importance of
the noneconomic consequences of authorship for scientists undermines one of the key
assumptions of the instrumental/pragmatic approach: that streamlining the exploitation
of the copyright interest is more important than ensuring that creators are recognised as
authors. Indeed, most of the cases that have been brought to court concerning the
authorship of scientific articles appear to have been motivated by the need to ensure
accurate attribution of a work rather than a concern to recoup royalties120.
The processes by which large scientific collaborations create articles differs
significantly from the romantic author model. A great level of organisation is required,
and a division of labour is essential. The efforts of a number of highly specialised
individuals, with considerable autonomy in relation to their own tasks, are coordinated
with a view to achieving a common goal. The specialist knowledge or skills of many
contributors means that in a real sense the final product could not exist without most, if
not all, the contributions. Few, if any, contributors have the expertise to be held
responsible for the work as a whole, but many are incentivised to contribute by the
prospect of authorship credit. The ICMJE guidelines, which restrict authorship claims
to dominant contributors, respond poorly to this creative model. They allow authorship
to gravitate towards those with institutional authority, eliding the contributions of junior
scientists. The particle physics approach is better adapted to the way that large groups
120
Weissman v Freeman (n102); Noah v Shuba (n99); Anya v Wu (n93).
171
create. Its inclusive, non-hierarchical attribution practices reward many contributors and
internal consensus-based processes allow for the collective management of
responsibility. The downside of this approach is that it results in long author lists which
convey too little information about the origin of an article and tend to divorce
authorship from expression. It would not be appropriate for copyright to adopt either of
these approaches, because both clash with copyright’s core notion of authorship as
understood in terms of the origin of the protected expression.
Yet, whilst it is important to maintain a core independent legal notion of au-
thorship in copyright law, this concept ought to be applied in a way that takes into
account the peculiar nature of large group collaboration. Despite the hazards of
incorporating social norms, it is important that copyright law remains connected to
creative communities’ own understandings about authorship in order to maintain its
credibility as an important tool for encouraging and regulating creativity121. Weissman
v Freeman demonstrates the value of an independent legal standard of authorship,
which can be implemented in a way that is sensitive to the context of creativity. The
current application of copyright’s joint authorship test, with its impoverished
understanding of collaboration, assumes an integrated notion of authorship that lacks
sophistication in determining authorship in large collaborations, which often require a
division of labour. The instrumental/pragmatic approach is difficult to apply as it is
hard to identify one or two dominant authors. Attempts to do so are likely to reinforce
power dynamics which are already thought to be unsatisfactory in science because they
divorce authorship from the act of creation.
121
Communication & Society 179.
172
I would suggest that the joint authorship test already contains tools which might
be used to gauge the value of contributions to large collaborative projects in a
contextual way. In applying the joint authorship test, judges have tended to treat the
requirement that a contribution be ‘significant’ together with the requirement that it be
of the ‘right kind’122. Despite their flaws, social norms might provide a useful metric
for determining which contributions are considered to be significant within the con-
text of a particular collaborative project. This would imply taking a qualitative ap-
proach to the requirement for a significant contribution that incorporates contextual
considerations. The requirement for a common design, not comprehensively
elaborated in UK case law, might provide a conceptual apparatus to capture the
different nature of creative processes in large collaborations. In this way, the common
design requirement might also provide a helpful benchmark in determining whether a
contribution to a joint project is ‘significant’. In the context of science, for example,
the design of a study in accordance with the scientific method and the interpretation of
results, are likely to be ‘significant’ contributions to the expression.
Most cases on joint authorship turn upon judicial characterisation of the con-
tributions, particularly, whether a putative author has made the right kind of
contribution123. For these purposes, preparatory contributions and the contribution of
ideas tend not to count 124 . This might suggest that many valuable contributors to
scientific collaborations, such as those involved in planning, controlling, and directing
experiments; carrying out experiments; designing software or machines used in
122
See 2.2.3.
123
Hadley v Kemp [1999] EMLR 589 (Ch); Beckingham v Hodgens [2002] EWHC 2143 (Ch), [2002]
EMLR 45; Bamgboye v Reed (n48).
124
Robin Ray v Classic FM [1998] FSR 622 (Ch); 4.4 above.
173
experiments; and proof-reading papers, do not make the right kind of contributions.
Yet, as argued in the previous Chapter, the expression of the work ought not to be
conflated with the fixation of that expression, as what is required is ‘something
approximating penmanship’ 125 . Some recent case law has seemed to take a more
restrictive approach, privileging contributions to fixation over less tangible
contributions to the expression126. The example of scientific collaborations, however,
indicates the need to take a flexible approach to expression that takes into account the
necessity of divisions of labour in large collaborations127. Other case law supports this
approach128. Although copyright law ought to be wary of granting authorship to those
who have no more than ‘signed’ a work, lending their authority to it, sometimes when
‘authority’ is combined with control over the creative process that results in a
contribution which is evident in the expression, evidence of a relevantly authorial
contribution may be found129.
The requirement that a contribution be of the right kind has been used as a way of
enforcing copyright’s core notion of authorship130. In this way the joint authorship test
might avoid the pitfalls of particle physics collaborations’ over-inclusive approach, by
filtering contributions through copyright’s core notion of authorship, which favours
125
Cala Homes v Alfred McAlpine Homes [1995] EWHC 7, [1995] FSR 818.
126
In Ray v Classic FM (n124) 636, Lightman J describing Cala Homes as likely to be exceptional. See
also the discussion in L Zemer, ‘Contribution and Collaborations in Joint Authorship: Too Many Mis-
conceptions’ (2006) 1(4) JIPLP 283. See discusssion at p23 above.
127
Zemer, ibid also argues for such an approach.
128
Cala Homes (n125); Heptulla v Orient Longman [1989] 1 FSR 598 (HC of India); Donoghue v Allied
Newspapers (n94).
129
130
Fylde Microsystems (n94) a technician who tested and debugged software was not a joint author of
that software, because his work was more akin to proof-reading than authorship.
174
intellectual/creative contributions over mechanical ones and focuses on contributions
evident in the expression (broadly construed). Courts have sometimes tended to adopt a
restrictive interpretation of the joint authorship test, favouring those contributors who
appear ‘dominant’ to the exclusion of other contributors131. This resembles the approach
taken by biomedical journals in selecting a few authors who appear more worthy than
others. This strategy is poorly adapted to the way in which large collaborations work
together to create. The temptation to reduce the number of authors might be explained
by instrumental concerns that arise from one of the consequences of joint authorship:
joint ownership. Yet, authorship is a question that is conceptually distinct from
ownership. Designations of authorship also usually entail moral rights, which have
consequences that are quite separate from copyright’s economic rights.
Although the conventional notion of scientific authorship differs from copyright
law’s concept of authorship, both are challenged by the activities of large
collaborations. This Chapter has suggested that the ways in which scientific commu-
nities have responded to the authorship challenges posed by large scale collaboration
might offer insights for copyright law. I have argued that the joint authorship test ought
to be applied with an appreciation that large group authorship is a special kind of
authorship. This requires taking into account the process of collaboration and not simply
trying to identify a small number of dominant contributors. As collaborations grow in
size, there is a necessary division of labour. Even when a collaboration is hierarchically
organized, often no one contributor may be held responsible for its creative products.
For these reasons, the joint authorship test ought to be sensitive to the relative
importance of contributions within the particular context of a collaboration.
131
Hadley v Kemp (n123); Brighton v Jones (n94).
175
It is also important that the joint authorship test does not merely reproduce
power relations, but tells us something about who is responsible for the creation of the
expression of a work. For copyright, authorship ought to remain a question of law and
not an empty vessel for unquestioningly incorporating external conceptions of
authorship that might be unstable and contested within a creative community. Yet, the
joint authorship test is flexible enough to allow it to be more sensitive to the creative
processes of large collaborations than might commonly be thought. The joint
authorship test, for example, might better allocate authorship within large
collaborations by (i) taking an expansive, qualitative view of which contributions are
‘significant’ with reference to the joint project (common design); and balancing this
by (ii) applying the restrictive filter of copyright’s core understanding of authorship.
176
Chapter 6: Film
Films are, perhaps, the archetypal example of collective authorship as they tend to be a
product of the efforts of a large number of specialised contributors. Unlike the other
case studies considered in this thesis, the authorship of films has been specifically
addressed by the CDPA. Films are to be treated as works of joint authorship 1, with the
producer and the principal director taken to be their authors2. This approach has been
adopted as a response to the high risk, high investment nature of the film business. In
this context, there appears to be good reasons to concentrate copyright ownership. But
is the statutory designation of authorship to two particular contributors an effective
solution to the problem of determining the authorship of works created by a large
collaboration?
The legislative approach mirrors the film industry’s historical tendency to hold
one or two powerful players, often the director and/or producer, ‘responsible’ for a film.
Yet, this view of authorship has constantly been under siege from different interest
groups3. The recent trend is towards a more inclusive notion of authorship in line with
the reality of the film-making process. Whilst economic control of a film tends to be
streamlined through the assignment of rights by most contributors via contract; social
perceptions of authorship are dispersed in long credits that make it impossible to tell
1
CDPA s10. Unless the producer and the principal director are the same person.
2
CDPA s9(2)(ab).
3
P Decherney, Hollywood’s Copyright Wars: From Edison to the Internet (Columbia UP 2012) 90.
177
exactly where it begins or ends4. The film industry uses this ambiguity about authorship
strategically as a way of encouraging contributors – by showing that, great or small,
their contributions matter. In this way, as with some of the previous case studies, the
economic and reputational incidents of authorship are separated5. Contracts, as well as
being a vehicle to concentrate and redistribute copyright ownership in the production
company, also have an important role to play when it comes to the noneconomic
consequences of authorship, eg attribution.
Although the provisions on film copyright were intended to streamline
determinations of the authorship of a film, the reality is far more complex. Films which
are cinematographic works might also be dramatic works whose authorship is
determined according to the joint authorship test6. This means that such works have
many potential authors, and films are not insulated from the problems of applying the
joint authorship test to works of collective authorship discussed in previous Chapters.
Matters are further complicated by the fact that there are likely to be many different
layers of copyright or underlying rights embodied in some films (each with potentially
different authors). Indeed, I suggest that a restrictive approach to the application of the
joint authorship test is likely to increase pressure on judges to recognise copyright in
underlying works in order to provide remedies to creators7.
4
Screen credit seems to be part way between a designation of authorship and mere acknowledgement of a
contribution. This can be seen by the controversial position of the ‘film by…’ or ‘possessory’ credit, to
which screenwriters’ guilds have consistently objected. C Fisk, ‘The Role of Private Intellectual Property
Rights in Markets for Labor and Ideas: Screen Credit and the Writers Guild of America, 1938-2000’
(2012) 32 Berkeley J of Employment and Labour L 215, 256-258.
5
Chapter 3 (Wikipedia) and Chapter 5 (scientific collaborations).
6
Norowzian v Arks (No 2) [2000] FSR 363 (CA).
7
This may seem counter-intuitive given that such an approach tends to be motivated by a desire to reduce
the number of creators with copyright interests in the same subject matter (2.3.2).
178
This Chapter begins with an overview of the contested nature of film authorship;
the trend towards a more inclusive approach; and the pragmatic use of authorship credit
in the film industry. The second section considers the complex layers of copyright that
might subsist in a feature film, and the different rules that are used to determine the
authorship of each layer. The third section looks at the use of private ordering solutions
to resolve authorship questions and, in particular, the role of collective bargaining in
strengthening the bargaining power of groups of contributors. The final section
considers the insights for copyright law that arise from this case study.
Although in the CDPA ‘film’ might include many different works (eg, amateur
video, news reports, footage of sporting events, surveillance footage), in this Chapter I
focus on commercial feature films, because they are the clearest examples of collective
authorship.
6.1 Authorship Dynamics: The Pragmatic Value of Authorship
In this section, I consider the dynamics of film authorship. Dominant views about the
authorship of films have been constantly under siege as the film industry has grown and
as power has shifted between different industry players. Historically, authorship of a
film was seen as residing with the most powerful industry players, such as producers or
directors 8 . In modern times, the industry/social understanding of authorship is more
ambiguous, covering shifting sands and often employed pragmatically in the promotion
of films. Authorship is atomised or fragmented in film credits for pragmatic reasons: to
8
In this context, authorship tends to be understood in terms of reponsibility for the creative content of a
film. In particular, because of their control over the production process, these figures have tended to be
held responsible for whether a film is ‘good’ and also whether it is commercially successful.
179
encourage and reward a broad range of contributors. Although film credits are long,
they are not meaningless. In fact, credit is a matter of great importance to industry
players9.
Feature film production shares a number of characteristics with the other case
studies of collective authorship. A large number of individuals are often involved in
making a film. There is usually a division of labour that is organised hierarchically10.
Most contributors have very specialised skills. They might include producers, directors,
cinematographers, scriptwriters, actors, composers, visual effects technicians, set
designers, costume designers, managers, microphone operators, location scouts, and
many more. The types of contributions made to a film vary and include creative
contributions, technical support, management, promotional work, logistics, funding, etc.
Many contributors have a significant amount of control and autonomy over decisions
within their sphere of responsibility. Most contributors work collaboratively with other
contributors11. Some players have an important role in giving feedback and approving
or contesting the contributions of others. The actions of contributors are guided by
industry practices, instructions from senior players, their own aesthetic sensibilities,
their view of the common project, and their skills base12.
9
C Fisk, ‘Credit Where It’s Due: The Law and Norms of Attribution’ (2006) 95 Georgetown LJ 49, 80,
109.
10
J Naremore ‘Authorship’ in T Miller and R Stam (eds) A Companion to Film Theory (Blackwell
Publishing 2005) there is a mix of industrialised, technical, theatrical and artisanal practices.
11
The screenplay, for example, is a highly collaborative undertaking that might evince the contributions
of a number of key players. P Bloore, Managing Creativity and Script Development in the Film Industry
(Routledge 2013) 10.
12
Bloore (n11) 10 describing film production as an ‘industrial collaboration of different creative and
commercial agendas’.
180
Film resists traditional, romantic, notions of authorship13. In a real sense, a film
could not be made without the contributions of all of these individuals. Yet, historically,
authorship of a film has often been seen as residing with a few powerful players,
typically producers, studios and directors14. These contributors all tend to have a great
degree of control over the film-making process, particularly, the ability to grant or
restrict the creative freedom of other contributors and even to replace them. The
producer’s and the studio’s claim to authorship is justified by their investment in, and
overall control of, the film making process and personnel 15 . The director’s claim is
based on their control over the creative aspects of the process such as the positioning of
cameras, directing the actors, etc16. The precise amount of control exercised by these
figures varies from production to production.
Most other contributors are credited in a way that resembles acknowledgements
rather than authorial by-lines. Screen credits are governed by complicated rules
negotiated by the various talent guilds. The extent to which other contributors ought to
be considered authors is obfuscated because film credits do not indicate where
authorship begins or ends17. This ambiguity is strategic, as it allows a wide range of
13
A Notaro, ‘Technology in Search of an Artist: Questions of Auteurism/Authorship and the
Contemporary Cinematic Experience’ (2006) 57 The Velvet Light Trap 86, 87.
14
Decherney (n3) 89.
15
FJ Dougherty, ‘Not A Spike Lee Joint? Issues in the Authorship of Motion Pictures Under US
Copyright Law’ (2001) 49 UCLA LRev 225, 282-4, 311-13.
16
ibid 311-13.
17
Decherney (n3) 90: ‘Hollywood authorship exists as a spectrum – one that is constantly in flux and
always under siege’. The most significant contributors (the producer, the director, major actors, the
screenwriter, the composer, etc) are often mentioned in opening credits as well as closing credits.
181
contributors to be rewarded and acknowledged 18 . Authorship is atomised, so each
contributor is credited as the author of their specific contribution 19 . In respect of
authorial works, copyright has a mechanism for recognising that contributors might
have different shares of responsibility for the creation of a work, in the possibility that
joint authors might be entitled to unequal shares of ownership of the copyright
subsisting in their joint work20. Yet, in the case of film copyright, the designation of
joint authorship to the producer and the principal director appears to preclude the
possibility of such a nuanced approach to authorship21.
A possessory authorship credit to the director is often a powerful tool in the
marketing of a film, for example, ‘A Stanley Kubrick Film’ (The Shining) or ‘A film by
Quentin Tarantino’ (Pulp Fiction)22. This sort of credit has a role akin to a trade mark in
that it suggests something about the quality or character of a film to potential
theatregoers23. ‘A Steven Spielberg film’ such as Jurassic Park, for example, might be
expected to have spectacular special effects. The value of leaving the question of film
18
Although those listed in the opening credits and at the beginning of the closing credits are considered
to have made contributions of a more authorial-type, than those listed towards the end of the closing
credits.
19
They might also be celebrated for their particular contribution in award ceremonies and trade
magazines.
20
Based on the quantity and the quality of their contributions to the work, see n121, p27. There are no
provisions, however, for a particular order in the attribution of authorship.
21
Although where a film is also a dramatic work this approach might be implemented.
22
A Martin, ‘Possessory Credit’ (2004) 45 (1) Framework 95. The possessory credit does not always
refer to director: for example, in the case of ‘Tim Burton’s The Nightmare Before Christmas’ Tim Burton
was a writer and producer, not the director. Possessory credits have also sometimes been used to refer to
the writer of a work upon which a film is based, for example, ‘Bram Stoker’s Dracula’.
23
On the importance of the label of ‘author’ outside the copyright context, see p36.
182
authorship undetermined is so significant that these sorts of possessory credits have
long been resisted by other stakeholders, particularly writers24.
In contrast to the other case studies, film production is usually characterised by
high costs and high risks25. The film industry is often seen as dominated by a few major
(Hollywood) studios. The studio structure facilitates the funding of films and mitigates
the risks of film production. Studios can rely on a few high profit earning films or ‘tent-
poles’ to finance the majority of films26. Preferred locations for filming and producing
films change from year to year according to the availability of favourable regulatory
regimes. In the UK, commercial film production is dominated by independent film
productions which usually require funding to be amassed from a variety of public and
private sources 27 . These sorts of productions rely on a delicate coordination of
interests28, and even critically successful independent production companies can face
24
Dougherty (n15) 293; Fisk (n4).
25
Film production in the UK is precarious and often unprofitable. See: R Murphy, ‘Postscript: A Short
History of British Cinema’ in R Murphy (ed) The British Cinema Book (3rd edn, Palgrave Macmillan
2009) 417, 423.
26
J Garon, ‘Content, Control, and the Socially Networked Film’ (2010) 48 U of Louisville LRev 771,
777-778.
27
A significant proportion of film funding comes from government schemes and the distribution of Lotto
funds. The BFI website provides information about available funding: <http://www.bfi.org.uk/film-
industry>. See also: ‘A Future For British Film: It begins with the audience…’, Film Policy Review,
Department for Culture Media and Sport, 2012, 40. On the use of subsidies to promote creativity (instead
of intellectual property rights): S Breyer, ‘The Uneasy Case For Copyright: A Study of Books,
Photocopies and Computer Programs’ (1970) 84(2) Harvard LRev 281; J Love and T Hubbard ‘The Big
Idea: Prizes to Stimulate R&D for New Medicines’ (2007) 82 Chicago-Kent LRev 1519. Different
concerns relate to the production of short films, which can often be produced on a shoe string budget.
28
Channel 4, BBC Films and the UK Film Council generally support films only as partners in
complicated co-production deals that sometimes take years to set up: R Murphy, ‘ Postscript: A Short
History of British Cinema’ in R Murphy (ed) The British Cinema Book (3rd ed, Palgrave Macmillan 2009)
423.
183
chronic financial difficulties29. Where a number of different companies are involved in
financing and distributing a film, rights management can be a complex affair30.
As risk is such a dominant feature of the industry, studios and production
companies generally seek ownership of all intellectual property rights that might arise
as a result of their investment31. Production companies consolidate their ownership of
rights through contracts with all the various contributors to ensure they are unimpeded
in their exploitation of the final product – so they can recoup their investment. The need
to streamline distribution was an important factor in the reforms which lead to the
designation of the producer as the author of a film in the UK32. Indeed, the addition of
the principal director was an afterthought, primarily implemented to comply with the
EU Duration Directive33. Since the first moving pictures, views on film authorship have
been influenced by power struggles between different categories of contributors to
determine the content of copyright law with respect to film 34 . Lobbying has been
particularly important in this context. The effective lobbying of groups representing the
interests of producers and directors goes some way to explaining the special protection
they are awarded under the CDPA.
29
J Barnett, ‘Hollywood Deals: Soft Contracts For Hard Markets’ Duke LJ (forthcoming, available at
SSRN: <http://ssrn.com/abstract=2118918>) 7-8.
30
Complicated arrangements to share revenues from distribution can result in disputes, see M Rimmer,
‘Heretic: Copyright Law and Dramatic Works’ (2002) 2(1) QUT LRev 131, 137-9.
31
Report from the Commission to the Council, the European Parliament and the Economic and Social
Committee on the question of authorship of cinematographic or audiovisual works in the Community of 6
December 2002, COM (2002) 691.
32
P Kamina, Film Copyright in the European Union (CUP 2004) 138.
33
L Bently and B Sherman, Intellectual Property Law (OUP 2009) 124.
34
Rimmer (n30); Decherney (n3) 7-10, 89.
184
Although copyright law provisions on film authorship appear to be in the
producer’s favour, the significant economic investment that many films require has led
production companies to be cautious of relying upon the uncertainties of copyright to
regulate the authorship and ownership of films. Decherney carefully outlines how the
film studios have used contract to stay one step ahead of copyright law35. This may
suggest that copyright law’s provisions have little practical effect, yet, as I argue in 6.3,
they are an important source of default standards36. These standards are particularly
important for those in weaker bargaining positions as they may help them to gain
recognition and fair remuneration for their contributions37. Additionally, copyright law
is important, as not everything is comprehensively anticipated with contracts,
particularly in the early stages of a project when greater flexibility is needed38.
As the film industry has transformed over time, different players have enjoyed
prominence as authors. In the marketplace and in the eyes of the general public, film
authorship has often been linked to financial clout and/or control of the creative
process39. This was particularly the case at the height of the production of studio films
by the ‘majors’ or the ‘big five’ Hollywood studios40. Films were originally sold on the
35
Decherney (n3) 99-101.
36
On the significance of legal rules in providing bargaining endowments, see: R Mnookin and L
Korhauser, ‘Bargaining in the Shadow of the Law: The Case of Divorce’ (1979) 88 Yale LRev 950, 968.
37
This is a more favourable outcome than granting them non waivable rights, as it still allows for the
consolidation of copyright ownership (via contract) in the hands of the more powerful player, who is
likely to best placed to ensure the most effective exploitation of the work. Contributors have an incentive
to enter into such contracts, because most cannot exploit their copyright interest separately, see p159.
38
Often much is left to an oral agreement or handshake at these stages and with ‘key talent’: Barnett
(n29).
39
Charlie Chaplin was presented as a romantic author figure in his times: Decherney (n3) 67-76
40
From 1928 to 1949 these were: Loew’s/MGM, Paramount, Fox (which became 20th Century-Fox after
a merger in 1935), Warner Bros and RKO.
185
basis of the novelty of the technology. Initially, producers of that technology, such as
Thomas Edison, were the most important figures in the film industry41. With the rise of
the studio system, some corporate executives took on the role of ‘impresario’ and were
highly visible. At that time, screenwriters and other contributors were paid employees
of studios on long-term contracts and so were seen more as delegates than authors42. As
auteur theory came to prominence, there came to be a greater appreciation of the role of
the director, particularly certain directors who had a signature style, such as, Alfred
Hitchcock43.
Since its early days, film production has also been closely linked to the cult of
celebrity. In some situations celebrity names are used in a way that implies quasi-
authorship. The ‘star system’ particularly encouraged credit to gravitate towards the
celebrities whose names brought attention to movies. Agents might use their client’s
star power as a bargaining chip to negotiate credit for the star as an ‘executive
producer’. Credit is so significant that stars might accept less remuneration for a better
credit. Investment in films might be shored up by the commitment of a famous star that
has a certain pulling power in attracting audiences. Whilst it is true that one might
garner some information about the likely quality of a film from the names of certain
stars – such as Tom Cruise, Charlie Chaplin or Anthony Hopkins – star names are
generally employed somewhat cynically for their ability to assure ‘bums on seats’.
41
Decherney (n3) Ch 1.
42
J Schwab, ‘Audiovisual Works and the Work For Hire Doctrine in the Internet Age’ (2011) 35
Columbia JL & Arts 141, 147.
43
P Watson, ‘Cinematic Authorship and the Film Auteur’ in J Nelmes (ed) Introduction to Film Studies
(5th ed, Routledge 2012) 142, 148-152.
186
Whilst stars are sometimes held responsible for box office flops44, on other occasions
they are able to successfully distance themselves from films by arguing that they lacked
control over the end product45. Stars are rarely seen as responsible as an author of the
overall film in the same way studios or directors might be (unless they also had some
involvement in producing, writing or directing it).
In the film industry, groups compete for status. Those viewed as artists have
tended to be granted more prestige than those who are considered technicians or
craftsmen. Yet, no one is clearly the author of a film. After the dissolution of the studio
system in the 1950s, contributors tended to be contracted to work on a film by film
basis46. This led to an improvement in status for ‘creative’ contributors, partly because
they appeared more like authors than those who made mechanical or technical
contributions 47 . Since this time, contributors’ attempts to vie for more control,
remuneration, and credit have tended to be linked to claims of responsibility for
important creative elements in a film. Less powerful contributors have also begun
drawing attention to the creative aspects of their work in an attempt to improve their
status. The screenwriters’ guild, for example, stressed the creative nature of writers’
contributions as part of their bargaining to improve the terms of collectively bargained
agreements48. Those arguing for an improvement in conditions for visual effects artists
44
C Rosen, ‘‘Transcendence’ is a Real Bad Flop for Johnny Depp’, The Huffington Post, 21 April 2014
<http://www.huffingtonpost.com/2014/04/21/johnny-depp-transcendence-flops_n_5184999.html>; G
Macnab, ‘Johnny Depp’s New Film Transcendence is Yet Another Expensive Flop – So Has The Star
Lost His Mojo?’ The Independent, 30 April 2014.
45
For example, K Finbow ‘Nicole Kidman: ‘I had not control over Grace of Monaco’ Digital Spy, 27
May 2014.
46
Schwab (n42) 147.
47
Decherney (n3) 89-90.
48
ibid.
187
also stress their effect on the overall aesthetic of the film49. Thus, authorship is used as a
tool to improve a contributor’s bargaining position with more powerful industry
players.
Authorship remains a greatly disputed issue for film studies today. It is not
necessary to give a comprehensive account of film scholarship here, but suffice to make
the following points. Early philosophy of film struggled with the question of whether
film could be recognised as an art form50. As a popular cultural form, cinema appeared
too vulgar to be considered on the same level as opera, theatre or poetry51. The idea of
the director as auteur was first suggested by François Truffaut, who became one of the
central directors in the French New Wave 52 . He argued that the only films worth
valorising as ‘art’ were those in which the director had direct control over production
and screen writing as well as in relation to the actors’ performances 53. His ideas were
picked up by film theorist Andrew Sarris and auteur theory grew in importance within
the context of film studies54. Limiting the number of ‘authors’ of a film made it look
49
E Rome, ‘Oscars: Visual Effects Artists Protest Outside Dolby Theatre’ Entertainment Weekly, 25 Feb
2013 <http://insidemovies.ew.com/2013/02/25/oscars-visual-effects-protest-life-of-pi/> citing Bill
Westenhofer’s comments in support of fellow striking visual effects artists: ‘Visual effects is not just a
commodity that’s being done by people pushing buttons… Life of Pi shows that we’re artists not just
technicians’.
50
T Wartenberg, ‘Philosophy of Film’ in EN Zalta (ed) The Stanford Encyclopedia of Philosophy (2014)
<http://plato.stanford.edu/archives/fall2014/entries/film/>. Copyright law has similarly struggled with
whether to treat film as an authorial or entrepreneurial work, see 6.2.
51
ibid.
52
ibid. Truffaut’s manifesto ‘A Certain Tendency of the French Cinema’ published in Cahiers du Cinéma
January 1954 (No 31) translated and reproduced in B Keith Grant (ed), Auteurs and Authorship: A Film
Reader (Blackwell 2008) 9.
53
Wartenberg (n50).
54
In 1962 Andrew Sarris called the approach, referred to as la politique des auteurs, the auteur theory: A
Sarris, ‘Notes on the Auteur Theory in 1962’ in Keith Grant, ibid 35. See also A Astruc, ‘The Birth of the
Avant Garde: Le Caméra Stylo’, L’Écran français (30 March 1948).
188
less like a product of technicians and more like a work of art, the product of a romantic
author: the director. By adopting literature’s romantic author trope as the dominant
paradigm for understanding film, theorists paved the way for the acceptance of film as a
respectable art and this allowed for the growth of film studies as a discipline55.
Auteur theory has since been largely discredited because of its neglect of other
important contributions to a film56. The strategies by which the number of authors of a
film was restricted appeared artificial and highly constructed 57 . Yet, auteurism
succeeded in cementing the authorial role of the director in public understanding 58 .
Now, a director’s reputation might sell a film almost as effectively as the reputation of
its stars59. It also profoundly affected Hollywood’s view of its own past 60. Auteurism is
not universally accepted in the film community (it is now primarily seen as a marketing
tool)61. Instead, film-making is widely acknowledged to be a cooperative, collaborative
activity62. The vast majority of contemporary film scholarship also favours an inclusive
approach in which multiple authorship is embraced as a better model than auteurism63.
55
Naremore (n10) 10-11.
56
It was thought to have distorted the way that films were understood. Wartenberg (n50) describes the
theory as ‘clearly flawed’. See also: A Martin, ‘Possessory Credit’ (2004) 45 (1) Framework 95.
57
B Gaut, ‘Film Authorship and Collaboration’ in R Allen and M Smith (eds) Film Theory and
Philosophy (OUP 1997) 149.
58
Fisk (n4) 257, 276 describes it as a threat to screenwriters in addition to the work for hire doctrine in
the US.
59
Notaro (n13) 87.
60
Naremore (n10) 16.
61
Watson (n43) 157-161 discussing Quentin Tarantino as a contemporary case for auteurism (within the
context of an overall argument in favour of a pragmatic approach to film authorship).
62
Rimmer (n30) 135-6.
63
Gaut (n57) 149; P Livingston, ‘Cinematic Authorship’ in Allen and Smith (n57) 132; P Livingston, Art
and Intention (OUP 2005); P Livingston, ‘On Authorship and Collaboration’ (2011) 69(2) J of Aesthetics
and Art Criticism 221; P Sellors, ‘Collective Authorship in Film’ (2007) 65 J of Aesthetics and Art
189
Although a discussion of individual directors’ styles remains a relevant project,
contemporary film scholarship pays more attention to the context in which films are
produced (influenced by cultural studies and semiotic analysis). The debate on film
authorship remains, however, a lively topic in aesthetics where the focus is on how to
conceptualise the authorial group responsible for creating a film 64. This view of the
dynamics of film authorship suggests that an inclusive approach to the joint authorship
test would be more aligned with the prevailing social norms65.
6.2 The Subsistence of Copyright
In this section I argue that copyright law fails to provide adequate means of determining
the authorship of a film because it struggles to conceptualise the subject matter of
protection66. Whilst from one angle a film is an industrial product, a recording best
treated as an entrepreneurial work 67 ; from another it is a highly creative work best
treated as an authorial work. The current hybrid approach to the treatment of films in
the UK reflects this ambiguity. In so doing it eschews the certainty that the introduction
of (first fixation) film copyright was expected to provide. I argue that the recognition of
Criticism 263; S Bacharach and D Tollefsen, ‘We Did It: From Mere Contributors to Coauthors’ (2010)
68(1) J of Aesthetics and Art Criticism 23; S Bacharach and D Tollefsen, ‘We Did It Again: A Reply to
Livingston’ (2011) 69(2) J of Aesthetics and Art Criticism 225; DH Hick, ‘Authorship, Co-Authorship,
and Multiple Authorship’ (2014) 72(2) J of Aesthetics and Art Criticism 147.
64
Whether by the intentional contribution of artistically relevant features (Gaut); according to various
types of shared intention (Livingston, Sellors); by membership of an artistic group that shares a joint
commitment (Bacharach & Tollefsen); or in terms of responsibility for the form of the expression (Hick).
Citations above.
65
See 8.2 on the limits of incorporating social norms in legal decision-making.
66
Copyright law also struggles to define the boundaries of the literary work(s) on Wikipedia, see 3.2.1.
67
Barnett (n29) 7: ‘A Hollywood studio… is primarily a vehicle for coordinating the inputs required to
assemble a film project and financing, promoting and distributing films produced by internal production
divisions and independent production entities’.
190
additional layers of copyright flows from the imperative to protect creators who would
otherwise be excluded by the narrow view of the joint authorship of a film. I begin by
briefly introducing the history of copyright protection for film. Then, I consider the
potential complexity of copyright interests in a film (6.2.2). The next sections look at
the subsistence of copyright in a film as a first fixation (6.2.3); and as a dramatic work
(6.2.4). I end this section with a comment on the use of pragmatic reasoning in a US
decision on the authorship of a film, which has been much commented upon:
Aalmuhammed v Lee.
6.2.1 A Brief Historical Note
The way that copyright law has dealt with film has evolved over time. The 1911 Act
provided that the individual frames of a film could be protected as photographs. In
addition, a dramatic work was expressly defined to include ‘any cinematograph
production where the arrangement or acting form or the combination of incidents
represented give the work an original character’ 68 . In 1952, the Gregory Report
recommended that film be protected as a distinct type of work, in large part due to the
practical worry that too many people might claim authorship of a film, resulting in a
complexity of claims 69 . The Gregory Report considered that films bore more
resemblance to industrial products than they did to authorial works70. The perceived
complexity of the collective product led to the recommendation of a fixation-only form
68
1911 Act s35(1). Before then, individual frames of a film could be protected as a series of photographs
under the Fine Arts Copyrights Act 1862 and the dramatic content of a film could potentially be protected
under the Dramatic Copyright Act 1833.
69
‘Report of the Board of Trade Copyright Committee’ (Her Majesty’s Stationery Company, October
1952) (‘Gregory Report’) 37.
70
ibid 34 for this reason a relatively short duration of protection was recommended.
191
of protection. Accordingly, the subsequent 1956 Act protected cinematograph films as a
first fixation, entrepreneurial-type work and explicitly removed any protection for film
as an authorial work. Copyright was owned by the maker (the person who would now
be considered to be the ‘producer’ under the CDPA)71. This history is relevant because
it shows a tendency to prefer to treat films as industrial products, and thus, to award
copyright to the entrepreneur who took initiative and invested in the project. In this way
the law was shaped to fit industry practice, rather than existing copyright principles on
creative subject matter72.
It seems counter-intuitive to recognise films as the product of creative
collaboration, but then to deny copyright protection to potential film authors simply on
the basis that there are too many of them73. This tension between two conceptions of a
film (as an industrial product, but also as a creative work) might explain the hybrid
protection that films enjoy today that includes mechanical protection of the fixation
(‘film’); and more traditional protection of the cinematographic work (‘dramatic
work’)74. As a compromise and concession to Europe, eventually the director was also
considered to be an author of a film. Film copyright reflects a historical view of film as
a craft and the policy concern to reward investment. However, the true complexity of
film authorship is reflected by judicial development of the law in the acknowledgement
71
1956 Act s13(10).
72
M Handler, ‘Continuing Problems with Film Copyright’ in F Macmillan (ed) New Directions in
Copyright Law: Volume 6 (Edward Elgar 2007) 173, 177. In the Gregory Report this was partly
explained by the fact that no single creator of a film could be found. This comment appears to confirm
the fears of some scholars, that copyright law is unduly influenced by a romantic notion of authorship,
see 2.4.
73
ibid.
74
The distinction is probably best understood by reference to the distinction made in the 1956 Act
between (authorial) ‘works’ and ‘other subject matter’ (entrepreneurial works).
192
that a film might be a dramatic work (likely to have more authors); and in the
recognition that individual contributions might constitute separately protectable
copyright works. I argue that these developments are a response to the evidently
authorial characteristics of most films and the underlying imperative in copyright law to
protect creators.
6.2.2 The Complexity of Film Copyright
Determining the subsistence of copyright in film is a complicated matter. There are
likely to be a number of overlapping copyright interests, as contributors may make
separate copyright works which are used in a film. Indeed, according to Kamina, the
main problem is the diverse nature of the numerous different contributions made to a
film, rather than simply the large number of people who have contributed 75 . These
contributions can cover the whole spectrum of types of work protected by copyright and
may be made at different points in the production process. Some may exist
autonomously so as to be capable of being exploited independently of the film (eg,
props and costumes); whereas others may be inseparable from the film (eg, editing).
Some autonomous elements will not be protected by copyright, as they do not fall
within any of the categories for copyright works (for example, mise-en-scene).
Therefore, fictional characters will not be afforded copyright protection 76 , although
drawings of cartoon characters in animated film may be protected as artistic works 77.
75
Kamina (n32) 130. Lucasfilm v Ainsworth [2008] EWHC 1878, [2009] FSR 103.
76
Kamina (n32) 103 summarises the authorities: Kelly v Cinema Houses Ltd [1928-35] Macq Cop Cas
362 (cf dicta of May LJ in O’Neill v Paramount [1983] CAT 235).
77
King Features Syndicate v O and M Kleeman [1941] AC 417 (HL) (Popeye); Mirage Studios v
Counter-Feat Clothing [1991] FSR 145 (Ch) (Teenage Mutant Ninja Turtles).
193
Facial makeup may not be permanent enough to attract copyright 78 . There may be
copyright in the music used as a musical work (or musical works) and also as a sound
recording. The actors are likely to have performance rights in respect of their
performances 79 . The screenplay is a dramatic work and the final film might be
considered to be a derivative (dramatic) work of that screenplay80.
Prior to Lucasfilm v Ainsworth, the general position appeared to be that sets,
costumes or props may be artistic works81. In Shelley Films v Rex Features there was
held to be a serious issue to be tried as to whether copyright subsisted in costumes and
prostheses as works of artistic craftsmanship and it was held plainly arguable that
copyright could subsist in a film set as a work of artistic craftsmanship82. In Creation
Records Justice Lloyd could ‘readily accept that a film set does involve craftsmanship’,
distinguishing it from an assembly of ‘objets trouves’ (he made no comment on whether
it might be considered ‘artistic’) 83 . In Lucasfilm (HC) Justice Mann thought that a
Storm Trooper helmet was plainly a work of craftsmanship, but that it was not artistic84.
The careful arrangement of a set might be considered to be a compilation, although it
78
Merchandising Corporation of America v Harpbond [1983] FSR 32 (CA).
79
See Pt II CDPA.
80
It is unlikely to be seen as a literary work given that this category of work explicitly excludes anything
that might be considered a dramatic work (s3(1) CDPA).
81
Cf Lucasfilm (n75).
82
[1994] EMLR 134, 143 (Mann QC sitting as Deputy High Court Judge): ‘In principle this would seem
correct [that copyright can subsist in a film set as a work of artistic craftsmanship] since, if the set is
imaginatively conceived and implemented overall as a work of artistic craftsmanship, it cannot matter
that it happens to be made up of numerous, perhaps many thousands, of components in some of which,
when considered separately, copyright might not exist, provided the effect and intent overall is artistic’.
83
[1997] EMLR 444, 449.
84
Lucasfilm (n75). For similar reasons that it was held that the helmet was not a sculpture, as discussed in
the next paragraph. The issue was not considered on appeal: Lucasfilm v Ainsworth [2009] EWCA Civ
1328, [2010] Ch 503.
194
may not be a collage85. Although a compilation is a type of literary work (and a film set
might not seem to be naturally characterised as a literary work), this interpretation is
plausible given the definition of a literary work as including anything that is ‘written’86.
In the CDPA, ‘writing’ is defined very broadly to include ‘any form of notation or code,
whether by hand or otherwise and regardless of the method by which, or medium in or
on which, it is recorded’87.
Lucasfilm v Ainsworth, however, might seem to suggest that props will only be
considered to be artistic works in rare cases. In that case, the Supreme Court affirmed
lower court decisions that a Storm Trooper helmet from the film Star Wars was not a
sculpture88. The court stressed that it was the film itself that was the ‘work of art’, and
that, as a step in the production process – the stormtrooper helmet was necessarily
utilitarian, rather than artistic 89 . This was despite the fact that it contributed to the
artistic effect of the finished film. Although the Supreme Court did not interfere with
the trial judge’s findings on this point, it did express some hesitation in accepting them
85
In Creation Records (n83) the arrangement of a scene was held to be too ephemeral to constitute a
collage. The court also thought that the placement of objects (without any sticking or gluing) precluded
finding it to be a collage. The court did not consider whether the scene might be protected as a
compilation.
86
CDPA s3(1).
87
This argument does not apply in relation to musical works on the score because a literary work is
defined as excluding musical works. An editor’s selection and arrangement of the scenes is also unlikely
to be able to be protected as a compilation because it is not relevantly written in any notation or code.
88
Lucasfilm v Ainsworth [2011] UKSC 39, [2012] 1 AC 208. At first instance (n75), Mann J held that the
helmet was not a work of artistic craftsmanship (not challenged on appeal). He set out a multifactor test
for determining whether a work constitutes a ‘sculpture’ that was accepted by the UKCA and the UKSC.
89
ibid [44]: ‘But it was the Star Wars film that was the work of art that Mr Lucas and his companies
created. The helmet was utilitarian in the sense that it was an element in the process of production of the
film’.
195
given the ‘imagination that went into the concept’ of the helmets 90. In this way, the
court seems to consciously leave space for the protection of highly imaginative and
artistic props, sets, or costumes as artistic works as long as they satisfy the multi-factor
test identified by Justive Mann at first instance.
Oddly, the trial court considered that the preliminary paintings that the Storm
Trooper costumes were based upon (but that were not included in the film) might be
artistic works 91 . By extension, this suggests that storyboards might be protected as
artistic works. Yet, it is difficult to see why the helmet is a utilitarian object on the basis
that it is a step in the production process; whereas the paintings it is based upon are
artistic works. The main difference is incorporation into the final product (the film)92.
The best way of understanding this strained reasoning, is that the court is construing the
boundaries of the ‘work’ so as to avoid overlapping copyright interests. If copyright
law’s imperative to incentivise and reward creators is to be taken seriously, surely the
corollary of refusing to recognise copyright subsisting in component parts of a film,
ought to be a broader approach to construing the authorship of the film as a whole93.
Indeed, Lucasfilm follows the recognition that some films will also be dramatic works
90
ibid. Even Mann J, at first instance, had accepted that the helmet was one of the most abiding images of
the film.
91
They were considered to be original graphic works. This appears to have been conceded by Mr
Ainsworth (n75) [89].
92
Another explanation might be that a helmet has a non-artistic purpose (protecting the head), which a
painting does not. Yet, this ignores the fact that this helmet was designed so as to suggest various
qualities of the character of a Storm Trooper. The creator’s aesthetic purposes in making the Storm
Trooper helmet seem to be impliedly accepted in the court’s reasoning that a replica helmet in a twentieth
century war film would be less likely to be a sculpture (n88) [44].
93
The Gregory Report (n69) 36-7 relied upon the comparable reasoning (in the opposite sense) when it
justified the narrow definition of film authorship which it proposed on the basis that many contributors to
a film would have separate copyrights in the underlying works they had created.
196
in Norowzian v Arks (No 2)94. Yet, in stressing the utilitarian nature of the helmet, the
reasoning of the court in Lucasfilm obscures more than it reveals. This is essentially an
aesthetic judgment, although it is not entirely clear which parts of the evidence
specifically supported it95. Surely the decision would have had more secure foundations
if the court had connected its reasoning to the common design of contributors to the
film 96 . As a step in the production of the film, the creation of the helmet might
potentially entitle its creator to joint authorship of the film as a dramatic work (although
it is unlikely to count as a significant contribution of the right kind). The court could
have held that, as a component of a joint work that is ‘not distinct’, there is no separate
copyright in the helmet.
The complex issue of the underlying rights in a film is further complicated by
CJEU’s suggestion in Infopaq that parts of a work that reflect the author’s own
intellectual creation may be separately protected as copyright works. On this view, a
particular scene of a film (or component parts) might be separate copyright works
where they are the intellectual creation of its author. This is at odds with UK decisions
that stress the necessity of objectively determining the scope of the work, as copyright
should not be seen as a ‘legal millefeuilles’ in which a claimant might try to claim
various layers of protection by asserting that small parts of a work are independent
works, thus making it easier to establish that an infringer has copied a substantial part of
94
Considered at 6.2.4 below. More contributors are likely to count as authors of a dramatic work than do
as authors of a film.
95
J Pila, ‘The “Star Wars” Copyright Claim: An Ambivalent View of the Empire’ (2012) LQR 15, 17
arguing that the court’s approach was correct insofar as the multi-factor test focused on the intent of the
creator and the view of society with respect to the nature of the creation; but that the suggestion that
objects created as a step in any production process would not be artistic works is more problematic.
96
There is some space for consideration of these factors in the multi-factor test for a sculpture adopted by
Justice Mann (n75).
197
the work97. In Chapter 3, I argued that this approach is problematic in the context of
Wikipedia and the same appears true here for film98.
In addition to the potential complexity of underlying rights in component parts of
a film; many films are likely to enjoy dual protection as a whole99. A film which is a
cinematographic work may be protected as a dramatic work, and the recording of that
film (or first fixation) is potentially protected by film copyright. The next two
subsections consider the authorship of both these types of work.
6.2.3 Film as a First Fixation
Film copyright is a mechanical-type of copyright that subsists in the recording of
images and sound100. Copyright is acquired by the act of first recording, irrespective of
any personal labour or skill101. Copyright will subsist in the film as long as it is not
simply copied 102. The definition of a film is broad, and includes, ‘recording on any
medium from which a moving image may by any means be produced’103. In Spelling
Goldberg v BPC Publishing there were said to be three characteristics of film: (i) a
sequence of images; (ii) recorded on material; (iii) capable of being shown as a moving
97
IPC Media v Highbury-Leisure Publishing [2004] EWHC 2985, [2005] FSR 20 [23] (Laddie J),
approved in Coffey v Warner/Chappell Music [2005] EWHC 449 (Ch), [2005] FSR 34 (Ch) [10]
(Blackburne J).
98
As argued above, p48.
99
Bently and Sherman 60: they ‘occupy a space between authorial and entrepreneurial works’. The
Rental and Related Rights Directive and the Duration Directive also distinguish between
‘cinematographic works’ and related rights in mere fixations, called ‘films’ or ‘videograms’.
100
Norowzian v Arks (No 1) [1998] FSR 394 (Ch).
101
CDPA s5B(4). Laddie et al [3.100], 125.
102
This means that it must not be copied from another film.
103
CDPA s5B.
198
picture104. The soundtrack is treated as part of the film, but this does not affect any
copyright subsisting in the film soundtrack as a sound recording 105 . Although film
copyright arises easily without the usual requirement of an original contribution, it
subsists only in the fixation and, therefore, is only infringed by literal copying (taking
the actual images or sounds recorded in the film) 106 . This means that there is no
infringement of copyright in a film to transcribe it, to perform it as a play, or even to
reshoot it scene by scene (although this may infringe underlying works included in the
recording). A film is presumed to be a work of joint authorship, with its authors deemed
to be the producer and the principal director107.
The producer is the ‘person by whom the arrangements necessary for making the
film are undertaken’108. The producer is not the person who makes the recording in the
literal sense of the one holding the camera. Direct organisational control over the
process of production is required, thus, the producer is the person who organises,
coordinates and controls the production of the work 109 . It is not enough just to
commission or provide finance for a film (otherwise a bank might be an author), but
104
[1981] RPC 283.
105
CDPA s5B(2) and (5).
106
Norowzian v Arks (No 1) (n100). Section 17(4) provides that the copying of a film includes ‘making a
photograph of the whole or any substantial part of any image forming part of the film – so the infringing
use need not be as a ‘moving image’. See: Spelling Goldberg Productions v BPC Publishing [1981] RPC
283 (CA) (a single frame); Football Association Premier League v QC Leisure (No 2) [2008] EWHC
1411 (Ch) (four frames of a video stream not a substantial part); R v Higgs [2008] EWCA 1324, [2009] 1
WLR 73 ([9] Jacob LJ, obiter, a single frame ‘probably’ infringement). There is no adaptation right in
relation to film (s21(1)).
107
Unless these are the same person, CDPA s9(2)(ab).
108
CDPA s178.
109
Adventure Films v Tully [1993] EMLR 376.
199
these contributions could be relevant when combined with other contributions110. This
definition is significant because it means that investors will not automatically be
considered authors unless they have ‘made arrangements necessary for the making of
the film’, which implies some input in coordinating the film or setting the production in
motion. There is no definition of ‘principal director’, but this is likely to be the person
who has creative control over the making of a film 111 . It might be problematic to
determine the authorship of films which do not appear to naturally have ‘producers’ or
‘directors’, such as home video112. It will generally be easy to identify these figures in
the case of feature films, although this depends upon the facts113.
In Slater v Wimmer, for example, Wimmer planned and performed a skydive over
Mount Everest, which he arranged for Slater to film114. Both parties subsequently used
the footage without the consent of the other. Unusually, there was no written agreement
and no express term in the oral contract as to the ownership of copyright in the film, so
the court was left to decide the matter according to the statutory rules on film
authorship. Judge Birss held that Wimmer was the producer as it was ‘his project’ – he
had made all the arrangements for the dive to be filmed and had paid the costs of
filming (including Slater’s travel expenses). Slater was the principal director as ‘he
decided what to film and how to do it’ – this included controlling the settings of the
110
Beggars Banquet v Carlton [1993] EMLR 349; Century Communications v Mayfair Entertainment
[1993] EMLR 335. This may mean that a commissioner might be considered an author.
111
Slater v Wimmer [2012] EWPCC 7 [72]; Laddie et al [7.41] suggest that the use of the word ‘principal’
limits the scope for recognition of those who have made a lesser level of creative input to the film.
112
Handler (n72) 186-7; Kamina (n32) 140.
113
Slater v Wimmer (n111) [80].
114
[2012] EWPCC 7.
200
camera and choosing the relevant angles115. The court did not imply any terms requiring
the transfer of intellectual property rights from director to producer noting that although
Wimmer had reimbursed Slater’s expenses, he did not pay any additional fee or salary
for his work as a cameraman116. This was seen as indicative of a need for Slater to
retain control over the footage, ie a copyright interest in the film.
Slater and Wimmer were held to own copyright in the film as tenants in common
in equal shares117. This finding meant that both had infringed the rights of the other in
the film. Bonadio characterises this as an ‘unpleasant lock-in situation’ of the sort that
presents a serious obstacle to the exploitation of rights in the film in the absence of a
prior written agreement on intellectual property rights118. The decision also highlights
the value to individual creators of having copyright authorship as a bargaining chip,
which might be used to help them to ensure fair remuneration. The court implicitly
recognises this in the reasoning given for failing to imply a transfer of rights. This lock-
in situation may seem ironic, given that the default rules on film authorship are
designed to avoid exactly such a situation119. In this way the decision elevates fairness
over efficiency, providing an important reminder that it is creators (authors) who are at
the centre of copyright law.
115
ibid [73].
116
ibid [95].
117
ibid [89].
118
E Bonadio, ‘Joint Ownership of Films in the Absence of Express Terms’ (2012) 7(7) JIPLP 493.
119
This situation may have been more satisfactorily resolved had the case been decided under US law,
which provides each joint owner has the right to licence uses of the work without the consent of the other
joint owners: See MB Nimmer and D Nimmer, Nimmer on Copyright (Matthew Bender 2011)
§6.10[A][1][a] or [3-36].
201
The designation of the producer and the principal director as the authors of a film
is justified by the need to have few owners so that films might be efficiently exploited,
and investment recouped. Similar concerns underlie the treatment of employee works,
except in this case the CDPA provides for presumptive employer ownership of
copyright in works created by employees within the course of their employment 120. Yet
employees are still the authors of their work and might explicitly agree to retain their
ownership of copyright121. The stronger approach to film might be explained by the
need to create a sense of certainty given the large capital investment and high risk
nature of the industry. In fact, the reason is more likely to be related to the powerful
lobbying of the film industry for special protection. The complex layers of copyright
interests that subsist in many films, however, likely undermine this sense of certainty122.
Section 13B of the CDPA provides that film copyright expires 70 years from the
last to die of: the principal director; the author of the screenplay; the author of the
dialogue; or the composer of music specially created for and used in the film. This is an
unusual state of affairs, because copyright tends to adopt the author as the point of
reference for the duration of protection. These contributors are not recognised as
authors, for reasons of expediency: the need to concentrate rights in the hands of the
producer and director in order to facilitate exploitation of the film. Perhaps this
provision reflects an acknowledgement of the significant creative contributions made by
these contributors. In addition, the producer as an orchestrator (rather than a creator)
120
CDPA s11(2).
121
The different treatment of film might seem to be related to the moral rights consequences of
authorship. But these have been modified in the case of employee works (CDPA s79(3), s82); a similar
solution could have been adopted for film.
122
This may not have much of a practical impact, given that the film industry relies primarily upon
contract to deal with issues of copyright ownership, 6.3 below.
202
does not have moral rights in relation to the film123. These provisions seem to inch
towards an acknowledgment of the creative contributions to a film and the fact that the
creative aspects of the reality of film production tend to clash with the notion that it is
an industrial product. The case law has gone much further as the next section
demonstrates.
6.2.4 Film as a Dramatic Work
Film copyright leaves any underlying rights intact, thus preserving the possibility that a
film may also be a dramatic work124. A dramatic work includes a work of dance or
mime, but otherwise the statutory definition is open-ended125. Case law establishes that
a dramatic work is a ‘work of action, with or without words or music that is capable of
being performed before an audience’126. There must be ‘sufficient unity’ to a dramatic
work for it to be capable of being performed127, which is likely to exclude computer
games128, news reels129 and sports matches130. A film might be a recording of a dramatic
123
Unlike the director, the producer’s more creative co-author (CDPA s77(1), s80(1)).
124
T Rivers, ‘Norowzian Revisited’ (2000) EIPR 389 suggests this was probably not the intention of the
legislature. On the debate about overlapping categories compare I Stamatoudi, ‘“Joy” for the Claimant:
Can a Film Also be Protected as a Dramatic Work’ (2000) 1 IPQ 117 and R Arnold, ‘Joy: A Reply’
(2001) 1 IPQ 10. Films which came into existence before the 1956 Act came into force were, and
continue to be, dramatic works or photographs only.
125
CDPA s3(1)(d).
126
Norowzian (No 2) (n6) 367 (Nourse LJ).
127
Green v Broadcasting Corporation of New Zealand [1989] 2 All ER 1056 (PC).
128
Nova Productions v Mazooma Games [2006] EWHC 24 (Ch) [116] this point not challenged on
appeal: [2007] EWCA Civ 219; [2007] RPC 25 [3].
129
Laddie et al [3.117], 137 also suggesting that security footage may not be a dramatic work, although
the ‘the latest 24 hour Internet shows’ may be.
130
Copinger et al [3-39] consider that although a football match is not an intellectual creation (C-403/08
Football Association Premier League v QC Leisure [2012] 1 CMLR 29 [98]), a film of a sports event
may be a dramatic work where it has been filmed in such a way as to increase its impact.
203
work and it might also be a dramatic work itself following Norowzian v Arks (No 2)131.
This is significant, because a person recreating the dramatic work may be sued for
infringement, which would not be the case in relation to the film copyright132.
Norowzian v Arks (No 2) concerned ‘Joy’ a short film made using a technique
called jump-cutting so that it featured a person who appeared to perform a sequence of
movements that in reality could not be physically performed by an actor133. The film
was the inspiration for a commercial which depicted a man dancing with the same
surreal effect in front of a pint of Guinness. The commercial did not infringe the film
copyright in ‘Joy’ because no part of it had been directly, photographically reproduced.
At first instance, Justice Rattee held that a film could not be a dramatic work per se, but
it could record a dramatic work (although he thought that in this case it did not, as the
dance depicted was incapable of being performed)134. The Court of Appeal disagreed on
the first point, and held that a film was capable of being a dramatic work per se 135. For
these purposes, ‘capable of being performed’ included performance by artificial means,
such as the screening of a film. The Court of Appeal held that the film itself was a
dramatic work, but the judges differed in their reasoning136. Lord Justice Nourse came
131
Affirmed in Dramatico Entertainment v British Sky Broadcasting [2012] EWHC 268, [2012] RPC 27
[63].
132
Norowzian (No 1) (n100).
133
n6.
134
It is hard to make sense of this part of the reasoning because it seems to imply that the underlying
dramatic work embodied in the dance had somehow disappeared during the editing of the film. Rivers
(n124).
135
Stamatoudi (n124) criticises the decision suggesting that a film ought not to be seen as a dramatic
work per se. Cf. Arnold (n124).
136
A film could also be a recording of a dramatic work. In this case there may have been a recording of
the dance in the initial film or ‘rushes’, however, this was not relevant as the claim concerned the
particular surreal ‘dance’ evident in the film. As a result of such extreme editing the original dramatic
work could no longer be said to be recorded in the film (n6) 210.
204
to this conclusion in light of the broad definition of dramatic work in the Act, which he
thought would often, although not always, include a film. Lord Justice Buxton, on the
other hand considered the need to interpret the CDPA consistently with the UK’s
obligations under the Berne Convention, which required all cinematographic works to
be included within the Act’s definition of a dramatic work even in cases where the
natural meaning of this term might not embrace the particular film in question. As
Berne equates cinematographic works with other authorial works, he reasoned that they
ought to be given the same level of protection137.
The court in Norowzian v Arks (No 2) was likely to have been, at least, partly
motivated to recognise the existence of a dramatic work in film because film copyright
provides thin protection and fails to recognise and reward creators 138. Thus, the court
sidesteps the legislature’s solution to concentrate authorship in the hands of one or two
contributors seeking an approach that better reflects creative realities. In this light, it
should be no surprise post-Norowzian to see a resistance in Lucasfilm to separating out
individual contributors to the joint creative process. In the absence of an authorial-type
protection for the cinematographic work in a film, the only other way to reward creative
contributors is to find that they created distinct copyrightable works – which seems to
be the approach taken prior to Lucasfilm139.
As most films are the product of editing, they will almost certainly be dramatic
works distinct from any other dramatic works underlying them, such as the screenplay.
137
Arnold (n124).
138
A Barron, ‘The Legal Properties of Film’ (2004) 67 MLR 177, 207 citing the judge’s comment that
Joy was a ‘striking example of the film director’s art’.
139
As was anticipated in the Gregory Report (n69) 36-7, see text at n93.
205
In this way, a film may be a dramatic work and also a record of another pre-existing
dramatic work (the screenplay). The joint authorship test must be applied to determine
the authorship of these dramatic works. Although this will depend in large part upon the
specific context of a film, generally the editor, the screenwriter and the director will
have strong claims to be joint authors (where they are responsible for originating the
action portrayed)140. Whether or not any of the other contributors might count as joint
authors is more uncertain (for reasons explained more fully at 2.3 above); and will
depend upon the approach taken to the application of the joint authorship test.
Most contributors to a film that is a dramatic work are likely to satisfy the
requirement for collaboration or common design 141 . The scriptwriter might not be
collaborating where the script has been written before the concrete idea of the film is set
in motion. In this case, the scriptwriter will be the author of the screenplay, but not a
joint author of the resulting cinematographic work. There also may not be collaboration
present where a person is contracted to provide certain contributions (eg, costumes or
props) with little knowledge of the use their works will be put to. In some cases these
may be protectable as distinct works.
Another requirement is that a joint author’s contribution must not be distinct.
Although this point rarely causes issues in the case law on joint authorship, it is
surprisingly difficult to pin down the meaning of the word ‘distinct’. A contribution is
not distinct merely because it is identifiable. Instead, the question is whether the work
140
Copinger et al 238 [4-22] suggesting that the authors of the screenplay and any underlying works are
not authors of the dramatic work – rather, they are authors of their separate literary or dramatic works.
141
Beckingham v Hodgens [2003] EWCA Civ 143, [2003] EMLR 18.
206
142
would be different in character without the contribution . Another way of
conceptualising this is to consider whether the contribution is integral to work 143 .
Probably the best way to make sense of this requirement is to consider the function of
this part of the test – which is to distinguish between a case involving the collocation of
different works by different authors; and a joint work, that is, one work, by two or more
authors. In this way the individual authors of encyclopaedia entries (where each has
written a single entry, unaided by the other authors) are not together, joint authors of the
encyclopaedia as a whole. Where a work is separately exploitable this might seem to
suggest that it is ‘distinct’, although separate exploitability is probably not a
requirement144. There appears to be a bias toward treating works separately whenever
this is possible, which reinforces the idea that copyright is influenced by a romantic
notion of authorship (or at least a preference for avoiding group authorship). This
results in the recognition of many layers of copyright protection, which does not avoid
the problem of many rightholders that the instrumental/pragmatic approach seeks to
rein in.
The screenplay is likely to be distinct from the dramatic work embodied in the
film 145 . Kamina suggests that those who contribute and arrange the music; the art
director and those who help design costumes, props, etc. would not be considered
authors of the resultant dramatic work because their contributions are separate. This
142
In other words does the contribution form an integral part of the work. See: Beckingham v Hodgens
(n141) – although the violin part of the song could be identified separately in terms of musical notation, it
was not ‘distinct’ because the part was dependent upon what was already there. It would have sounded
odd and lost its meaning on its own.
143
Fisher v Brooker [2009] UKHL 41; [2009] 1 WLR 1764; [2009] ECDR 17.
144
Kamina (n32) 144, 146 commenting on the 1956 Act.
145
By analogy with Chappell v Redwood Music [1980] 2 All ER 817, [1981] RPC 337 (HL).
207
really comes down to the question of how to draw the boundaries of the ‘work’, which
is a difficult question to pin down. If props are not considered to be artistic works
because of their role as utilitarian objects in the film production process, surely the
corollary of this is that their creators ought to be considered potential authors of the
dramatic work to which they contribute. This would suggest that such contributions
ought not to be generally regarded as ‘distinct’. That said, there is precedent (pre-
Lucasfilm) that suggests that the contribution of costumes is not sufficient to establish
authorship of a dramatic work in the case of a theatre production146. Similarly, Copinger
et al suggest sets do not constitute part of a dramatic work147.
This most significant hurdle is the requirement that a joint author must make a
significant contribution of the right kind. Contributions of investment alone will not be
sufficient. Certain pre-expressive contributions may not count, as they will not be
evident in the expression of the work – for example, set or costume design drawings.
These may, however, be separate copyright works (as in Lucasfilm). In relation to the
contribution being of the right kind it has been held that ‘interpretation’ is not an
authorship-type contribution, which may exclude the contributions of actors 148 . A
performance is not a separately copyrightable work, but an actor may have performance
rights in relation to his or her performance 149 . Similarly, ‘proof-reading’ is not an
146
This conclusion might still be justified on the basis that making costumes is not a significant
contribution of the right kind.
147
(n130) [3-37].
148
Hadley v Kemp [1999] EMLR 589 (Ch) 646. Cf. D Free, ‘Beckingham v Hodgens: The Session
Musician’s Claim to Music Copyright’ (2005) 1(3) Entertainment and Sports LJ 93 available at
<http://www.warwick.ac.uk/go/eslj/issues/volume1/number3>.
149
Tate v Fullbrook [1908] 1 KB 821 (CA).
208
authorship-type contribution150. This might be problematic for producers who suggest
small changes or visual effects technicians who clean up the film by removing minor
errors. The most problematic aspect of the exclusion of proof-reading is that it seems
that even a significant amount of time and effort will not suffice (as in Fylde
Microsystems)151. It will not be sufficient to suggest the concept or idea for the film152.
Norowzian establishes that the use of an editing technique is also considered to be an
idea (thus not protectable).
Joint authors are those who originated the protectable elements of the dramatic
work153. Yet, it is difficult to precisely determine the sort of skill, labour and judgment
that will be required, although it is a fairly low standard. Copinger et al suggest a person
will be an author of a dramatic work where he or she adds something of dramatic
significance, for example, choice of camera angle, lighting or editing154. The greatest
challenge is working out what counts as a ‘dramatic’ contribution. The similarities
between film and photography are great. The case law on photographs suggests that an
author makes a relevant contribution when they make decisions in relation to lighting,
camera angle, etc155. Does this also apply to dramatic works? On the other hand, in the
context of a cinematographic film, there are reasons to suppose that a court may be
150
Fylde Microsystems v Key Radio Systems [1998] FSR 449 (Ch).
151
ibid. Although the case might be distinguished on the basis that it concerned a literary, rather than a
dramatic work.
152
Tate v Fullbrook (n149). In Tate v Thomas [1921] 1 Ch 503: contribution to scenic effects and the
stage ‘business’ were not of the right kind, and other contributions suggesting the name of the play and a
few catch lines and words were too negligible.
153
Rivers (n124) suggesting that it is problematic that most of these cases concern the very different
context of the theatre.
154
(n130) 93, [3-39]. This appears to be similar to the approach taken to determining the existence of an
artistic work in Lucasfilm (n75).
155
Painer v Standard Verlags [2012] ECDR 6.
209
more concerned with contributions that have an impact on the visual and emotive
character of the work. Following this reasoning, there is a strong case that an editor will
be an author – as they cut shots together to create the dramatic feel and flow to the
action of a film. Similarly, those responsible for the dialogue and the acting directions
would also potentially have a good claim to joint authorship. By the same token, any
music created to accompany the action and add to its dramatic quality might entitle its
creator to joint authorship (as long as it is not held to be distinct). Kamina suggests that
there needs to be some sort of story for there to be dramatic work, yet this is probably
unduly restrictive. The dramatic quality of a film dramatic work relates more naturally
to the aesthetic aspects of work (its look and feel) as well as its content.
Although one would think that film producers would be inclined to argue against
the proposition that many of the other contributors are authors of the dramatic work in a
film, the opposite may be true in some cases. Where strong contractual arrangements
provide for the assignment of any copyright interest owned by film contributors,
production companies may wish to argue that many contributors are authors so as to
prolong the duration of copyright (which will be seventy years from the death of the last
surviving joint author)156. In cases where copyright ownership is less clear, those who
have heavily invested in the production will have a strong incentive to argue for few
authors of the dramatic work.
156
CDPA s12(2), s12(8).
210
6.2.5 The Pitfalls of Pragmatic Reasoning
Although US law is beyond the scope of this thesis, a brief digression to consider the
US case Aalmuhammed v Lee is worthwhile, because it provides an interesting example
of the influence of pragmatic concerns about joint ownership on determinations of the
authorship of a film157. In that case, Jefri Aalmuhammed claimed to be a joint author of
Malcolm X, a film directed, produced and written by Spike Lee (with others). Mr
Aalmuhammed had assisted Denzel Washington prepare for the film; reviewed the
script suggesting extensive revisions to ensure historical accuracy; directed the actors
on occasions; created two additional scenes; translated for subtitles; supplied his voice
for voiceovers; edited parts of the film in post-production and provided technical help
in relation to the location. Although he had asked for a credit as a co-writer, he was
listed as ‘Islamic Technical Consultant’ far down the list of credits.
Circuit Judge Kleinfeld acknowledged that Mr Aalmuhammed had made
substantial contributions, but held that he was not a joint author of the film. In coming
to this decision the judge considered that the test for joint authorship ought to be
applied in a much stricter manner in the case of a multi-author work. In addition to the
usual requirement of having made an original copyrightable contribution, a joint author
must have ‘super intended the work by exercising control’; there must be objective
manifestations of shared intent to be co-authors; and the audience appeal of the work
must turn upon the contributions such that the share of each in its success cannot be
appraised158. The judge specified that ‘control in many cases will be the most important
157
202 F3d 1227 (9th Cir, 2000).
158
ibid [22].
211
factor’ 159 . The judge was explicitly concerned with facilitating the exploitation of
highly collaborative works. In particular, he worried that contributors might hold a film
entrepreneur to ransom, or that film entrepreneurs might be discouraged from
consulting with others for fear of losing control of the work160. The judge considered
that this test ‘would generally limit authorship to someone at the top of the screen
credits, sometimes the producer, sometimes the director, possibly the star, or the
screenwriter – someone who has artistic control’161.
Dougherty has criticised this decision as a missed opportunity to formulate a test
for the authorship of joint works that fairly rewards contributors for their
contributions 162 . The court implies that there is a higher standard of authorship for
motion pictures, appearing to imply that only one person might be considered an
author163. The case falsely equates creative control with sole authorship, in a way that
seems likely to mean that few of the most collaborative types of works will be
considered joint works 164 . This approach required the court to deny many of the
fundamental concepts of copyright authorship165. Although the court’s intention was to
provide clarity, in departing from established precedent it seems to leave the law in a
more uncertain state. Dougherty considers the control standard (understood in terms of
159
(n157) 1235.
160
(n157) [27-28]. The concern in the US relates to the risk of licences being granted too readily, rather
than to hold-ups (because a joint owner might licence a work without the permission of the other joint
owners), see n119.
161
(n157) 1233. This is much more explicit reasoning than can be found in the UK caselaw, see 2.2.
162
Dougherty (n15).
163
ibid 277.
164
M LaFrance, ‘Authorship, Dominance, and the Captive Collaborator: Preserving the Rights of Joint
Authors’ (2001) 50 Emory LJ 193.
165
Dougherty (n15) 275 argues that the court significantly departed from precedent.
212
the ability to accept or reject a contribution) to be particularly unhelpful because
although control is an important factor in determining whether there is a work-for-hire,
it is irrelevant to authorship166. This approach creates the possibility that a film might
have no authors where the person who has control does not also contribute the required
creative expression167. Dougherty considers that the complications that might arise from
a finding of joint ownership would have been better dealt with by reconsidering the
prevailing interpretations of the consequences of joint ownership168. This argument is
supported by the findings of this thesis that distorting the application of the joint
authorship test to reduce the number of potential joint owners is likely to result in side
effects that are substantially worse than the disease it aims to cure.
A recent case, Google v Garcia, hints at some of the strange results that might
follow from this view of joint authorship. In that case, the Judge Kozinski held that a
claim that an actress had copyright in her performance in a short film was likely to
succeed 169 . He did not impose the higher standard for originality developed in
Aalmuhammed because her claim was in respect of her performance and not in relation
to the joint ownership of the film as a whole. She was held to have granted the
producer an implied licence to reproduce her performance in the anticipated film,
although this licence did not extend to the offensive short film that was in fact produced
because it differed so radically from anything that she could have imagined when cast.
Although Judge Kozinski recognised the creative contribution of the actress this is an
166
ibid 279. Control is also an important factor in distinguishing contributions to fixation from
contributions to original expression (so a scribe is not an author).
167
Dougherty, ibid 280 considers that this may mean that commercial motion pictures may rarely be held
to be joint works.
168
ibid 281, 306, 319.
169
Note the equivocation in the amended opinion filed 11 July 2014 at 11.
213
extremely unsatisfactory decision as it involves creating an additional layer of copyright
protection. A restrictive approach to the joint authorship test results in real creators
being denied their rights. In this case the court has sought to compensate for this by
creating new levels of protection. Yet this appears to be an inadequate response that
does not fully capture the collaborative nature of a film. The fragmentation of copyright
in film in this way has also been apparent in the UK law, discussed above (6.2.2). A
restrictive approach to authorship that flies in the face of creative realities tends to push
judges to fragment the copyright work in order to provide a remedy for creators.
These decisions show the difficulties judges have in serving the two contradictory
purposes of concentrating rights to ensure efficient exploitation and protecting the
interests of those who have made creative contributions to a work.
6.3 Private Ordering
The practical problems of determining the authorship of a film are elided by the
combination of the employer’s presumptive ownership of copyright in the works
created by his/her/its employees and the use of (express or implied) contracts to transfer
copyright from creators to producers/exploiters170. The significant costs involved in the
production and distribution of films means that it makes sense to concentrate rights in
this way171 . Given the complexity of copyright interests that may subsist in a film,
production companies rely upon contracts with all participating parties to ensure that
they are assigned ownership of intellectual property rights in any copyrightable
170
Dougherty (n15) 238, 317-18, 327-33; Garcia v Google (Unreported, Docket Number: 12-57302) (9th
Cir, 2014) 12.
171
Schwab (n42) 149.
214
contributions made to a film. Under the classic studio system that prevailed in
Hollywood from the 1920s to the late 1940s most ‘talent’ was employed under long-
term contracts with a studio. Nowadays, contributors tend to be hired as independent
contractors rather than employees, so the importance of dealing with copyright
ownership issues by contract is heightened172.
Contracts tend to be very effectively used to regulate ownership issues in relation
to film173. Indeed, this is one of the reasons given in the European Commission’s 2002
opinion that there is no need to further harmonise film protection174. The Commission
stressed both the practical need to place rights in the hands of the producer and the need
to respect the basic principles of author’s rights protection (granting copyright to
creators)175. Contract provides a helpful way of ensuring both imperatives. The default
entitlement under copyright law is important because it increases the ability of creative
contributors to gain recognition for their contributions and helps them to improve their
bargaining position in other negotiations (conditions, remuneration, etc)176. There is an
incentive for contributors to agree to transfer ownership rights to producers as ensuring
broad and effective exploitation also serves the interests of contributors177. Indeed, it is
172
G Poll, ‘Harmonization of Film Copyright in Europe’ (2002) 50 J of the Copyright Society of the USA
519, 532.
173
G Dworkin, ‘Authorship of Films and the European Commission Proposal for Harmonising the Term
of Copyright’ (1993) EIPR 151, 152 noting that this is the reason for the lack of litigation. Bently and
Sherman (n33) 124.
174
Report on the question of authorship of cinematographic or audiovisual works in the Community
(n31) 691: ‘In practice, potential difficulties in exploitation of the works that arise due to the fact that
there may be more than one author, are overcome by contractual arrangements.’
175
ibid 4.
176
Dworkin (n173) 153.
177
EI Obergfell, ‘No Need For Harmonising Film Copyright in Europe?’ (2003) 4 European Legal Forum
199, 200.
215
rarely possible for contributors to a film to monetise their contributions separately from
the film178. This incentive to cooperate is often incorporated into film contracts that link
payment of a part of copyright compensation with the success of a film. Uncertainty
about the scope of copyright interests and conflict of laws issues that arise with
transnational productions provide a strong incentive to compel producers to ensure that
issues related to the ownership of intellectual property are settled in advance179.
Screen credit tends to be governed by collectively bargained agreements with the
various talent guilds180. This allows some of the benefits of authorship to be shared
more widely than copyright law appears to allow, and it helps contributors improve
their bargaining position181. There are a number of different guilds that represent many
contributors to a film182. The guilds have a number of collective bargained agreements
with standard terms that reflect a compromise between the various parties. These
agreements cover issues related to working conditions, pay and credit. There is a guild
that represents most of the major contributors to film: writers, production crew, actors,
musicians, cinematographers, producers, directors, independent production companies,
editors, etc183. There are craft guilds that represent location scouts, camera technicians,
178
Schwab (n42) 149.
179
Obergfell, ibid 201 disagrees with the Commission’s opinion that there is no need to harmonise on this
point for this reason.
180
C Fisk, ‘The Jurisdiction of the Writers Guild to Determine the Authorship of Movies and Television
Programs’ (2010) 48 English Language Notes 15, <http://ssrn.com/abstract=1694043> 8.
181
Fisk (n4) 250.
182
ibid 221.
183
These include: Writers Guild of Great Britain; Broadcasting, Entertainment, Cinematograph and
Theatre Union; EQUITY; Musician’s Union; British Society of Cinematographers; Production Guild;
Directors Guild of Great Britain; Producers Alliance for Cinema and Television; Guild of British Film
and Television Editors.
216
stunt action co-ordinators, etc184. The Guilds organise collective strike action to protect
the interest of their members, for example, the 2008 Writers Guild of America strike on
the issue of the calculation of residual payments for the reuse of writers’ work in new
media185. The rules of the Writers Guild can provide a valuable source of information
about the relative value of different writers’ contributions186. The Guild procedure is
limited, however, in that its effectiveness may be undermined by external powers as
occurred with the blacklisting of certain writers during the Cold War187.
The film case study shows a splitting of the financial and the reputational
incidents of authorship188. Whilst studios have fought to retain control and ownership of
films, they have been prepared to support a complicated system of self-regulation of the
credit associated with a film. Authorship credit is significant because it can affect, for
example, how contributors get paid, their future employment prospects, how they are
regarded within the community, and how films are interpreted by critics 189 . Fisk
analyses the bureaucratisation of the process of determining authorship credit by the
Writers Guild of America set out in the Screen Credits Manual190. The processes of the
Guild are determined democratically by its members. The Guild’s view of authorship in
large part ‘treats authorship as an historical fact reflecting degrees of creative
184
For a list of film industry trade union bodies see Copinger et al [26-246].
185
Fisk (n180) 13 arguing that this is because they are fairly transparent, arise as a result of participatory
processes, are fairly equitable and provide due process. On the usefulness and limitations of incorporating
social norms in legal determinations, see 8.1 and 8.2 below.
186
Fisk (n9).
187
Fisk (n4) 231.
188
As in the case of Wikipedia (Chapter 3) and scientific collaborations (Chapter 5).
189
Fisk (n180).
190
ibid 7.
217
contribution to be deduced based on the significance of the various writers’ work’191.
Guilds also seek to restrict the number of people that can be credited as a writer so that
the value of credits is not too diluted192. The Guild offers an arbitration service, much
like the Writer’s Guild of Great Britain. These rules represent a trade-off between the
benefits of attributing authorship to contributors and the benefits of concentrating the
financial and reputational rewards/punishments with those whose contributions are the
most significant 193 . This self-government of credit resembles the practices of large
particles physics collaborations (see 5.2.2). Similarly, it results in people being listed as
authors on works that have been substantially changed following their contribution194.
A crucial difference with the case of particle physics is the limit on the number of
credited writers, which has been adopted to enhance the status of writers vis-a-vis
directors 195 . This may result in a list that is under-inclusive from a copyright
perspective.
Film studios rely on a mix of soft and hard contracts to govern relationships
among the multiple parties that supply inputs to any film project196. Oral contracts are
most frequently used with high-value talent such as actors and directors, even where
191
ibid 8.
192
Dougherty (n15) 285, n322.
193
Fisk (n180) 11; (n4) 244.
194
Fisk (n4) 218 gives the example of Robert Towne who won the Oscar and got sole credit for the
screenplay for Chinatown even though the owner of the film had changed the ending to one that he had
not written and did not want.
195
ibid 244.
196
See Barnett (n29) for a justification of the economic rationales for doing so.
218
large sums are at stake197. Unsigned ‘deal memos’ can be used as the basis of an oral
contract, especially where it is necessary to act while an idea is ‘hot’198. Oral contracts
might be used for a number of reasons. They are likely to be expedient, as they are not
too costly in terms of time or momentum199. As the film industry is fairly close-knit,
parties may feel uncomfortable demanding a written and signed agreement, as this may
imply distrust of the other party. This will be particularly important in relation to
powerful individuals who may have a great degree of clout and a sensitive ego 200. Oral
contracts might also be relied upon out of habit, as parties are likely to defer to the norm
where it is too difficult or costly to work out whether an oral or written contract is
best201. This explains why some contracts are not formalised, even where this might be
relatively easily achieved.
Barnett provides an interesting analysis of the use of ‘soft contracts’ in
Hollywood, which he describes as: ‘a mix of legal and reputational governance situated
between the standard alternatives of short-term contracting governed solely or primarily
by law and repeat-play relationships governed solely or primarily by reputation’ 202 .
Parties use strategic ambiguity as to the existence of a contract to manage the ‘chicken-
197
GM McLaughlin, ‘Oral Contracts in the Entertainment Industry’ (2001) 1(1) Virginia Sports and
Entertainment LJ 101, 103, 118 noting that Charlton Heston did not sign a complete contract prior to
filming 60 of his films.
198
ibid 119, 126.
199
ibid 127.
200
ibid 130.
201
E Posner, ‘Norms, Formalities, and the Statute of Frauds: a Comment’ (1996) 144 U Pennsylvania
LRev 1971, 1974.
202
Barnett (n29) 19 argues that they are adopted as ‘a prudent response to a challenging environment
characterized by three major features: (i) high holdup risk and project risk, (ii) high specification and
enforcement costs, which reduce the efficacy of formal contract; and (iii) positive but weak reputational
constraints, which demand some recourse to formal contract’.
219
and-egg problem’ caused by the multiple inputs, severe uncertainty and extended timing
of a film 203. A studio or outside investor may not be willing to commit until a star is
signed up; the star may not be willing to commit until the investor is signed up; and the
distributor may not be willing to commit until both the star and investor are signed up,
etc 204 . To address this problem studios use time-limited option contracts which
condition performance obligations on the occurrence of certain events; or they use a
mix of uncertain communications which constitute an ‘unsigned deal’205. Soft contracts
are most prevalent between studios or other production entities and higher-value ‘star’
talent (such as, actors and directors). They tend to be effective because of the
availability of reputational sanctions for unreasonable behaviour. These are particularly
effective in small close-knit industries where parties may need to deal repeatedly with
the same players, as is the case with the film industry where a few major studios
dominate206. In this context, a bad reputation might significantly harm a contributor’s
career progression.
Although the film industry manages authorship issues very well through a mixture
of hard and soft contracts, there are limitations to private ordering. Parties in weaker
bargaining positions are not always able to adequately protect their interests. This can
be seen by the need to develop collective bargaining organisations. Their endurance and
success is a testament to the need they respond to.
203
ibid 9.
204
ibid 10.
205
ibid.
206
McLaughlin (n197) 130.
220
Private ordering has been very successful at regulating authorship in film. It has enabled
a separation of the economic and reputational incidents of authorship. Both the success
of private ordering, and the importance of the noneconomic incidents (ie, reputational)
incidents of authorship demonstrate that the assumptions that the instrumental approach
relies upon are mistaken207. Private ordering provides a flexible solution that can be
tailor-made to the context. In the case of film, for example, authorship is dispersed in
film credits so as to provide an incentive and reward contributors in a gradated manner.
Private ordering can compensate, in part, for the uncertainty in the law.
Copyright law, nonetheless, has an important normative role to play in providing
good authorship standards 208 . Copyright law sets the default position from which
contributors must bargain and, thus, it influences the behaviour of parties in
negotiations on issues of credit, remuneration and ownership209. These rules shift the
dynamics in negotiations when they favour one party over another as bargaining occurs
in the shadow of the law. Large motion picture companies are not known for their
generosity to relatively minor participants in the film-making process210. In this context,
it may not be advisable to leave it to private ordering to remedy the defects in film
copyright 211 . As has been argued in the preceeding Chapter, copyright law has a
207
See further: 8.4.3.
208
Handler (n72) 182-192 on the importance of this role for copyright law in the context of film.
209
Decherney (n3) discusses the impact that copyright law has had on business models in the US film
industry over time.
210
Dougherty (n15) 325. See, for example, the dispute in Aalmuhammed v Lee (n157) and the facts of
Slater v Wimmer (n111).
211
Handler (n72).
221
particularly important potential role to play in protecting the interests of creators who
are in a less favourable bargaining position.
In the commercial film industry, authorship historically tended to gravitate to
powerful figures with control over the film-making process. The legislative solution to
the problem of film authorship appears to adopt this industry custom. This view of
authorship has been contested in the film industry and in film scholarship as
unnecessarily narrow, eliding the important contributions of many others. This can be
seen in the activities of the guilds, particularly, the Writers’ Guild’s strong reaction to
the use of possessory credits. This resistance is mirrored in film theory that has
generally abandoned auteurism in favour of more inclusive approaches.
Film studies takes a normative approach to film authorship which considers the
reality of creativity, rather than the pragmatic approach taken when the aim is to
facilitate exploitation. Although the copyright law’s purposes differ from those of film
studies, there are good reasons for copyright to favour a normative approach over the
pragmatic one. Pragmatic solutions preferable to the parties can be achieved through
private ordering. The best role for copyright law to play in this context is to provide a
default minimum standard to aid creators in their negotiations with investors and
disseminators who tend to be in a better bargaining position. The norm of holding the
director and the producer solely responsible for a film does not provide a good
authorship standard because it is not representative and substantively fair212.
212
J Rothman, ‘The Questionable Use of Custom in Intellectual Property’ (2007) 93 Virginia LRev 1899
and 8.2 below.
222
Yet, the legislative solution to the problem of determining the authorship of a
film, primarily responds to the practical problem that film producers need to be in
possession of all relevant rights to be able to exploit a film commercially. This has led
to laws being designed to facilitate this task and to minimise disruption to commercial
practices, seemingly with little regard to protecting creators213. Yet, Michael Handler
points out that there is little evidence of a causal relationship between particular legal
designations of authorship and/or ownership of film copyright and the costs involved in
managing rights for the exploitation of films214. Copyright is often accused of being too
preoccupied with protecting disseminators, publishers and others who invest in creative
works, rather than those who actually create them. Film copyright seems to provide
support for this view.
The pragmatic legislative approach does not avoid uncertainty by designating two
contributors as the only authors of a film215. The restrictive interpretation of the work
favoured by the Gregory Report to avoid the problem of collective authorship has been
reversed in light of the realities of film (Norowzian). I argue that this has occurred, in
part, because the legislative solution clashes with copyright law’s imperative to protect
creators. In response to this imperative, judges have found two potential safety valves
by recognising that: (i) contributors might own copyright in their separate contributions;
and (ii) many films will embody or record dramatic works (which will have true
creative authors). In light of the contextual approach to the joint authorship test which I
213
Handler (n72) 177 describing film fixation copyright as a ‘triumph of pragmatism’.
214
ibid 183. This is likely to be due to the fact that issues of authorship and ownership generally tend to
be successfully managed with private ordering.
215
See Slater v Wimmer (n111), although this is probably an unusual case. Most of the uncertainty is due
to the complexity of underlying works.
223
argue for in this thesis, the common design of the joint authors of the dramatic work can
be used as a guide to determining which of these two valves is most appropriate in
respect of a particular contribution.
One might suggest that the copyright law position does not matter given that the
film industry appears to be able to rely very successfully on contract216. Yet, sometimes
contracts will not be agreed as in Slater v Wimmer and Aalmuhammed. Copyright has
an important role to play in acknowledging and rewarding creators. Instead of
concentrating on the possibility of unreasonable claims of joint authorship by relatively
minor contributors, it would be better to focus on providing copyright protection to all
of the creators who merit that protection. This reasoning is particularly compelling
considering that the current legislative solution does not actually prevent a lock-in
situation (see Slater v Wimmer). Private ordering might also be affected by power
imbalances. Although this might sometimes be addressed by collective bargaining,
there remains an important role for copyright law in providing a minimum standard of
protection for creators. This can help improve the quality of private ordering and
provide recourse for creators where no written agreement exists.
The treatment of film provides an example of the tension between two competing
conceptions of it: (i) as an industrial product which is made by many contributors
organised by a few entrepreneurs; and (ii) as a work of art which is the product of one
or two romantic authors. Modern conceptions of film authorship in film studies
scholarship and in the film industry increasingly resist this false dichotomy (to different
degrees). Copyright law appears caught in the dichotomy, conceptualising film in part
216
Dworkin (n173) 151, 152.
224
as an industrial product (the fixation); and in part as a work of art in the many layers of
copyright protection for individual contributions (which I argue would be the likely
corollary of a restrictive approach to the application of the joint authorship test to the
dramatic work). Instead of striving to choose between these two conceptions, copyright
law ought to take a third way and embrace an inclusive, contextual approach to
determining the authorship of a film as a dramatic work.
The instrumental/pragmatic approach to the joint authorship test is likely to lead
to authorship gravitating to the most powerful players, or those with the most control. In
discussing Aalmuhammed I suggested the dangers of using control as a proxy for
authorship. Screenwriters, for instance, provide a very significant creative input, yet
notoriously lack any control217. Proper recognition of the range of collaborators who
actually create a film as joint authors of that film is clearly preferable to the copyright
thicket that results from attempts to accommodate their legitimate interests by creating
new layers of copyright. The common design of the contributors to a film ought to be
the focus of the application of the joint authorship test and provides a better point of
reference for determining whether a contribution is ‘significant’ enough to make the
contributor a joint author; and also whether the contribution is ‘distinct’ (thus possibly a
separate copyright work).
217
Dougherty (n15) 294.
225
Chapter 7: Characteristics of Collective Authorship and the
Role of Copyright Law
This Chapter draws upon the prior case studies to discern a number of characteristics of
collective authorship and features of the treatment of such authorship by copyright law.
These are dealt with in five sections as outlined below.
The first section of this Chapter sets out the special features of the creative
process in collective authorship. When groups work together to create there tends to be
a division of labour with creative control shared among a number of contributors. This
makes any attempt to identify one or two authors a difficult, highly artificial task;
therefore, a more inclusive approach to determining authorship better reflects the
creative realities of collective authorship. Collective authorship groups also rely upon
social norms to coordinate and regulate their creative activities. These social norms
provide a useful source of information to help answer questions of fact that relate to the
creative process.
The second section explores the different meanings of ‘authorship’ in each
collective authorship group. In light of the diverse meanings of authorship, the factual
specificity of the joint authorship test in copyright law appears to be a feature, not a
defect. Despite these different meanings of authorship, in each case, authorship shares a
core feature in that it signifies responsibility for what is considered valuable about the
work. As what is considered valuable depends upon the particular group, the
significance and nature of a contribution is best understood contextually in relation to
226
the common design of contributors. Authorship is often associated with belonging to a
particular group and usually entails reputational benefits in that context. These benefits
are frequently more important to creators than ownership of the copyright interest in a
work of collective authorship. As the attribution of authorship may be affected by
power dynamics within authorial groups; copyright law provides an important
independent minimum standard for the legal recognition of authorship.
The third section sets out two gaps between copyright law and the reality of
collective authorship. First, copyright law assumes that authors desire control over
reproductions of their works or financial benefits, whereas contributors to works of
collective authorship are more concerned about the reputational consequences of
authorship. This undermines the incentive rationale for copyright law and reinforces the
importance of maintaining separate concepts of ‘authorship’ and ‘ownership’. Second,
the current approach to the joint authorship test is a poor fit for collective authorship
because there is not enough emphasis given to the collaboration limb of the test.
The fourth section considers private ordering mechanisms that have been adopted
to bridge the gap between copyright law and the realities of collective authorship. I
consider the benefits, as well as the limitations, of relying on private ordering. The
success of private ordering reveals that the instrumental/pragmatic approach is based
upon the mistaken assumption that finding that there are many authors of a work will
cause logistical problems that impede the exploitation of the work. This analysis also
reveals that although copyright can rely on private ordering to sort out many of the
practical problems that might arise in cases of collective authorship, there remains an
important role for copyright law. In particular, copyright provides an important source
227
of default standards that can be used to bolster the bargaining position of authors with
other more powerful players.
Finally, the fifth section further investigates the argument that copyright law has
an important role to play in the regulation of collective authorship, finding that its role
sometimes appears complex. Copyright law provides an important source of standards
and copyright law’s rules about authorship influence creative communities, although
not always in straightforward ways.
7.1 The Nature of Collective Authorship
Collective authorship differs from individual authorship, and even from collaborations
involving only two or three people. This is because collective authorship requires: (i) a
division of labour; (ii) the responsibility for the creative or intellectual content of the
work to be shared among many contributors; and (iii) social norms which regulate the
creative process. In this section, I argue that the special characteristics of collective
authorship provide two important insights for copyright law. First, the search for one or
two authors (the instrumental/pragmatic approach) requires one to completely disregard
the reality of the creative process in cases of collective authorship; therefore, an
inclusive approach to the application of the joint authorship test is preferable. Second,
social norms provide a useful source of information about the creative process that can
help answer questions of fact and therefore, improve the quality of decisions on joint
authorship.
The first characteristic of collective authorship – a division of labour – arises by
necessity from the fact that a large number of people contribute to the work.
228
Contributors to works of collective authorship often make highly-specialised, diverse
contributions. In large scientific collaborations no one contributor is likely to have
sufficient expertise to understand every aspect of the collaboration’s work. Similarly,
films require contributions from a broad range of professionals, artisans and
technicians. This division of labour may be structured hierarchically as is the case for
many commercial feature films; or it may develop in an organic, ad hoc way as it does
on Wikipedia. As in the case of Wikipedia there may be fluidity in the division and
distribution of labour; or the division of labour may be fixed as it is in indigenous
communities, where the nature of a contribution to an artistic work is limited by factors
such as descent, initiation and permission from elders.
Although a division of labour is bound to arise when more than one person
creates a work, the division of labour is particularly important for large groups. The
particular ways in which large groups share the labour that goes into creating a work
reveals much about how they ‘author’ works. The large numbers of contributors means
that the division of labour has two other important characteristics: contributors tend to
have significant autonomy in relation to the form of their contribution; and the group
must adhere to rules which help coordinate and regulate the creative process. Because
of these characteristics, the creative process of a group is relevantly different from the
creative process of an individual creator. In other words, it is not generally accurate to
conceive of collective authorship as involving many contributors acting as metaphorical
arms, mechanically implementing one person’s vision.
229
When many people work together to create, for practical reasons, each contributor
usually has some amount of control over their own contribution1. This tends to result in
the diffusion of creative control among many contributors. This is even more likely to
be the case where contributors are highly specialised, as they are in film and science. In
these cases, some amount of creative control must be delegated in order for contributors
to effectively perform their roles in the collaboration 2 . Even when contributors are
organised in a strict hierarchy, they often retain significant scope for creative or
intellectual input in relation to their own contribution3. This is because directions from
above require some interpretation or personal choices in their implementation, except in
the case of simple mechanical tasks.
The degree of creative control of each contributor is greatest in the case of
Wikipedia. Anyone can add, change or delete content from Wikipedia (although their
contribution, in turn, can easily be reverted or changed by another contributor). Even
where creative freedom is most strictly limited, as it is for Indigenous artists bound to
observe customary law, there is scope for the artist to add their own personal touches.
Indeed, the case law recognises that the creation of Indigenous art involves enough
skill, labour and judgment or intellectual creation to satisfy copyright law’s demands
1
Where large numbers of authors are involved it is often impossible, or at least impractical, for one
person to control the activities of each contributor. Thus, Wikipedia, for instance, relies on the collective
wisdom of a group of regular contributors that share common views about the nature of the project.
2
The head of a scientific research project, for example, may not have the knowledge or expertise to
understand the work of all its contributors. This means that he or she can only direct the activities of other
contributors in a general sense and must leave a large measure of ‘creative control’ in the hands of each
contributor.
3
One might take the example of film. Although a producer might express the desire that a costume reflect
a particular aesthetic, the costume designer might take inspiration from a variety of sources and come up
with costumes that reflect his or her view of how that aesthetic is best realised.
230
for originality4. In cases of collective authorship, authorship is distributed widely, that
is, many contributors make contributions of an ‘authorship-type’ in the copyright sense.
This remains the case even where it is possible to identify some contributors who have
more control over the creative process than others, such as, a director or film producer.
The fact that many contributors have some creative control over their contributions
provides support for an inclusive approach to determining the authorship of such works
for copyright purposes.
In contrast to the dominant position in copyright law, reflected in cases such as
Hadley v Kemp, the view that the work of a large collaboration might be authored by a
single individual that has exercised a controlling influence on the creative process is
considered outmoded in most of the non-legal scholarship reviewed in this thesis.
Historians of science have revealed the hidden technicians, whose valuable inputs were
erased from records of their work 5 ; and the important contributions of the many
scientists united behind one sacralised individual, eg the Alvarez Group 6 . These
practices no longer sit well with modern attribution practices in science which are much
more inclusive 7 . Similarly, film theorists have argued that auteurism is a fallacy,
exposing the artificial strategies used to prove that the director is the sole author of a
film. Even the most controlling film director’s directions are brought to life by a myriad
4
The question of the originality of Indigenous artworks was laid to rest in: Milpurrurru v Indofurn (1994)
54 FCR 240 (Fed. Ct. of Australia).
5
S Shapin, ‘The Invisible Technician’ (1989) 77 American Scientist 554.
6
P Galison, ‘The Collective Author’ in M Biagioli and P Galison (eds), Scientific Authorship: Credit and
Intellectual Property in Science (Routledge 2003) 329, 329-330; H Gusterson, ‘The Death of the Authors
of Death: Prestige and Creativity among Nuclear Weapons Scientists’ in M Biagioli and P Galison (eds),
Scientific Authorship: Credit and Intellectual Property in Science (Routledge 2003) 281, 282 makes a
similar point. He gives the example of the atomic bomb which was credited to Robert Oppenheimer
despite the key roles of a number of other scientists.
7
See Chapter 5.
231
of other contributors who bring their own perspective to bear on the interpretation and
implementation of those directions. Most modern film theorists accept that films are a
product of multiple authors, although they disagree on the best way of identifying those
authors. Despite the different meanings of the concept of authorship for different
creative groups and different academic disciplines, the scholarly literature is fairly
united in the trend away from the collapsing of authorship into singular romantic author
figures8. Copyright scholars are similarly critical of the influence of the romantic author
construct, but this critique has yet to be translated into concrete proposals for reform of
the joint authorship test9.
The fact that many contributors have some control over the creative, aesthetic or
intellectual content of the work results in a need to coordinate behaviour. This need is
solved by contributors observing the prevailing social norms of their creative
community; or, where these are inadequate, by developing social norms better adapted
to the group project. The Wikipedia community, for example, has developed its own
social norms to help coordinate behaviour and foster the community spirit that is
important in motivating contributors. Social norms establish the benefits, and police the
responsibilities, of authorship within the collective authorship group. They also embody
the group’s shared views about the creative process 10 . Social norms help determine
what distinguishes a mere contributor who might be acknowledged, from an author who
can be held responsible for the work.

8
This is no doubt due at least in part to the seminal works of Roland Barthes and Michel Foucault on the
death of the author: R Barthes, ‘The Death of the Author’ in S Heath (trans) Image, Music, Text (Fontana,
1977) 142; M Foucault, ‘What is an Author?’ in D Bouchard (ed) Language, Counter-Memory, Practice:
Selected Essays and Interviews by Michel Foucault (Cornell UP 1977).
9
See 2.4.
10
Norms have been described as the grammar of social interactions: C Bicchieri, The Grammar of
Society: the Nature and Dynamics of Social Norms (CUP, 2006).
232
Social norms can provide answers to the factual questions that form part of the
joint authorship test. Social norms provide a useful metric for assessing how the value
of an individual’s contribution is understood by the other contributors and even by its
target audience. They also give an insight into contributors’ shared views about the
creative process. Thus, they indicate when collaboration is present and provide
information about the common design of contributors. Social norms provide a
transparent basis for assessing the existence of collaboration and the significance of a
particular contribution for the purposes of the joint authorship test that does not require
judges to make value judgements about the aesthetic merit of the work, or of
contributions to that work. Social norms vary substantially from collaboration to
collaboration (compare, for example, biomedical science and particle physics
collaborations)11. By considering the social norms that apply to a particular collective
authorship group, the joint authorship test can be applied in a flexible way that is
sensitive to the creative context.
The application of social norms that regulate the creative process reveals an
important aspect of the nature of a work of collective authorship. Contributors make
their own contributions to the work but they also have a role to play in creating,
observing and enforcing social norms which govern the creative process. Such a work
would not exist but for the involvement of all the contributors in making their
contributions, but also by their actions and attitudes which establish the norms which
govern the group as a whole. In this way, a work of collective authorship is best
11
5.2.
233
conceptualised as greater than the sum of individual contributions12. When understood
from this angle, the search to find one or two controlling minds who might be identified
as the authors of a work of collective authorship seems to fundamentally miss the point
of how large groups work together in the creative process. Indeed, widespread
application of the instrumental/pragmatic approach is likely to undermine the credibility
of copyright law by making it seem disconnected from the realities of creativity13. In
the eyes of contributors to works of collective authorship, the selection of only one or
two authors, at best, appears arbitrary; and at worst, rewards powerful players at the
expense of true creators14.
7.2 The Different Meanings of Authorship for each Collective
Authorship Group
This section employs the case studies to enhance our understanding of what it means to
be called an author in the different cases of collective authorship (ie, how authorship is
attributed and what this signifies in each context). Four propositions emanate from this
analysis:
(i) authorship has a different meaning in each collective authorship group;
(ii) authorship signifies responsibility for the work (determined according to
community-specific criteria);
(iii) authorship is a status that signals belonging to a particular community; and
12
This maps onto the conception of a work of joint authorship assumed by the CDPA (p19). For this
reason, collective authorship cannot be seen as one of intellectual property’s ‘negative spaces’. For the
literature on ‘negative spaces’: see p8, n37.
13
See 8.1 and 8.4.3.
14
For example, the risk of authorship gravitating to producers and directors at the expense of other
contributors to a film, see Chapter 6.
234
(iv) the attribution of authorship can be affected by power dynamics.
I consider each proposition in turn, commenting upon its significance for copyright law.
7.2.1 Authorship Has a Different Meaning in Each Collective Authorship
Group
The dynamics of creativity are very different in each case study. The concept of
authorship has a different role to play in each creative community that is linked to the
community’s values. This means that authorship signifies different things in different
creative contexts. A contributor must satisfy different requirements to count as an
author; and there are different consequences (ie benefits and responsibilities) of being
credited as an author in each context. As the different meanings of authorship have been
considered in some detail in the previous Chapters, here, I only set out a few brief
points.
In Chapter 3, I argued that rather than being a space of non-authorship 15 , the
Wikipedia community has created its own concept of authorship that is tailor-made to
the project’s needs. On Wikipedia, authorship is an inclusive and community-orientated
notion built on sharing. It is a dynamic concept that is defined in terms of participation
in a project with altruistic goals. ‘Good’ Wikipedians (who adhere to its goals and
policies) receive rewards in the behind-the-scenes areas of Wikipedia; whereas ‘bad’
contributors might have their contributions ‘reverted’ or even be banned from
contributing.
15
As C Sunstein, Infotopia: How Many Minds Produce Knowledge (OUP 2006) 153 appears to suggest.
235
Australian Indigenous communities have a very different understanding of
authorship. For Indigenous people, the authorship of art is an act of religious
significance expressing and maintaining a relationship to the land. It is an important
way of preserving and transmitting significant cultural beliefs. Indigenous artists are
custodians of cultural knowledge. Although the ritual knowledge depicted in an artwork
belongs to the community as whole, individual artists are held responsible for
disrespectful uses of the artwork even where he or she had no control over, or
knowledge of, the use16.
In science, publications are the main means through which new claims are
presented to the scientific community to be evaluated, retested and eventually accepted
into the canon of scientific knowledge17. Authorship entails a credit value as a marker
of scientific achievement; and often also a responsibility value as the author is expected
to adhere to the requirements of the scientific method and other ethical obligations. The
responsibility value establishes the trustworthiness of particular claims, which helps
support science’s epistemological claim to produce knowledge or ‘truth’. Although
there is a common idea of the scientific author, there are different interpretations of who
fits the bill. This is evident by the fact that attribution practices vary significantly from
discipline to discipline and even from collaboration to collaboration.
The concept of authorship in film has changed over time as it has been constantly
under siege from different interest groups18. Historically, authorship tended to gravitate
16
p85.
17
5.1.
18
6.1.
236
to those who could exercise the most creative control, or to the entrepreneurs who have
invested in the project. In modern times, authorship is used as a marketing tool
(particularly in relation to directors with distinctive styles), because of its trade mark-
like function. Some actors’ names are used in a way that appears to imply quasi-
authorship, but actors are rarely held responsible for the overall film in the same way as
a director or a producer19. Although there is a trend towards a more inclusive approach
to film authorship, the question of authorship remains hotly contested. The atomisation
of contributions in film credits provides a useful compromise, as it gives the impression
that there are grades of authorship without defining the precise boundaries of the
20
concept . These credits have significant reputational consequences within the
industry21.
In light of the wide range of meanings of authorship and the different attribution
practices in collective authorship groups, the factual specificity of the joint authorship
test is an important feature. The factual specificity of the joint authorship test allows it
to be flexible to the dynamics which operate in different creative contexts. The
differences between the case studies also provides support for a contextual approach to
determining authorship, that is more likely to correspond to the creative realities of
collective authorship.
19
Film stars are often able to successfully distance themselves from ‘bad’ films by arguing that they
lacked ‘control’ over the end product. See: K Finbow ‘Nicole Kidman: ‘I had no control over Grace of
Monaco’, Digital Spy, 27 May 2014.
20
Chapter 6.1.
21
Indeed, some contributors will accept less remuneration for a better credit.
237
7.2.2 Authorship Signifies Responsibility for the Work
Although authorship has different meanings in different creative contexts, it has a stable
core feature: in every case authorship signifies responsibility for what is seen as
valuable about a work. What is valuable depends upon the authorial group and tends to
include many different contributions to the creative process. This suggests that the
common design of the authorship group might be a useful gauge of the relative
significance of a particular contribution. Further, the question of whether a contribution
is made to the ‘ideas’ or to the ‘expression’ of a work is best determined in light of the
creative context.
Whether a contributor is responsible for what is valuable about a work tends to be
assessed with reference to their role in the creative process (and not by reference to the
final form of the work considered in isolation). So, for example, creative contributions
to film are more highly valued than technical or mechanical contributions. The value of
a particular contribution to a scientific article is assessed by reference to the process-
orientated requirements of the scientific method as well as the ongoing ethical
requirements embodied in Merton’s norms. Similarly, a contribution to Wikipedia is
assessed by reference to its adherence to a set of community-developed policies and
norms that are crafted to reflect the project’s aims. Indigenous art is one step in a
process of handing down cultural knowledge to the next generation. The artist is
responsible for the faithful and respectful depiction of ritual knowledge; and subsequent
uses made of his or her artwork. This focus on process, suggests that it would be
unrealistic to try to assess the significance of a particular contribution to a work of
collective authorship without regard to the common design of the contributors. This
provides support for refocusing the joint authorship test towards the collaboration limb
238
and for a more contextual approach to determining the significance of a particular
contribution to a joint work.
The focus on value created in the creative process best fits with Justice Laddie’s
broad approach to the idea/expression dichotomy in Cala Homes. Justice Laddie’s
approach pays less attention to the mechanical process of fixation and concentrates on
the concepts, emotions and other content embodied in the form of the expression. In
Chapter 4, I argued that the court’s narrow approach to the idea/expression dichotomy
in Bulun Bulun might be the result of a cultural bias in favour of those contributors who
most closely resemble the romantic author 22 . Cultural biases against non-Western
modes of creativity seep into the decision because of the court’s acontextual approach
to the application of the idea/expression dichotomy. A broader, more inclusive
approach would have allowed for the recognition of the communal origin of artwork,
which made the artwork meaningful from an Indigenous point of view23. In light of the
special features of collective authorship identified in the first section of the Chapter, a
broader, contextual approach to the idea/expression dichotomy also better reflects the
process of creativity, which involves sharing responsibility for the creative or
intellectual content of the expression.
In some cases, the view that authorship signifies responsibility for the work has
resulted in a gradation of authorship. On film credits, producers and directors are
usually listed first so they probably count as authors. A contributor probably stops
22
Chapter 4, p75.
23
The contributions of the Indigenous community are evident in the expression of the artwork in the
same way that the design director’s contributions were evident in the drawings made by the draftsmen in
Cala Homes v Alfred McAlpine Homes [1995] EWHC 7, [1995] FSR 818. See Chapter 4, p77.
239
counting as an author towards the end of the list where one might find, for example, the
caterers and logistical support. The film industry uses this ambiguity to its advantage, as
a way of rewarding and incentivising contributors with credits that designate their
responsibility for specific aspects of the production process, whilst avoiding selective
distribution of the label ‘author’ 24 . In scientific collaborations, the order in which a
contributor is listed in the by-line communicates information to the scientific reader
about their importance and their role in the creative process. The first and last positions
are particularly prestigious. Copyright has a mechanism for recognising the different
shares of responsibility that contributors may have for the creation of a work, in the
possibility that contributors might be held to own unequal shares of a copyright interest
based on the quantity and the quality of their contributions to the work25.
As the dynamics of authorship vary greatly, the case studies suggest that it would
be inaccurate to evaluate a contributor’s contribution in isolation from its context in a
particular collaboration. This suggests that the collaboration limb ought to be given a
greater role in the application of the joint authorship test26.
7.2.3 Authorship Signals Belonging to a Particular Community
The attribution of authorship is used to signal belonging to, or an elevated status within,
a particular community. In many contexts authorship is used as a reward for
24
Although those who make contributions which are recognised to be ‘creative’ tend to find it easier to
establish that they are authors than those whose contributions are regarded as purely technical (6.1). The
connection between authorship and creativity is reflected in the CDPA both in the definition of the author
as the ‘creator’ of a work and also in the more generous scope of copyright awarded to authorial works as
opposed to entrepreneurial works (2.1).
25
There are no provisions, however, for a particular order in the attribution of authorship.
26
As argued at p19.
240
accumulated labour to a particular project (for example, scientific collaborations,
Wikipedia and film). Authorship signals belonging and status most strongly in the case
of Indigenous art. The whole process of the creation of Indigenous artworks (and the
preparation for their creation by initiation rites, etc) promotes community cohesion and
helps socialise young members of the group into a particular worldview. All members
of the tribe have an ongoing connection to the artwork and artists have a special place in
the hierarchy of access to cultural knowledge as custodians of ritual knowledge.
Authorship indicates belonging to a particular joint endeavour and is
inextricably linked to community norms27. The reputational consequences of authorship
within authorial communities are often far more important to contributors than their
copyright interest. Indeed, in the context of collective authorship, authorship is
sometimes consciously distanced from remuneration (Science, Wikipedia, Indigenous
art) and from control (Wikipedia). This view of authorship contrasts starkly with the
instrumental view of copyright law that assumes authors are motivated by the ability to
control reproductions of their works in order to recoup royalties (see 7.3 below). This
aspect of collective authorship also reinforces the need to keep the concepts of
authorship and ownership separate in copyright law, given the important ramifications
of authorship other than copyright ownership, for example, the right to attribution.
Furthermore, it provides an additional reason for an inclusive approach to the joint
authorship test, as the reputational value of authorship credit might often be more
important to contributors than ensuring the efficient exploitation of the copyright
interest.
27
For example, positive contributions to Wikipedia can build the esteem of a particular contributor
among the community of regular contributors; collectively such contributions help constitute the culture
of sharing which sustains Wikipedia (3.1). As to the constitution and content of such norms in collective
authorship groups, see 7.1.
241
7.2.4 The Attribution of Authorship Can Be Affected by Power Dynamics
The attribution of authorship in collective authorship groups may be affected by power
dynamics within those groups. This is particularly apparent in film and scientific
authorship. In the context of biomedical science, there are ongoing controversies about
inappropriate attribution practices, which tend to favour powerful players. Senior
scientists and heads of laboratories often ask to be added to this list of authors on an
article where they could not be reasonably considered responsible for the scientific
work in it. Sometimes these figures are added without their knowledge as a way of
increasing the likelihood of publication of the article. These practices are mirrored in
film where authorship tends to gravitate to the most powerful players such as the
director (with creative control) or the producer (with financial might and possibly also
creative control). The endurance of possessory credits, despite strong resistance from
many film contributors, is a testament to the ability of powerful players (in this case
directors) to claim more than their authorship share.
As demonstrated by the US case of Weissman v Freeman discussed in Chapter 5,
copyright law has the potential to act as a bulwark against such power dynamics by
ensuring that authorship is only granted to a person who has made a creative or
intellectual contribution that is evident in the expression of the work. In this way
copyright law can provide a source of authorship standards that are independent from
the power dynamics that operate in collective authorship groups. In the context of film
(Chapter 6), I argued that copyright might provide an important default position that
weaker parties can use to improve their bargaining power in contractual negotiations on
authorship issues such as credit or remuneration. Thus, copyright law has an important
242
role to play in supporting positive social norms and undermining negative ones that
may not be compatible with copyright law’s core notion of authorship.
7.3 The Gap between Copyright Law and Creative Realities
There is a gap between copyright law and the creative reality of collective authorship.
This gap is twofold. First, copyright law’s assumptions about the motivations for
authorship in most cases do not correspond with the primary motivations for creation in
cases of collective authorship. Second, the relative insignificance of the collaboration
limb in the application of the joint authorship test does not fit the creative realities of
collective authorship.
The fact that contributors to works of collective authorship are often motivated by
intrinsic reasons suggests that an instrumental view of copyright is incomplete. In the
different contexts of Wikipedia, science, and Indigenous art, the primary motivation for
creativity tends to be unrelated to the ability to extract royalties. Wikipedians contribute
to Wikipedia for fun, out of boredom, for altruistic reasons, or because they might gain
esteem within the community of regular contributors. In science, recognition by
academic peers and reputational benefits are key motivating factors for authorship.
Indeed, scientists consciously distance themselves from financial incentives because
disinterestedness is one of the foundational norms of science. Although the sale of
artwork can be an important lifeline for Indigenous communities, the creation of art is
first, and foremost, a matter of religious significance. Even in the context of film, the
most commercially significant case study, some contributors will accept less
remuneration in return for a better credit.
243
In some cases a sharing economy is an important factor in motivating
contributions28. The success of Wikipedia is often attributed to a sense of community
that is sustained by many individual acts ‘donated’ to the shared project. Indigenous art
is seen as belonging to the community and scientific articles are presented as gifts to the
scientific community. Even in the case of film, the possessory credit (which suggests
that a particular person is the author) is widely seen as an illegitimate power grab,
which undermines the sharing of authorship or quasi-authorship among many
contributors in the otherwise extremely inclusive film credits.
Credit is highly valued because of the reputational benefits that it might bring.
Contributors to works of collective authorship often value credit more than financial
benefits or control. The importance of credit is evident in the fact that particle physics
collaborations tend to establish authorship protocols to govern how credit will be
distributed before any measurements are taken. Similarly, scientific journals record
much academic debate about which contributors in large collaborative projects deserve
to be credited and the order in which they ought to be credited. This suggests that there
are good reasons to ensure that authorship and ownership are treated as distinct
concepts in the CDPA. This is because authorship leads to moral rights as well as
economic rights. The experience of collective authorship suggests that in determining
who counts as an author it is better to ensure that enough creators can access the right of
attribution, than to restrict authorship on the basis that this might streamline the
28
Sharing economies often operate in opposition to the market economy (eg, Wikipedia and science).
The opposition between these two different systems of exchange, might also explain arguments that
Indigenous peoples risk losing an important dimension of their traditional cultural expressions by trying
to fit them into Western categories of protection for cultural property which are premised upon its
commodification (4.4). On Copyright and the commodification of culture, see: A Barron, ‘The Legal
Properties of Film’ (2004) 67 MLR 177; F MacMillan, ‘The Cruel C: Copyright and Film’ (2002) EIPR
483.
244
economic exploitation of the work. This is further supported by the fact that ownership
issues tend to be successfully managed with private ordering (see 7.4 below).
The attribution of authorship is far more nuanced in creative communities than it
appears to be in copyright law. Copyright tends to assume that authorship is like an
on/off switch: you are either an author or you are not. In film and science, the order in
which contributors are listed is of great significance and affects the reputational value
associated with the credit. In this way, it is possible to have different degrees of
authorship according to the significance of the contribution to the whole. The current
approach to the joint authorship test lacks an appropriate framework for taking these
subtleties into account. In the following Chapter, I suggest a contextual approach to the
joint authorship test that would allow for the relative value of particular contributor’s
contribution to be taken into account when determining whether they qualify as a joint
author. At present, copyright law’s joint authorship test does not give enough weight to
the collaboration limb of the test. I argue that the common design shared by the
contributors can provide a valuable metric for determining the significance of a
particular contribution in the context of the activities of the collective authorship group
as a whole.
The thesis forms part of the growing body of literature that suggests that the
incentive justification for copyright law is inaccurate, or at least, incomplete 29 . The
incentive theory is often attacked on the basis that many creators do not rely upon
29
J Silbey, The Eureka Myth: Creators, Innovators, and Everyday Intellectual Property (Stanford Press
2014); D Zimmerman, ‘Copyright as Incentives: Did We Just Imagine That?’ (2011) 12 Theoretical
Inquiries in L 29; N Elkin-Koren, ‘Tailoring Copyright to Social Production’ (2011) 12 Theoretical
Inquiries in L 309; K Raustiala and C Sprigman, ‘The Piracy Paradox: Innovation and Intellectual
Property in Fashion Design’ (2006) 92 Virginia LRev 1687, 1762-65.
245
copyright law’s incentives. The activities of these creators are sometimes referred to as
‘IP’s negative spaces’, because they often relate to the works that do not fit within
copyright law’s rules of subsistence (for example, the activities of magicians, chefs,
comedians, etc) 30 . The case studies considered in the thesis provide a more direct
challenge to the incentive theory, because they concern copyright law’s heartland as
they involve works that fall neatly within the copyright law’s rules of subsistence. The
case studies also are not confined to small creative niches, but rather, concern activities
of significant cultural and economic importance. In light of the limitations of the
incentive justification for copyright law, this thesis provides support to the arguments of
some that the incentive story can only provide a shaky foundation for the copyright
expansionism that has occurred in its name31.
The case of collective authorship demonstrates that copyright law seems
disconnected both from the motivations for creativity and from the ways in which
authorial groups value their own creative activities32. In light of this disconnect, it is
unsurprising that most collective authorship groups turn to private ordering mechanisms
to regulate their own creative practices. The next section considers the benefits and
limitations of relying upon private ordering to regulate authorship issues. The following
section builds upon this analysis and looks at the complex ways in which copyright law
30
Rosenblatt (n12) provides an overview of the literature, drawing together common themes to construct
a theory of IP’s negative spaces. See also p8, n37.
31
Zimmerman (n29) 58.
32
This finding is supported by empirical scholarship: CJ Buccafusco, ‘On the Legal Consequences of
Sauces: Should Thomas Keller’s Recipes Be Per Se Copyrightable?’ (2007) 24 Cardozo Arts &
Entertainment LJ 1121; D Oliar and C Sprigman, ‘There’s No Free Laugh (Anymore): The Emergence of
Intellectual Property Norms and the Transformation of Stand-up Comedy’ (2008) 94 Virgina LRev 1787;
E Fauchart and EA von Hippel, ‘Norm-Based Intellectual Property Systems: The Case of French Chefs’
available at <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=881781>.
246
affects collective authorship groups: in particular, its role as a source of ‘good’
authorship standards, and a bargaining chip for creators in private ordering negotiations.
7.4 Private Ordering: Bridging the Gap between Copyright Law and
Creative Reality
In this section I consider the private ordering mechanisms that collective authorship
groups have adopted to bridge the gap between copyright law and creative realities.
Private ordering is a fairly broad term that can be understood as referring to a variety of
regulatory mechanisms which do not originate from the state33. In the case studies, the
attribution of authorship and its consequences have frequently been affected by such
mechanisms. These have included: journal guidelines, contracts, collectively bargained
agreements, protocols, codes of conduct, copyleft licences and social norms. In general,
the attribution and consequences of authorship tend to be fairly successfully managed
by collective authorship groups with private ordering mechanisms. This ought not to be
surprising, as Ellickson and others have shown that people often rely primarily on
norms and customs to govern their behaviour (rather than law)34.
I begin by considering the most successful examples of the regulation of
authorship issues with private ordering (Wikipedia, particle physics collaborations and
film). Then, I consider the least successful attempts at private ordering (Indigenous art
33
MJ Radin and RP Wagner, ‘The Myth of Private Ordering: Rediscovering Legal Realism in
Cyberspace’ (1998) 73 Chicago-Kent LRev 1295 noting that many such mechanisms rely upon the law of
property or the law of contract. S Schwarcz, ‘Private Ordering’ (2002) 97 Northwestern U LRev 319,
324.
34
In his seminal work on the social norms that operate among farmers and ranchers in Shasta County,
California, Ellickson suggested that legal theorists persistently overestimate the role of law. R Ellickson,
Order Without Law: How Neighbours Settle Disputes (Harvard UP 1991) 137-8, 280-81. See also E
Posner, Law and Social Norms (Harvard UP 2009).
247
and biomedical science collaborations). Finally, I compare the benefits and the
limitations of private ordering. This analysis reveals two main insights for copyright
law. In particular, the success of private ordering measures suggests that the
instrumental/pragmatic approach to the joint authorship test is based upon the mistaken
assumption that groups of creators cannot themselves effectively organise the
exploitation of a work. Further, copyright law has an important role to play in
minimising the limitations of private ordering and, in particular, it might provide a
bulwark against the effects of power imbalances in collective authorship groups.
7.4.1 Successful Examples of Private Ordering
The most successful examples of the regulation of authorship with private ordering
mechanisms are Wikipedia, particle physics collaborations and film.
Copyleft licences are crucial to the success of Wikipedia, because they facilitate
the highly iterative process of creation that it relies upon 35 . Each contributor, when
submitting their contribution to the site, must agree to allow anyone to reproduce,
change or delete their contribution for free at any time in the future. This provides
subsequent contributors with a defence to potential future claims of copyright
infringement in respect of previous contributors’ contributions. Copyleft licences allow
for the constant re-working or refactoring of Wikipedia in real time (without the need to
pay royalties or ask permission). The licences alleviate the uncertainty about who might
count as an author by treating all contributors equally. They allow the community to tap
into some of the values of copyright law’s notion of authorship whilst discarding others.
35
3.3.
248
They also provide for ways of attributing authorship that are better suited to the internet
age36.
Copyleft licenses also have a more subtle role to play on Wikipedia. The
association of copyleft licences with the Creative Commons and free software
communities bolsters the sharing ethic which motivates many contributors to
Wikipedia. This creates a community spirit that encourages more positive than negative
contributions to Wikipedia. The role of copyleft licenses as an embodiment of
Wikipedia’s social norms is far more significant than their role as legal instruments.
The most significant force governing the social incidents of authorship are these social
norms, which are developed and implemented by regular contributors. Contributors
create and enforce these norms when they engage in the behind-the-scenes parts of
Wikipedia (for example, awarding each other barnstars, reverting the contributions of
vandals, or participating in the resolution of disputes).
In the context of science, protocols developed by particle physics collaborations
have been effective in regulating the complex issues that arise in relation to authorship
in large collaborations 37 . These protocols enshrine mechanisms of ensuring group
responsibility for the scientific content of collectively authored articles. Preliminary
versions of articles, for example, may be posted on an internal webpage where all
members of the collaboration may comment on them. Mechanisms for determining the
attribution of authorship vary depending upon the context, for example, attribution
practices differ for journal articles and conference papers. These protocols have enabled
36
For example, by including a hyperlink to the Wikipedia page concerned, see p56.
37
5.2.2.
249
particle physics communities to use credit liberally as a reward for accumulated labour,
whilst ensuring that the group takes responsibility for the scientific content of articles.
As these protocols are often determined before most of the research has been
undertaken, they shape the expectations of all contributors who join the collaboration
from the outset.
In film, private ordering mechanisms also tend to be very successful 38 . Most
authorship issues appear to be quite well managed with contracts. This is important
given the high risk, high investment nature of the film business. The ability of
contributors to increase their bargaining power by joining guilds has helped to improve
the quality of contractual negotiations. Collective bargaining by the talent guilds has
been a significant force in the negotiation of terms for members on issues of pay,
conditions and credit. Collective bargaining agreements promote cooperation and even
up the unequal bargaining power between contributors and production companies. Of
course, those who are not represented by a guild may struggle to negotiate favourable
terms, as can be seen in the visual effects sector. Contributors in more favourable
bargaining positions sometimes rely on soft contract measures, using option agreements
and unexecuted contracts as a way of promoting cooperation, whilst maintaining
flexibility. In doing so the parties rely upon the prevailing industry social norms,
specifically, they expect the possibility of reputational sanctions to deter unreasonable
conduct by other parties.
38
6.3.
250
7.4.2 Less Successful Examples of Private Ordering
The least successful examples of the regulation of authorship issues with private
ordering are Indigenous art and biomedical sciences.
Within ‘traditional’ Indigenous communities, customary law is effective at
regulating the creative process and the consequences of authorship 39. Yet, as customary
law is not enforced by the Australian legal system it is ineffective vis-à-vis outsiders.
Most scholars agree that copyright law cannot provide adequate protection for the range
of interests that Indigenous communities seek to protect in their artwork. Indigenous
communities have sometimes turned towards private ordering as a way of addressing
these gaps. Codes of conduct and protocols have only tended to be successful in
regulating the conduct of those who are already motivated to respect Indigenous
communities’ interests. There are limits to the effectiveness of relying upon contract,
because Indigenous artists are likely to be in a relatively weak bargaining position.
Although there are some examples where contract has been successfully used40, as for
codes of conduct, this approach is most likely to work with those who are already
motivated to respect Indigenous cultural interests. Even where favourable contractual
terms can be agreed, they do not burden third parties to the contract. Attempts to adopt
a collective trade mark have also been unsuccessful. Although the main reason that this
scheme was abandoned was because it was poorly drafted and ineffectively publicised;
it is unlikely that a market-based solution would be effective as consumers may not
39
4.3 above.
40
M Rimmer, ‘The Bangarra Dance Theatre: Copyright and Indigenous Culture’ (2000) 9(2) Griffith
LRev 274.
251
make the effort to seek out and pay more for genuine Indigenous art (particularly in the
tourist market)41.
Biomedical journals’ have adopted guidelines that attempt to restrict the number
of contributors who can be credited as authors on articles to those who have made
contributions at a number of different stages of the process (the triple lock formula)42.
These appear to have been fairly unsuccessful, because they do not reflect the reality of
collective authorship. It may be impossible for one person to completely understand, let
alone be held responsible for, every part of a complex study. Restricting authorship to a
few contributors fails to recognise the nature of scientific collaborations which
necessarily involve a division of labour wherein many individual contributors have a
significant degree of autonomy in relation to their own contribution. Further, credit can
be a powerful way of facilitating collaboration, encouraging and rewarding contributors
for participating in a joint project. In light of these realities, another reason for the lack
of success of the guidelines is that there is no genuine consent to them by the people to
whom they apply43. In addition, the guidelines are ineffective at addressing the power
dynamics which may operate in biomedical collaborations. Less powerful contributors
may not be in a position to resist requests for ‘guest authorship’ by more powerful
41
This approach, however, has been more successful in other jurisdictions. See: M Annas, ‘The Label of
Authenticity: A Certification Trade Mark for Goods and Services of Indigenous Origin’ (1997) 3(90)
Aboriginal Law Bulletin 4.
42
See 5.2.1.
43
Private ordering is generally seen as undesirable where there is no genuine consent. This explains the
existence of legislation restricting the use of contracts of adhesion and consumer protection law in
general. See also JH Reichman & JA Franklin, ‘Privately Legislated Intellectual Property Rights:
Reconciling Freedom of Contract with Public Good Uses of Information’ (1999)147(4) U of
Pennsylvania LRev 875.
252
players; and they may even find themselves made ‘ghost authors’ on publications to
which they have made substantial contributions44.
7.4.3 The Benefits and Limitations of Relying Upon Private Ordering
As these examples suggest, there are benefits and limitations to relying upon private
ordering mechanisms to regulate authorship issues. The most obvious benefits are
flexibility and efficiency45. Private ordering might also have intrinsic benefits, such as,
promoting creators’ autonomy, self-determination and liberty. There are limits to the
desirability and effectiveness of private ordering mechanisms where there are
imbalances of power; where they are applied top-down to those who have not consented
to them; and when there are public interests at stake. This analysis of private ordering
leads to a number of insights for copyright law: (i) private ordering can be a desirable
way of regulating authorship; (ii) public structuring of private ordering is necessary to
minimises the limitations of such mechanisms; and (iii) the general success of private
ordering mechanisms in regulating issues relating to the exploitation of collective
authorship works suggests that the instrumental/pragmatic approach is based upon
flawed assumptions.
The most significant benefit of private ordering is that it provides flexibility46.
This includes the ability to adapt regulatory strategies over time and to changing
44
Worse still, pharmaceutical companies may be able to undermine the spirit of the guidelines by
selectively applying them, see p93.
45
M Lemley, ‘Intellectual Property and Shrinkwrap Licenses’ (1995) 68 Southern California LRev 1239
outlining the benefits of contract.
46
J Macey, ‘Public and Private Ordering and the Production of Legitimate and Illegitimate rules’ (1997)
82 Cornell L Rev 1123; S Dusollier, ‘Sharing Access to Intellectual Property Through Private Ordering’
253
circumstances. Particle physics collaborations, for example, each have their own
protocols and committees to deal with authorship issues. Protocols might be amended,
and committees might act, to adapt practices to changing circumstances. There are
usually different rules of attribution for different contexts and dispute resolution
procedures. Some methods of private ordering offer more flexibility than others.
Because of their standard form, for example, copyleft licences have been criticised for
lacking flexibility and interoperability with other copyleft licences 47. Yet, it is worth
remembering that the primary role of copyleft licences is in supporting the social norms
that govern creativity on Wikipedia (and those social norms are very flexible)48.
Private ordering allows communities to tailor their responses to regulatory
challenges49. In this way private ordering might be better able to reflect the concerns of
the relevant players, than judges and legislatures who are confined to a much smaller
regulatory tool box50. Public ordering is more restricted in its approach to regulatory
challenges, especially where it needs to be generalisable as in the case of legislation51.
The CDPA, for example, provides broad categories of works for which there is a single
definition of authorship and a single joint authorship test52. For example, although a
(2007) 82(3) Chicago-Kent LRev 1391 noting the use of private ordering to expand or change rights,
which has been observed in the case of Wikipedia.
47
Copyleft licences are difficult, if not impossible, to amend: RW Gomulkiewicz, ‘Open Source License
Proliferation: Helpful Diversity or Hopeless Confusion?’ (2009) 30 Washington U J of L and Policy 261.
48
On the usefulness of relying upon social norms to regulate behaviour: 8.1 below.
49
Dusollier (n46).
50
W Fisher, ‘Property and Contract on the Internet’ (1998) 73 U Chicago-Kent LRev 1203, 1252.
51
M Madison, ‘Legal-ware: Contract and Copyright in the Digital Age’ (1998) 67 Fordham LRev 1025,
1137.
52
Whether or not the definition is unitary is up for debate. Indeed, some cases appear to suggest that the
category of work affects the type of contribution required to count as an author, eg Sawkins v Hyperion
Records [2005] EWCA 565, [2005] 1 WLR 3281; Lucasfilm v Ainsworth [2011] UKSC 39, [2012] 1 AC
254
literary work may be a novel, a scientific journal article, a page of Wikipedia, a
computer program or a database, the same definition of ‘work’ and ‘author’ will apply
to it. Private ordering allows for more varied conceptualisations of authorship and
consequences of authorship than the CDPA’s one-size-fits-all regime.
The case studies show a very wide range of different approaches to regulating the
attribution and the consequences of authorship with private ordering. In this way,
private ordering allows for the flourishing of a greater diversity of creative models,
some of which may be impeded by the ordinary operation of copyright law (eg,
Wikipedia). These models can provide lawmakers with a source of inspiration and, over
time, they might provide valuable information on the merits of a range of different
approaches to the regulation of creativity.
Where the interests of the relevant parties are not represented in the private
ordering process, private ordering does not tend to work well. This can be seen from the
unsuccessful top-down attempts of biomedical journals to change the attribution
practices of large collaborations 53 . By involving members of the collaboration in
decision-making, particle physics groups have been able to regulate authorship issues
more successfully than biomedical journals 54 . Copyleft licences are successful at
regulating authorship on Wikipedia even though they reflect the will of strong norm
208, etc. It is uncertain whether this will still be the case as some recent European cases suggest that the
closed list categories of work ought to be irrelevant to copyright protection (particularly Case C-393/09,
Bezpecnostn´ı softwarováasociace v Ministerstvo Kultury [2011] ECDR 3, [2011] FSR 18; and Case C-
403/08 & C-429/08 Football Association Premier League v QC Leisure [2012] 1 CMLR 29, [2012] ECDR
8). The current European jurisprudence suggests a unitary definition of authorship, see Case C-5/08
Infopaq v Danske Dagblades Forening [2009] ECR I-6569. Yet, the type of work in question may still
affect the sorts of choices which indicate that it is the author’s own intellectual creation, see p15, n43.
53
5.2.1.
54
5.2.2.
255
entrepreneurs55. Although copyleft licences are a top-down regulatory mechanism, the
crucial difference is that they are a good fit for the Wikipedia project and they do not
favour one category of contributor over another56. Again, it is worth remembering that
copyleft licences are effective on Wikipedia primarily because of their role in
supporting the social norms that govern authorship on Wikipedia. These social norms
arise as a result of consensus-driven measures that are created, and consented to, by the
community of regular contributors (and thus are adequately representative)57.
In some situations it may be more efficient to allow groups to regulate themselves
via private ordering 58 . Indeed, much of the literature on social norms stresses their
ability to ensure efficiency and welfare maximisation. Social norms may reduce the cost
of regulation in terms of capital and labour. The parties themselves are in the best
position to evaluate the advantages of different regulatory arrangements. Private
ordering, thus, increases the possibility of finding consensual solutions that reflect the
preferences of all the parties 59 . In addition, private ordering allows for different
55
F Marrella and CS Yoo, ‘Is Open Source Software the New Lex Mercatoria?’ (2006–7) 47 Virginia J
of International L 807. This may be changing as copyleft licences become increasingly prevalent and
widely accepted in particular domains, such as self-publishing on the internet.
56
Cf. J Rothman, ‘The Questionable Use of Custom in Intellectual Property’ (2007) 93 Virginia LRev
1899, 1971 who considers them favourably because they have ‘aspirational goals’. Some are concerned
about their tendency to enshrine a view which might devalue authorship, making free access to works the
politically correct way for an author to exercise his or her rights. Thus, Dusollier compares copyleft
licences to other social constructions which can operate to undervalue labour: S Dusollier, ‘The Master’s
Tools v the Master’s House: Creative Commons v Copyright’ (2005) 29 Columbia J of L and Arts 271,
288.
57
This is a feature of norms which makes them amenable to incorporation in judicial decision-making,
see: 8.2 and 8.3.
58
Macey (n46) 1140.
59
R Mnookin, ‘Divorce Bargaining: The Limits on Private Ordering’ 18(4) J of L Reform 1015, 1018.
256
standards of authorship in different contexts, which makes them flexible in a low cost
way60.
The more tight-knit a group, the better they are able to collectively manage
authorship issues with private ordering mechanisms61. For this reason, particle physics
collaborations have been more effective at regulating authorship issues than biomedical
collaborations. Wikipedia is created by a large disparate group of volunteers located all
over the world – yet, is also very successful at private ordering. The architecture of the
wiki software and the use of automated tools to perform repetitive maintenance tasks
are crucial in this respect because they provide tools to facilitate the development and
enforcement of consensus-based policies. Although many proponents of private
ordering claim that it is most likely to work in close-knit communities, it seems likely
that the key is iterative transactions – which would explain why Wikipedia works so
well62.
Private ordering, particularly contract, has sometimes been supported on the basis
that it promotes individual liberty because it increases choice63. Private ordering is also
thought to have the intrinsic benefits of promoting autonomy and self-determination64.
60
On the value of the flexibility offered by private ordering see Fisher (n50) 1211-1212, 1237-1239.
According to Mnookin (n59) 1018 private ordering might avoid resort to litigation thus reducing public
and private costs as well as minimising emotional upset (taking the example of divorce settlements).
61
Social norms are generally seen as more effective at regulating tight-knit communities: Ellickson (n34)
156-66.
62
T Sagy, ‘What’s So Private About Private Ordering’ (2011) 45(4) L and Society Rev 923 (arguing
there is an important relationship between private ordering and social hierarchy).
63
Mnookin (n59) 1018.
64
K Bowrey, ‘Alternative IP?: Indigenous Protocols, Copyleft and New Juridifications of Customary
Practices’ (2006) 6 Macquarie LJ 65.
257
This is most likely to be the case where self-regulation arises as a result of a fair,
transparent, representative procedure to people who have consented to it65. Where the
group concerned is marginalised, as in the case of Indigenous communities, if self-
regulation bolsters self-determination it is likely to be an especially desirable means of
regulation66. Yet, private ordering has not proved an effective means for Indigenous
communities to enforce their interests in art. This is because the application of
customary law within an Indigenous community does nothing to promote self-
determination in relation to the wider Australian community which does not support
customary law with legal sanctions. This reveals another characteristic of private
ordering: it is most effective when supported and structured by public ordering67.
Private ordering is highly susceptible to imbalances of power and other defects of
the market68. This potentially limits the usefulness of relying on private ordering unless
players are in a reasonably equal bargaining position69. Thus, Indigenous artists have
found contract to be a relatively unhelpful tool, given they tend to be in a weaker
65
Mnookin (n59) 1019 argues that capacity is a limit to private ordering, as private ordering presupposes
rational, well-informed, self-interested participants. The possibility of temporary diminished capacity or
‘transactional incapacity’ is discussed by R Eisenberg, ‘The Bargain Principle and its Limits’ (1982) 95
Harvard LRev 741, 763.
66
Private ordering may be seen to promote cultural autonomy: G Barzilai, Communities and Law:
Politics and Cultures of Legal Identities (U of Michigan P 2003) 13.
67
G Dinwoodie, ‘Private Ordering and the Creation of International Copyright Norms: The Role of
Public Structuring’ (2004) 160 J of Institutional and Theoretical Economics 161.
68
Mnookin (n59) 1017, 1024-1031 arguing that relative bargaining power is one reason to limit private
ordering (he argues that there are two others: capacity concerns and externalities).
69
In some situations, for a variety of reasons, people might not be thought to have the capacity to
properly consent to private ordering. See the private ordering literature especially in the area of family
law (eg Mnookin (n59)). There do appear to be cases involving Indigenous art where capacity to
understand the full implications of a contract could be called into question – as may have been the case in
Yumbulul v Reserve Bank of Australia [1991] FCA 332, 21 IPR 481 (Fed. Ct. of Australia), see p67 and
also Arts Laws’ experiences, p69. There are a number of examples of Indigenous artists selling their
artwork in circumstances of undue influence or unconscionable conduct, see J Oster, ‘Proposal for a
Regulated Indigenous Art Industry’ (2006) 6 (18) Indigenous L Bulletin 21.
258
bargaining position than those with whom they are contracting. Different types of
contributors to films have organised themselves in talent guilds to promote their
interests and shape contractual negotiations on issues of pay, credit and working
conditions 70 . Collective bargaining is an effective way of ensuring solutions to the
regulation of authorship issues that adequately reflect the interests of different parties,
because it can help redress the imbalances of power between them. Law also has a
significant role to play in redressing power imbalances to ensure that private ordering
functions effectively 71 . There are many examples of legal rules that exist to shape
private ordering in this way, for example, the contra proferentum rule in contract law;
equitable doctrines, such as, unconscionability; and consumer protection law.
Another well-recognised limit of private ordering is that it is indifferent to the
interests of third parties or outsiders72. Further, private ordering often arises without a
thought for important public interests73. Contract law shapes private ordering in order to
constraint its potential limitations74. Thus, it imposes limits on the freedom to contract
to promote economic public order (eg, free competition) and protective public order
(eg, to protect weaker parties to transactions) 75 . Similarly, fundamental rights often
70
Where there is a less tight-knit community, formal measures might help.
71
R Mnookin and L Korhauser, ‘Bargaining in the Shadow of the Law: The Case of Divorce’ (1979) 88
Yale LRev 950, 968 arguing that legal rules create bargaining endowments or ‘bargaining chips’. This
affects negotiations because it modifies the parties’ best alternative to a negotiated agreement (BATNA).
R Fisher and W Ury, Getting to Yes: Negotiating Agreement Without Giving In (Houghton Mifflin 1981)
45 explain the role of the BATNA concept in negotiation.
72
Lemley (n45) 1285.
73
N Elkin-Koren, ‘What Contracts Cannot Do: The Limits of Private Ordering in Facilitating a Creative
Commons’ (2005) 74 Fordham LRev 375, 1197-99; Mnookin (n59) 1017, 1031-35.
74
Mnookin (n59) 1031.
75
L Guibault, Copyright Limitations and Contracts: An Analysis of the Contractual Overridability of
Limitations on Copyright (Kluwer Law International 2002) 123.
259
trump private ordering 76 . Private ordering might sometimes conflict with other
important policy goals that are embodied in legislation 77 . The interaction between
copyright law and contract law, particularly, the extent to which private ordering can be
used to avoid copyright law’s standards, is a matter that has caused much scholarly
debate78. On the question of authorship the situation is relatively clear: authorship is
ultimately a legal question, not a matter of agreement79. When it comes to determining
the authorship of a copyright work, the CDPA provides a definition that embodies a
balance of the policy goals that copyright law seeks to achieve: a contribution of
creative or intellectual choices to the protected expression80.
Private ordering is often thought to lack the legitimacy of the legislative process,
generally, on the basis that it does not represent the interests of outsiders (third parties
and the public)81. Sometimes, however, private ordering does not even represent the
interests of all those to whom it applies82. For these reasons, despite the benefits of
private ordering in terms of efficiency, dynamism and contextualism, there remains an
important role for law in helping to address some of the limitations of private ordering
and protecting important public interests83. Eric Posner argues that there is an important
76
Guibault, ibid.
77
Mnookin (n59) 1016 stressing the importance of considering fairness and other societal considerations.
78
Eg, Elkin Koren (n73); Madison (n51); Guibault (n75).
79
This point is discussed in Copinger et al [4-38]. See also Samuelson v Producers Distributing [1932] 1
Ch 201, (1932) 48 RPC 580, 586 and Wiseman v George Weidenfeld [1985] FSR 525.
80
See 2.1.
81
J Weinberg, ‘ICANN and the Problem of Legitimacy’ (2000) 50 Duke LJ 187, 191, 257. Schwarcz
(n33) 324.
82
A good example is contracts of adhesion, see Radin and Wagner (n33) 1311-1313.
83
Radin and Wagner (n33) 1317 stressing the importance of a baseline of due process and public policy
limits on contractual private ordering.
260
place for law in enhancing good social norms and undermining bad ones84. To this end
Fisk suggests that community practices on the attribution of authorship ought to be
assessed in relation to six criteria: transparency, participation, equality, due process,
efficiency and substantive fairness85.
Overall, issues relating to the ownership of the copyright subsisting in a work of
collective authorship generally appear to be successfully managed with private
ordering. Most contributors to films, for example, sign contracts with the production
company at the outset. Powerful players sometimes use options and unexecuted
contracts, relying upon social norms and default legal rules as a disincentive for
unreasonable behaviour. This gives them increased flexibility and helps individuals
manage the large risks involved in film projects. Some collective authorship groups are
unconcerned with remuneration and thus are able to resolve ownership issues very
easily. Wikipedians, for example, all agree to copyleft licences that remove any
possibility of ownership conflicts, because they agree not to exercise their rights to
control or remuneration, as a condition of participation in the project.
Scientific authors often assign or exclusively licence their copyright interest to
journal publishers as a condition of publication. This concentrates ownership rights in
one entity. Even where this is not the case as, for example, with open source publishing,
scientists are often happy to allow reproductions of their work because the reputational
value of authorship increases on wide distribution. Royalties for individual scientific
84
Posner (n34).
85
C Fisk, ‘Credit Where It’s Due: The Law and Norms of Attribution’ (2006) 95 Georgetown LJ 49.
These factors have much in common with the vectors that Rothman uses to assess whether the custom
ought to be incorporated into a copyright decision: J Rothman, ‘The Questionable Use of Custom in
Intellectual Property’ (2007) 93 Virginia LRev 1899, see further 8.2 and 8.3 below.
261
journal articles are likely to be negligible (or nil in the case of open source publishing).
Conflicts on issues of remuneration are unlikely to arise because this conflicts with
science’s norm of disinterestedness. Disputes amongst scientific authors are far more
likely to occur on matters relating to the attribution or false attribution of authorship
than they are on ownership issues86.
From these examples it appears clear that the issue of coordinating ownership and
the exploitation of works is not as problematic as the instrumental/pragmatic approach
assumes it to be. There is a possible objection that might be made to this point. I have
argued that an Indigenous community might be considered to be joint authors of an
artist work made in accordance with customary law if a more inclusive, contextual
approach were taken to the application of the joint authorship test. If this were the case,
the community would be able to impede the exploitation of the work and, indeed, they
would be very likely to seek to do so where a proposed reproduction would be offensive
or cause cultural harm. In light of the special spiritual significance of Indigenous art,
and the urgent need to ensure the continuation of Indigenous culture, this restriction
should be seen as a tolerable limit on copyright law that serves policy goals, equally
valuable as copyright law’s own goals. Further, this is the most desirable outcome given
that Indigenous communities are best placed to determine when offensive uses are
being made of their cultural artefacts.
Whilst private ordering can be a good way of bridging the gap between copyright
law and creative realities, there are limitations to this approach. Private ordering is not
86
The few legal cases on scientific authorship all arise from concerns about attribution: see: Noah v
Shuba [1991] FSR 14 (Ch); Weissmann v Freeman 684 FSupp 1248 (SDNY, 1988); 868 F2d 1313 (2nd
Cir, 1989); Anya v Wu [2004] EWCA Civ 755.
262
very effective where there are imbalances of power between parties and where there are
broader public interests at stake. These limitations of private ordering indicate that there
remains an important role for copyright. In particular, copyright can provide a source of
authorship standards that are independent from imbalances of power within the creative
authorship group. Copyright law’s default standard can help shape private ordering to
further important policy goals, particularly, copyright law’s purposes to reward and
incentivise creators. Therefore, there is an important role for copyright law in shaping
private ordering87.
The figure of the author is at the heart of copyright law. The author has the right
to first ownership of any copyright subsisting in his or her work. This right ought to be
granted to all those who qualify as authors because of their role in creating the work.
This cornerstone of copyright law is significantly undermined by adopting an
instrumental/pragmatic approach that requires one to ignore the existence of a number
of authors, granting authorship rights only to a chosen few. This clearly results in a less
satisfactory result (because it is inflexible and acontextual); and an unfair result (some
authors are treated as more worthy than others).
Private ordering and the law might interact in two main ways. First, private
ordering might be within a range of valid alternative regulatory mechanisms that the
law tolerates (or even upholds)88. Second, private ordering may be taken into account in
the application of legal rules, as part of the matrix of relevant facts. In this section I
87
Private ordering often presupposes some public ordering: Radin and Wagner (n33) 1296; S Dusollier,
‘The Master’s Tools v the Master’s House: Creative Commons v Copyright’ (2005) 29 Columbia J of L
and Arts 271; (n46).
88
Or even actively encourages: Mnookin (n59) 1036 argues that the primary function of the legal system
at the time of divorce is to facilitate private ordering.
263
have been mostly concerned with the first type of interaction, as I considered how
collective authorship groups use private mechanisms to bridge the gaps between their
creative reality and copyright law’s provisions. Some collective authorship groups, for
example, have used private ordering mechanisms to implement a gradation of
authorship credit, which provides for more possibilities than the CDPA’s binary
author/non-author approach to attribution. This has been very useful in encouraging and
rewarding contributors to scientific collaborations and films. The private regulation of
attribution in this way is permitted by copyright law in the sense that it is not
prohibited 89 . It may even be upheld as a result of the operation of statutory
presumptions on authorship. This is a useful example of how the legal regime leaves
space for parties to resolve certain issues in a manner that better suits their needs via
private ordering 90 . There is an important role for public ordering in shaping these
private ordering mechanisms 91 . In relation to the gradation of authorship credit,
copyright law provides a basic structure for private ordering practices by providing that
certain contributors (in certain circumstances) have a legally enforceable right to be
included (the attribution right); and certain people have the right to not be included (the
right to object to false attribution). The next section of this Chapter further considers
the role of copyright law in collective authorship.
89
L Bently and L Biron, ‘Discontinuities Between Legal Conceptions of Authorship and Social Practices:
What, if anything, is to be done?’ forthcoming in M van Eechoud (ed), The Work of Authorship
(Amsterdam UP 2014) 237.
90
L Guibault, ‘Wrapping Information in Contract: How Does it Affect the Public Domain?’ in L Guibault
and PB Hugenholtz (eds), The Future of the Public Domain: Identifying the Commons in Information
(Kluwer Law International 2006) 87, 103 noting that public ordering governance regimes have an
important symbolic role in that they embody the idea that certain activities are so important that they
ought not to be manifested purely in private transactions. Thus, she argues that privatization regimes that
undercut that symbolic role are presumptively offensive.
91
Dinwoodie (n67).
264
The following Chapter considers the second way in which private and public
ordering may be enmeshed. In that Chapter I consider the extent to which social norms
ought to be taken into account in the application of the joint authorship test. Social
norms, it must be observed, are a form of private ordering. An important distinction
ought to be made at this point. Although social norms might be evidenced by private
ordering mechanisms, such as protocols or contracts; those mechanisms are not
coterminous with the social norms. In the following Chapter I suggest an inclusive,
contextual approach to the application of the joint authorship test, which takes into
account the social norms which operate to regulate creativity in cases of collective
authorship. I also consider the limits of such an approach. But first, I review the
complex role that copyright law currently plays in the regulation of collective
authorship.
7.5 The Role of Copyright Law and its Concepts
Copyright law forms part of the background upon which collective authorship groups
organise their creative activities. Rosemary Coombe argues that law and culture relate
to one another in an ‘ongoing mutual rupturing’92. Indeed, culture appears to influence
law. Some argue that copyright has been influenced by the literary trope of the romantic
author. In Chapter 2, I argued that copyright decisions necessarily incorporate aesthetic
values, in so doing judges are likely to be influenced by prevailing cultural values93.
The opposite is also true, that is, that copyright law influences cultural practices. The
92
R Coombe, ‘Contingent Articulations: A Critical Cultural Studies of Law’ in A Sarat and T Kearns
(eds) Law in the Domains of Culture (University of Michigan Press 1998).
93
Roland Barthes defined a text as ‘a tissue of quotations drawn from the innumerable centres of culture’:
‘The Death of the Author’ in S Heath (trans) Image-Music-Text (Hill and Wang 1977) 142, 146. See
Chapter 4 for a critique of the influence of Western notions of authorship in copyright decisions on
Indigenous art.
265
example of collective authorship reveals that copyright law does not always affect
cultural practices in straightforward ways. Copyright can modify the bargaining
positions of contributors in private ordering processes. Copyright can provide a source
of good authorship standards (although these might sometimes be reinterpreted). In
other cases copyright’s normative force is reshaped and repurposed in order to better
suit the needs of the collective authorship group. Murray, Piper and Robertson suggest
that intellectual property is often appealed to in creative communities in ‘pragmatic
ways, opportunistically and instrumentally’94. Their suggestion appears to be borne out
in the case studies on collective authorship.
Copyright law is sometimes appealed to for its expressive function, that is, its
capacity to provide an authoritative pronouncement on authorship. The attribution of
the label of ‘author’ in copyright law has significance for the public at large because
Intellectual Property can serve as a valuable source of ‘default rules, policy and public
interest safeguards’ 95 . Copyright pronouncements on authorship have a symbolic
meaning that can support or detract from certain contributors’ abilities to assert that
they are authors. Copyright law also provides an important default position that can
help or hinder contributors in negotiations with other contributors as well as with
distributors.
94
L Murray, S Tina Piper and K Robertson, Putting Intellectual Property in its Place: Rights Discourses,
Creative Labor, and the Everyday (OUP 2014) 14.
95
J Pila, ‘Authorship and e-Science: Balancing Epistemological Trust and Skepticism in the Digital
Environment’ (2009) 23 Social Epistemology 1; RC Dreyfuss, ‘Collaborative Research: Conflicts on
Authorship, Ownership and Accountability’ (2000) 53 Vanderbilt LRev 1161; R Coombe The Cultural
Life of Intellectual Properties: Authorship, Appropriation and the Law (Duke UP 1998) particularly the
Introduction.
266
Copyright is clearly a matter of great importance in the film industry as a tool for
investors to recoup their investment96. Production companies and other investors carve
up the distribution and licensing rights to a film97. Hollywood studios are probably best
conceived of as intellectual property management companies rather than film
producers. Yet, interestingly most of the language in the film industry seems to avoid
use of the term author. The relevant textbooks speak of rights management, residuals
and ‘credits’98. Contributors generally sign contracts which involve the assignment of
copyright in their contributions to the producer or production company even where they
negotiate some share of resulting profits. The default positions provided by copyright
might offer important bargaining chips which can increase the bargaining power of
contributors. This is because there is a strong preference for the resolution of
intellectual property issues via contract rather than in accordance with the default
position provided by the law99.
The preference for private ordering is most likely due to the size of the investment
involved in making a film and ‘tentpole’ model of many production company
businesses which rely on revenue from a few big successes to sustain a number of
ongoing projects which may not make a profit 100 . As films increasingly rely upon
merchandising and toy sales to boost profits there may be an increase in disputes
brought by the creators of costumes, etc where they have not effectively assigned their
96
6.3.
97
6.3.
98
Eg, P Bloore, The Screenplay Business: Managing Creativity and Script Development in the Film
Industry (Routledge 2012); G Doyle, Understanding Media Economics (2nd edn, Sage 2013); D Gray,
Setting up and Managing a Production Company (Medialex Publications 2006).
99
This might seem odd given the amount of lobbying on the issue of film copyright.
100
p111.
267
copyright interest to the producers (eg, Lucasfilm)101. It seems that judges will normally
be prepared to imply the transfer of copyright in such circumstances, especially where
the contributor has been adequately remunerated (in a way that represents payment for
the rights as well as for the labour)102.
Even where copyright law is seen as irrelevant, or an obstacle to the joint project
(eg, Wikipedia), it has a valuable, but complicated, role to play. Wikipedia’s culture of
sharing both relies upon, and is partly constituted by, copyleft licences. The fact that
copyleft licences rely upon copyright law for their effectiveness and as a source of
standards means that the shadow of copyright law still casts itself over the norms which
operate to create and regulate this alternative system of self-regulated creativity. The
Wikipedia community taps into some of the values of copyright’s notion of authorship,
whilst discarding others. Contributors retain copyright in their contributions and in
choosing to forgo many of the benefits of that copyright interest in the name of the
Wikipedia project, they help constitute the sharing norms which sustain the community.
Despite the fact that Wikipedia’s model of creativity seems to resist copyright law’s
notion of authorship, the importance of copyright is something that is stressed on
Wikipedia, and a good deal of effort is spent flagging and removing copyright
infringing content. There is even a barnstar for this activity. In some ways the fact that
contributors are seen to be contributing something of value bolsters Wikipedia’s sharing
economy.
101
6.2.2.
102
The opposite suggestion was a way not to imply such a right in the case of Slater v Wimmer, see p123.
268
In a sense, the use of copyleft licences might have the effect of appearing to
amplify who might be considered an author, as every contributor agrees to such a
licence regardless of how great or small their contribution may be. Copyleft licences
establish a presumption that in order to copy or modify a contribution subsequent
contributors must follow the terms of the licence. This may not be strictly true from a
legal point of view (as bare licences copyleft licences likely operate only as a defence to
a claim of copyright infringement, which will not arise if copyright does not subsist in
the contribution in the first place, as may be the case for trivial contributions). More
contributors enjoy the right to be attributed as authors under the terms of the copyleft
licences than they would under moral rights provisions. In this way, copyleft licences
expand authorship rights in one sense (attribution) whilst contracting them in another
(control).
In appealing to copyright law as a source of standards, scholars and key players
within creative communities do not always appear to have a clear understanding of the
actual provisions of the law. At the very least, sometimes, a selective view of copyright
law is presented. In the biomedical science context, for example, scholars have
appealed to copyright law as an authoritative source of support for propositions to
restrict the attribution of authorship to fewer contributors. Sean Seymore argues that the
authorship crisis in the biomedical sciences might be addressed by realigning journal
practices with (US) copyright norms 103 . Seymore’s view relies upon a selective
understanding of US copyright law focusing upon the requirement for ‘some creative
spark’. Yet he may overstate the level of intellectual or creative input that this
requirement implies, as it is generally accepted that fairly mundane works reach this
103
See p93.
269
standard. Seymore implies that the ICMJE guidelines are in line with copyright norms,
however, they might be both under-inclusive and over-inclusive from a copyright point
of view. The trend in some joint authorship decisions to favour contributors who have
exercised control over the creative process might actually exacerbate the practices of
assigning guest authorship to powerful players. Whilst Seymore’s approach may have
the result of reducing the number of people who could be attributed as authors, it is by
no means clear that the authors chosen would align with those who ought to be
considered responsible for the article from a scientific point of view. Further, copyright
can provide no guidance on the order in which authors are attributed, which is a matter
of considerable significance for scientists (as both authors and readers).
Although Seymore elides the potential mismatch between copyright’s concept of
authorship and science’s concept of authorship, his analysis does suggest an important
basic component of copyright authorship which seems attractive. That is, the fact that
copyright law identifies the creator of the expression as the author. In this way
copyright law provides support for the normative view that those responsible for
creating the work qua copyright work are authors. The difficulties of establishing
scientific responsibility are exacerbated as attribution is increasingly distanced from the
actual creation of the work. Indeed, many of the most problematic practices involve the
attribution of those who have had very little to do with the creation of the work (gift
authorship, coercion authorship, mutual support authorship), or the failure to attribute
those who were closely connected to its creation (ghost authorship). In this respect
copyright law’s notion of authorship might provide a good source of standards.
270
This appeal to copyright norms might appear surprising in the context of science,
because many of science’s assumptions are opposite to copyright law’s assumptions.
Copyright law assumes that authors want control over their works, which may allow
them to extract royalties for uses of it. On the contrary, science values openness and
economic disinterestedness. From this point of view, it might seem that science and
copyright law have little in common. Yet, there is some convergence between these two
worlds on the issue of authorship. Copyright is necessary for scientists to be able to
publish their work in reputable journals 104. Publication is important for science, and
copyright facilitates this by allowing publishers to profit from publications without fear
of free-riding. Whilst royalties might not be a relevant concern for scientists, the
accurate attribution of authorship to the creators of the work is important because
authorship is a locus for credit and responsibility within the scientific community.
Although copyright law has an important role to play as a source of good
authorship standards, the dominant approach to the joint authorship test leaves it ill-
equipped to play this role. In Chapter 1, I observed that many copyright scholars
suggest that the joint authorship test might not be well-adapted to collective authorship.
The analysis of the case studies have borne this out, revealing that the
instrumental/pragmatic approach is based upon a number of misassumptions about
collective authorship and risks, denying many genuine authors the rights to which they
would otherwise be entitled under the CDPA. The emphasis on the authorship limb, at
the expense of the collaboration limb, means that the joint authorship test lacks the tools
to make sense of a work such as Wikipedia. Under the current approach Wikipedia
104
Cf the open science movement which is gaining increasing momentum, particularly, in response to the
perceived expensiveness and unfair commercial practices of some of the larger scientific publishers. See:
A Swan, ‘Open Access and the Progress of Science’ 95 American Scientist 198; L Guibault and C
Angelopoulos, Open Content Licensing: From Theory to Practice (Amsterdam UP 2011), Ch 6.
271
might be conceived of in an unreal way as a series of derivative works (many of which
will lack originality), instead of a joint work authored by many contributors with a
common design 105 . The acontextual approach to the joint authorship test caused the
Federal Court of Australia in Bulun Bulun to elide the role of the community in the
creation of the expression of the artistic work. The court focused on Bulun Bulun’s role
in fixing the artwork; ignoring the community’s role in the selection of the subject
matter of the painting, determining how that subject matter was depicted, the techniques
used to depict it, and its ongoing control over subsequent uses of the artwork. Indeed,
the community may have made a more significant contribution to the expression of the
artwork, than the design director made to the drawings prepared by technical draftsmen
in Cala Homes.
Although collective authorship groups are generally able to regulate authorship
effectively with private ordering, there remains an important role for copyright law: (i)
as a source of good authorship standards that are independent from the power dynamics
which operate in creative communities; and (ii) as a method of promoting copyright’s
policy goals. In order to be able to effectively regulate collective authorship, the joint
authorship test ought to be applied in a way that appreciates the nature of collective
authorship and the context in which it occurs. This might go some way towards
addressing the problem identified by Jessica Litman that although people tend to buy
into copyright norms, they do not tend to translate them into rules in the same way that
the copyright statute provides106.
105
There are similar difficulties in construing the boundaries of the work in the case of film. A broader
approach to the application of the joint authorship test to the dramatic work, might make it easier to avoid
a thicket of underlying rights.
106
J Litman, ‘Copyright Noncompliance (or why we can’t “just say yes” to licensing)’ (1996) 29 NYU J
of Intl L and Politics 237, 238.
272
7.6 Summary
In summary, the case studies have revealed a number of insights about collective
authorship:
 Collective authorship is a different process to individual authorship;
 Social norms have an important role to play in motivating and regulating
authorship;
 Responsibility for the creative, aesthetic or intellectual content of a work is shared
among many contributors;
 Authorship has different meaning, and is attributed differently, in different contexts;
 The attribution of authorship can be affected by power dynamics
 Attribution is often more important to creators than financial benefits or control;
 Authorship issues are often successfully managed with private ordering
mechanisms;
 There are limits to the desirability of relying on private ordering mechanisms alone;
and
 Copyright law norms matter to creative communities (although perhaps not in
straightforward ways).
In light of these insights, a number of points must be borne in mind when applying the
joint authorship test to cases of collective authorship. These are as follows:
(i) Authorship has important ramifications for creators other than copyright
ownership.
(ii) The factually specific application of the joint authorship is a feature, rather
than a defect, because it allows the test to be flexible to different creative
contexts.
273
(iii) Social norms provide a handy source of information about the creative
process (particularly in relation to what contributors consider to be a valuable
contribution and other shared assumptions).
(iv) An inclusive approach to the application of the joint authorship test reflects
the creative realities of collective authorship better than the
instrumental/pragmatic approach.
(v) The joint authorship test ought to be applied in a way that is sensitive to the
creative context.
(vi) ‘Authorship’ in copyright law is an important legal standard that is
independent from the power dynamics that operate in authorial groups.
(vii) There are good reasons to be cautious about how social norms are taken into
account in legal decision-making.
In light of these points, the next Chapter proposes an inclusive and contextual approach
to the joint authorship test. In so doing the Chapter considers both the usefulness and
the limitations of considering the social norms which apply in various creative
communities when judges apply the joint authorship test.
274
Chapter 8: An Inclusive, Contextual Approach to the Joint
Authorship Test
Copyright law is said to be undergoing a credibility crisis, because it seems

1
disconnected to modern creative practices . Anne Barron has suggested that
‘…copyright discourse is perpetually haunted by the problem of reconciling the often
conflicting demands of certainty and legitimacy’ 2. This thesis has shown that when it
comes to collective authorship, current copyright law appears to fail on both accounts.
The instrumental/pragmatic approach sacrifices legitimacy for certainty, concentrating
authorship in the hands of a few dominant contributors in order to streamline the
exploitation of a copyright work. This approach distances authorship both from the
creative context and from copyright’s own notion of the ‘author’. Legitimacy concerns
lie behind the judicial preoccupation with aesthetic neutrality. This preoccupation tends
to lead to a lack of analytical clarity in judicial reasoning, which makes it extremely
difficult to tell how the joint authorship test might be applied to works of collective
authorship. Nevertheless, it is virtually impossible to apply the joint authorship test
without reference to aesthetic criteria. When such criteria are latently considered both
the certainty and the legitimacy of the law might be compromised. This thesis argues
that copyright law has an important role to play in the regulation of collective
authorship. However, in order to effectively play this role, the creative realities of
collective authorship should be taken into account in the application of the joint
1
N Elkin-Koren, ‘Tailoring Copyright to Social Production’ (2011) 12 Theoretical Inquiries in L 309,
310.
2
A Barron, ‘The Legal Properties of Film’ (2004) 67 MLR 177, 204-5.
275
authorship test. The joint authorship test contains the tools and the inbuilt flexibility to
do this. With this in mind, I suggest a recalibration of the joint authorship test to restore
the balance between the conflicting demands of certainty and legitimacy.
This Chapter proposes an approach to the application of the joint authorship test
that is inclusive (of all those who have made a more than de minimis contribution of
creative choices to the expression) and contextual (in that it takes the context of
creativity into account). I argue that such an approach is best suited to the realities of
collective authorship and copyright law’s purposes to encourage and reward creativity.
Instead of adopting the instrumental/pragmatic approach, judges ought to consider the
social norms that apply within collective authorship groups. These social norms provide
information that is highly pertinent to a determination of whether there is
‘collaboration’ and whether a contributor has made a ‘significant’ contribution to the
work. They provide a useful point of reference that does not require judges to make
subjective value judgements on aesthetic matters. Yet, there are some dangers in
uncritically incorporating social norms in judicial decision-making. Thus, I suggest a
three-prong framework for filtering norms based on their certainty, representativeness,
and policy implications. In addition, the requirement that a contribution be of the ‘right
kind’ is an important safety mechanism that ensures that authorship is only granted to
those who have made a more than de minimis creative or intellectual contribution to the
protected expression (filtering out contributions solely of an investment or orchestration
type).
This Chapter begins by considering why social norms might be relevant to
judicial decision-making in this context. I argue that judges should take social norms
276
into account in their decisions for three reasons: (i) they provide important information
about the facts of a case; (ii) the credibility of copyright depends upon its ability to
correspond with reality at least in a minimal sense; and (iii) people are more likely to
obey copyright law where it appears to be fair and legitimate. The second section
considers the limits of incorporating social norms in judicial decision-making. I argue
that social norms should not be simply deferred to; rather, they should be carefully
assessed. The third section suggests a framework for assessing when norms ought to be
incorporated into judicial decision-making. The fourth section revisits the three
critiques of the joint authorship test (see Chapter 2). The fifth section proposes an
inclusive, contextual approach to the application of the joint authorship test bearing in
mind both the usefulness, and limitations, of incorporating social norms in copyright
cases. It also elaborates on how this approach might work in practice by providing a
few examples.
8.1 The Relevance of Social Norms
Social norms have been extensively studied in the social sciences, particularly, by
anthropologists, sociologists and economists3. The body of literature on law and social
norms has recently been growing4. In the legal context, scholars have argued that the
state might more effectively achieve certain regulatory outcomes by choosing to ignore,
strengthen or undermine particular social norms 5 . Some take an optimistic view of
3
C Bicchieri & R Muldoon, ‘Social Norms’, Stanford Encyclopaedia of Philosophy <http://plato.stanford.
edu/entries/social-norms/>.
4
Posner suggests that this scholarship grew out of the Legal Realist scholarship of the early 20 th century.
E Posner, ‘Introduction’ in E Posner (ed), Social Norms, Nonlegal Sanctions, and the Law (Edward Elgar
2007) ix.
5
Posner (n4).
277
social norms, arguing that they can be a more efficient and flexible way of regulating
behaviour than legal rules6. It may also be considered important to uphold norms in
order to fulfil parties’ expectations or to promote their interests in autonomy7. Others
are more pessimistic, focusing on the dangers of relying on social norms that cause
social costs or that conflict with fundamental values, such as distributive justice. These
groups disagree on whether it is best for law to be imposed by the state from the top
down or to grow from the bottom up by enforcing social norms8. Ultimately, this is
likely to depend upon the particular issue to be regulated. Even where a top-down
approach is preferred, it is important to have an understanding of the social norms that
operate in a given community in order to understand the actual effects of legal rules on
that community9.
At first blush, copyright law might appear too standardised to fit the vast range of
creative works to which it is intended to apply10. A single definition of authorship and
of joint authorship applies across all categories of work. In Chapter 2, however, I
showed that there is, in fact, a considerable amount of flexibility built into copyright
law. Although judges have sought to take the facts of the case into account, they have
6
E Posner, Law and Social Norms (Harvard UP 2000); R Ellickson, Order Without Law: How Neighbors
Settle Disputes (Harvard UP 1991) argues that social norms are welfare maximising.
7
M Albertson Fineman, The Autonomy Myth: A Theory of Dependency (The New Press 2004); L
Guibault, Copyright Limitations and Contracts: An Analysis of the Contractual Overridability of
Limitations on Copyright (Kluwer Law International 2002); R Mnookin, ‘Divorce Bargaining: The
Limits on Private Ordering’ 18(4) J of Law Reform 1015, 1018; G Barzilai Communities and Law:
Politics and Cultures of Legal Identities (U of Michigan P 2003) 13.
8
R Cooter, ‘Normative Failure Theory of Law’ (1997) 82 Cornell LRev 947.
9
E Posner, ‘The Regulation of Groups: The Influence of Legal and Nonlegal Sanctions on Collective
Action’ (1996) 63 U of Chicago LRev 133, 193.
10
On the possibility that social norms might alleviate the problem of a one-size-fits-all intellectual
property regime see: D Oliar and C Sprigman ‘There’s No Free Laugh (Anymore): The Emergence of
Intellectual Property Norms and the Transformation of Stand-up Comedy’ (2008) 94 Virgina LRev 1787,
1839-40.
278
not tended to explicitly consider the appropriate sources of those ‘facts’ or the point at
which they ought to be incorporated in the application of the joint authorship test.
Given that the application of the joint authorship test depends upon the facts of the case,
it is important to accurately establish those facts. The case law provides little guidance
on this point. Further, current copyright scholarship rarely elaborates on which aspects
of the facts ought to be relevant in determining the authorship of a work11.
The previous case studies show that collective authorship groups share a number
of characteristics12. One of these is the role of social norms in governing the creative
process. Contributors adopt social norms as a response to the need to solve coordination
problems that are likely to develop as a result of the large number of contributors, the
division of labour, and the necessary diffusion of creative control amongst contributors.
These social norms might be created by the group or they might come from external
sources (eg, predominant norms in the particular discipline or genre)13. They provide a
good source of information about the creative reality and shared assumptions of creators
on a particular project. Social norms can provide answers to factual questions relevant
to determining the authorship of works created by large groups. They provide
information about whether contributors are collaborating in the relevant sense, because
they offer insights on how the project is understood by contributors. They also provide
11
This is likely to be due to the fact that much copyright scholarship remains similarly committed to the
exclusion of aesthetic considerations from copyright decision-making on questions of authorship. See
further, Chapter 2.
12
See 7.6.
13
This is not to suggest that they are explicitly created; rather, they are the unexpected/unplanned result
of individuals’ interactions: Bicchieri and Muldoon (n3).
279
information about the value of a particular person’s contribution to the project, which is
of particular relevance in determining the significance of their contribution14.
As argued in Chapter 2, it is very likely that judges already take social norms into
account in their decision-making. This can be seen in the case law which implies
requirements for a contribution of a ‘musical’ or ‘artistic’ kind, etc15. It can also be seen
in the acceptance of musicologist evidence that performance is not a contribution of an
authorship kind16. Yet, judges rarely set out their view of the facts in terms that make it
clear which aesthetic perspectives they consider to be relevant to the case in issue. The
cases are also ambiguous about the particular parts of the joint authorship test to which
such perspectives might be relevant. Later in this concluding Chapter I set out a
framework for the incorporation of social norms in answering questions of fact that
arise in the application of the joint authorship test.
The problem with current copyright jurisprudence is this failure to engage
explicitly with social norms in judicial reasoning17. I argue that this is due, in part, to an
anxiety about aesthetic neutrality18. Yet, there is no violation of the principle that a
judge should not be influenced by his or her judgment of the aesthetic merits of the
work, in taking aesthetic considerations into account when determining the authorship
of the work. Indeed, explicitly taking social norms into account may be the best way of
14
This requirement ought to be a qualitative one: 2.2.3.
15
Sawkins v Hyperion Records [2005] EWCA 565, [2005] 1 WLR 3281; Interlego v Tyco Industries
[1989] AC 217 (PC).
16
Hadley v Kemp [1999] EMLR 589 (Ch).
17
Rothman discusses this problem in ‘The Questionable Use of Custom in Intellectual Property’ (2007)
93 Virginia LRev 1899, 1902.
18
2.3.3.
280
ensuring aesthetic neutrality, because it appeals to the aesthetic views of the relevant
creative community and not the judge’s own subjective views. Doing so would also
improve the quality and the clarity of decisions. The failure to explicitly engage with
social norms has reduced the visibility of important contextual factors that weigh in
favour of an inclusive definition of joint authorship. This enhances the appeal of the
instrumental/pragmatic approach which presents itself as a simple and efficient answer.
However, this thesis argues that the instrumental/pragmatic approach is rarely likely to
lead to the right answer when it comes to determining the authorship of a work created
by a large group of people19.
There are at least two other reasons to explicitly take social norms into account in
judicial decision-making. The first is that the credibility of copyright law depends on its
ability to correspond, at least in a minimal sense, with reality20. One might ask how
copyright law can claim to be the principal legal mechanism for the regulation of
creativity, if it is out of touch with the realities of creative communities. The
designation of authorship must ultimately be a legal question because of the central role
of this concept in the scheme of the CDPA. However, that legal determination must
both relate, and seem relevant, to the creative realities of creators. Indeed, it is virtually
impossible to make a decision on joint authorship without latently considering such
factors. It is vital for the ongoing credibility of copyright law that these considerations
are dealt with openly. This is also likely to enhance the quality of decision-making by
19
Indeed, this approach is likely to result in side effects that are significantly worse than the disease it
aims to cure. It distorts the application of the joint authorship test, distancing it both from the reality of
creativity and from copyright’s notion of the author.
20
L Bently and L Biron, ‘Discontinuities Between Legal Conceptions of Authorship and Social Practices:
What, if anything, is to be done?’ forthcoming in M van Eechoud (ed), The Work of Authorship
(Amsterdam UP 2014) 237 argue that realigning legal and social authorship norms is important for the
sake of copyright law’s legitimacy. See also: J Tehranian, ‘Infringement Nation: Copyright Reform and
the Law/Norm Gap’ (2007) Utah LRev 537.
281
inviting more explicit and thoughtful consideration of the validity and usefulness of
particular norms21.
Secondly, as an instrumental matter, the public is more likely to obey laws that
they consider to be fair and legitimate22. All law relies on a certain amount of self-
enforcement and co-operation from the community to ensure its effectiveness in
regulating behaviour23. Psychologists have found promoting self-compliance to be more
effective than deterrence factors24. The perceived fairness of a law appears to be the
single most important factor in shaping compliance25. In order to promote compliance
with intellectual property law, prominent psychologist Tom Tyler suggests trying to
understand the public’s feelings about what is fair in order to tap into those feelings26. A
significant element of the perceived fairness of copyright law is the way in which the
21
M Madison, ‘Legal-ware: Contract and Copyright in the Digital Age’ (1998) 67 Fordham LRev
1025,1137-8 argues that exposing the legal system directly to community norms ensures that they are
reconciled with copyright policy and also that, over time, judicial decisions produce relatively stable
bodies of rules that reflect community standards of fairness.
22
Tom Tyler’s work is very influential in this field, see especially: Why People Obey the Law (Princeton
UP 2006).
23
Tyler (n22) 271. Also, D Kahan, ‘Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms Problem
(2000) 97 U of Chicago LRev 607 suggesting that authorities tend to hold back from enforcement of laws
that greatly contradict social norms.
24
Tyler (n22) 270. Some of the literature is outlined in: J Tankebe, ‘Viewing Things Differently: The
Dimensions of Public Perceptions of Police Legitimacy’ (2013) 51 Criminology 103; T Tyler and J
Jackson, ‘Popular Legitimacy and the Exercise of Legal Authority: Motivating Compliance, Cooperation
and Engagement’ (2014) 20(1) Psychology Public Policy and L 78.
25
Even more important than the likelihood of being caught and punished. Tyler, ibid.
26
T Tyler, ‘Compliance with Intellectual Property Law: A Psychological Perspective’ (1996) 29
International L and Politics 219, 227. Indeed, he suggests that it is particularly important to consider how
to promote compliance with intellectual property laws, at a time when non-compliance appears to be a
serious issue with the rise of file sharing and the proliferation of copying on the internet (at 219). See also
J Litman, ‘Copyright Noncompliance (or why we can’t “just say yes” to licensing)’ (1996) 29
International Law and Politics 237.
282
beneficiaries of copyright are viewed 27 . Ensuring that the ‘authors’ protected by
copyright law correspond to common expectations, at least in a minimum sense, is a
simple step towards promoting compliance. As Jessica Litman succinctly puts it in
discussing copyright licensing, ‘[p]eople don’t obey laws that they don’t believe in’ 28.
Inadequate and irrelevant legal provisions are likely to be ignored or avoided.
Wikipedia, for example, adroitly sidesteps the potentially restrictive and uncertain
29
concept of authorship in copyright law with copyleft licences . The
instrumental/pragmatic approach is likely to clash with what most contributors to works
of collective authorship consider ‘fair’. Granting authorship to a few dominant players
and denying it to other valuable contributors disconnects authorship from the creation
of the expression30. The ineffectiveness of the ICJME journal guidelines for biomedical
collaborations, for example, shows that a regulatory strategy that is disconnected from
creative realities is unlikely to be successful31. Where the law accords with public views
about what is fair, it can also have an important symbolic function which, in turn,
promotes compliance32.
27
This explains lobbyists’ frequent appeals to the figure of the deserving author when arguing for
copyright law reform: see L Bently, ‘R v Author: From Death Penalty to Community Service’ (2008)
32(1) Columbia J of L & the Arts 1.
28
Litman (n26) 239; M Schultz, ‘Fear and Norms and Rock & Roll: What Jambands Can Teach Us about
Persuading People to Obey Copyright Law’ (2009) Berkeley Technology LJ 651 also argues that
copyright law ought to be realigned with social norms in order to promote compliance.
29
3.3.
30
This clashes with the understanding, common in all the creative communities considered in the case
studies, that authorship signifies responsibility for what is considered valuable about the work: 7.2.2.
31
5.2.1.
32
Litman points out that ‘…laws that we keep around for their symbolic power can only exercise that
power to the extent that people know what the laws say’ (n26) 237. Thus, it is also important to promote
analytical clarity in judicial decision-making.
283
The perceived legitimacy of the law can also promote compliance33. Thus, people
are more likely to obey laws that they believe serve reasonable social purposes and that
are not simply efforts to create profits for special interest groups, such as large
corporations 34. This provides another good reason to favour an inclusive, contextual
approach to the joint authorship test over the instrumental/pragmatic approach 35. The
latter approach appears to unfairly advantage powerful players, some of whom, such as
film producers, already constitute special interest groups that have been able to secure
greater protection of their interests through lobbying36. The concept of the author as a
creator of protected expression is at the heart of copyright law’s claims to legitimacy37.
The instrumental/pragmatic approach distances the beneficiaries of copyright protection
from copyright law’s raison d’être: to reward and encourage authors38. Thus, there is a
good instrumental reason to ensure that copyright law’s concept of authorship is in
touch with creative realities (so as to appear ‘fair’ to those to whom it applies); and
33
Tyler (n26) 224 discusses two strategies which his study finds to be more effective in promoting
compliance than deterrence. One is based on legitimacy (‘one’s feelings that one ought to obey the law’),
the other is based on morality (‘an individual’s personal feelings about what is right or wrong’). He
suggests that the legitimacy-based strategy is more difficult to implement than the morality-based
strategy, given the recent decline in public respect for legal authorities in the US (Tyler, 230). A similar
trend is likely to exist in the UK: D Bloomfield, K Collins, C Fry and R Munton, ‘Deliberation and
Inclusion: Vehicles for Increasing Trust in UK Public Governance’ (2001) 12 Environment and Planning
C: Government and Policy 501, 508-9; C Pettie, P Seyd, P Whiteley, Citizenship in Britain: Values,
Participation and Democracy (CUP 2004).
34
Tyler, ibid 233.
35
The existing copyright literature on this point focuses on the need to recognise the interests of users as
well as the powerful interest groups that tend to influence the legislative process. C Jensen, ‘The More
Things Change, the More They Stay the Same: Copyright, Digital Technology, and Social Norms (2003)
56(2) Stanford LRev 531; Litman (n26).
36
Jenson, ibid 540.
37
W Cornish, ‘The Author as Risk-Sharer’ (2002) 26 Columbia J of L & the Arts 1, 12: copyright law
derives most of its legal and moral force from the act of creating. See also, RR Kwall, The Soul of
Creativity: Forging a Moral Rights Law for the US (Stanford UP 2009) on the morality of copyright law.
38
Some argue that copyright law is primarily a mechanism for disseminators, publishers and other
intermediates to recoup their costs, thus, ensuring the widespread distribution of creative works (rather
than their production). This view requires one to ignore the fact that the author is put before the
disseminator according to both the letter and the spirit of CDPA. Thus, the conceptual divide between
authorship and ownership. See further: Chapter 2.
284
reflects the underlying purpose of copyright law to incentivise and reward creators of
protectable expression (in order to enhance the ‘legitimacy’ of the law).
8.2 The Dangers of Deferring to Social Norms
In the previous section, I argued that there are a number of good reasons for judges to
take social norms into account in the application of the joint authorship test. Few, if
any, theorists suggest that social norms ought to be completely disregarded in legal
decision-making. Most counsel for caution in incorporating social norms, industry
practices, and customs in legal decision-making for three main reasons. First, the
incorporation of social norms in judicial decisions risks colonising independent legal
standards 39. Second, it may be difficult to identify the content of social norms with
certainty. Third, the content of a particular social norm might be cause for concern on
the basis of its lack of reliability, efficiency, desirability, or fit with the law. In this
section I consider how these three concerns relate to one another. In light of these
concerns, I argue that in applying the joint authorship test social norms should be
considered carefully.
Social norms might not be sufficiently stable or reliable. This is particularly likely
to be the case where the content of the norm is unclear or contested. The case study of
science, for example, reveals how norms on attribution have varied over time and
between disciplines40. Legal rules ought to satisfy certain minimum standards of clarity,
39
Rothman (n17) 1908.
40
5.1
285
openness, and predictability41. This is particularly the case for copyright law, where the
provisions on authorship provide the basis for the ownership of property rights. The
scope of such rights and their owners must be capable of being identified without too
much uncertainty. This dictates the consideration of social norms in the application of
the joint authorship test only where the content of the norm is fairly certain, and the
norm is widely accepted and stable. According to these criteria, for example, the
practice of listing the head of a laboratory as an author on a biomedical science journal
where that person has made no contribution to the research is unlikely to qualify
because it is too contested42.
Some warn that the extolled virtues of social norms in terms of efficiency are not
universally applicable 43. Although inefficient social norms should not be adopted as
solutions to normative problems (ie, answers to legal questions), the inefficiency of a
norm does not diminish its value as a source of information about the realities of the
creative process. Indeed, I have argued that social norms are primarily relevant for
reasons other than their efficiency. One of the most important reasons is the fact that
taking into account the realities of creativity might enhance the credibility of copyright
law, which is valuable for intrinsic as well as instrumental reasons.
41
J Raz, ‘Rule of Law and its Virtue’ in J Raz, The Authority of Law: Essays on Law and Morality
(Clarendon Press 1979) 210.
42
See 5.2.1.
43
Rothman (n17); E Posner, ‘Law, Economics and Inefficient Norms’ (1996) 144 U Pennsylvania LRev
1697; A Katz, ‘Taking Private Ordering Seriously’ (1996) 144 U Pennsylvania LRev 1745, 1749.
286
Some argue that social norms do not actually represent the parties’ expectations.
Indeed, social norms are not explicitly made or consented to 44 . Social norms are
particularly unlikely to be representative where there are imbalances of power within
the relevant group. Norms which develop in these situations are unlikely to represent a
satisfactory balancing of the interests of all parties. This can be seen in the relative
ineffectiveness of relying upon contracts and codes of conduct to protect Indigenous
peoples’ interests in art45. There are good policy reasons not to incorporate unequal or
unrepresentative social norms in legal decision-making46.
Jennifer Rothman has provided a leading critique of the incorporation of social
norms in intellectual property decisions47. Rothman is particularly concerned about the
incorporation of industry practices in determining what might count as a ‘fair’ use,
because these practices are unlikely to be representative of public interests, which ought
to weigh heavily in the fair use analysis 48. However, there may be less reason to be
concerned about the incorporation of social norms in the application of the joint
authorship test, where the question is which contributors are joint authors. In this case
the most relevant parties are the other contributors. Where a norm is stable and
uncontested amongst an authorial group, it is likely to be appropriately representative of
44
J Gardner, ‘Some Types of Law’ in D Edlin (ed), Common Law Theory (CUP 2007) 51; see also
Bicchieri and Muldoon (n3).
45
See 4.3.1 and 4.3.3.
46
On the importance of equality as a substantive value in a theory of justice, see, for example, E
Chemerinsky, ‘In Defense of Equality: A Reply to Professor Westen’ (1983) 81 Michigan LRev 575,
585-91.
47
Rothman (n17) 1908, 1902 is concerned that custom has ‘tremendous influence on IP law’; see also J
Rothman, ‘Copyright, Custom and Lessons from the Common Law’ in S Balganesh (ed) Intellectual
Property and the Common Law (CUP 2013) 230.
48
Rothman (n17) 1937, especially the interests of users.
287
those contributors. Still, even in this context a norm should be rejected where it
conflicts with copyright law’s purposes or other policy goals49.
Questions which are essentially legal in nature ought to be answered
independently by judges guided by the policy and purposes of the CDPA. As the
concept of originality helps determine the scope of protection, much of the analysis on
this point must be an independent legal evaluation. Yet this only accounts for the
requirement in the joint authorship test that a contribution be of the ‘right kind’. There
is more scope for incorporating social norms in answering factual questions which go
towards establishing whether a contributor is a joint author. Therefore, concerns about
the incorporation of social norms are less compelling in relation to determinations of
the significance of a contribution or the existence of a common design among
contributors 50 . Although generally sceptical about the incorporation of norms in
intellectual property decisions, even Rothman admits that they are valuable evidence of
parties’ intentions and of what is generally done in an industry51.
Authorship cannot be a purely legal inquiry, because it needs to adapt across
industries, in very different contexts, and over time. This need for flexibility supports
recourse to social norms in answering factual questions. Nevertheless, not all social
norms can provide appropriate answers to these questions. Theorists tend to be highly
critical of the desirability of norms that develop in situations where there is great
49
One such policy goal might be substantive equality among contributors who qualify as authors.
50
These ought to be treated as factual questions: 2.2.4.
51
Rothman (n17) 1937.
288
inequality of bargaining power between parties52. In addition, social norms ought not to
be adopted where they may result in third party or societal costs; or harm other
important policy interests. These objections counsel caution in the way in which norms
are incorporated in judicial decision-making.
The next section provides a framework for assessing the desirability of norms.
Following a restatement of the critiques of the joint authorship test raised in Chapter 2,
the final section presents an inclusive, contextual approach to the application of the
joint authorship test that attempts an appropriate incorporation of social norms in
reaching legal determinations.
8.3 A Framework for Considering Social Norms
Although some of the literature on law and social norms is relatively new, the idea that
it is important for judges to have regard to customary practice is quite old. Indeed,
recourse to custom as a source of legal rules is at the heart of the common law
method53. From the early days of the common law, judges developed rules to consider
whether it was appropriate to recognise custom as law54. According to Blackstone, a
custom was a good candidate to be incorporated into the common law if it satisfied the
following requirements: antiquity, continuing peaceable use, certainty, reasonableness,
52
ibid 1956. Although powerful interest groups may also affect the legislative process, Rothman
considers that these groups will have a more pronounced impact on custom which does not develop in the
context of open debate and public commentary. See also: Posner (n43) 1709 who provides a critique of
the reliance on norms as a source for legal rules on the basis that the asymmetrical position of parties is
likely to result in inefficient norms.
53
Rothman (n47).
54
Since the sixteenth century it has been necessary to prove the antiquity, continuity, certainty and
reasonableness of a custom. A Loux ‘The Persistence of the Ancient Regime: Custom, Utility and the
Common Law in the Nineteenth Century’ (1993) 79 Cornell LRev 183, 189.
289
compulsoriness and consistency55. These requirements might be boiled down to two
central concerns relating to the proof of the existence and the reasonableness of the
custom 56 . These requirements filter out the aspects of custom that are potentially
arbitrary and burdensome to non-community members 57 . The need to ensure that
illegitimate or undesirable customs do not spread and are not upheld by the law was
recognised from these early days58.
This thesis does not suggest that courts ought to adopt social norms and enforce
them as if they were legal rules. Rather, social norms provide an important source of
contextual information that may help answer questions of fact in the application of the
joint authorship test 59 . As part of the factual matrix which existed at the time of
creation, social norms provide important information about the existence and
parameters of a common design; and what constitutes a significant contribution in light
of that common design. In this section, I suggest a three-prong framework to guide the
incorporation of social norms in the application of the joint authorship test, drawing
upon the limits to the incorporation of custom that the common law has recognised60.
55
W Blackstone, Commentaries, Volume 1, 76-78.
56
Loux (n54) 186. Halsbury’s Laws of England (5th edn, LexisNexis 2012) Volume 32, 1(3)(i)6
describes the requirements in the following terms: ‘To be valid, a custom must have four essential
attributes: (1) it must be immemorial; (2) it must be reasonable; (3) it must be certain in its terms, and in
respect both of the locality where it is alleged to obtain and of the persons whom it is alleged to affect;
and (4) it must have continued as of right and without interruption since its immemorial origin’.
57
H Smith, ‘Community and Custom in Property’ (2009) 10(1) Theoretical Inquiries in Law 5, 10.
58
Smith (n57) 11.
59
Rothman (n17) 1968 argues that custom should only be relevant in relation to questions about what the
parties intended or what is generally done. She thinks it should not be used to answer normative questions
(ie, what ought to be done).
60
Rothman (n17) sets out a detailed and compelling argument for the assessment of custom according to
these principles. She takes these rules as an inspiration for the 6 vector approach that she suggests to
guide the incorporation of custom in intellectual property decisions. Her vectors include: certainty,
290
This framework, thus, addresses the three potential defects of social norms noted above:
their lack of certainty; the influence of power imbalances on their development; and the
fact that they develop without regard for public interests and the interests of third
parties. I suggest that the desirability of a social norm might be ascertained by
considering: (i) its certainty; (ii) its representativeness; and (iii) the policy implications
of adopting it. This framework requires proof that the social norm exists in a stable
form and that it does not enshrine values that might be repugnant to the law’s overall
purposes.
(i) Certainty
A number of problems may arise in discovering the relevant norms and in dealing with
changing norms. For this reason, the requirement for certainty is important. Courts
ought not to incorporate social norms into legal decisions unless they are identifiable,
stable and widely accepted61. The ICJME guidelines discussed in Chapter 5 might not
quite reach this level of certainty because they are not widely accepted62. On the other
hand, Indigenous customary laws on the authorship of art will satisfy the requirement of
certainty63. Social norms may change over time. When it comes to determinations of
joint authorship, judges ought to inquire into the social norms that existed at the time
the work was created. Proof of certainty provides a measure of transparency and
motivation, representativeness, application of the custom (against whom), application of the custom (for
what proposition), implications of the custom. See also Rothman (n47).
61
It does not make much sense to apply the requirement of immemorial existence in this context,
although as a general matter the stability of the norm over time ought to be required. The length of time
might vary depending on the characteristics of the creative industry in question.
62
5.2.1.
63
Australian courts have often accepted evidence on these matters in copyright cases, see 4.2.
291
stability, which attenuates concerns that the incorporation of social norms might
potentially inject an element of arbitrariness to joint authorship cases64.
Although it might be said that powerful parties will more easily be able to prove
the existence of a social norm that favours them65, this is no more likely to be the case
here than in relation to any other aspect of an intellectual property case. The use of
expert evidence on issues of copyright subsistence (including authorship) is already
common66. Judges are very accustomed to, and adept at, determining industry practice
in other contexts, for example, in contract cases67. There is no reason to suppose that
discovering the social norms that operate in creative communities will be any more
challenging in this context than it is in others. In any case, getting joint authorship right
is surely worth the price of any problems of proof that may arise from the explicit
consideration of social norms. The case studies discussed in this thesis show that the
benefits the instrumental/pragmatic approach seeks are not likely to be worth the price
paid: a fundamental disconnect between copyright law and the reality of collective
authorship68.
64
C Fisk, ‘Credit Where It’s Due: The Law and Norms of Attribution’ (2006) 95 Georgetown LJ 49
considers these to be important characteristics of desirable private ordering regimes.
65
Rothman (n17) 1959, also noting that formal written statements of social norms, such as guidelines, are
more likely to be developed by powerful players and are easy to prove, yet these written statements may
not reflect actual practice.
66
See: L Bently, ‘Authorship of Popular Music in UK Copyright Law’ (2009) 12(2) Information
Communication & Society 179 who has criticised the over-reliance on the evidence of classically-trained
musicologists who have a particular point of view that may make them ill-suited to comment on genres
such as popular music.
67
S Carter ‘Custom, Adjudication and Petrushevsky’s Watch: Some Notes from the Intellectual Property
Front’ (1992) 78 Virginia LRev 129, 130 (although he expresses some scepticism about judicial expertise
to do so in exercises of what he calls ‘judicial anthropology’).
68
See 2.3.2, 7.1 and 7.2.2.
292
(ii) Representativeness
Custom has its greatest legitimacy when applied to the community that generated it69.
The common law has always held that customs should only apply to those who
demonstrate consent to them. The requirement for consent might seem to underlie
Blackstone’s requirements for the custom’s continuity, and existence since time
immemorial; and that it be peaceable (undisputed) and compulsory 70 . In this sense,
‘consent’ was a legal fiction, particularly given the requirement of the immemorial
existence of the custom 71. Social norms arise as an unplanned result of individuals’
interactions72. In light of the absence of active consent to a social norm, I prefer the
term ‘representativeness’. Under this second prong, I suggest that norms that are
developed with the input and participation of contributors are to be preferred to top-
down norms. Similarly, courts should not incorporate customs that only serve one
party’s interest73.
This part of the framework provides a bulwark against undesirable norms which
can result from imbalances of power within creative communities. In the case study of
scientific collaborations, it has been observed that power imbalances can negatively
impact prevailing social norms. Thus, senior scientists are able to exert a great deal of
control over the attribution of authorship, in ways which distance authorship from the
69
Millar v Taylor (1769) 4 Burr 2303, 2368 (Yates J): custom is ‘lex loci, the law of the place’.
70
Rothman (n47) 243-244.
71
Loux (n54) 208. Chapman v Smith (1754) 2 Ves Sen 506, 510 (Lord Hardwicke LC) for legal purposes
this date was fixed at the year 1189.
72
See Bicchieri and Muldoon (n3); and Gardner (n44) 60-3 noting that custom is not expressly made and
not intentionally made.
73
Rothman (n17) 1970-4; Fisk (n64) 74-6.
293
creation of the work and even from scientific responsibility for it 74. Likewise, some
directors have used their superior bargaining power to obtain a possessory credit on a
film seemingly implying that they are its sole author75. The literature on social norms
supports the observation that power dynamics within groups can prevent community
norms from being ideal in terms of efficiency and fairness76. The need to protect weaker
parties is also recognised in many parts of the law and commentators have made similar
suggestions in slightly different contexts77. Fisk, for example, suggests that attribution
regimes ought to be assessed on the basis of equality, substantive fairness and
participation 78 . And, Rothman argues that before incorporating a custom in an
intellectual property decision, judges must consider those to whom it would apply and
the ‘motivations’ of a custom79. Collectively bargained agreements on screen credit and
particle physics authorship protocols provide evidence of social norms that are
representative80.
(iii) Policy Implications
The court should take policy considerations into account when considering whether to
incorporate a particular social norm. The requirement of reasonableness was one of the
74
5.2.1.
75
See pp135, 179, 181.
76
Smith (n57) 14.
77
This might be seen, for example, in the contra proferentum rule in contract law, or in the equitable
doctrine of unconscionability.
78
Fisk (n64) 74-6.
79
Rothman (n17) 1970-4.
80
See 5.2.2 and 6.3.
294
most important restrictions on the incorporation of custom into the common law81. Even
where a custom was widely accepted and long-standing, it was not appropriate for
judges to defer to it as a proxy for what is reasonable82. This allowed judges to assess
the substantive fairness of a custom and also to consider public interests. In the previous
section, I observed that one of the key limitations of social norms is that they are
developed without regard for the interests of third parties and other public policy
concerns. The third prong of this framework assures the primacy of copyright law’s
policy goals and filters out undesirable norms 83 . This involves the independent
evaluation of whether the end result of the adoption of a particular norm into law would
be objectively desirable84. It also has the added benefit of allowing the incorporation of
social norms to evolve, rather than being locked in by courts85.
8.4 Revisiting the Critiques of the Joint Authorship Test
Before suggesting an approach to the application of the joint authorship test that is
better suited to collective authorship, I briefly return to the three critiques of the
application of the joint authorship test (discussed in Chapter 2)86.
81
Halsbury’s Laws of England (5th edn, Lexis Nexis 2012) Volume 32, 1(3)(iii)9.
82
Rothman (n47) 244.
83
Rothman (n17) 1977.
84
ibid 1976.
85
ibid.
86
2.3.
295
8.4.1 The Factual Specificity of the Joint Authorship Test
In Chapter 2, I suggested that the factual specificity of the joint authorship test makes it
difficult to predict the outcome of a particular case, especially where no similar cases
have been previously decided87. In light of the very different dynamics of creativity in
collective authorship groups, the factually specific nature of the joint authorship test is a
very useful feature because it gives the test the flexibility to adapt to the creative
context. In this context, it is important to ensure that that the ‘facts’ are correctly
understood. The social norms which regulate creativity in collective authorship groups
provide a good source of information about the creative process (as long as their
limitations are borne in mind).
8.4.2 The Preoccupation with Aesthetic Neutrality
In Chapter 2, I argued that the principle that judges should not be affected by their own
view of the aesthetic merits of the work in making copyright decisions has led to a
general anxiety about aesthetic neutrality88. It is very difficult to make judgments about
who should be an author without some regard to aesthetic considerations. Given the
inescapable relevance of aesthetic considerations, social norms might provide a handy
point of reference that avoids the dangers of judges making subjective decisions about
the aesthetic merits of a work. Collective authorship groups themselves are best placed
to determine the relative value of each contributor’s contribution to the whole. By
taking into account the social norms that operate in particular creative contexts, judges
87
2.3.1.
88
2.3.3. J Cohen, ‘Creativity and Culture in Copyright Theory’ (2007) 40 U of California Davis LRev
1151, considers this anxiety to be reflected in copyright scholarship, leading to a preference for
abstraction and a propensity to ignore the context of creativity.
296
can define relevant aesthetic criteria which help establish joint authorship, whilst
devolving any questions of aesthetic merit to the collective authorship group. Adopting
social norms as a point of reference for aesthetic matters is likely to result in better
quality decisions that more accurately reflect creative realities. The first section of this
Chapter suggests that an open, transparent, contextual approach such as this may
enhance the credibility of copyright law as a tool to encourage, reward and regulate
creativity89. This may have the knock on effect of enhancing compliance.
8.4.3 The Instrumental/Pragmatic Approach
The case studies suggest that adopting the instrumental/pragmatic approach to the
application of the joint authorship test is misguided, because that approach divorces the
test from both copyright’s core concept of authorship and the reality of collective
authorship.
The instrumental/pragmatic approach divorces the joint authorship test from
copyright’s concept of authorship by implying that a higher standard applies in the case
of joint works. This risks undermining copyright law’s goal to reward and incentivise
creators of protected expression, because one or two contributors will receive too much
of an authorship share, whilst many more will receive no authorship share at all. Worse
still, authorship is likely to gravitate to the most powerful or dominant contributors,
who tend to be orchestrators and investors, rather than creators. Although these figures
are very important to the creative process, the CDPA already provides for their
protection with specially adapted entrepreneurial copyrights (that belong to ‘makers’ as
89
8.1.
297
distinct from ‘authors’), and with the presumption that copyright in works created by
employees during the course of their employment is owned by their employer unless
terms to the contrary have been agreed90. The provisions of the CDPA make it clear that
‘author’ is intended to imply creator, not orchestrator or investor 91 . It is clearly
problematic that the instrumental/pragmatic approach to the joint authorship test
requires judges to distance themselves from the core meaning of authorship in copyright
law, applying a more restrictive approach in the case of joint works. Not only does this
approach result in too few authors, it also results in the wrong contributors being
granted too much of an authorship share.
Furthermore, the instrumental/pragmatic approach divorces the joint authorship
test from creative realities. The nature of collective authorship is that creative
responsibility is shared amongst a large number of contributors. The
instrumental/pragmatic approach requires judges to construct a fiction about how a
work is created in order to elevate the contributions of one or two individuals and deny
the contributions of the other contributors. These legal fictions, based on instrumental
concerns rather than legal principles, are likely to lead to unpredictability in the case
law92.
The instrumental/pragmatic approach also relies on the assumption that many
joint owners will struggle to effectively exploit a work. The case studies, however,
indicate that this may not be a significant concern, as private ordering can often be very
90
Even here there is provision for express terms to trump investor rights.
91
See 2.1.
92
2.3.2.
298
effectively used to regulate authorship issues such as attribution or copyright
ownership. The Wikipedia community has adopted creative commons licences that
virtually eliminate some of the coordination problems that can arise from multiple
ownership, whilst retaining the right to attribution, which is clearly important to its
contributors 93 . These licences also provide for workable ways of implementing the
attribution requirement in the digital environment94.
Contract can be a particularly valuable tool for resolving ownership issues which
might result in cases of multiple ownership, because it can be flexible and context-
specific. The benefits of relying upon contract include efficiency, flexibility, and
intrinsic benefits such as promoting autonomy and self-determination. The example of
film demonstrates that contract can be a very effective means of dealing with issues that
might arise in relation to the copyright ownership of works of multiple authorship95. In
some cases such issues are unlikely to arise in the first place, as authors have no interest
in controlling future uses of their work or because a sharing economy is a more
important factor in authorship than the ability to charge royalties (eg, science and
Wikipedia)96.
There are limitations on the effectiveness of relying upon contracts where there is
no capacity to agree, where parties have unequal bargaining power, and where there are
societal costs. Therefore, contract is not well-adapted to deal with Indigenous art, as
93
3.3.
94
For example, it is considered sufficient acknowledgment to provide a hyperlink to the relevant page
(see p52).
95
6.3.
96
3.1, 5.1.
299
Indigenous artists may not have equal bargaining power with art dealers and museums
who wish to purchase their work97. A contract is also limited to those who are parties to
it and cannot affect the conduct of third parties. A precondition to reliance on contract
to deal with ownership issues is that it requires parties to anticipate difficulties and take
the step of actually entering into a contract. Indigenous artists may not realise the need
to ensure explicit protection of their cultural interests98. Where authorship takes on its
main significance in a sharing economy, contributors may not care to deal with
ownership issues that might arise. In these cases copyright provides an important source
of default rules. Yet, as has been considered above, these rules will only be helpful
where they reflect creative realities. Based on the analysis of the case studies, this thesis
suggests that an inclusive approach to joint authorship is preferable. This approach
would remain faithful to the same core notion of authorship that exists for individually
authored works. Such an approach more accurately reflects the reality of collective
authorship99, and provides less powerful contributors with a bargaining chip that may
improve the overall quality of contractual negotiations100. The example of film shows
how certain creative players have sought to be recognised as ‘authors’ or ‘creatives’ as
a means of improving their bargaining position in contractual negotiations over pay and
credit101.
97
The Bangarra Dance troupe are a notable exception: M Rimmer, ‘The Bangarra Dance Theatre:
Copyright and Indigenous Culture’ (2000) 9(2) Griffith LRev 274.
98
Yumbulul v Reserve Bank of Australia [1991] FCA 332, 21 IPR 481 (Fed. Ct. of Australia) seems to
provide an example of this, p67 above.
99
These groups require the input of a large number of specialised contributors, only some of whom
undertake completely mechanical, non-creative tasks.
100
Dominant contributors are unlikely to need the bargaining help they receive from the application of
the instrumental/pragmatic approach. See RC Dreyfuss, ‘Collaborative Research: Conflicts on
Authorship, Ownership and Accountability’ (2000) 53 Vanderbilt LRev 1161, 1207.
101
p139.
300
The instrumental/pragmatic approach elides the concepts of authorship and
ownership in an unhelpful way. By focusing solely on the economic consequences of a
designation of joint authorship (ie, joint ownership), the instrumental/pragmatic
approach ignores the fact that a designation of joint authorship is also likely to lead to
moral rights. The right to be attributed as an author is particularly valued in a number of
the creative communities considered 102 . Authorship and ownership are separate and
conceptually distinct questions.
There are other better ways of dealing with the potential unsuitability of the rules
of joint ownership than taking a restrictive approach to the joint authorship test103. The
best option would be legislative amendment by, for example, including a presumption
that only a majority of owners need to consent to license a work in cases involving a
large number of owners (unless all owners agree to the contrary) 104. The ‘majority’
might be calculated in terms of ownership share, which case law establishes may be
unequal among the contributors. At the very least, it would be helpful to introduce a
provision that unless the contrary is explicitly agreed to, one joint owner cannot
unreasonably withhold consent to licence a work105.
102
The importance of attribution within collective authorship groups raises the question of whether the
current provisions on the right of attribution offer too little protection (particularly in light of the
requirement to assert the right).
103
The question of joint ownership is not within the scope of this thesis, so I only provide a few brief
suggestions on this point.
104
This would be to adopt an intermediate position between the UK and the US approach. In the US any
owner might consent to license the work (subject to a duty to account for profits), for more see: M
LaFrance, ‘Authorship, Dominance, and the Captive Collaborator: Preserving the Rights of Joint
Authors’ (2001) 50 Emory LJ 193; MB Nimmer and D Nimmer, Nimmer on Copyright (Matthew Bender
2011) §6.10[A][1][a].
105
The absence of a solution on this issue is considered problematic by D Vaver Intellectual Property
Law (Irwin Law 2011) 122.
301
The rules on joint ownership largely arise from case law, applying the rule of
statutory interpretation that the singular implies the plural106. Given the absurdity of the
results of this interpretation in cases of large group authorship, there may be grounds
for a different judicial interpretation of the statute in cases of collective authorship. It is
clearly unworkable to allow one of a hundred joint authors to hold up the exploitation
of the work. In this context, it would certainly be plausible for a court to adopt a
different and more nuanced approach to the application of the rules on joint
ownership 107 . Failing this, a solution may be also found by implying terms into
contracts or by implying licences allowing certain uses. This will usually be supported
by reasons of business efficacy. Many large collaborations have evolved models of
collective decision-making to deal with these challenges (eg, particle physics
protocols). Judges should strive to uphold these consensually developed strategies for
managing collective authorship whenever reasonably possible.
8.5 An Inclusive and Contextual Approach to the Joint Authorship Test
Building on the insights from the case studies, I propose an approach to the application
of the joint authorship test that is:
 adapted to the ways in which collective authorship groups work together to
create (Chapter 7);
 sensitive to the relevance of social norms (8.1); but also
 mindful of the dangers of uncritically incorporating social norms (8.2).
106
s10(3).
107
The potential flexibility of the rules on joint ownership can be seen in a case where on the specific
facts a husband and wife were said to own copyright in commissioned wedding photos as joint tenants
rather than as tenants in common, contrary to most of the case law: Mail Newspapers v Express
Newspapers [1987] FSR 90 (Ch).
302
What do the case studies reveal about the best way of applying the joint authorship test
to works of collective authorship? They reveal, first, that the joint authorship test should
be applied contextually. Authorship should be a flexible concept that can adapt to
creative practices that change over time and that differ between creative communities.
This is the best way to make a one size fits all copyright regime work in a credible way.
Second, they reveal that the test should be inclusive, rather than restrictive. At the very
least, the authorship requirement should not be stricter in the case of joint authorship
than it is for individual authorship. This best reflects the realities of collective
authorship and provides protection for those contributors who have made significant
contributions to the expression but who may find themselves in weaker bargaining
positions.
An inclusive, contextual approach to the joint authorship test can be achieved in
the following way. First, judges should take social norms into account in determining
whether contributors are collaborating. Second, in light of the common design of the
contributors, the question of whether or not a contribution is significant ought to be a
qualitative inquiry that takes into account the social norms which operate to govern the
creative activities of the contributors to the work. Overall, this approach entails a shift
in emphasis in the application of the test from an abstract consideration of whether an
individual contribution is ‘authorship’ to focus on the collaborative nature of the joint
work. This involves explicitly considering the prevailing social norms which regulate
authorship in a particular context. This is not to say that judges should simply adopt an
authorial community’s view of who is an author. But, rather, that social norms provide
a valuable source of information about the collaborative nature of a joint work, which
may help assess the contributions made to it.
303
Given the dangers of the uncritical incorporation of social norms expressed at 8.2,
there is a need to exercise caution in the incorporation of norms when applying the joint
authorship test. I suggest a two-fold filter. First, norms should be assessed according to
the three-prong framework suggested above, based on their certainty, representativeness
and the policy implications of adopting them. Second, social norms ought to feed into
the legal test for authorship (requiring a contribution of the right kind). This requires a
more than de minimis contribution of creative choices which are evident in the
expression. This means that pure investment will never be enough; and creative control
alone will not be enough – unless it is coupled with creative choices evident in the
expression. In the rest of this section, I explain this approach in further detail, offering
examples of how it might work in practice.
Whether a contributor has made a significant contribution is best determined in
light of the specific collaboration concerned. In Chapter 2, I argued that it makes more
sense to adopt a qualitative, rather than a quantitative, approach to the question of
whether a contribution is significant108. One of the main objections to the adoption of a
qualitative approach is the possibility of subjective judgments of aesthetic merit seeping
into the decision-making 109 . This risk can be minimised by explicitly incorporating
social norms at this point, devolving the decision of which sorts of contributions are
significant to the best source of information about the relative value of contributions, ie:
the social norms governing creativity in the particular creative context. This approach
avoids the pitfalls and analytical obfuscation associated with an acontextual approach to
108
2.2.3.
109
p82.
304
determining the relative value of particular contributions to a joint project110. On this
view, a statistician who analyses the results of scientific experiments might make a
‘significant’ contribution to a journal article reporting the results of that study 111 .
Similarly, a Wikipedia contributor might make a significant contribution when they edit
a page so as to make it comply with the Neutral Point of View policy112.
Whether or not there is collaboration or a common design is also best assessed
with regard to the shared assumptions of the creators. In Chapter 2, I argued that this
part of the joint authorship test has the important function of distinguishing joint works
from a series of derivative works113. The question of whether a highly iterative work
such as Wikipedia is a series of derivative works or a number of joint works may seem
quite a challenging question to resolve on its face. A consideration of the social norms
which operate in this context, however, makes the question significantly easier to
answer. From this point of view, it is apparent that Wikipedia is not a series of
individually authored derivative works, but its pages are joint works. The main reason
for this is that the social norms establish collaboration amongst the regular
contributors114.
110
An example of an acontextual approach can be seen in the ‘elephant test’ for artistic works (you know
one when you see one) suggested by the Court of Appeal in Lucasfilm v Ainsworth [2009] EWCA Civ
1328 [77] and regarded with scepticism by the Supreme Court: Lucasfilm v Ainsworth [2012] 1 AC 208
[47].
111
5.1 (the norms of science) and p115 (the significance of the work of statisticians).
112
p51.
113
2.2.2.
114
See further, Chapter 3, p50.
305
Although there is no need for a subjective intention to be a joint author115, as
argued in Chapter 2, the requirement of collaboration necessarily requires some
intention to work together 116 . It is difficult to see how one might accidentally or
unwittingly collaborate with another person. Social norms provide important
information about the shared assumptions of contributors to creative works117. These
assumptions are particularly relevant when they satisfy the requirements of the three-
prong framework suggested at 8.3118. Such norms are pertinent in determining whether
there can be said to be a common design especially where they are stable, fairly
uncontested and representative.
An understanding of the nature of the common design shared by contributors may
also help shed light on the significance of various contributions. So, for example, in the
context of a scientific collaboration: involvement in the design of a study in accordance
with the scientific method is likely to be a significant contribution in light of the
common design of contributors to produce research that will be accepted as valid by the
relevant scientific community. By the same token, a screenwriter who writes
humourous dialogue is also likely to have made a significant contribution to a film in
light of the common design of contributors to make a romantic comedy119.
When it comes to considering whether the contribution is of the right kind (ie,
authorship) this needs to be done partly with an eye to the reality of creativity. This
115
Beckingham v Hodgens [2003] EWCA Civ 143, [2003] EMLR 18.
116
2.2.2.
117
7.1.
118
The three prongs are: certainty, representativeness and policy implications.
119
p139, p152.
306
information should feed into copyright’s own legal notion of authorship, which requires
a more than de minimis contribution of creative choices which are evident in the
expression (Chapter 2). The important role of the concept of authorship in the CDPA
means that there is a normative dimension to the authorship requirement120. This part of
the test is a legal question that is common to joint works and single author works. At
this stage in the analysis, therefore, the purposes of the CDPA and the limits of
copyright protection need to be taken into account. The idea/expression dichotomy is
one of the principles that have been used to police the boundaries of the scope of
copyright protection. The categories of work have also sometimes been used in this
way121. This provides a stable legal core to the joint authorship test122.
In addition to ensuring that the purpose of copyright law is fulfilled and that other
policy goals are served, there is another reason for the need to have a legal dimension to
the joint authorship test: to ensure that this important question is not completely
subordinated to market forces. The legal dimension of the joint authorship test provides
protection for creators in weaker bargaining positions. In this way it might, for example,
help contributors to a film ensure adequate remuneration for their contributions 123 .
Indeed, a central justification for the copyright regime itself is premised on the need to
protect authors who are in a position that makes them vulnerable to free-riding due to
120
2.1.
121
The requirement that a contribution be relevant to the category of work, ie, that it be ‘musical’ or
‘artistic’ in nature. The ongoing relevance of this limitation on the scope of copyright protection may now
be in doubt following the Court of Justice of the European Union’s opinions in Case C-393/09,
Bezpecnostn´ı softwarováasociace v Ministerstvo Kultury [2011] ECDR 3, [2011] FSR 18 and Case C-
403/08 & C-429/08 Football Association Premier League v QC Leisure [2012] 1 CMLR 29, [2012] ECDR
8. But see, p15 n43.
122
Bently and Biron (n20) note that this is important because authorship forms the basis upon which
property rights are granted.
123
This is inherent in the court’s reasoning in Slater v Wimmer, see p148.
307
the nature of their creations, which are costly to create, but inexpensive to copy. There
is a natural relationship between copyright and contract law. The scheme embodied in
the CDPA assumes that authors will contract with publishers and other disseminators
(assigning or licensing their copyright interest) and that they will grant licences to other
users of their works. In order for this scheme to be effective, there needs to be a bare
minimum level of equality of bargaining power124. An inclusive default approach to the
joint authorship test serves this purpose by providing an incentive for more powerful
parties to define issues of ownership at the outset by potentially raising the costs of not
doing so. This is what many film contributors have tried to achieve by joining guilds
which have the power to make more favourable collective agreements with producers
on their behalf, than those contributors would be able to negotiate individually125. An
inclusive approach is also preferable even where there is no inequality of bargaining
power, or where contributors are not concerned about the ownership of copyright,
because it better reflects the reality of collective authorship126.
Adopting an inclusive, contextual approach to the joint authorship test means, for
example, that a scientific author who deletes important inaccuracies in an article might
count as an author127. This contribution is seen as significant according to the scientific
community and it also counts as a contribution of creative choices that is evident in the
124
There are many examples of important legal principles that are premised on the need to protect those
in weaker bargaining positions: consumer protection laws; equitable rules on unconscionable conduct; the
contra proferentum rule, etc.
125
7.4.1, p159, p163.
126
7.1.
127
In a case involving a scientific article, the Polish Supreme Court held that one might become a joint
author by deleting content which is factually incorrect (Case II CSK 527/10, 22 June 2010): T Targosz,
‘Authorship By Deletion’, Kluwer Copyright Blog, <http://kluwercopyrightblog.com/2011/11/02/author
ship -by-deletion-supreme-court-june-22-2010-ii-csk-52710/> dated 2 November 2011.
308
expression. Recourse to the Mertonian norms of science is also likely to pass the three-
prong framework suggested at 8.3 above128.
To take an example from the case law, this approach, may have allowed for the
recognition of communal authorship in a case such as Bulun Bulun 129 . In light of
customary law, it seems quite clear that there was a common design between Bulun
Bulun and his community, as represented by its Elders: the transmission and
preservation of sacred cultural knowledge. In light of this common design, the
contribution of the community was significant. The apprenticeship of the artist and the
ongoing control of the community over subsequent uses of the artwork attest this. The
Federal Court of Australia held that the community had contributed unprotectable ideas.
As has been argued in Chapter 4, the community contributed to the expression
(although not to the fixation of the work). Once the context of creativity is taken into
account, there seems to be little distinction between Bulun Bulun and Cala Homes in
which a design director was held to be a joint author of drawings made by draftsmen
under his direction130. Indeed, the decision in Bulun Bulun seems best explained as a
result of a Western bias toward individual authorship 131. Thus, it fails to reflect the
reality of the authorship of art in the Indigenous community concerned.
128
p105.
129
4.2.
130
p93.
131
p94. See also Bowrey’s critique of the decision: ‘The Outer Limits of Copyright Law – Where Law
Meets Philosophy and Culture’ (2001) 12 L and Critique 75.
309
The two-fold filter suggested ensures that undesirable norms are not uncritically
adopted into law132. For example, the attribution practices that many have criticised in
biomedical science would not be accepted because they are not stable and uncontested.
Furthermore, gift authorship and guest authorship will not meet the legal requirement
for a contribution of the right kind to the expression of the work. Instances of irrational
discrimination will also fall foul of the last part of the framework: the implications of
adopting the social norm133.
8.6 Conclusion
In popular culture, peppered throughout the case law, and in much of the scholarship
considered here, one might discern a concept of the ‘author’ as puppeteer or
orchestrator. Although the romantic author has been convincingly exposed as a myth
and pronounced ‘dead’, its ghost still seems to haunt copyright law today. The romantic
author has taken on a modern guise in the pragmatic insistence on the need to
incentivise those who take risks and choose to invest in creative projects 134 . This
approach tends to divorce copyright from creative realities and also from its own core
notion of authorship. Although historically, it has been expedient to be able to hold one
person responsible for a collective work in some circumstances, this does not accurately
reflect the way in which groups actually work together to create. This has been
132
The first filter is the requirement that the contribution be of the right kind. The second filter is the
three-prong framework for assessing norms suggested at 8.3 above.
133
Social discrimination can affect attribution practices, see for example, M Terrall, ‘The Uses of
Anonymity in the Age of Reason’ in M Biagioli and P Galison (eds), Scientific Authorship: Credit and
Intellectual Property in Science (Routledge 2003) 91 discussing the historical use of anonymity in
relation to works that pushed the boundaries of propriety at the time of publication or for reasons related
to the public status of the author (eg, female scientists).
134
A Casey and A Sawicki, ‘Copyright in Teams’ (2013) 80 U of Chicago LRev 1683 use the theory of
the firm to argue for the concentration of IP rights with managers.
310
appreciated in disciplines where romantic author figures have been prominent in the
past, such as science and film, but which now have moved towards more inclusive
approaches to authorship.
The concept of authorship in copyright law is not about control; rather, it is about
creative contribution to the expression of a work. From a copyright perspective,
authorship is about more than who signs a work135, it is about the person who creates
it 136 . This thesis reaffirms the importance of maintaining separate concepts of
authorship and ownership in copyright law. The importance of the noneconomic
functions of authorship to authors, and the influence of sharing economies in motivating
their activities, casts the limitations of the incentive justification for copyright law in
sharp relief. The fact that the case studies considered here concern works which fall
within copyright law’s heartland adds gravity to this critique, and suggests that the
incentive story can only provide a shaky foundation for the copyright expansionism that
has occurred in its name. There are important instrinsic and social dimensions to
creativity, which require a more nuanced, contextual approach to the implementation of
copyright law. In particular, an appreciation of the creativity reality of collective
authorship suggests a need to critically assess the way in which instrumental reasoning
shapes the development of copyright law more broadly.
The experience of collective authorship suggests that in determining who counts
as an author, it is better to ensure that enough creators can access the right of
attribution, than to restrict authorship on the basis that this might streamline the
135
J Ginsburg, ‘The Author’s Name as a Trade Mark: A Perverse Perspective on the Moral Right of
“Paternity”?’ (2005) 23 Cardozo Arts and Entertainment LJ 379.
136
CDPA s9.
311
economic exploitation of the work. This is further supported by the fact that ownership
issues tend to be successfully managed with private ordering. Private ordering allows
for the flourishing of different creative models, some of which would be impeded by
the ordinary operation of copyright law 137 . Private ordering also allows creators to
separate the economic and reputational incidents of authorship, allocating them to
contributors in ways that best support their needs. Although the case studies
demonstrate the natural interaction between private ordering and copyright law, they
also indicate the need for some regulation of this relationship. To this end, there is an
important role for copyright as a source of good authorship standards. This is enhanced
by the interaction between legal standards and community standards that takes places in
the application of the joint authorship test (partly a legal, and a partly a factual,
question).
This thesis has argued for an inclusive, contextual approach to the joint authorship
test that would result in better quality decisions that correspond to creative realities and
do not simply replicate power dynamics in creative communities. Authorship should be
a flexible concept that can adapt to creative practices that differ over time, from place to
place, and between creative communities138. By considering the social norms that apply
to a particular collective authorship group, the joint authorship test can be applied in a
flexible way that is sensitive to the creative context and that would allow the one size
fits all copyright regime to work in a credible way. This approach might also help
counteract the effect of cultural biases and elitist tendencies in the application of the
137
These might provide a valuable source of information for copyright policy makers about the merits of
different approaches to the regulation of creativity.
138
The lack of international harmonisation of copyright on this point reflects the fact that the concept of
authorship is partly contingent on local conditions.
312
authorship standard, by devolving the assessment of aesthetic criteria to creators rather
than judges.
The proposal that judges ought to take social norms into account in applying the
joint authorship test does not require a paradigm shift. Indeed, the case law suggests
that judges already take social norms into consideration when applying the test,
although they tend to do so latently. I have argued that there is value in a more explicit,
and cautious, consideration of the social norms that regulate collective authorship
groups, in the application of the joint authorship test 139 . As mass collaboration is
becoming increasingly common, if copyright law is to retain its credibility as the
primary tool for regulating creativity, it must come to terms with the challenge of
determining the authorship of works of collective authorship. It is hoped that the
approach proposed in this thesis would not only bring the joint authorship test in line
with the reality of collective authorship, but also provide a useful analytical framework
to promote greater clarity in judicial decision-making.
139
G Dinwoodie, ‘Private Ordering and the Creation of International Copyright Norms: The Role of
Public Structuring’ (2004) 160 J of Institutional and Theoretical Economics 161 on the value of the
public structuring of private ordering.
313
Bibliography
—, ‘Creative Nation’, Commonwealth Cultural Policy Statement, October 1994
—, Dracula (1992) (Film)
—, Halsbury’s Laws of England (5th edn, LexisNexis 2012)
—, Jurassic Park (1993) (Film)
—, Pulp Fiction (1994) (Film)
—, ‘Report of the Board of Trade Copyright Committee’ (Her Majesty’s Stationery

Company October 1952) (‘Gregory Report’).
—, ‘Report from the Commission to the Council, the European Parliament and the
Economic and Social Committee on the question of authorship of cinematographic or
audiovisual works in the Community of 6 December 2002’ COM (2002) 691
—, ‘Report of the Royal Commission into Aboriginal Deaths in Custody’, tabled in the
Australian Federal Parliament on 9 May 1991
—, The Nightmare Before Christmas (1993) (Film)
—, The Shining (1980) (Film)
—, ‘Stopping the Rip-Offs’ Issue Paper, Australian Federal Government, October 1994
<http://www.ag.gov.au/Publications/Pages/StoppingtheripoffsOctober1994.aspx>
—, ‘The 7.30 Report’, ABC, 19 August 2010

<http://www.abc.net.au/7.30/content/2010/s2988038.htm>
—, ‘Uniform requirements for Manuscripts Submitted to Biomedical Journals’ (1997)

277 JAMA 928
Adeney E, ‘Unfixed Works, Performers’ Protection, and Beyond: Does the Australian
Copyright Act Always Require Material Form?’ (2009) IPQ 77
Adler A, ‘Against Moral Rights’ (2009) 97 California LRev 263
Albertson Fineman M, The Autonomy Myth: A Theory of Dependency (The New Press
2004)
Annas M, ‘The Label of Authenticity: A Certification Trade Mark for Goods and
Services of Indigenous Origin’ (1997) 3(90) Aboriginal L Bulletin 4
Arnold R, ‘Joy: A Reply’ (2001) 1 IPQ 10
314
Astala M, ‘Comment, Wronged by a Professor? Breach of Fiduciary Duty as a Remedy
in Intellectual Property Infringement Case’ (2003) 3 Houston Bus & Tax LJ 31
Astruc A, ‘The Birth of the Avant Garde: Le Caméra Stylo’, L’Écran français, 30
March 1948
Australian Senate Standing Committee on Environment, Communications, Information

Technology and the Arts, ‘Indigenous Art – Securing the Future: Australia’s Indigenous
Visual Arts and Craft Sector’, June 2007 (‘Senate Report’)
Ayres R, ‘The Wandjina Case Illustrates the Lack of Protection for Indigenous Culture’,
Art+Law, Issue 3, September 2010
Bacharach S and Tollefsen D, ‘We Did It Again: A Reply to Livingston’ (2011) 69(2) J
of Aesthetics and Art Criticism 225
Bacharach S and Tollefsen D, ‘We Did It: From Mere Contributors to Coauthors’
(2010) 68(1) J of Aesthetics and Art Criticism 23
Barlow JP, ‘The Economy of Ideas: Selling Wine Without Bottles on the Global Net’
<https://projects.eff.org/~barlow/EconomyOfIdeas.html>
Barnett J, ‘Hollywood Deals: Soft Contracts For Hard Markets’ Duke LJ (forthcoming)
available at SSRN: <http://ssrn.com/abstract=2118918>
Barron A, ‘Copyright and the Claims of Art’ (2002) 4 IPQ 368
Barron A, ‘Copyright Law and Musical Practice: Harmony or Dissonance?’ (2006)

15(1) Social and Legal Studies 25
Barron A, ‘The Legal Properties of Film’ (2004) 67 MLR 177
Barthes R, ‘The Death of the Author’ in S Heath (trans) Image, Music, Text (Fontana,
1977) 142
Barzilai G, Communities and Law: Politics and Cultures of Legal Identities (U of

Michigan Press 2003)
Bazerman C, ‘Emerging Perspectives on the many dimensions of scientific discourse’

in Martin JS and Veel R (eds), Reading Science: Critical Functional Perspectives on
discourses of Science (Routledge 1988) 15
Beebe B, ‘Intellectual Property and the Problem of Aesthetic Progress’, Ninth Annual
International Intellectual Property Lecture, Emmanuel College, University of
Cambridge, 11 March 2014
Behrendt L, Cunneen C, Libesman T, Indigenous Legal Relations in Australia (OUP

2009)
315
Bently L, ‘Authorship of Popular Music in UK Copyright Law’ (2009) 12(2)
Information Communication & Society 179
Bently L, ‘Copyright and the Death of the Author in Literature and Law’ (1994) 57
MLR 973
Bently L, ‘R v Author: From Death Penalty to Community Service’ (2008) 32(1)

Columbia J of L & the Arts 1
Bently L and Biron L, ‘Discontinuities Between Legal Conceptions of Authorship and

Social Practices: What, if anything, is to be done?’ forthcoming in van Eechoud M (ed),
The Work of Authorship (Amsterdam UP 2014) 237
Bently L and Sherman B, Intellectual Property Law (OUP 2009)
Bhopal R et al, ‘The Vexed Question of Authorship: Views of Researchers in a British

Medical Faculty’ (1997) 314 (7086) British Medical J 1009
Biagioli M and Galison P (eds), Scientific Authorship: Credit and Intellectual Property
in Science (Routledge 2003)
Biagioli M, ‘Documents of Documents: Scientists’ Names and Scientific Claims’ in

Riles A (ed), Documents: Artifacts of Modern Knowledge (U of Michigan Press 2006)
129
Biagioli M, ‘Rights or Rewards? Changing Frameworks of Scientific Authorship’ in

in Science (Routledge 2003) 260
Bicchieri C & Muldoon R, ‘Social Norms’ in Stanford Encyclopaedia of Philosophy

<http://plato.stanford.edu/entries/social-norms/>
Bicchieri C, The Grammar of Society: the Nature and Dynamics of Social Norms (CUP
2006)
Birnholtz J, ‘When Authorship Isn’t Enough: Lessons from CERN on the Implications
of Formal and Informal Credit Attribution Mechanisms in Collaborative Research’
(2008) 11(1) Journal of Electronic Publishing
<http://dx.doi.org/10.3998/3336451.0011.105>
Biron L and Cooper E, ‘Copyright and Multiple Authorship: Insights from the
Philosophy of Art’, HERA Working Paper, 21 June 2013 (on file with the author)
Blackstone W, Commentaries, Volume 1
Bloomfield D, Collins K, Fry C and Munton R, ‘Deliberation and Inclusion: Vehicles

for Increasing Trust in UK Public Governance’ (2001) 12 Environment and Planning C:
Government and Policy 501
316
Bloore P, Managing Creativity and Script Development in the Film Industry (Routledge
2013)
Bloore P, The Screenplay Business: Managing Creativity and Script Development in the
Film Industry (Routledge 2012)
Bonadio E, ‘Joint Ownership of Films in the Absence of Express Terms’ (2012) 7(7)
JIPLP 493
Bowrey K, ‘Alternative IP?: Indigenous Protocols, Copyleft and New Juridifications of

Customary Practices’ (2006) 6 Macquarie LJ 65
Bowrey K, ‘The Outer Limits of Copyright Law – Where Law Meets Philosophy and
Culture’ (2001) 12 L and Critique 75
Bowrey K, Law and Internet Cultures (CUP 2005)
Bowrey K and Anderson J, ‘The Politics of Global Information Sharing: Whose

Cultural Agendas Are Being Advanced?’ (2009) 18 Social and Legal Studies 479.
Boyle J, ‘Second Enclosure Movement and the Construction of the Public Domain’
(2003) 66 L and Contemporary Problems 33
Boyle J, Shamans, Software and Spleens: Law and the Construction of the Information
Society (Harvard UP 1996)
Bracha O, ‘The Ideology of Authorship Revisited: Authors, Markets, and Liberal

Values in Early American copyright’ (2008) 118 Yale LJ 186
Breyer S, ‘The Uneasy Case For Copyright: A Study of Books, Photocopies and
Computer Programs’ (1970) 84(2) Harvard LRev 281
Bruns A, Blogs, Wikipedia, Second Life, and Beyond: from Production to Produsage
(Peter Lang 2008)
Buccafusco CJ, ‘On the Legal Consequences of Sauces: Should Thomas Keller’s
Recipes Be Per Se Copyrightable?’ (2007) 24 Cardozo Arts & Entertainment LJ 1121
Burk D, ‘Research Misconduct: Deviance, Due Process, and the Disestablishment of

Science’ (1995) 3 George Mason U Independent LRev 305
Burkitt D, ‘Copyrighting culture – the history and cultural specificity of the Western
model of copyright’ (2001) IPQ 146
Burns Coleman E, Aboriginal Art, Identity and Appropriation (Ashgate 2005)
Carter S, ‘Custom, Adjudication and Petrushevsky’s Watch: Some Notes from the
Intellectual Property Front’ (1992) 78 Virginia LRev 129
Casey A and Sawicki A, ‘Copyright in Teams’ (2013) 80 U of Chicago LRev 1683
317
Chander A and Sunder M, ‘The Romance of the Public Domain’ (2004) 92 California
LRev 1331
Chemerinsky E, ‘In Defense of Equality: A Reply to Professor Westen’ (1983) 81

Michigan LRev 575
Chon M, ‘New Wine Bursting From Old Bottles: Collaborative Internet Art, Joint
Works, and Entrepreneurship’ (1996) 75 Oregon LJ 257
Clark G and Davies I, ‘Mediation – When is it not an appropriate dispute resolution

process?’ (1992) 3(2) Australian Dispute Resolution J 70
Claxton L, ‘Scientific Authorship Part 2: History, recurring issues, practices, and

guidelines’ (2005) 589 Mutation Research 31
Cohen J, ‘Creativity and Culture in Copyright Theory’ (2007) 40 U of California Davis

LRev 1151
Commonwealth Department of Communications, Information Technology and the Arts,

‘Report of the Contemporary Visual Arts and Crafts Inquiry’, June 2002 (‘Myer
Report’)
Conley T, ‘End Credits’ in Biagioli M and Galison P (eds), Scientific Authorship:

Credit and Intellectual Property in Science (Routledge 2003) 360
Coombe R, ‘Contingent Articulations: A Critical Cultural Studies of Law’ in Sarat A

and Kearns T (eds), Law in the Domains of Culture (U of Michigan Press 1998)
Coombe R, ‘Fear, Hope and Longing for the Future of Authorship and a Revitalized
Public Domain in Global Regimes of Intellectual Property’ (2003) 52 DePaul LRev
1171
Coombe R, The Cultural Life of Intellectual Properties: Authorship, Appropriation and

the Law (Duke UP 1998)
Cooter R, ‘Normative Failure Theory of Law’ (1997) 82 Cornell LRev 947
Corbett S, ‘Creative Commons Licences, the Copyright Regime and the Online
Community: Is there a Fatal Disconnect?’ (2011) 74(4) MLR 503
Cornish W, ‘The Author as Risk-Sharer’, (2002) 26 Columbia J of L & the Arts 1
Cornish W, Llewelyn D and Aplin T, Intellectual Property: Patents, Copyright, Trade

Marks and Allied Rights (8th edn, Sweet & Maxwell 2013)
Craig C, Copyright, Communication and Culture: Towards a Relational Theory of

Copyright Law (Edward Elgar 2011)
318
Davis P, ‘An Authorship Accelerator’
<http://scholarlykitchen.sspnet.org/2008/12/08/hep-authorship> dated 8 December
2008
Decherney P, Hollywood’s Copyright Wars: From Edison to the Internet (Columbia UP

2012)
Department for Culture Media and Sport, ‘A Future For British Film: It Begins With
The Audience…’ Film Policy Review, 2012
Derclaye E, ‘Database Sui Generis Right: What is a Substantial Investment? A

Tentative Definition’ (2005) 36(1) Intl Rev of Intellectual Property and Competition L
2
Derclaye E, ‘Databases Sui Generis Right: Should We Adopt the Spin Off Theory?’
(2004) 26(9) EIPR 402
Derclaye E, ‘Do Sections 3 and 3A of the CDPA violate the Database Directive? A
Closer Look at the Definition of a Database in the UK and its Compatibility with
European Law’ (2002) 24(10) EIPR 466
Derclaye E, ‘The Court of Justice Interprets the Database Sui Generis Right for the first
time’ (2005) 30(3) Eur LR 420
Derclaye E, ‘What is a Database? A Critical Analysis of the Definition of a Database in

the European Database Directive and Suggestions for an International Definition’
(2002) 5(6) JWIP 981
Dietz A, ‘The Concept of Authorship Under the Berne Convention’ (1993) 155 Revue
Internationale du Droit d’Auteur 3
Dinwoodie G, ‘Private Ordering and the Creation of International Copyright Norms:

The Role of Public Structuring’ (2004) 160 J of Institutional and Theoretical Economics
161
Diver A, ‘“A Just War”: Protecting Indigenous Cultural Property’ (2004) 6(4)
Indigenous L Bulletin 7
Dodson M and Barr O, ‘Breaking the Deadlock: Developing and Indigenous Response
to Protecting Indigenous Traditional Knowledge’ (2007) 11(2) Australian Indigenous
LR 19
Dougherty FJ, ‘Not A Spike Lee Joint? Issues in the Authorship of Motion Pictures
Under US Copyright Law’ (2001) 49 UCLA LRev 225
Doyle G, Understanding Media Economics (2nd edn, Sage 2013)
Drahos P with Braithwaite J, Information Feudalism: Who Owns the Knowledge

Economy? (Routledge 2002)
319
Drassinower A, ‘From Distribution to Dialogue: Remarks on the Concept of Balance in
Copyright Law’ (2009) 34(4) J of Corporation L 991
Dreyfuss RC, ‘Collaborative Research: Conflicts on Authorship, Ownership and

Accountability’ (2000) 53 Vanderbilt LRev 1161
Dusollier S, ‘Open Source and Copyleft: Authorship Reconsidered?’ (2003) 26

Columbia J of L and Arts 281
Dusollier S, ‘Sharing Access to Intellectual Property Through Private Ordering’ (2007)

82(3) Chicago-Kent LRev 1391
Dusollier S, ‘The Master’s Tools v the Master’s House: Creative Commons v

Copyright’ (2005) 29 Columbia J of L and Arts 271
Dworkin G, ‘Authorship of Films and the European Commission Proposal for

Harmonising the Term of Copyright’ (1993) EIPR 151
Ebersbach A, Glaser M and Heigl R, Wiki: Web Collaboration (Springer 2006)
Eisenberg R, ‘Proprietary Rights and the Norms of Science in Biotechnology Research’

(1987) 97 Yale LJ 177
Eisenberg R, ‘The Bargain Principle and its Limits’ (1982) 95 Harvard LRev 741
Eliades D, ‘Power in Mediation – Some Reflections’ (1999) 2(1) ADR Bulletin 4
Elkin-Koren N, ‘Tailoring Copyright to Social Production’ (2011) 12(1) Theoretical

Inquiries in L 309
Elkin-Koren N, ‘What Contracts Cannot Do: The Limits of Private Ordering in

Facilitating a Creative Commons’ (2005) 74 Fordham LRev 375
Ellickson R, Order Without Law: How Neighbors Settle Disputes (Harvard UP 1991)
Ellison D, ‘Unauthorised Reproduction of Traditional Aboriginal Art’ (1994) 17

UNSWLJ 327
Endicott T, ‘Questions of Law’ (1998) 114 LQR 292
Epstein R, ‘Academic Fraud Today: Its Social Causes and Institutional Responses’
(2010) 21 Stanford L & Policy Rev 135
Everard D, ‘Code of Conduct for Indigenous Art’, Art+Law, June 2009, available at
<http://www.artslaw.com.au/artlaw/archive/2009/09CodeOfConduct.asp>
Fauchart E and von Hippel EA,‘Norm-Based Intellectual Property Systems: The Case
of French Chefs’ available at
<http://papers.ssrn.com/sol3/papers.cfm?abstract_id=881781>
320
Field R, ‘Mediation and the Art of Power (Im)balancing’ (1996) 12 QUTLJ 264
Finbow K, ‘Nicole Kidman: ‘I had no control over Grace of Monaco’, Digital Spy, 27
May 2014
Fisher R and Ury W, Getting to Yes: Negotiating Agreement Without Giving In

(Houghton Mifflin 1981)
Fisher W, ‘Theories of Intellectual Property’ in S Munzer (ed), New Essays in the Legal
and Political Theory of Property (2001) <http://www.tfisher.org/publications.htm>
Fisher W, ‘Property and Contract on the Internet’ (1998) 73 U Chicago-Kent LRev

1203
Fisk C, ‘The Role of Private Intellectual Property Rights in Markets for Labor and
Ideas: Screen Credit and the Writers Guild of America, 1938-2000’ (2012) 32 Berkeley
J of Employment and Labour L 215
Fisk C, ‘Credit Where It’s Due: The Law and Norms of Attribution’ (2006) 95
Georgetown LJ 49
Fisk C, ‘The Jurisdiction of the Writers Guild to Determine the Authorship of Movies
and Television Programs’ (2010) 48 English Language Notes 15, available at SSRN:
<http://ssrn.com/abstract=1694043>
Foucault M, ‘What is an author?’ in D Bouchard (ed) Language, Counter-Memory,

Practice: Selected Essays and Interviews by Michel Foucault (Cornell UP 1977)
Fox M, Ciro T and Duncan N, ‘Creative Commons: an Alternative, Web-based

Copyright System’ (2005) 16(6) Entertainment LR 111
Fox PS, ‘Preserving the Collaborative Spirit of American Theatre: The Need for a ‘Joint
Authorship Default Rule’ in Light of the Rent Decision’s Unanswered Question’ (2001)
19 Cardozo Arts & Entertainment LJ 497
Free D, ‘Beckingham v Hodgens: The Session Musician’s Claim to Music Copyright’

(2005) 1(3) Entertainment and Sports LJ 93 available at
<http://www.warwick.ac.uk/go/eslj/issues/volume1/number3>
Fugh-Berman AJ, ‘The Haunting of Medical Journals: How Ghostwriting Sold “HRT”’
(2010) 7(9) PLoS Med e1000335
Galison P, ‘The Collective Author’ in Biagioli M and Galison P (eds), Scientific

Authorship: Credit and Intellectual Property in Science (Routledge 2003) 329
Gardner J, ‘Some Types of Law’ in Edlin D (ed), Common Law Theory (CUP 2007) 51
Garfield E, ‘Giving Credit Only Where it is Due: The Problem of Defining Authorship’
(1995) 9(19) The Scientist 13
321
Garnett KM, Davies G, Harbottle G, Copinger WA and Skone James EP, Copinger and
Skone James on Copyright (16th edn, Sweet & Maxwell 2011)
Garon J, ‘Content, Control, and the Socially Networked Film’ (2010) 48 U of Louisville
LRev 771
Gaut B, ‘Film Authorship and Collaboration’ in Allen R and Smith M (eds), Film
Theory and Philosophy (OUP 1997) 149
Gibson J, ‘Justice of Precedent, Justice of Equity: Equitable Protection and Remedies

for Indigenous Intellectual Property’ (2001) 6(4) Australian Indigenous L Reporter 1
Ginsburg J, ‘The Author’s Name as a Trade Mark: A Perverse Perspective on the Moral
Right of “Paternity”?’ (2005) 23 Cardozo Arts and Entertainment LJ 379
Ginsburg J, ‘The Concept of Authorship in Comparative Copyright Law’ (2003) 52 De

Paul L Rev 1063
Goldstein P, Copyright: Principles, Law and Practice (Little Brown & Co 1989)
Golvan C, ‘Aboriginal Art and Copyright: An Overview and Commentary Concerning

Recent Developments’ (1999) EIPR 549
Golvan C, ‘Aboriginal Art and Copyright: The Case for Johnny Bulun Bulun’ (1989)
11(10) EIPR 346
Golvan C, ‘Aboriginal Art and the Protection of Indigenous Cultural Rights’ (1992)
14(7) EIPR 227
Gomulkiewicz R, ‘Conditions and Covenants in License Contracts: Tales from a Test of

the Artistic License’ (2009) 17(3) Texas Intellectual Property LJ 335
Gomulkiewicz R, ‘General Public License 3.0: Hacking the Free Software Movement’s
Constitution’ (2006) 42(4) Houston LRev 1015
Gomulkiewicz R, ‘Open Source License Proliferation: Helpful Diversity or Hopeless

Confusion?’ (2009) 30 Washington UJ of L and Policy 261
Gonzalez AG, ‘Viral Contracts or Unenforceable Documents? Contractual Validity of

Copyleft Licences’ (2004) 26(8) EIPR 331
Gøtzsche PC et al, ‘Ghost Authorship in Industry-Initiated Randomised Trials’ (2007) 4

PLoS Med e535
Gøtzsche PC et al, ‘What Should Be Done To Tackle Ghostwriting in the Medical

Literature?’ (2009) 6(2) PLoS Med. e1000023
Graber C and Taubner G, ‘Art and Money: Constitutional Rights in the Private Sphere?’
(1998) 18 OJLS 61
322
Graber C, ‘Aboriginal Self-determination vs the Proprietisation of Traditional Culture:
The Case of Sacred Wanjina Sites’ (2009) 13(2) Australian Indigenous L Report 18
Gray D, Setting up and Managing a Production Company (Medialex Publications 2006)
Griffin J, ‘The Changing Nature of Authorship: Why Copyright Law Must Focus on the
Increased Role of Technology’ (2005) IPQ 135
Guibault L, ‘Wrapping Information in Contract: How Does it Affect the Public

Domain?’ in Guibault L and Hugenholtz PB (eds), The Future of the Public Domain:
Identifying the Commons in Information (Kluwer Law International 2006) 87
Guibault L, Copyright Limitations and Contracts: An Analysis of the Contractual

Overridability of Limitations on Copyright (Kluwer Law International 2002)
Guibault L and Angelopoulos C, Open Content Licensing: From Theory to Practice

(Amsterdam UP 2011)
Gusterson H, ‘The Death of the Authors of Death: Prestige and Creativity among Nu-
clear Weapons Scientists’ in in Biagioli M and Galison P (eds), Scientific Authorship:
Credit and Intellectual Property in Science (Routledge 2003) 281
Halbert DJ, Intellectual Property in the Information Age: The Politics of Expanding
Ownership Rights (Quorum 1999)
Handler M, ‘Continuing Problems with Film Copyright’ in Macmillan F (ed), New

Directions in Copyright Law: Volume 6 (Edward Elgar 2007) 173
Hargreaves I, ‘Digital opportunity: A Review of Intellectual Property and Growth for

HMG’ (2011) available at: <http://www.ipo.gov.uk/ipreview>
Hawkins C, ‘Stopping the Rip Offs: Protecting Aboriginal and Torres Strait Islander
Cultural Expression’ [1995] Aboriginal L Bulletin 3
Henley M, ‘Jacobsen v Katzer and Kamind Associates – an English Legal Perspective’

(2009) 1(1) IFOSS L Rev 41
Hick DH, ‘Authorship, Co-Authorship, and Multiple Authorship’ (2014) 72(2) J of

Aesthetics and Art Criticism 147
Horton R, ‘The Signature of Responsibility’ (1997) 350 Lancet 5
Huang T, ‘Gaiman v McFarlane: The Right Step in Determining Joint Authorship for
Copyrighted Material’ (2005) 20(1) Berkeley Technology LJ 673
Jackson MW, ‘Can Artisans be Scientific Authors? The Unique Case of Fraunhofer’s
Artisanal Optics and the German Republic of Letters’ in Biagioli M and Galison P
(eds), Scientific Authorship: Credit and Intellectual Property in Science (Routledge
2003) 113
323
Janke T, ‘Briefs: Copyright – The Carpet Case’ [1995] Aboriginal L Bulletin 13; (1995)
3(72) Aboriginal L Bulletin 36
Janke T, Beyond Guarding Ground: A Vision for a National Indigenous Cultural

Authority (Terri Janke & Co 2009)
Janke T, Minding Culture: Case studies on intellectual property and traditional cultural
expressions (WIPO 2003)
Janke T, Our Culture: Our Future A Report on Australian Indigenous Cultural and
Intellectual Property Rights (Michael Frankel & Co and Terri Janke 1998)
Jaszi P, ‘Towards a Theory of Copyright: The Metamorphoses of Authorship’ (1991) 2

Duke LJ 455
Jaszi P, ‘On The Author Effect: Recovering Collectivity’ in Woodmansee M and Jaszi
P (eds) The Construction of Authorship: Textual Appropriation in Law and Literature
(Duke U Press 1999) 29
Jensen C, ‘The More Things Change, the More They Stay the Same: Copyright, Digital
Technology, and Social Norms (2003) 56(2) Stanford LRev 531
Johnson P, ‘“Dedicating” Copyright to the Public Domain’ (2008) 71(4) MLR 587
Jones AH, ‘Can Authorship Policies Help Prevent Scientific Misconduct? What Role
for Scientific Societies?’ (2003) 9 Science and Engineering Ethics 243
Kahan D, ‘Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms Problem (2000)
97 U of Chicago LRev 607
Kamina P, Film Copyright in the European Union (CUP 2004)
Katz A, ‘Taking Private Ordering Seriously’ (1996) 144 U Pennsylvania LRev 1745
Kaufman R, ‘After 5 Years, Heated Controversy Persists in Science Copyright Case’

The Scientist, 14 September 1992 at: <http://www.the-
scientist.com/?articles.view/articleNo/12483/title/After-5-Years-Heated-Controversy-
Persists-In-Science-Copyright-Case/>
Keen A, The Cult of the Amateur: How Today's Internet is Killing our Culture
(Doubleday/Currency 2007)
Keith Grant B, Auteurs and Authorship: A Film Reader (Blackwell 2008)
Kleinart S and Neale M, The Oxford Companion to Aboriginal Art and Culture (OUP
2000)
Kwall RR, ‘Author-Stories: Narrative’s Implications For Moral Rights And Copyright's
Joint Authorship Doctrine’ (2001) 75 Southern California LRev 1
324
Kwall RR, ‘The Author as Steward “for Limited Times”: A Review of “The Idea of
Authorship in Copyright” (Lior Zemer, Ashgate Press, 2007)’ (2008) 88(3) Boston U
LRev 685
Kwall RR, The Soul of Creativity: Forging a Moral Rights Law for the US (Stanford
UP 2009)
Lacey LJ, ‘Of Bread and Roses and Copyright’ [1989] Duke LJ 1532
Laddie H, Prescott P and Vitoria M, The Modern Law of Copyright and Designs (4th
edn, LexisNexis 2011)
LaFrance M, ‘Authorship, Dominance, and the Captive Collaborator: Preserving the

Rights of Joint Authors’ (2001) 50 Emory LJ 193
Lastowska G, ‘The Trade Mark Function of Authorship’ (2005) 85 Boston U LRev

1171
Lemley M, ‘Intellectual Property and Shrinkwrap Licenses’ (1995) 68 Southern

California LRev 1239
Lemley M, ‘Romantic Authorship and the Rhetoric of Property’ (1997) 75 Texas LRev
873
Lessig L, Free Culture: How Big Media uses Technology and the Law to Lock Down
Culture and Control Creativity (Penguin Press 2004)
Lessig L, Remix: Making Art and Commerce Thrive in the Hybrid Economy (Penguin
Press 2008)
Lessig L, The Future of Ideas: The Fate of the Commons in a Connected World
(Vintage Books 2002)
Lipton J, ‘Wikipedia and the European Union Database Directive’ (2010) 26 Santa
Clara Computer & Technology LJ 631
Litman J, ‘Copyright Noncompliance (or why we can’t “just say yes” to licensing)’
(1996) 29 NYU J of Intl L and Politics 237
Livingston P, ‘Cinematic Authorship’ in Allen R and Smith M (eds), Film Theory and
Philosophy (OUP 1997) 132
Livingston P, ‘On Authorship and Collaboration’ (2011) 69(2) J of Aesthetics and Art
Criticism 221
Livingston P, Art and Intention (OUP 2005)
Lopez Romero T, ‘Sui Generis Systems for the Protection of Traditional Knowledge’
(2005) 6 International L: Revista Colombiana de Derecho Internacional 301
325
Loren LP, ‘Building a Reliable Semicommons of Creative Works: Enforcement of
Creative Commons Licenses and Limited Abandonment of Copyright’ (2007) 14
George Mason LRev 271
Loshin J, ‘Secrets Revealed: How Magicians Protect Intellectual Property without Law’
in Corcos C, Law and Magic: A Collection of Essays (Carolina Academic Press 2008)
123
Love J and Hubbard T, ‘The Big Idea: Prizes to Stimulate R&D for New Medicines’
(2007) 82 Chicago-Kent LRev 1519
Loux A, ‘The Persistence of the Ancient Regime: Custom, Utility and the Common
Law in the Nineteenth Century’ (1993) 79 Cornell LRev 183
Macey J, ‘Public and Private Ordering and the Production of Legitimate and
Illegitimate rules’ (1997) 82 Cornell L Rev 1123
Mackay E, ‘Indigenous Traditional Knowledge, Copyright and Art – Shortcomings in

Protection and an Alternative Approach’ (2009) 32(1) UNSWLJ 1
MacMillan F, ‘The Cruel C: Copyright and Film’ (2002) EIPR 483
Macnab G, ‘Johnny Depp's new film Transcendence is yet another expensive flop - so
has the star lost his mojo?’ The Independent, 30 April 2014
Madison M, ‘Legal-ware: Contract and Copyright in the Digital Age’ (1998) 67

Fordham LRev 1025
Mainous III AG, Bowen MA, Zoller JS, ‘The importance of interpersonal relationship
factors in decisions regarding authorship’ (2002) 34 Family Medicine 462
Manton B, ‘Russian Ice Dancers Should Re-think their Routine’ Sydney Morning
Herald, 21 January 2010 <http://www.smh.com.au/opinion/politics/russian-ice-dancers-
should-rethink-their-routine-20100121-mnwj.html>
Marrella F and Yoo CS, ‘Is Open Source Software the New Lex Mercatoria?’ (2006–7)
47 Virginia J of International L 807
Martin A, ‘Possessory Credit’ (2004) 45 (1) Framework 95
Matheson A, ‘How Industry Uses the ICMJW Guidelines to Manipulate Authorship –

And How They Should Be Revised’ (2011) 8(8) PLoS Med. 1001072
McCausland S, ‘Protecting Communal Interests in Indigenous Artworks after the Bulun

Bulun Case’ (1999) 4(22) Indigenous L Bulletin 4
McHenry L, ‘Of Sophists and Spin-Doctors: Industry-Sponsored Ghostwriting and the

Crisis of Academic Medicine’ (2010) 8(1) Journalology 129
326
McLaughlin GM, ‘Oral Contracts in the Entertainment Industry’ (2001) 1(1) Virginia
Sports and Entertainment LJ 101
McSherry C, ‘An Uncommon Controversy’ in C McSherry, Who Owns Academic

Work: Battling for Control of Intellectual Property (Harvard UP 2001) 68
Merges R, Justifying Intellectual Property (Harvard UP, 2011)
Merges R, ‘Locke for the Masses: Property Rights and the Products of Collective
Creativity’ (2008) 36(4) Hofstra LRev 1179
Merges R, ‘The Concept of Property in the Digital Era’ (2008) 45(4) Houston LRev
1239
Merton RK, The Sociology of Science: Theoretical and Empirical Investigations (Chi-
cago UP, 1973)
Mnookin R and Korhauser L, ‘Bargaining in the Shadow of the Law: The Case of
Divorce’ (1979) 88 Yale LRev 950
Mnookin R, ‘Divorce Bargaining: The Limits on Private Ordering’ 18(4) J of L Reform

1015
Moglen E, ‘Free Software Matters: Enforcing the GPL, I’

<http://emoglen.law.columbia.edu/publications/lu-12.pdf>
Morphy H, Ancestral Connections: Art and an Aboriginal System of Knowledge (U of

Chicago Press 1991)
Murphy R, ‘Postscript: A Short History of British Cinema’ in R Murphy (ed) The

British Cinema Book (3rd edn, Palgrave Macmillan 2009) 417
Murray LJ, Tina Piper S and Robertson K, Putting Intellectual Property in Its Place:
Rights Discourses, Creative Labor, and the Everyday (OUP 2014)
Naremore J, ‘Authorship’ in Miller T and Stam R(eds) A Companion to Film Theory

(Blackwell Publishing 2005) 9
Netanel NW, ‘Copyright and Democratic Civil Society’ (1996) 106 Yale LR 283
Nimmer D, ‘Copyright in the Dead Sea Scrolls: Authorship and Originality’ (2001) 38
Houston LRev 1
Nimmer MB and Nimmer D, Nimmer on Copyright (Matthew Bender 1998, 1976 and
2011 editions)
Notaro A, ‘Technology in Search of an Artist: Questions of Auteurism/Authorship and

the Contemporary Cinematic Experience’ (2006) 57 The Velvet Light Trap 86
327
O’Reilly T, ‘What is Web 2.0?’
<http://www.oreillynet.com/pub/a/oreilly/tim/news/2005/09/30/what-is-web-20.html>
O’Sullivan M, ‘The Pluralistic, Evolutionary, Quasi Legal Role of the GNU General
Public Licence in Free/Libre/Open Source Software (FLOSS)’ (2004) 26(8) EIPR 340
Obergfell EI, ‘No need for harmonising film copyright in Europe?’ (2003) 4 European
Legal Forum 199
Oliar D and Sprigman C, ‘There’s No Free Laugh (Anymore): The Emergence of

Intellectual Property Norms and the Transformation of Stand-up Comedy’ (2008) 94
Virgina LRev 1787
Ong B, ‘Originality from Copying: Fitting Recreative Works into the Copyright
Universe’ (2010) 2 IPQ 165
Oster J, ‘Proposal for a Regulated Indigenous Art Industry’ (2006) 6 (18) Indigenous
LBulletin 21
Parker RA and Bergman NA, ‘Criteria for Authorship for Statisticians in Medical
Papers’ (1998) 17(20) Statistics in Medicine 2289
Perry M and Margoni T, ‘Ownership in complex authorship: a comparative study of

joint works’ (2012) 34(1) EIPR 22
Pettie C, Seyd P, Whiteley P, Citizenship in Britain: Values, Participation and

Democracy (CUP 2004)
Phillips J, ‘Authorship, Ownership and Wikiship: Copyright in the Twenty-First

Century’ (2008) 3 (12) JIPLP 788
Phillips R, ‘Researchers’ Objective is to Get the Job Done’ (1997) 315 British Medical
J 747
Pila J, ‘An Intentional View of the Copyright Work’ (2008) 71(4) MLR 535
Pila J, ‘Authorship and e-Science: Balancing Epistemological Trust and Skepticism in

the Digital Environment’ (2009) 23 Social Epistemology 1
Pila J, ‘The “Star Wars” Copyright Claim: An Ambivalent View of the Empire’ (2012)
LQR 15
Poll G, ‘Harmonization of Film Copyright in Europe’ (2002) 50 J of the Copyright

Society of the USA 519
Posner E (ed), Social Norms, Nonlegal Sanctions, and the Law (Edward Elgar 2007)
Posner E, ‘Introduction’ in Posner E (ed), Social Norms, Nonlegal Sanctions, and the
Law (Edward Elgar 2007)
328
Posner E, ‘Law, Economics and Inefficient Norms’ (1996) 144 U Pennsylvania LRev
1697
Posner E, ‘The Regulation of Groups: The Influence of Legal and Nonlegal Sanctions
on Collective Action’ (1996) 63 U of Chicago LRev 133
Posner E, Law and Social Norms (Harvard UP 2009)
Posner E, ‘Norms, Formalities, and the Statute of Frauds: a Comment’ (1996) 144 U
Pennsylvania LRev 1971
Posner R, Law and Literature (3rd edn, Harvard UP 2009)
Price M and Pollack M, ‘The Author in Copyright: Notes for the Literary Critic’ in
Woodmansee M and Jaszi P (eds) The Construction of Authorship: Textual
Appropriation in Law and Literature (Duke U Press 1999) 439
Radin MJ and Wagner RP, ‘The Myth of Private Ordering: Rediscovering Legal
Realism in Cyberspace’ (1998) 73 Chicago-Kent LRev 1295
Rafaeli S and Ariel Y, ‘Online Motivational Factors: Incentives for Participation and
Contribution in Wikipedia’ in Barak A (ed), Psychological Aspects of Cyberspace:
Theory, Research, Applications (CUP 2008)
Rahmatian A, Copyright and Creativity: The Making of Property Rights in Creative

Works (Edward Elgar 2011)
Raustiala K and Sprigman C, ‘The Piracy Paradox: Innovation and Intellectual Property
in Fashion Design’ (2006) 92 Virginia LRev 1687
Raymond E, ‘The Cathedral and the Bazaar’

<http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar/>
Raz J, ‘Rule of Law and its Virtue’ in J Raz, The Authority of Law: Essays on Law and
Morality (Clarendon Press 1979) 210
Reichman JH & Franklin JA, ‘Privately Legislated Intellectual Property Rights:

Reconciling Freedom of Contract with Public Good Uses of Information’ (1999)147(4)
U of Pennsylvania LRev 875
Rennie D, Yank V and Emanuel L, ‘When Authorship Fails: A Proposal to Make Con-
tributors Accountable’ (1997) 278 JAMA 579
Reynolds W and Moringello J, ‘Survey of the Law of Cyberspace: Electronic

Contracting Cases 2008-2009’ (2009) The Business Lawyer 65 available at
<http://ssrn.com/abstract=1437162>
Rheinberger H, ‘“Discourses of Circumstance”: A Note on the Author in Science’ in

in Science (Routledge, 2003)
329
Richardson W, Blogs, Wikis, Podcasts, and Other Powerful Web Tools for Classrooms
(Corwin Press 2006)
Ricketson S, ‘The Concept of Originality in Anglo-Australian Copyright Law’ (1991)

9(2) Copyright Reporter 1
Ricketson S, ‘People or Machines? The Berne Convention and the Changing Concept
of Authorship’ (1991) 16 Columbia J of L and the Arts 1
Rimmer M, ‘Australian Icons: Authenticity Marks and Identity Politics’ 2004(3)

Indigenous LJ 139
Rimmer M, ‘The Bangarra Dance Theatre: Copyright and Indigenous Culture’ (2000)
9(2) Griffith LRev 274
Rimmer M, ‘Heretic: Copyright Law and Dramatic Works’ (2002) 2(1) QUT LRev 131
Rivers T, ‘Norowzian Revisited’ (2000) EIPR 389
Rome E, ‘Oscars: Visual Effects Artists protest outside Dolby Theatre’ Entertainment
Weekly, 25 February 2013 <http://insidemovies.ew.com/2013/02/25/oscars-visual-
effects-protest-life-of-pi/>
Rosati E, ‘Originality in a Work, or a Work of Originality: The Effects of the Infopaq

Decision’ (2011) 33(12) EIPR 746
Rose M, ‘Copyright and its Metaphors’ (2002) 50 UCLA LRev 1
Rose M, ‘The Author as Proprietor: Donaldson v Beckett and the Geneology of Modern
Authorship’ (1988) 23 Representations 51
Rose M, Authors and Owners: The Invention of Copyright (Harvard UP 1993)
Rosen C, ‘‘Transcendence’ is a real bad flop for Johnny Depp’, The Huffington Post, 21
April 2014 <http://www.huffingtonpost.com/2014/04/21/johnny-depp-transcendence-
flops_n_5184999.html>
Rosenblatt E, ‘A Theory of IP’s Negative Space’ (2011) 34(3) Columbia J of L and the
Arts 317
Ross M, ‘Authority and Authenticity: Scribbling Authors and the Genius of Print in
Eighteenth Century England’ (1992) 10 CAELJ 495
Rothman J, ‘Copyright, Custom and Lessons from the Common Law’ in Balganesh S
(ed) Intellectual Property and the Common Law (CUP 2013) 230
Rothman J, ‘The Questionable Use of Custom in Intellectual Property’ (2007) 93

Virginia LRev 1899
330
Sackville R (Justice), ‘Legal Protection of Indigenous Culture in Australia’ (2003)
11(2) Cardozo J of International and Comparative L 711
Sagy T, ‘What’s So Private About Private Ordering’ (2011) 45(4) Law and Society Rev
923
Sanger L, ‘WHO SAYS WHAT WE KNOW: On the New Politics’, The Edge,
<http://www.edge.org/3rd_culture/sanger07/sanger07_index.html>
Sarris A, ‘Notes on the Auteur Theory in 1962’ in Keith Grant B (ed) Auteurs and
Authorship: A Film Reader (Blackwell 2008) 35
Saunders D, Authorship and Copyright (Routledge 1992)
Schwab J, ‘Audiovisual Works and the Work for Hire Doctrine in the Internet Age’
(2011) 35 Columbia JL & Arts 141
Schwarcz S, ‘Private Ordering’ (2002) 97 Northwestern U LRev 319
Scott T, ‘Changing Authorship System Might be Counterproductive’ (1997) 315 British

Medical J 744
Sell S, Private Power, Public Law: The Globalisation of Intellectual Property Rights
(CUP 2003)
Sellors P, ‘Collective Authorship in Film’ (2007) 65 J of Aesthetics and Art Criticism

263
Seymore S, ‘How does my work become our work? Dilution of authorship in scientific
papers, and the need for the academy to obey copyright law’ (2006) 12 Richmond J of
L & Tech 11
Shapin S and Schaffer S, Leviathan and The Air-Pump: Hobbes, Boyle, and The
Experimental Life (Princeton UP 1985)
Shapin S, ‘How to be Antiscientific’ in Labinger JA and Collins H (eds), The One

Culture? A Conversation about Science (Chicago UP 2001)
Shapin S, ‘The Invisible Technician’ (1989) 77 American Scientist 554
Shapin S, A Social History of Truth: Civility and Science in Seventeenth-Century

England (Chicago UP 1994)
Sherman B, ‘What is a copyright work?’ (2011) 12(1) Theoretical Inquiries in L 99
Schultz M, ‘Fear and Norms and Rock & Roll: What Jambands Can Teach Us about
Persuading People to Obey Copyright Law’ (2009) Berkeley Technology LJ 651
Silbey J, The Eureka Myth: Creators, Innovators, and Everyday Intellectual Property
(Stanford Press 2014)
331
Simone D, ‘Copyright or Copyleft?: Wikipedia as a Turning Point for Authorship’
(2014) 25(1) Kings LJ 102
Simone D, ‘Recalibrating the Joint Authorship Test: Insights from Scientific

Collaborations' (2013) 26(1) Intellectual Property J 111
Sismondo S, ‘Ghosts in the Machine: Publication Planning in the Medical Sciences’

(2009) 39 Social Studies of Science 171
Smith H, ‘Community and Custom in Property’ (2009) 10(1) Theoretical Inquiries in L

5
Smith J, ‘Gift Authorship: A Poisoned Chalice?’ (1994) 309 British Medical J 1456
Spence M, Intellectual Property (OUP 2007)
Spence M and Endicott T, ‘Vagueness in the Scope of Copyright’ (2005) 121 LQR 657
Stamatoudi I, ‘“Joy” for the Claimant: Can a Film Also be Protected as a Dramatic
Work’ (2000) 1 IPQ 117
Steinberg NS, ‘Regulation of Scientific Misconduct in Federally Funded Research’

(2000) 10 South California Interdisciplinary LJ 39
Stern S and Lemmens T, ‘Legal Remedies for Medical Ghostwriting: Imposing Fraud
Liability on Guest Authors of Ghostwritten Articles’ (2011) 8(8) PLoS Med e1001070
Stokes A, ‘Authorship, Collaboration and Copyright a View from the United Kingdom’
(2002) Entertainment LRev 121
Strandburg K, ‘User Innovator Community Norms: At the Boundary Between

Academic and Industry Research’ (2009) 77 Fordham LRev 101
Sunstein C, Infotopia: How Many Minds Produce Knowledge (OUP 2006)
Swan A, ‘Open Access and the Progress of Science’ 95 American Scientist 198
Tankebe J, ‘Viewing Things Differently: The Dimensions of Public Perceptions of

Police Legitimacy’ (2013) 51 Criminology 103
Tapscott D and Williams A, Wikinomics: How Mass Collaboration Changes Everything

(Atlantic Books 2008)
Targosz T, ‘Authorship by Deletion’, Kluwer Copyright Blog,

<http://kluwercopyrightblog.com/2011/11/02/authorship-by-deletion-supreme-court-
june-22-2010-ii-csk-52710/>
Tehranian J, ‘Infringement Nation: Copyright Reform and the Law/Norm Gap’ (2007)
Utah LRev 537
332
Terrall M, ‘The Uses of Anonymity in the Age of Reason’ in Biagioli M and Galison P
2003)
Todd A, ‘Painting a Moving Train: Adding “Postmodern” to the Taxonomy of Law’

(2008) 40(1) U of Toledo LRev 105.
Truffaut F, ‘A Certain Tendency of the French Cinema’ (1954) 31 Cahiers du Cinéma

translated and reproduced in Keith Grant B, Auteurs and Authorship: A Film Reader
(Blackwell 2008) 9
Tyler T and Jackson J, ‘Popular Legitimacy and the Exercise of Legal Authority:
Motivating Compliance, Cooperation and Engagement’ (2014) 20(1) Psychology Public
Policy and L 78
Tyler T, ‘Compliance with Intellectual Property Law: A Psychological Perspective’

(1996) 29 International L and Politics 219
Tyler T, Why People Obey the Law (Princeton UP 2006)
Valiela I, Doing Science: Design, Analysis, and Communication of Scientific Research

(OUP 2001)
van Eechoud M, ‘Along the Road to Uniformity – Diverse Readings of the Court of
Justice’s Judgments on Copyright Work’ (2012) 1 JIPITEC 60
Vaver D, Copyright Law (Irwin Law, 2000) 76
Vaver D, Intellectual Property: Copyright, Patents and Trade-Marks (2nd ed, Irwin Law
2011)
Versteeg R, ‘Defining “Author” for Purposes of Copyright’ (1996) 45 American U

LRev 1323
Vesterman W, ‘The Death of the Scientific Author: Multiple Authorship in Scientific

Papers’ (2002) 8 Common Knowledge 439
Wacha J, ‘Taking the Case: Is the GPL Enforceable?’ (2005) 21 Santa Clara Computer
& High-Technology LJ 451
Wager E, ‘Authors, Ghosts, Damned Lies, and Statisticians’ (2007) 4(1) PLoS Med e34
Walshok M, Knowledge without Boundaries: What America’s Research Universities

Can Do for the Economy, the Workplace, and the Community (Jossey-Bass 1995)
Wartenberg T, ‘Philosophy of Film’ in Zalta EN (ed) The Stanford Encyclopedia of

Philosophy (2014) <http://plato.stanford.edu/archives/fall2014/entries/film/>
333
Warwick A, ‘“A very hard nut to crack” or Making Sense of Maxwell’s Treatise on
Electricity and Magnetism in Mid-Victorian Cambridge’ in Biagioli M and Galison P
2003) 133
Watson P, ‘Cinematic Authorship and the Film Auteur’ in Nelmes J (ed) Introduction to
Film Studies (5th ed, Routledge 2012) 142
Weatherall K, ‘Culture, Autonomy and Djulibinyamurr: Individual and Community in

the Construction of Rights to Traditional Designs’ (2001) 64 MLR 215
Weinberg J, ‘ICANN and the Problem of Legitimacy’ (2000) 50 Duke LJ 187
Weinburg AM, ‘Impact of Large-Scale Science on the United States’ (1961) 134 (3473)
Science 161
Wendland W, ‘Intellectual Property, Traditional Knowledge and Folklore: WIPO’s

Exploratory Program – Part 1’ (2002) International Review of Intellectual Property and
Competition L 485
Wiseman L, ‘The Protection of Indigenous Art and Culture in Australia: the Labels of
Authenticity’ (2001) EIPR 14
Woodmansee M, ‘On the Author Effect: Recovering Collectivity’ in Woodmansee M

and Jaszi P (eds) The Construction of Authorship: Textual Appropriate in Law and
Literature (Duke U Press 1999) 15
Woodmansee M, ‘The Genius and the Copyright: Economic and Legal Conditions of
the Emergence of the “Author”’ (1983-1984) 17 Eighteenth Century Studies 425
Woodmansee M and Jaszi P (eds) The Construction of Authorship: Textual

Appropriation in Law and Literature (Duke U Press 1999)
Zemer L, ‘Contribution and Collaborations in Joint Authorship: Too Many Mis-

conceptions’ (2006) 1(4) JIPLP 283
Zemer L, ‘Is Intention to Co-author an Uncertain Realm of Policy?’ (2007) 30(4)

Columbia J of L and the Arts 611
Zemer L, The Idea of Authorship in Copyright (Ashgate 2007)
Ziman J, Of One Mind: The Collectivisation of Science (American Institute of Physics

Press 1995)
Ziman J, Real Science: what it is, and what it means (CUP 2000)
Ziman J, Science in Civil Society (Imprint Academic 2007)
Zimmerman D, ‘Copyright as Incentives: Did We Just Imagine That?’ (2011) 12(1)

Theoretical Inquiries in L 29
334
Websites
All websites last accessed on 30 September 2014.
<http://en.wikipedia.org/wiki/Big_Science>
<http://en.wikipedia.org/wiki/Main_Page>
<http://en.wikipedia.org/wiki/Wikipedia:Copyrights>
<http://en.wikipedia.org/wiki/Wikipedia:Copyrights>
<http://en.wikipedia.org/wiki/Wikipedia:Text_of_the_GNU_Free_Documentation_License>
<http://en.wikipedia.org/wiki/Wikipedia:Etiquette>
<http://en.wikipedia.org/wiki/Wikipedia:Five_pillars>
<http://en.wikipedia.org/wiki/Wikipedia:List_of_policies_and_guidelines>
<http://en.wikipedia.org/wiki/Wikipedia:Ownership_of_articles>
<http://en.wikipedia.org/wiki/Wikipedia:Barnstars>
<http://en.wikipedia.org/wiki/Wikipedia:Text_of_Creative_Commons_Attribution-
ShareAlike_3.0_Unported_License>
<http://en.wikipedia.org/wiki/Wikipedia:Welcoming_committee/Welcome_to_Wikipedia>
<http://public.web.cern.ch/public/en/About/Global-en.html>
<http://www.afc.gov.au/filminginaustralia/indigproto/fiapage 9.aspx>
<http://www.alexa.com/siteinfo/wikipedia.org#>
<http://www.artslaw.com.au/Indigenous/IndigenousArtistAndArtCentreAgreement.asp>
<http://www.australiacouncil.gov.au>
<http://www.bfi.org.uk/film-industry>
<http://www.creativecommons.org
<http://www.culture.com.au/exhibition/niaaa/labelqa.htm>
<http://www.indigenousartcode.org>
<http://www.melboume.vic.gov.au>
<http://www.oed.com>
<http://www.wikipedia.org>
335

Copyright and Collective Authorship

Uploaded by

Copyright:

Available Formats

Copyright and Collective Authorship

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Copyright and Collective Authorship

Uploaded by

Copyright:

Available Formats

What is the thesis about?

What is the thesis about?

What are the four case studies analyzed?

What are the four case studies analyzed?

COPYRIGHT AND COLLECTIVE

SUPERVISOR: PROF GRAEME DINWOODIE

Trinity Term 2014

COPYRIGHT AND COLLECTIVE AUTHORSHIP

DANIELA SIMONE, WORCESTER COLLEGE

DPHIL, TRINITY TERM 2014

I was fortunate to have the opportunity to present my work in progress at a number of

I owe a particular debt of gratitude to a number of intellectual property scholars whose

Most importantly, I am grateful to all of my family. My parents’ selfless love created

Table of Abbreviations ...................................................................................................... i

Chapter 1: Introduction .................................................................................................... 1

Chapter 2: Authorship and Joint Authorship ................................................................. 16

Chapter 3: Wikipedia ...................................................................................................... 65

Chapter 5: Scientific Collaborations ............................................................................. 139

Chapter 6: Film ............................................................................................................. 177

Bibliography ................................................................................................................. 314

1911 Act – Copyright Act 1911

1956 Act – Copyright Act 1956

Copinger et al – KM Garnett, G Davies, G Harbottle, WA Copinger and EP Skone

Cornish et al – W Cornish, D Llewelyn and T Aplin, Intellectual Property: Patents,

Duration Directive – Directive 2006/116/EC of the European Parliament and of the

InfoSoc Directive – Directive 2001/29/EC of the European Parliament and of the

IPQ – Intellectual Property Quarterly

JIPLP – Journal of Intellectual Property Law and Practice

Myer Report – Commonwealth Department of Communications, Information

Rental and Related Rights Directive – Directive 2006/115/EC of the European

Senate Report – Australian Senate Standing Committee on Environment,

Abraham Moon v Thornber [2012] EWPCC 37, [2013] FSR 17 23, 24

Aboriginal Sacred Sites Protection Authority v Maurice; Re the Warumbingu Land

relevance of copyright law standards for the regulation of collaborative creativity.

1.1 Copyright Law and Collective Authorship

crisis of legitimacy 1 . In recent decades, the successful lobbying of entertainment

distributing creative works, than it is towards rewarding and incentivising authors. At

the same time, non-compliance with copyright law is becoming increasingly

themselves as an ‘ethical alternative’ to copyright, are also gaining popularity. As

non-compliance may appear unsurprising 2 . Indeed, psychologists have demonstrated

creating a work of collective authorship and, in particular, analyses the dynamics of

collective authorship’ refers to any work that is created by many contributors6.

Large scale collaboration is becoming increasingly widespread and economically

significant. This is due, in large part, to advances in digital and communications

technology. In their best-selling book, Wikinomics, Don Tapscott and Anthony

is adapted to the challenge of regulating collaborative works10. If copyright law is to

with this challenge.

literary trope which conceives of authors as ‘solitary geniuses’, has influenced

copyright law leaving it ill-adapted to collaborative creativity11. Others offer a more

adapted to collaborative creativity. Instead, it investigates the underlying assumption

1.2 Method and Structure of the Thesis

with an increased interest in economic analysis and empirical investigations among

intellectual property scholars. Although valuable, these frames of reference do not

law within creative communities. I embrace an interdisciplinary approach that draws

abstract approach of previous copyright scholarship14.

amateur (Wikipedia) examples. In each case, authorship is driven by different

motivations, from largely commercial motivations (Film), to religious and spiritual

provide examples of a range of different ways in which issues of authorship might be

In order to investigate how best to determine the authorship of works of collective