iTheories of Film

il \

'^m'' *14
 CHILIS SUT^L/FFB

m m Cinema One

23 Theories of Filiri
 f; ^ iTTfT- - "L - r

r.t
, , i
, I • * ;'i ;
I 'v
* . -4 •ifcr!

*♦

I • '
Theories of Film
Andrew Tudor

London

Seeker and Warburg in association with th^

British Film Institute
The Cinema One Series is published by
Martin Seeker & Warburg Limited
14 Carlisle Street, Soho Square, London WIV 6NN
in association with the
British Film Institute, 81 Dean Street, London WIV 6AA

General Editors
Penelope Houston and David Wilson (Sight and Sound)
Christopher Williams (Educational Advisory Services Department)

Theories of Film by Andrew Tudor

first published by Seeker & Warburg 1974
Copyright © 1974 Andrew Tudor
Second Impression 1975

SEN 436 09936 5 (hardcover)

SEN 436 09937 3 (paperback)

Designed by Michael Farrell

Filmset in Photon Times 12 pt. by

Richard Clay (The Chaucer Press), Ltd, Eungay, Suffolk
and printed in Great Britain by
Fletcher & Son, Ltd, Norwich
Contents

Bloyv-Up (Michelangelo Antonioni)

1. Introduction

2. Eisenstein: Great Beginnings

3. The Problem of Context: John Grierson

4. Aesthetics of Realism: Bazin and Kracauer

5. Critical Method: Auteur and Genre

6. Epilogue

Bibliography

Acknowledgements

Cover: A Fistful ofDynamite (Sergio Leone)

 >?! '•!
f,,

Hitchcock: Vertigo and Frenzy (top); Hawks: Bringing Up Baby

1: Introduction

Recent years have seen a developing interest in serious study of

the cinema. The range and number of close analyses of the
work of particular directors and studies of special aspects of
film has increased, it sometimes seemed, day by day. In the face
of this burgeoning obsession it is thus a little odd that film
theory, classically defined by Eisenstein, has remained much as
it was: a little practised and barely reputable pursuit. Though
we may now take seriously a Howard Hawks or an Alfred
Hitchcock,' to 'theorize' about film still smacks of over-
intellectualism. It is one thing to recognize the need to reflect
self-consciously on the critic's 'tools of the trade'; it is another
to indulge. In some part this is a consequence of the traditional
Anglo-Saxon mistrust of such a suspiciously European pursuit.
English thinking, and not only on film, has seemingly always
preferred to keep its feet firmly planted on the ground, a policy
which has not been without its rewards. But on the debit side it
has also led to a situation whereinthe larger intellectual environ
ment has barely felt the need to address itself to the cinema.
Even now our orthodox academic institutions lack a context in
which to study the major new art of the twentieth century
b See, for example, Robin Wood, Hitchcock's Films, A. Zwemmer Ltd.,
London, 1965, and Howard Hawks, Seeker and Warburg/B.F.I.'
London, 1968.
as anything more than some sort of sociological curiosity.
Academic aesthetics has concerned itself with more 'profound'
and 'genuine' arts.
This belief that film aesthetics is not quite respectable has
had variops consequences. As I shall go on to amplify, such
antagonism pressured attention away from Eisenstein's 'scien
tific' and general interest in a theory of film and toward a
defensive need to justify the new 'art'. Much of the effort which
might have gone toward extending our understanding of the
new medium has instead been poured into the bottomless pit of
aesthetic respectability. The proposition that Film was indeed
Art haunted a whole generation of critics and was to set a limit
on their ambitions and on their contribution. Because of such
outside social and intellectual pressures - the intellectual
antagonism was socially supported - we have inherited a cul
ture of film with a vacuum at the centre. Seventy-five years of
history have left us with no unified body of knowledge on
which to draw; no consistent set of terms to employ. Other
than in the precariously surviving specialist journals film
criticism remains even now a fundamentally dilettantish pur
suit. The critic of one English Sunday paper recently implied
that Hitchcock had only been in the business some eighteen
years in which time he had made less than a dozen films. The
film under review had already received publicity as his fiftieth
feature! In magnitude this is atypical; in trend it is all too
common. ^ ,
It can hardly be the intention of this book to fill such a gap ,
even in its theoretical aspect. Optimistically assuming that such
changes do come about, they will be many years inthe making.
But this book does ask that we give some portion of our
expanding energies to the particular area of development very
broadly defined as 'theory' of film. That we not only take films
seriously; we look likewise to the history of critical writings.
Elia Kazan, in a bitter moment, accused film critics of being
eunuchs; of writing about what they were unable to do. To
 press the analogy further is to recollect that the eunuchs
frequently played a crucial role in their world and, while I
would not wish to defend all against Kazan's charge, i would
surely wish to defend some. Perhaps this survey of some of the
chief eunuchs will demonstrate quite why. Atleast it might help
inthe task ofdefinition; in understanding where we are,how we
got there, and what we might need to get any further. It is to
this preliminary aim of clarification that I direct the exegesis
and criticism which follow. To build on such a basis is the task
of another day.

Theories of Theories
Most obviously we have to begin by asking what constitutes a
theory of film. In a sense this whole book offers an answer. But
in a more limited way there is some use in putting the question
specifically. In particular, what has traditionally been meant by
the phrase 'theory of film'? Now there are two well-known
books of precisely this title, but neither of them is clear as to
the 'theory' in question. The one, by Bela Balazs, is a collection
of general and often rather dated fragments only tenuously
interlinked. The other, by Siegfried Kracauer, is a massively
desperate attempt to justify realist cinema as aesthetic perfec
tion. The one thing they do have in common - in retrospect the
lowest common denominator of their 'theories' - is a desire to
make general statements about tJie cinema. To advance
propositions which transcend particular films and thence apply
to film in general. Like Eisenstein, Balazs expresses it at its
strongest in his avowed interest in the 'intrinsic laws of
development' of the art, though in practice, also like Eisenstein,
he leaves such laws unspecified.
Minimally, then, the expression 'theory of film' has been
applied to any attempt to make general assertions about the
medium. Much everyday usage corresponds to this meaning.
Any critic concerned to talk about characteristics of the
medium as well as about particular movies has been labelled,
for good or ill, a theorist, a terminology which is clearly far
too vague. While generalization is undoubtedly a necessary
component of 'theory', all generalizations do not therefore
constitute theories. A theory is not simply the sum total of our
general knowledge ofa subject; its functions are not exhausted
in the presentation of ad hoc sets of assertions linked only in
thatthey all apply to film. It has, in its way, a creative character
of its own. By explicitly linking together its various com
ponents we are made aware of relationships and regularities
which would not otherwise be apparent. We are able to clarify,
for example, the links between conceptions of film editing and
of film acting, and their joint consequences for the process of
film communication. The classic Russian arguments about
typage and montage involve just such links.
This enables us to further pin down the notion of film theory.
It is not simply a question of the generality of our statements,
of their status as presumptive 'laws' which always hold true.
There is also an issue of method involving the systematization
of our thought. The film theorist is distinguished from the film
essayist (who might also make general statements) by his stress
on the systematic. To theorize is, of necessity, to invoke the
criterion of logical consistency, and so logically to interrelate
various diverse elements into theories. As a body of such
theory is formed, each stage in its development gives us a new
vantage point on our subject. It provides a different pair of
spectacles, a different 'theoretical framework', much as, for
example, the collective propositions of Newtonian mechanics
historically offered a new perspective on the dynamics of mov
ing bodies. At root, our every act of observation invokes some
implicit framework; to theorize is to make such a framework
explicit so that it might be explored for cracks. Given all this
there is still obviously a range of meanings of 'theory'. At one
end there is the minimal demand that writers on film render
explicit their assumptions. At the other extreme a maximal
demand that we must work toward formulating a general and
systematic body of empirically tested knowledge about film; in
effect, a science of film. Recent years have seen a fairly wide
spread critical demand for 'minimal' theory: the breakdown in
communication between different 'schools' has made such a
need evident to a wide range of interested parties. This has
inevitably led to a more philosophically developed discussion
of cinema. On the other hand, 'maximal' theory has become
increasingly hived off into outside specializations. From the era
of Eisenstein, whose intention it was to create a mighty, over
arching, scientific theory of film, we have come to the age of
the specialist. Eisenstein attempted to invoke his own psy
chology, sociology, and film 'linguistics'. For him the impetus
came from a central concern with film. Nowthe psychologists,
sociologists, and linguists fragment the subject into its dis
ciplinary variants. For them the impetus is extra-cinematic. At
its most general, one task of 'maximal' theory could be to
reunite these elements.
It should be clear that I am using 'theory' in a fairly general
way. It would obviously be possible to limit the term quite
drastically; in some circumstances it might even be essential.
Thus, 'theory' could be conceived as part of a process of
hypothesis formulation, testing, and, if necessary, refor
mulation on the basis of empirical materials. In short, the
application of 'scientific method' to the task of expanding our
knowledge of film. But most people do not limit 'theories of
film' in this way, and this book is primarily concerned with
what is and only partially with what ought to be. Much of what
has passed as theory of film has really been an attempt to lay
bare the assumptions and arguments which have underlain
certain critical practices. Bazin, Kracauer, the 'auteur theory',
are cases in point. They are elaborations of particular critical
'world-views', special frameworks for the analysis of films.
They canhardly be by-passed as notconstituting theory proper,
for we have much to learn from them.
This enlarged area of interest brings with it its own

Deep-focus cinematography: Citizen Kane ^

li

•k
particular problems. We are to be concerned with theoretical
frameworks employed for a range of purposes and deriving
from a number of disciplinary viewpoints. One difference, in
particular, needs attention. It is possible to study film in order,
primarily, to understand the empirical operation of the
medium, or, primarily, as a basis for making judgements of
quality. Evidently, the clearer our understanding the better the
factual basis on which our judgements are founded; equally
(believers in the myth of 'objective criticism' apart) our 'scien
tific' study could not be conducted in an evaluative vacuum.
But there is a definite distinction of aim between the two, and
their confusion can be, at the very least, misleading. Eisenstein,
as I shall discuss, is sometimes dismissed as guilty of aesthetic
monomania: of offering montage as the ultimate aesthetic
arbiter of taste. In fact, he was much more concerned with
montage as part of a theory of howfilm infact affected people,
than with making all-pervasive judgements of value. For the
want of terms I shall refer to the former interest as developing a
model of film, to the latter as developing an aesthetic of film.
The intertwining of such interests has elsewhere caused the
exact status of propositions to become confused. Bazin, for
example, argued that Citizen Kane was a film of high quality in
that it was a film of realism. Realism was an axiom of his
aesthetic position. But the statement which links this axiom
with the specific aesthetic judgement of Citizen Kane raises
problems. The realism of the film, Bazin argues, derives from
its use of deep-focus photography and minimal cutting. Such
techniques minimize fragmentation of the real world. The
trouble is that this could be a definition of realism as non-
fragmentation, or an assertion that films employing such
techniques are perceived as more real. The latter, unlike the
former, is open to empirical test, although Bazinuses it as a self-
evident aesthetic judgement. Thus, although there is nothing
inherently wrong with the argument, it does involve different
sorts of statements with consequent different criteria of
adequacy. Such shiftsshouldbe madeclear, thoughthey seldom
are.

This leads to a working distinction between theories which

are aimed principally at scientific comprehension of film - what
I have referred to as models of film —and theories aimed at
principally making evaluative judgements - what I have termed
aesthetics of film. Historically the emphasis in film writingshas
been in the aesthetic direction, in effect, with the provision of a
rationale for statements of the form T like this' or 'film should
be like that'. Clearly some body of aesthetic argument of an
abstract theoretical nature could be inferred from the work of
any critic, whether or not the resulting case was consistent or
even plausible. For example, Robin Wood's critical studies can
be shown to derive from the kind of moral arguments advanced
by the Leavisite school of literary criticism. To argue this is
not, as Alan Lovell has demonstrated, simply scholastic; it
enables us to better understand the basis on which the critic's
practice rests.^ Without at least mutual comprehension at this
level discussion becomes impossible. When there is consensus
about such assumptions they are usually left implicit. When
there is no such agreement - and we are far from it in film
criticism - self-consciousness about the critical process
becomes essential. Not that there is a formula solution to the
problem of judging movies. The exploration of theories of film
{models and aesthetics) is not designed to provide a universal
and ultimate standard for judging quality; as Pauline Kael says,
such 'objective standards' would make the critic unnecessary.
And even if they were possible, which they are not, they would
2. Alan Lovell, 'Robin Wood —A Dissenting View', Screen, 10, 2,
March/April, 1969.
Robin Wood, 'Ghostly Paradigm and H.C.F.: An Answer to Alan
Lovell', Screen, 10, 3, May/June, 1969.
Alan Lovell, 'The Common Pursuit of True Judgment', Screen, 11,
4/5, 1970.
John C. Murray and Ted Welch, 'Robin Wood and the Structural
Critics', Screen, 12, 3, Summer, 1971.
make life very boring. Looking at theories of film, rather,
is meant to help us in understanding film itself and, just as im
portant, in understanding why people - including ourselves -
think some films better than others.
In short, this book will be concerned with the systematic
theoretical element in both film models and film aesthetics.
Whether the intention is to develop knowledge or evaluation
(or frequently both) it is axiomatic that consistency is impor
tant. Without it, argument is no longer practicable; there is
more than one debate in film criticism which stems from
rapidly shifting and inconsistent assumptions. In my discussion
of particular thinkers one of my aims will thus be to explore the
consistency of the arguments they invoke. It would be unneces
sary to labour this point if it were true that such a criterion was
commonly accepted; however, Pauline Kael, in a much quoted
paragraph, has clearly missed the point. In her discussion of
the 'auteur theory' she asserts: 'It requires more care, more
orderliness to be a pluralist than to apply a single theory.'^ She
could hardly be more mistaken. To apply a single theory, of
whatever sort, is self-confessedly to attempt to be consistent.
To be a pluralist, as Miss Kael suggests, is to pick and choose,
'... the best [!] standards and principles from various systems
of ideas'. To be flexible, to bend with whatever wind is neces
sary, to leave consistency behind in the systems from which we
borrow. To continue to look at films in the ad hoc fashion with
which we are so familiar. This book is dedicated to the belief
that such an approach is the last thing we want to preserve.
Whatever the faults of theoretical self-consciousness, and
there are some, it does at least create a situation in which
communication is possible and in which positions can be
argued, expounded, or reformulated. As we have seen, it is
impossible to argue with Miss Kael, and above all theory is
about argument.
3. Pauline Kael, / Lost it at the Movies, Jonathan Cape, J..ondon, 1966,
p. 308.
The Development ofFilm Theory
Ultimately it would be useful to examine the range of assump
tions underlying all sorts of variations in critical practice. This,
however, is not intended here. My specific choice of theories
and theorists is dictated by two considerations: an attempt to
present a sketch of what I take to be the crucial elements in the
history of film theory, and an attempt to organize these
elements in an analytical context. In both cases the jumping-off
point lies in the work of Eisenstein. His remains the one major
attempt to coordinate a wide range of knowledge into a full-
scale theoretical framework, an attempt which is only now
being recognized for what it is. In a sense much of this book,
and much of film theory, is concerned with arguments which
served to push Eisenstein's contribution into the background.
His work was frequently drawn on as a Justification for certain
aesthetic postures, and its importance in toto was thus largely
ignored, misrepresented, and misunderstood. The standard
image of Eisenstein as an arch-priest of montage cinema
involves, at the very least, a misplaced emphasis, but an em
phasis which proved controversial enough to direct attention
away from Eisenstein's true interests.
To appreciate the consequences of this for film theory, we
must trace the development of a particular cluster of primarily
aesthetic themes. In the practical beginnings of film we find the
roots of what was to follow. Film developed from photography,
and, to the nineteenth-century thinkers, the characterizing
achievement of photography lay in its ability to reproduce
reality. The most efficacious of the highly valued realistpainters
could not hope to compare. So when the Lumiere brothers
made those first brief films, their subjects, not unnaturally,
came from their immediate physical environment. The train,
the baby, and the gardener with his hosepipe. But with almost
indecent haste the new invention also found a place with
Georges Melies, a magician who saw in film the new fount of
illusion. For him, monsters, planets, and space-travel. The
 •-v \ %
/
r

• -r

ti>iS
i|k

. '» li it
The 'art' of film: The Cabinet of Dr. Caligari (Robert Wiene); Mother (Pudovkin)
shoots growing from these seeds were to split film aesthetics in
two; the symbolic opposition between Lumiere and Melies
hovers perpetually over aesthetic debate.
The crucial polarity, then, became that between, on the one
hand, realism, naturalism, and the minimum interference of the
film-maker, and on the other hand, fantasy, expressionism, and
the formative influence of the film-maker. Which is not to say
that realism versus fantasy is identical to naturalism versus
expressionism; it was merely made to seem that way. One of
the earliest interests of film aesthetics- the attempt to establish
that film could indeed be justly called Art - was deeply
involved in this division. The silent films which were claimed as
'artistic' were those in which the 'creative' interference of the
artist was most evident. The painting-influenced designs and
'serious' subjects of German Expressionism were invoked as
evidence, as were the newly developed montage techniques of
the Russians. Caligari, Potemkin and Mother were used to
define the 'art' of film. Another source of artistic respectability
lay in the extravagant Freudian symbolism of the French
avant-garde. Films like Un Chien Andalou, The Seashell and
the Clergyman and Menilmontant were used to demonstrate
that the cinema could be just as experimental as its artistic
neighbours. By the time the silent era reached its culmination
aesthetic orthodoxy took the part of what Kracauer calls the
formative tendency'. In the generic imagery, Melies was on
top! And worse still, a bowdlerized version of Fisenstein
became an accepted Old Testament.
Fisenstein's major interest was in the workings of film
'language',and he conceived montageasa crucial elementin such
a process. But as we shall see his notion of montage was by no
means simple, and certainly never as simple as the version held
by the aesthetic orthodoxy. His analysis was complex to a
degree, requiring and making a number of conditioning
assumptions. The social and psychological contextsin which the
cinema operated were the primary focus ofthese assumptions.
Lumiere; A Boat Entering Harbour

In the former case Eisenstein's views were fairly straight Soviet

socialism; in the latter, a Pavlovian behaviourism. Either way,
this context was left largely undiscussed within the Eisenstein
canon. It is in later and otherwise unimportant side-tracks from
the main development, in particular the works of John
Grierson and Rudolf Arnheim, that the context problems were
re-raised. As far as the mainstream of aesthetic development
was concerned Eisenstein seemed irretrievably associated with
the 'Eilm as Art' school of the silent period. When this
tradition proved unable to cope with the coming of sound
(unlike Eisenstein himself), his much more interesting theories
sank with the rest. The pendulum swung toward Lumiere. The
documentarists of the 1920s, Renoir in the 1930s, and Italian
neo-realism in the immediate post-war years, were a progres
sion of 'realism'. The aftermath of war produced a cinema
Progressive realism: Renoir's La Marseillaise

properly fitted to a realist aesthetic-, the less the artist interfered

with the reality before his camera the better the film. Eisenstein,
a realist in some respects, could now be definitely proscribed as
a formalist (in this, at least, Stalinism had beaten the new
aesthetics some years earlier), and his models of film were
consistently interpreted as aesthetic dictates. First Andre
Bazin, and then Siegfried Kracauer, produced elaborate 'justifi
cations' for the resurgence of a realist cinema. The horns of the
dilemma were firmly fixed. Although compromise always
remained possible in principle, practice always seemed to
involve realism or formalism, non-interference or interference,
Lumiere or Melies. For a while film theory became obsessed.
The general form of this 'realism deadlock' has underlain a
number of aesthetic debates. The argument over the coming of
sound, for instance. Sound, seen by the traditionalists as the
great destroyer of an established visual art, was held up by the
new aesthetics as an asset to realism. And even when the
formalist schools finally accepted that sound was here to stay
they preferred to focus their interest on the 'creative use of
sound'. Traditionalist textbooks to this day retain chapters thus
labelled. And in much more recent years we have seen estab
lished an opposition between the 'natural' and uncluttered
American cinema and the 'aesthetic' and more complex
European cinema. The net result of such absolutist divisions
has been to press film theory into the ready-made moulds of
realism versus formalism, and to effectively prevent anything
but sterile polemic. In the end the only way of escaping the
polarity is to retreat entirely from the strictly aesthetic discus
sion on to the more neutral ground of meAodology. Recent
years have produced just such a development, a change in focus
from problems of evaluation to problems of descriptive inter
pretation.
The pattern of this 'historical' account should indicate my
interests. It seems to me that Eisenstein's work has become
overlaid with many years of aesthetic argument. Although he
by no means solves them, Eisenstein does raise, explicitly or by
default, a series of analytically crucial problems in the theory
of film. He attempts a beginning analysis of film language, his
work clearly demonstrates the need for a more extensive
discussion of the social and psychological context of film, he is
the interstitial figure in the realism debate, a debate which in
turn has pushed film theory back toward the problems of
method to which Eisenstein partly addressed himself. It is with
these problems that this account of film theory will primarily
be concerned.
Not surprisingly then, the first body of theory I shall discuss
is that proffered by Eisenstein himself. Chapter Two is entirely
devoted to an attempt to draw together some of the disparate
elements in his many essays. Two sets of problems grow out of
this. Eirst, the problem of context, both social and psycho-
logical. I shall briefly raise the question in Chapter Three in
relation to the work of John Grierson. Second, the realism
issue, most notably the arguments advanced by Bazin and
Kracauer. If Eisenstein is a pillar of film theory, these two are
clear candidates as aesthetic door-posts. Chapter Four will be
devoted to them. Chapter Five will be concerned with the
'retreat' from aesthetic theory and the beginning interest in the
methodology of film criticism. In particular, the theory of
genre and the so-called 'auteur theory'. And finally, in Chapter
Six I shall try to suggest the possible direction of development
of film theory.
Clearly what follows is a small selection from a large body
of film theory, and sometimes a highly critical account. Even
within the work of particular authors I have chosen to discuss
certain things and not others. Bazin, for instance, had much
wider interests than the argument for realism, while my ac
count of Grierson by-passes much of his educational thought. I
make no apology for this selectivity. My interest lies in the bits
and pieces of film theory and not the whole sweep of ideas on
film. I am writing at a time when there seems to be a very
considerable interest in the cinema, and when the art itself has
undergone some not unimportant changes. We are invited to
re-think the traditional postures of film theory, and, like
Thomas in Blow Up, we are perpetually reminded that there are
many ways of looking at the 'real' world. There is always an
invitation at such times to dismiss apparently discredited intel
lectual ancestors as no longer relevant, to start with a clean
sheet. However critical I may become in the pages that follow I
do believe that it would be to our loss were we totally to
dismiss the writers I shall consider here. We have much to
learn from both their qualities and their faults.
 2: Eisenstein: Great Beginnings

Of the classic writers on film Eisenstein is clearly the most

complex. Meyerhold, Freud, Pavlov, the Kabuki theatre, the
Commedia dell'Arte, flow in and out of his essays, bound to
gether,more or less,by ill-digested lumpsof orthodox historical
materialism. Often not orthodox enough, however, and at
least some of the stranger leaps in his thought can be accounted
for in terms of political pressures. For whatever else he may
have been, Eisenstein's thought was far too wide-ranging to be
constrained by Stalin's mock philosophy or even Engels'
universal and mechanical dialectic. His conception of the
relation of film to 'reality' went beyond the simplified world
of 'socialist realism'. His was the existence of the goat that
hungers for the grass beyond the end of its tether. In both
writings and films he was continually pulled back, sometimes
by others, sometimes, it seems, by himself. Bezhin
Meadow and the second part of Ivan the Terrible suffered
drastically in consequence, while the nationalistic and
highly regarded Alexander Nevsky is hardly his greatest
work. That other famous 'response to just criticism',
Shostakovitch's Fifth Symphony, is at least a resounding
artistic success!
Much of the problem, for us as well as for the Stalinist
establishment, is rooted in Eisenstein's unwillingness to base all
aesthetic judgements on one dominant criterion. In an essay
which brought about the closure of the journal in which it
appeared, he put it bluntly:
A high social appraisal must not serve as a shield, behind which with
impunity can be concealed poor editing or a low quality of enunciat
ing those words which, in any final accounting of a film, also deter
mine its value.'

In spite of this, however, it does seem that Eisenstein remained

a deeply committed socialist for most ofhis life. The 'romantic'
Bolshevism of Battleship Potemkin may well be representative.
His differences with the State arose from his view of how the
aims of socialist film were to be implemented. 'Socialist
realism' assumed that audiences would automatically respond
to the correct socialist subject matter, leading to, in the end,
pedestrian treatment of scenes from Soviet life. Eisenstein took
the view that audience responses were rather more complex
than this; that film could be used in such a way as to inspire an
audience with socialist ideas. Through 'pathos', through a unity
of form and content, through the formal language of film
making, it was possible to create a more efficacious means of
communication. The distinction is rather like that between the
speaking styles of the stereotyped university lecturer and the
fiery orator; between the wishy-washy romantic realism of
Ballad of a Soldier (so beloved by Western liberals of the early
1960s), and the equally romantic but immensely more power
ful Potemkin-, ultimately, between a mechanical and formula
Marxist aesthetic and a creative but heterodox Marxist aes
thetic.
But the political issues are the background to the tasks of
this essay. They simply make it rather difficult to answer some
of the questions we might sensibly raise about Eisenstein's film
theories. Eor instance, what are we to make of his declining
concern with dialectics? Was it simply that his version was too
1. Sergei Eisenstein, E///M Dobson, London, 1968, p. 152.
'idealistic' for the bowdlerized philosophy and aesthetics of his
period, or was he ultimately convinced that 'dialectical'
analysis was theoretically insufficient? To further complicate
the issue, considerable materials remain, as yet, untranslated
from the Russian. In particular, we must eagerly await trans
lations of the unfinished drafts of Montage, Direction, and Art
of Mise en scene. Even so, we can discern four main trends in
Eisenstein's thought from the essays available to us. First, a
moral commitment to socialism and to socialist aesthetics of
one form or another. Secondly, an analysis of the fundamental
formal elements of film and the manner of their relation: in the
main his approach to the 'language' of film through the concept
of montage. This is centrally concerned with the 'cells' of
cinematic expression and their interaction. Thirdly, a series of
ideas about what sort of cinema this language should be used to
create, discussed principally in relation to 'intellectual cinema'
and pathos. Lastly, and underlying the previous two, there is
Eisenstein's attempt to root his theories in an eclectic set of
anthropological and psychological assumptions as to the nature
of man. In this essay I shall be particularly concerned to
discuss the second and third of these trends, invoking the first
and last where it seems useful to do so.

Dialectics ofFilm
Although a loose idea of montage appears quite early, more
systematic analysis begins in the 1929 essay, 'A Dialectical
Approach to Film Form'.^ Here, Eisenstein argues for a
parallel between the method of thought termed 'dialectical
materialism', which arises from the 'projection of the dialec
tical system of things into the brain', and art, which arises from
the projection of the same dialectical system of things into the
creative process. Artistic creation develops from the inter
action ofcontradictory opposites; thedialectical process (thesis-
antithesis-synthesis) is the baseline on which the theory of
2. Sergei Eisenstein, Film Form, Dobson, London, 1951, pp. 45-63.

Socialist realism: Eisenstein's Bezhin Meadow •

f '0
 -I
-•

A".
:£
montage rests. The fundamental assumption is that all things in
the world are related in a dialectical manner, and so this
'universal' dialectic must find its place in film as much as
anywhere else. It is this rather extravagant claim that first
marks Eisenstein off from the other Soviet film directors. There
is, at least at fi.rst, a considerable pressure apparent in his work
to use the mould of 'dialectics' to hold his thoughts, accom
panied by an equally obvious tendency for them to overflow the
sides of the container.
There is a clear sense in which the dialectic provides a
rationale for Eisenstein's particular interpretation of the
Kuleshov experiments. One, in particular, seemed important to
Eisenstein, although the whole set developed a similar theme.
By splicing a strip of film showing a shot of Moszhukhin's
expressionless face to various other shots in turn (a bowl of
soup, a coffin, etc.) and then showing the results to an audience,
the experimentors succeeded in showing that the audience
believed the face to be expressing the appropriate emotion (e.g.
hunger, sadness, etc.) in each case. From this, it followed that
one of the important factors influencing an audience's response
to a film revolves around the juxtaposition of the shots
involved. In a word, montage. Now, thus far in film history
editing had been primarily dictated according to the narrative
needs of the film. A cut, when it came, was necessary to move
on to the next camera set-up. Only Griffith, most notably in
Intolerance, had begun to explore the immense formal possibili
ties of editing, so to the Russians the Kuleshov discovery had
all the marks of a fundamental insight.
Characteristically, Eisenstein took it to its most extreme
limits. Where Pudovkin, ever pragmatic, saw montage as a
process of 'building', of laying 'bricks' end-to-end, Eisenstein
tried to conceive it in a theoretically more sophisticated way.
For him it was from the 'collision' of independent shots that the
meaning arose in the minds of the audience. Now there is one
clear respect, which later proves to be' important, in which
 5*^ A

Battleship Potemkin; Ballad of a Soldier

Eisenstein's version is correct. The Kuleshov experiments
certainly suggest that a juxtaposition leads to something new in
the mind of the audience. It is not built brick by brick. We do
not have a conception of a face which is then modified by the
following shot; rather, we respond to the two shots as a whole.
The whole is different to the sum of the parts, this much can be
allowed to Eisenstein. But the rest of his conception is prob
lematic. Faced with the axiom that the basic elements of film are
the shot and the montage through which the various shots are
related, it remains to discover the exact manner of this relation;
to find the rules according to which the cinematic language
operates. In 1929 Eisenstein's first approximation is in terms
of dialectics - the creative 'collision' of opposites. But
such an identity of montage and dialectic is a narrow
conception indeed, andEisenstein always seems willing to admit
rather less narrowly defined forms. By 1938 the retrospec
tive version of the crucial early days' concept of montage no
longer involves 'collision'. Instead, it is '... that two film
pieces of any kind, placed together, inevitably combine into a
new concept, a new quality, arising out of thatjuxtaposition'.^
Indeed, The Film Sense is devoid of any mention of the
dialectic.
This is not a purely scholastic point. Eisenstein's desire, for
whatever reason, to employ dialectical conceptions has all sorts
of consequences. It leads to a stress on 'conflict' montage in his
earlier theoretical writings, a stress which drastically curtails
the applicability of the montage idea. Interestingly, once
Eisenstein embarks on his more detailed analyses of particular
films and sequences many of the 'dialectically' imposed con
straints disappear. The 1929 essay may have as its title 'A
Dialectical Approach to Film Form', but the approach, if
approach it be, barely survives the essay. Eisenstein soon turns
to more rewarding pursuits.

3. Sergei Eisenstein, The Film Sense,Faber, London, 1943, p. 14.

Aspects of Montage
Eisenstein initially distinguishes five types, or bases, of mon
tage: metric, rhythmic, tonal, overtonal, and intellectual. All
may exist simultaneously in any given film sequence. I shall
consider each in turn. Metric montage revolves around the
mechanical 'beat' of the cutting. The crucial basis for editing
lies in the absolute lengths of the strips of film, regardless of
what they portray. The strips are in constant proportionate
relations to one another. They may be shortened to increase
tension, but the proportionate relation is maintained. This is the
simplest method of all. Endless early film chases (and some not
so early) build up to their climax through accelerating metric
montage, and it is clear that Eisenstein sees metric montage by
itself as somewhat less than worthy. Pudovkin, he suggests in a
sideswipe, is the major exponent of the method. The second
type, rhythmic montage, takes into account part of the visual
content of the shots involved; in particular, the pattern of
movement within a shot. Eisenstein himself is initially much
concerned with the possibility of conflict between the metric
and the rhythmic types. Elence his own example from the
Odessa Steps sequence of Potemkin. The soldiers' feet as they
descend the steps set up a rhythm within the frame which is
not synchronized with the metric beat of the cutting. Thus the
film builds tension through the simple metric method and also
through violating that method at key points. Equally, however,
it is possible for the two methods to reinforce one another.
Otsep's A Living Corpse (a neglected classic of the 'extreme'
montage period) contains a very impressive gypsy dance
sequence where metric and rhythmic bases are mutually rein
forcing. The more frenetic the dancing movement within the
frame, the faster the metric montage. Similarly with many of
the passages in Ruttmann's Berlin.
These two aspects of the montage pattern -1 hesitate to call
them different methods since they are so often intertwined —are
very much part of modern editing stock-in-trade. In most films
 If

# *

"V
k'
1^' '

. JSiL£^ (

Madigan (Don Siegel)

which treat editing as anything more than something necessary
to facilitate the narrative, they can easily be found. The Wild
Bunch, as I shall suggest in more detail later, is full of such
cases. The 'Blue Danube' sequence in 2001: A Space Odyssey,
the pattern of footsteps in the long echoing corridors of Point
Blank, the climactic sequence of Masque ofthe Red Death, and
the tour-de-force at the end of Madigan, also offer cases. They
are the fundamentals of tempo in a work. And pursuing this
metaphor, Eisenstein's third type, tonal montage, has rather
more to do with melody.
Tonal montage is a little more complex. Eisenstein dis
tinguishes it from the rhythmic case in the following way:
In rhythmic montage it is movement within the frame that impels the
montage movement from frame to frame. Such movements within the
frame may be of objects in motion, or of the spectator's eye directed
along the lines of some immobile object.
In tonal montage, movement is perceived in a wider sense. The
concept of movement embraces all effects of the montage piece. Here
montage is based on the characteristic emotional sound of the piece —
of its dominant. The general tone of the piece.'*
The emotional sound, he continues, can be assessed accurately.
For the emotional conception 'more gloomy' we can find a
suitable 'light tonality', a suitable degree of illumination. For a
'shrill sound' we can find a relevant visual pattern of angles to
provide 'graphic tonality'. And, of course, in Eisenstein's later
work we also find considerable interest in colour tones. He
again uses Potemkin as an example. The 'fog sequence' mon
tage is based on tone. The pieces are assembled on the basis of
their light quality ('haze' and 'luminosity'). This is the basic
tonal dominant. Beneath this we can also find a rhythmic
dominant based on the tiny movements of the ships, the slight
movement of the water. In this way, the visual characteristics
of the shots set up the gentle, expectant mood of the sequence,
which is reinforced by the very slight in-frame rhythmic distur-
4. Sergei Eisenstein, Film Form, op. ait., p. 75.
bances. One must be careful, however, in succumbing to the
persuasiveness of Eisenstein's example. Although there is
evidently a case for singling out this form of montage it is
insufficiently specified. Even accepting that it has only to do
with plastic formal properties of the shot (that is, nothing to do
with particular narrative content), the sorts of rules of relations
between emotional and visual tones implied by the analysis are
not available even now. Thus, although such a category may
usefully direct our attention in the course of specific analysis,
to reach the level of Eisenstein's ambition requires a much
greater psychological knowledge, or a number of dangerously
extravagant assumptions. This latter is Eisenstein's strategy.
There are even more problems with overtonal montage,
Eisenstein's 'filmic fourth dimension'. Again the idea is
developed via the musical analogy, the montage being built on
the basis of the total stimuli offered by a shot rather than on
particular dominants. One includes in the montage rationale
the overtones and undertones of the shots. The basis for this
form of montage is somehow the totality of impression offered
by each shot, and it was on the editing table, Eisenstein claims,
that he discovered post hoc that what he had already edited of
The General Line did not fit into the orthodox categories he
had developed. In the 'creative ecstasy' of making the montage
he had edited the film on the basis of the whole range of stimuli
provided by the shots. It was only when the shots were seen in
movement that the new stimuli, the 'overtones', emerged. For
this reason examples are problematic:
... it [the visual overtone] cannot be traced in the static frame, just
as it cannot be traced in the musical score ... overtonal conflicts,
foreseen but unwritten in the score, cannot emerge without the
dialectic process of the passage of the film through the projection
apparatus, or that of the performance by a symphony orchestra.'
Even given this ambiguity, overtonal montage is far from clear.
5. Sergei Eisenstein, Film Form, op. cit., p. 69.
Is it really a similar sort of category to metric, rhythmic and
tonal? Or is it simply that the first three categories are ways of
breaking down the montage effects, while the latter attempts to
capture its overall characteristics? Eisenstein is not too clear,
but I would incline to the latter interpretation. Overtonal mon
tage is simply another way of talking about montage based on
the total effect of the piece; metric, rhythmic and tonal are
categories for analysing this process.
These four, or three, whichever the case may be, have both
physiological and emotional effects. Eisenstein implicitly
suggests whole series of hypotheses about their physical and
emotional consequences. But he is also concerned with the
possibility of formal montage characteristics having intellectual
consequences; with the possibility of intellectual montage. The
classic examples in Eisenstein's films, albeit simple ones, are
always taken from October. One commonly invoked case is the
sequence in which a series of images of various religious idols
are cut together, moving from the most modern Christian
representations to the most primitive native figure. The intent is
to pull back '... the concept of God to its origins, forcing the
spectator to perceive this "progress" intellectually'.® A further
obvious, and crude, example is the famous 'Kerensky climbing
the steps' sequence in October. The shots of him climbing are
intercut with titles ascending the scale of military rank and with
a mechanical peacock preening itself, the intention being to
make a direct intellectual point.
There are all sorts of problems associated with this classifi
cation, some of which I have already touched on. It is, for
instance, debatable as to whether overtonal factors can be
treated in the same way as the others. Such a conception seems
to encompass the whole range of response of an audience to a
film whieh, of course, is exactly what we are trying to analyse
through the various sub-categories of montage. We have no
way of capturing total response without employing some set of
6. Sergei Eisenstein, Film Form, op. cit., p. 82.
categories such as the three basic montage types. Overtonal
would thus become a redundant category, or at best a some
what ambiguous half-idea. Extensive problems also exist with
'intellectual' montage, finding their expression in some rather
peculiar psychological assumptions. Thus the alleged lack of
qualitative difference between intellectual and other forms of
montage is based on the fact that
... there is no difference in principle between the motion of a man
rocking under the influence of elementary metric montage [this refers
to a sequence in The General Line which induced some members of
the audience to rock from side to side with the metric pattern] and
the intellectual process within it, for the intellectual process is the
same agitation, but in the dominion of the higher nerve centres.''
Apart from any other consideration this omits the question of
the different effects which culture may have on intellectual as
opposed to physical response. Even if it is nottotally incorrect,
it is a much too simplistic conception, related basically to the
particular brand of Pavlovian behaviourism espoused by
Eisenstein. Inevitably these simpler ideas yield to the deepening
knowledge of the years.
Changing Ideas
Ten years sees the main development of these changes. His
crucial essay 'Word and Image' (also knovm as 'Montage in
1938') begins significantly:
There was a period in Soviet cinema when montage was proclaimed
'everything'. Now we are at the close of a period during which
montage has been regarded as 'nothing'. Regarding montage neither
as nothing nor everything, I consider it opportune at this juncture to
recall that montage is just as indispensable a component feature of
film production as any other element of film effectiveness.® (my
italics).
The reader could be forgiven for thinking that it was Eisenstein
7. Sergei Eisenstein, Film Form, op. cit., p. 82.
8. Sergei Eisenstein, The Film Sense, op. oil., p. 13.
 himself who once proclaimed montage everything, and that
conceiving it as a 'component feature' (rather than 'the nerve of
the cinema') is not something he might reasonably recall. Still,
such deviousness is intelligible when we remember that
Eisenstein was defending himself, as ever in these years, against
the frequent charges of formalism. What better way than to
deny that he ever elevated the formal aspects of montage to
solitary dominion? One should, he now admits explicitly, give
equal attention to both formal characteristics of film (montage)
and to content. His earlier works were specific researches into
one aspect of the whole; they should not be taken as anything
else. The aim now is to elaborate the principles whereby the
particular representations of each shot relate to the general
theme. Unfortunately, he never isolates these principles or what
he elsewhere refers to as the 'laws' of such a process. It is this
above all which makes The Film Sense an ultimately unsatis
factory thesis.
Meanwhile, the concept of montage itself undergoes the
generalization implicit in the altered position. In response to
the critic who might ask (and who probably did) what the
performance of an actor on a single uncut piece of film has to
do with montage, Eisenstein now replies —everything! We
must look for montage within the actor's performance. Now,
the conception of film acting that accompanied Eisenstein's
montage theory of the 1920s was typage: crudely, a playing
down of the importance of acting and as a corollary the use of
'types'. 'Actors' were used as elements in the frame but
moulded by the director through montage. The point was made
in the editing. But with a more general montage conception
more orthodox acting is permissible. Or perhaps if acting must
be made permissible then montage will have to be generalized.
Either way, as with Hitchcock, Eisenstein still seems to retain a
view of actors as 'cattle'. The physical contortions required of
Cherkassov in Ivan the Terrible exemplify the extent to which
he was more interested in the plastic properties of his actors

Sketch by Eisenstein for Ivan the Terrible-, and (right) Nikolai Cherkasov as Ivan ^
i
than in their histrionic talents. Nevertheless, the theory under
goes a generalization and montage becomes, effectively, a way
of formulating the relations of parts to wholes. A stage actor,
for example, creates his persona out of many detailed elements;
together they make up the character. A director creates a theme
through all the various partial representations of it. The
modified view can be expressed thus:
Before the inner vision, before the perception of the creator, hovers a
given image, emotionally embodying his theme. The task that con
fronts him is to transform this image into a few basic partial
representations which, in their combination and juxtaposition, shall
invoke in the consciousness and feelings of the spectator, reader, or
auditor, that same initial general image which originally hovered
before the creative artist.'

Instead of a particular 'dialectical' theory of the role of certain

forms of editing in film, the montage concept is now employed
to encompass a much wider range. All the multiform elements
that make up a film can be conceptualized within the new
scheme. At its bluntest the axiom is that parts make up wholes,
and wholes are rather different from the sum of their parts. Out
of this grows the idea of vertical montage - the general process
of realizing an effect through relating two separate parts of
the cinematic whole. The paradigm case for Eisenstein is in
the relation between sound 'pictures' and visuals, initially
developed in Alexander Nevsky and extended to include colour
in reference to Ivan the Terrible.
This phase of Eisenstein's writings provides a great deal of
fascinating detailed analysis of examples (throughout The Film
Sense) though relatively little completed theorizing. In many
ways the later essays are less specifically theories of film, and
more 'notes toward' a general conception of artistic experience.
At one level we find Eisenstein elaborating a set of very general
ideas about part-whole relations, the notion of synecdoche, and
9. Sergei Eisenstein, The Film Sense, op. cit., p. 33.
some scattered linguistic parallels. At another level, stimulating
accounts of particular sequences which do not bear a clear
relation to the more general formulations. Evidently his
frequent references to 'structural laws of process and rhythm'
and 'laws governing the construction of the form and com
position of art-works"" imply that such laws are the crucial
mediating factors. We can gratefully echo the sentiment, but
nowhere does Eisenstein offer a clear statement of these 'struc
tural laws'. Certainly the direction of his later work seems
potentially lucrative involving, as it does, the attempt to estab
lish a general method of analysing a whole rangeof'languages'.
Only the execution, perhaps understandably, remains wanting.
There is, of course, a further set of questions which I have not
yet properly touched upon. Much of Eisenstein's concern is with
the problems of analysing the formal language of the medium.
The questions he spends most time on have to do with the effects
of the formal ways by which the parts of cinemamay be related.
Montage, in both specific and general conceptions, is involved
with such issues. But there are also questions about what sort of
wholes these parts should make. What sort of cinema should
be created? These are the questions to which I shall now turn.

What Sort of Cinema?

Eisenstein's approach to such questions most often arose in
response to the repeated accusations of 'formalism'. Although
he remained very much committed to a Soviet socialist cinema,
to 'the liberation of the consciousness from all that represen
tational structure linked to the bourgeoisie'," he wasfrequently
in conflict with the Soviet establishment as to how such an aim
should be properly achieved. The earliest of his heterodox
conceptions was 'intellectual cinema'. A cinema whose prime
aim was the communication of ideas: 'a concrete sensual trans-
10. Sergei Eisenstein, The Film Sense, op. cit., p. 128, and Film Form, op.
cit., p. 130.
11. Sergei Eisenstein, Film Essays, op. cit., p. 98.
lation to the screen of the essential dialectics in our ideological
debates. Without recourse to story, plot, or the living man'.^^ In
the 1929 essay from which these linesare taken, 'Perspectives',
the idea is advanced with evangelical fervour. This, for
Eisenstein, was a fitting innovation in the realm of film to
match the greater innovation of the new proletarian society. A
cinema which directly imparted its intellectual content.
Inevitably there were pressures against such a view. It was,
they said, formalistic; it did not truly concern itself with the
masses. One might be tempted to add now, with that eternal
advantage of hindsight, that Eisenstein never really suggested
how to attain such a cinema anyway. However, he yielded, and
his writings suggest a mixture of genuine mellowing and
outside pressures. In his speech to the All-Union Creative
Conference of Workers in Soviet Cinematography in 1935, he
had this to say about the theory of'intellectual cinema':
This theory set before it the task of 'restoring emotional fullness to
the intellectual process'. This theory engrossed itself as follows, in
transmuting to screen form the abstract concept, the courseand halt
of concepts and ideas - without intermediary. Without recourse to
story, or invented plot, in fact directly —by means of the image-
composed elements as filmed. This theory was a broad, perhaps even
a too broad, generalisation of a series of possibilities of expression
placed at our disposal by the methods of montage and its combin
ations. The theory of intellectual cinema represented, as it were, a
limit, the reductio ad paradox of that hypertrophy of the montage
concept with which film aesthetics were permeated during the emer
gence of Soviet silent cinematography as a whole and my own work
in particular."
He then goes on to argue that the crucial 'dialectic' (the term is
now almost empty) in art is that involving both thematic-logical
thought and sensual thought. To place undue stress on one or
the other, as the intellectual cinema would stress the themato-
12. Sergei Eisenstein, Film Essays, op. cit., p. 46.
13. Sergei Eisenstein, Film Form, op. cit., p. 125.
 J-

Kino-nst: Strike (Eisenstein)

logical, is to fail. It is the unity of these forms of thought which

produces a true work of art.
The notion of unity is now increasingly prominent. Montage
is still important but within the context of, and unified with, the
other elements offilm. Hence, in 'The Structure of the Film',
the weight of discussion revolves round 'organic-ness' and
'pathos' in particular relation to Potemkin. A work is organic,
argues Eisenstein, where the laws by which the work is created
correspond to '... the law of structure in natural organic
phenomena'.The nature of this law remains unfortunately
obscure, although Eisenstein's passing reference to Engels'
Dialectics of Nature once more suggests the possibility of the
universal application of the dialectic. As ever in his later work,
Eisenstein sheers away and we are left with the familiar gap
14. Sergei Eisenstein, Film Form, op. cit., p. 160.
 H

%
'in «# K • •%
5^.** .--g

IT

Ik ^
P
T " i, 1\ • ^ #
9 %i
"i ft "
Sji*
V® *
•
••' ™* T Ml
between general statement and detailed analysis. There are
similar problems with 'pathos', the process whereby a spectator
is lifted 'out of himself, electrified, raised into ecstasy. How to
reach such intensity is again only partly clear, although
Eisenstein makes much of the parallel between montage tech
niques and the pattern of the creative process.
The strength of montage resides in this, that it includes in the creative
process the emotions and mind of the spectator. The spectator is
compelled to proceed along the selfsame creative road that the author
travelled in creating the image.
One element, then, is in the shared creative ecstasy produced
through the montage process. The spectator is also artist. But,
as a cursory look at his films would- suggest, Eisenstein was
also very much aware that this formal element was not the
whole story. Other factors would needs be invoked. What does
remain clear, however, is that 'pathos' is very much the aim of
Eisenstein's cinema. The goal is, in his unfortunate termin
ology, to ereate the 'pathetic' film!
This militant, fiery, pathos style, this 'kino-fist', is the style,
Eisenstein says, of Strike, Potemkin, Mother, and Arsenal. It
arises from the social militancy of the revolution and, in the
later case of Alexander Nevsky, from the fervour of Soviet
nationalism. But 'kino-fist' is not all, and Eisenstein is forced to
remark, in passing, a rather more prosaic means of attaining
'pathos'. This 'quieter' method is epitomized in the Maxim
Gorki trilogy, but it is clear that Eisenstein's sympathies lie
elsewhere. In the end they can be traced back to the Pavlovian
reflex psychology with which all of Eisenstein's thought is
permeated. Given the correct set of stimuli, there then follows
the correct 'pathetic' response. Modify this slightly with the
notion of organic unity (loosely formulated) of the elements of
film, and we know why Potemkin is experienced as intensely
moving all over the world. Even, Eisenstein suggests, the class
15. Sergei Eisenstein, The Film Sense, op. cit., p. 34.

Kino-fist: Strike
Prosaic pathos: The Childhood of Maxim Gorki (Donskoi)

enemy can be so moved. Behind the anthropological discus

sions, the arguments about modes of thought, there is the belief
that the montage techniques somehow resonate with something
fundamental to man's nature regardless of the cultural clothes in
which he is dressed. Ultimately Eisenstein takes a mechanistic
view of human nature. From this comes the proposition that
the unspecified laws of thought give rise to the equally
unspecified laws of cinematic structure. Perhaps the odd refer
ences to Engels are more than a simple justification in the
Marxist-Leninist canon.
It would, nevertheless, be difficult to deny that Eisenstein the
film-maker was extraordinarily successful in exploiting the
capacities of the medium for 'pathos'. And equally, those for
mal factors which he does single out as important in creating
'kino-fist' are not to be neglected. The problem is to avoid some
of the dressing in which the salad rests, and then to add
something more than radishes. It is because of these con
fusions, the imbalances of his ideas, and some of the barren
psychological and sociological assumptions with which he was
operating, that his work is not great theory. Is it, however, a
great beginning to the process of theorizing about film?

Great Beginnings?
One way of answering this question is to ask how his work can
be of use to us today. We have become accustomed to the idea
of Eisenstein as the arch-prophet of editing, particularly the
techniques of contrast and irony developed in October. For
criticism and appreciation of film this provides a useful view
point on one formal aspect of the medium. For the rest, the
argument might go, we must look elsewhere. This is evidently
only partly true, as Eisenstein was himselfaware. The charge of
creating, in montage, a monistic criterion for assessing a film
is an unfortunate exaggeration. His views on criticism were
rather less extreme. Thus, in his essay 'A Close-Up View' he
draws an analogy between the use of long shot, medium shot,
and close-up in film making and three aspects of film criticism.
A film can, and should, be looked at in these three ways. The
'long shot' which is concerned to explore the ideological (or
moral) correctness of a film. The 'medium shot', the view of the
normal spectator and primarily concerned with the 'living play
of emotions'. Finally, the 'close-up', which is concerned to
break down the film, analyse its parts and the manner of its
working. All these are essential elements for the consideration
of the critic; no singleone can justify accepting a film as good if
others are defective. All three approaches must be pursued.
The relation of this scheme to the rest of Eisenstein's theor
izing is obvious. The theorizing itself is the 'close-up' technical
analysis. The understanding attained in this way is directed
toward involving the ordinary spectator in the intense emotion
of the 'pathos' structure. Through this involvement 'the general
characteristics ofthe theme enter the spectator's consciousness en
passant. The generalized concept of the event is embedded in
the spectator's feelings.'The spectator grasps thematic con
tent through his emotional response to the film; form and
content are united. Eisenstein's tragedy was that the established
Stalinist position elevated the form-content distinction to a
fundamental law. The ideologically proper position must imbue
the whole film; Eisenstein's wish to grip the spectator in an iron
fist led to insufficiently obvious ideological rectitude. Reality
was there and Eisenstein wanted to mess with it. As we shall
presently discuss, that allegation is still being made. But it is to
his credit that he did not try to elevate some master belief to the
level of sole arbiter of aesthetic taste. Sometimes, in his more
polemical discussions of montage, he hovered near to it, but
always the position was finally rescued. His thinking at least
has the virtue of flexibility.
Still, other than as 'advice to young critics', what use is
Eisenstein's work? The answer I think can take two forms. One
involves a long-term investment in the ultimate usefulness of
his views of the languageof film, the creation of pathos, vertical
montage and the rest, as a partial basis for a more thorough
going theory of film. The other hasto do with using his thought
in the course of our analysis of particular films as and when it
proves useful. Clearly our understanding of the formal mon
tage elements of film can be facilitated by using Eisenstein's
views at least as signposts. The more a film (or sequence)
depends on montage techniques, then the more useful the
theories. Some directors, Howard Hawks, John Ford,
Michelangelo Antonioni, for instance, make virtually no use of
classic montage techniques. Since we are not elevating
Eisenstein's ideas into a system of aesthetic values this implies
nothing about the quality of their work. Others, Alfred
Hitchcock, Don Siegel, Sam Peckinpah, are variously depen
dent on montage for some elements of their cinema. This has
16. Sergei Eisenstein, Film Essays, op. cit., p. 151.
no consequences for the intensity of any 'pathos' that may
result: both Ford and Peckinpah can be as intensely moving as
any directors I know, using very different techniques. Indeed,
Peckinpah seems to me to be above all the modern montage
director, at least in his two epic films - the tattered Major
Dundee and The Wild Bunch. Let me illustrate some uses of
Eisenstein's work in exploring the formal determinants of the
intense effects of the final reel of The Wild Bunch.
The sequence I have in mind begins with Pike and the Gorch
brothers in the brothel. At this point the montage is solely
concerned to carry the simple narrative. There is no clear
metric or rhythmic pattern; the pace and visual tones are gentle
and subdued. This combines with two thematic references to
elsewhere in the film: the solitary guitar recalls Angel's playing
at an earlier camp, while the Mexican girl and her child locate
Pike firmly in the 'might have been' world of his past love. This
poignancy is reinforced in the formal technique, the 'vertical
montage' between music and visual tonality. The mood hangs
and flutters likethe briefly intercut bird on a string. Finally it is
broken, aurally and visually,by the first synchronized words of
the scene. For the last time comes Pike's ritual'Let's go!'; 'Why
not?' is the reply. Again the intercut bird, now panting on its
back. Like the ants and scorpions at the beginning, like the
peacock in October, this is a typically Eisensteinian imagistic
comment. The peaceful intermezzo is over; the bird is in its
final throes.
Collecting Dutch, they move to their horses. In narrative
terms we are still unaware of what they intend. But as they
form up and arm, in what irresistibly look like pre-determined
positions, the wide screen is suddenly dominated by the pattern
of the four men in a row. They set off to rescue Angel and there
follows a tour deforce. Every montage technique is involved at
one level or another. The basic tempo is laid down by the
metric cutting pattern, accelerating slightly as the sequence
progresses. This is the baseline of growing tension. Overlaid on

Four men in a row: The Wild Bunch ^

v

i K

'f
this is the rhythmic montage, cutting on the basis of movement
within the frame. Here there is mutual reinforcement between
the 'stepping out' of the four men (like the soldiers on the steps
in Potemkin and, incidentally, in many of the angles involved,
like the march to the corral in Gunfight at the OK Corral) and
the beat of the metric montage. From front or back four-shots
of the Bunch filling the screen the cut is either to side-angled
two-shots (one very notable one of Pike and Dutch) of head and
shoulders moving irresistibly on, or to various static groups of
Mapache's troops. This supplies a rhythmic montage contrast
between Bunch and troops, much as a similar technique did
with troops and populace on the Odessa Steps. Tension is built
and everything carries us with the Bunch. They become gar
gantuan. On top of this is added the tonal montage pattern, on
the basis of the clean visual line of the Bunch and the messy,
chaotic scatter of the troops. To match the visual pattern,
the troops are shot in a sort of dust-haze while the Bunch
remain optically clear. Thus the basic tempo and 'melody'
montage techniques screw us to an increasingly high emotional
pitch.
Over this basic pattern Peckinpah adds further elements of
vertical montage: in particular, the music. This is a combin
ation of the guitar and voices of the villager's farewell song and
the rhythmic side-drum first heard in the build up to the
opening massacre. In tempo it matches the metric and rhythmic
montage, thus providing a perfect combination of visual and
aural patterns, and its volume increases as the tension grows.
But the music itself is also based on a montage conflict. The
drum recalls the massacre while the song evokes the beauty of
the departure from the village. Embedded in it is the thematic
joining ofthe extreme violence ofthe Bunch tothe romantic and
idyllic appeal of the village. The music thus performs the
double function of supporting both the formal intensification of
'pathos' and the thematic development that, for the first time,
the Bunch is involved for 'altruistic' reasons. Through the total
The death of Angel

emotional involvement comes the theme, much as Eisenstein

envisaged in his discussions of 'pathos'.
At the climactic point all tempi cease. The four men once
more fill the screen. Everything hangs suspended because the
rhythm, resonating across several levels of the film, has been
cut off. In the ensuing exchange the metric pace again suddenly
accelerates at the point of Angel and Mapache's death. Again a
long, held shot, smilesof relief. Pike raises his pistol and shoots
the German. All hell breaks loose. What follows is principally
comprised of very rapid metric montage, the pattern of which
competes with the frequently slow-motion movement within
frame; a metric-rhythmic conflict. Occasionally the tempo is
completely broken to single out particular points: Pike's
'Bitch!' to the girl who shoots him; the Gorch brothers dying
endlessly in slow motion; Dutch and Pike exchanging looks
Breaks in the tempo;the girl shoots Pike;and (right) the Gorch brothers dying

behind the table. The impossible frenzy is finally stopped as

Pike and Dutch sink down to their deaths.
The fourth 'movement' begins in unearthly silence with, once
more, simple narrative montage. Shots are mostly long; only
Thornton's quiet removal of Pike's gun is in close-up. In
typically Eisenstein fashion Peckinpah cannot resist inter
cutting thebounty-hunters with thebuzzards. AsThornton seats
himself and the 'refugees' stream pasthim, the weight offormal
technique rests with the visual tone. Silhouetted buzzards, mute
colours, and dust. The cuts are imperceptible; the pace has
been slowed right dovra. The film fades away, pausing only to
recall the Bunch, jumping from Thornton and Sykes to the five
dead men on the basis of laughter, and from them to the
departure from the village once more. Perhaps this is part of
what Eisenstein meant by overtonal montage.
 •/ ' >'' •
1 '

This has only been a brief application of some of Eisenstein's

ideas relating to particular formal elements in film. The fact
that I have not considered, for example, the importance of our
already established ambivalent attitude to the Bunch does not
mean I am indulging in formalism. The idea is to illustrate the
usefulness of some of Eisenstein's ideas on aspects of film
'language', not provide a definitive analysis of the final reel of
The Wild Bunch. In a loose sense it does seem to me that the
ideas are useful. I say 'loose' because at this stage in our
understanding of film language all such analysis is bound to be
looser Eisenstein's basic conceptions of montage do provide us,
in my view, with a useful starting point for exploring certain
elements of cinematic 'linguistics'. Similarly, treated as a par
ticular mode of communication among others, his ideas on
'pathos' are of some interest, particularly in their stress on
Dutch and Pike

communicating theme through emotional involvement. But

partly because of the faults of the philosophical and psy
chological grounding of his thought, it is difficult to see his
contribution as the great overarching theory he hoped for. As
a theorist of cinema, Eisenstein is a man of brilliant ideas
embedded in an inconsistent and confusing background. Per
haps the as yet unpublished works will alter this necessarily
provisional conclusion. It would be nice to hope. But even if
they do not, it is certain that Eisenstein remains head-and-
shoulders above most of those who follow him. His great
beginnings have yet to be taken up.
3: The Problem of Context:
John Grierson

Eisenstein's great obsession was the language of film. He

brought a wide range of learning and a sharp intelligence to
bear on this element ofcinema. But to focus in such a way on a
specific partoffilm, to look indetail at the 'technical' workings
ofthe medium, required him to operate within a whole range of
assumptions. In other words, he assumed a certain context.
Take his discussions of montage 'rhythm'. It is now a common
place to assume that accelerated cutting raises the level of
emotional tension in an audience. In terms of behaviourist
psychology, that the stimulus provided by a particular pattern
of rhythm provokes a certain response on the part of the
audience. Now much of Eisenstein's analysis of montage rests
on such ideas of tempo, or, at least, on more complex versions
of this basic notion. This sort of suggestion about psy
chological response to a particular set of stimuli can be one (or
both) of two things: either a straight psychological assumption,
or part and parcel of a larger psychological body ofknowledge.
To explore the particular phenomena that interested Eisenstein
some such assumption was essential, and in his case he un
doubtedly felt it justified in terms of Pavlovian refiex psy
chology. Of course, if the assumption is demonstrably wrong,
or limited in particular ways, then serious questions must needs
arise about any theory of film employing it. Assumptions about
the psychological context of the cinema are necessary in our
theorizing; but for the sake of clarity they must be made
specific.
This need for context assumptions holds true for both
models and aesthetics. It is not possible to construct an aes
thetic system in a vacuum. At the very least a body of aesthetic
standards has some relation to the ways in which their author
conceives his world, his social life, and the role played by film
in this larger context. Bazin and Kracauer, the 'philosophical
realists', attempted to construct an aesthetic system which
would be in some part 'context-free'. They tried to ground their
criteria of evaluation in what they selected as the basic essence
of the medium itself. In effect, they argued that the medium has
a natural (and presumably eternal) affinity for particular forms,
specifically what Kracauer calls 'the redemption of physical
reality'. But of course there is no 'essence' of a medium apart
from our perception of it; the medium is not an objectively
given and isolated entity. It is part ofa physical, psychological,
social, and cultural context, and to talk about it at all is to make
a set of assumptions about such a context. As we shall see in
the next chapter, the characteristic arguments of the 'philoso
phical realists' are deeply problematic for all sorts of detailed
reasons. But a not inconsiderable part of the difficulties they
experience stems from the attempt to construct an aesthetic ina
vacuum. Their context assumptions remain implicit.
Unless the theorist's task is to be hopelessly complex, how
ever, he must call a halt somewhere. Imagine we are interested
in the basic'language' elements through which a film communi
cates, as was Eisenstein. To facilitate our analysis we make
certain assumptions about the characteristic response to any
particular configuration of 'terms' in this language. If an
audience sees a character in long shot, from above, and dwarfed
by his environment, it will conceive him as weak, isolated, and
lonely. This is the standard text-book point usually invoked
with reference to the later scenes in Xanadu in Citizen Kane.
This in turn depends upon a set of psychological assumptions
about the ways in which we emotionally respond to particular
perceptual situations, and such assumptions are themselves
grounded in a body of psychological theory and knowledge.
Obviously the theorist of film cannot be expected to turn aside
and devote a few years to these more general psychological
concerns. He takes them as given and gets on with his par
ticular tasks. Continuing the same example. The theorist may
make his psychological assumptions, but they could in turn be
conditioned by a further set of, say, sociological assumptions.
It might be the case, for example, that in our society people
respond to certain perceptual situations in the expected way.
But in another society different patterns of social life and social
structure alter the particular psychological response. Our
theories are then 'culture-bound'. Eisenstein, for example,
clearly believed that he had tapped some sort of fundamental
'human nature' in his assumptions about particular sets of
stimulus and response. Perhaps he did. And even if he didn't, it
might be that his particular assumptions enable us to get good
enough results to justify their use as limited approximations.
But they do exist, and although the theory of film is not to be
reduced to a psychology or sociology of film, the theorist must
be aware of the point at which he has said 'thus far and no
further'.
A corollary of this is that thereis some need to explore these
contextual factors further, in particular the psychological and
sociological. This means more than simply the development of
a psychology and sociology of film. It also suggests the need
for a less specialized awareness of the role of the social and
psychological context in our responses to, and evaluation of,
film. This much is clear from Eisenstein's work. Historically,
however, the mainstream of film theory left the question there
and with few exceptions the generation that followed Eisenstein
swallowed his assumptions whole. Film theory was developed
as if it were context-free. Until the realist debate came to
dominate the field only John Grierson and, to a lesser extent,
Rudolf Arnheim took up the problem unknowingly bequeathed
them by Eisenstein. Arnheim pursued his interests into areas
outside of our present scope and into the more general realms
of a psychology of art.' Grierson, on the other hand,developed
an aesthetic explicitly grounded in his more general views of
society and the role of the cinema within it. In so doing he
became the first major exponent of a socially derived theory of
film, a strand of thought which thenceforth continued to weave
in and out of the literature.^ In centring his aesthetic on a
morality of social responsibility he elevated one element in the
context of film to the pinnacle of aesthetic importance. It is this
original claim which still makes him interesting.
Grierson: Purposive Cinema
Grierson's theory, if such it is, was not formulated in the
abstract. It was not the fruit of an extended and systematic
reflection, but rather a set of loosely related ideas which grew
out of his experience in the documentary movement in
England, Scotland, and Canada. His writings are perhaps best
conceived of as weapons in a lengthy, single-minded, and prac
tical fight for what he saw as a socially responsible branch of
the cinema. Thus, in attempting to distil the essence of his
1. See, for example, Rudolf Arnheim, Film as Art, Faber and Faber,
London, 1958, and, among others, Arnheim, Art and Visual
Perception, Faber and Faber, London, 1956, and Towards a
Psychology of Art, Faber and Faber, London, 1966.
2. This 'socially conscious' tradition has been visible in a number of
contexts. In the 'commitment' arguments surrounding the position
developed by Lindsay Anderson: see, for example, 'British Cinema:
The Descending Spiral', Sequence, 1, 1949, and the famous 'Stand Up!
Stand Up!', Sight and Sound, 1956. In the concern with popular
culture in its social context: see, for example, Stuart Hall and Paddy
Whannel, The Popular Arts, Hutchinson, London, 1964. And, finally,
in the interesting work of one of the blacklistedwriters of the 'red scare'
period, John Howard Lawson, Film: The Creative Process, Hill and
Wang, New York, 1964.

Shanghai Express (Josef von Sternberg)

•i
argument it is always possible to overstress a view which may
only have been formulated thus for the immediate purpose of
debate. For instance, there is more than one occasion on which
Grierson denies that he would wish to extend his arguments
specifically in support ofdocumentary to apply to all films; his
is the pragmatic acceptance of more than one type of cinema.
But there is in his work a clear sense of the general desirability
of realism - 'fresh air and real people' - and social responsi
bility, the factors he developed as the keynote to documentary.
This belief that the medium should disport itself responsibly,
almost weightily, turns up frequently in his general film
criticism. Much though he admired the Joe Sternberg of The
Salvation Hunters and The Exquisite Sinner, the 'von' of
Shanghai Express becomes the maker of an empty film. To
Grierson it is '... a masterpiece of the toilette', an ineffective
disguise for the thinnest of themes. It is Sternberg's fall into
vacant aestheticism which prompts the lovely aphorism 'When
a director dies, he becomes a photographer.'^
The criticism becomes more pointed on the Hitchcock of the
1930s and, incidentally, on the 'aesthetic' critics of the period.
I believe the highbrows, in their praise of him, have sent Hitchcock
off in the wrong direction, as they have sent many another: Chaplin,
for example. They have picked out his clever little pieces, stressed
them and analysed them until they are almost everything in his
directorial make up. We have waited patiently for the swing of event
(preferably of great event) to come into his films, something that
would associate him more profoundly with the dramatic wants of
common people. Something serious, I am afraid, will have to happen
to Hitchcock before we get it.'*
'Great events' and the 'common people'. A seriousness of
purpose and a sort ofpopulism are the keys which run through
3. Forsyth Hardy (ed.), Grierson on Documentary, Collins, London,
1946, p. 38. (The current extended edition is published by Faber and
Faber. Page references to the firstedition.)
4. ibid, p. 51.
Drifters

Grierson's work. Not entirely unleavened with admiration for

the 'aesthetic' touch or with pleasure in the 'froth' of
Hollywood, but these are always secondary to the main con
cern. Technique must invariably be subjugated to purpose. The
founding film of the documentary movement, Drifters, though
obviously stylistically infiuenced by Eisenstein, ensured that
the style must always be harnessed to the dominant idea.
Grierson is a pragmatist of the people; his first concern is with
society, and the 'art of film' is a rather distant second. In his
own terms the documentary movement 'used' the aesthetes,
Cavalcanti and Flaherty. However great Grierson's admiration
for Nanook and Moana he remains unhappy with Flaherty's
Rousseauesque and romantic views. In his greatest moments of
evangelical fervour Grierson conceives himself as struggling
against the whole of individualist culture. It is very reluctantly
that he admits to still finding .. the greatest image in rhetoric
is the single man against his horizon, seeking his destiny',^ an
image which comes very close to typifying Flaherty's cinema.
Grierson is altogether rather uncomfortable with its in
carnation in Man of Aran, and although his essays revolve
principally round documentary, the arguments he invokes are
general enough to spread their wings throughout his writing.
In the end his is a case for an aesthetic of film, not simply a
justification for documentary.
His views on film grow out of his analysis of twentieth
century society; he thus has a practical social interest in film
theory. For him the cinema has a potentially great role to play
in the solution of twentieth century problems. Problems like
maintaining peace, increasing international understanding, and
maximizing opportunities for citizenship. The justification for
one form of film as against another, for realism as opposed to
fantasy, for the collective ethic as opposed to the individualis
tic, lies inthis social role. Ultimately Grierson's reference point
is democracy with a capital D. Mass media, he argues, can be
the binding which will hold our societies together. And the
cinema - this is, of course, before the rise of television - is
potentially the most powerful among them.
The particular form of his argument derives very much from
Lippmann's famous account of the failings of contemporary
education and democracy. Change, it runs, is all-pervasive in
the twentieth century. Even should he wish it, the individual is
no longer able to keep track of the changes, the problems, the
issues of his day. But the traditional conception of democracy
hinges on the 'informed citizen' as do traditional educational
techniques. Education is out of touch with the new demands of
these large societies, while the mystical participatory notion of
democracy has been left entirely behind by events. What is
required of the mass media is that they cover all sectors of
society in order that these various social elements can remain
5. Forsyth Hardy (ed.), Grierson on Documentary, p. 198.
 r I V

Housing Problems

in contact with one another. They should serve to reinforce

ideas of citizenship and of collective destiny. Like Drifters,
Night Mail and Industrial Britain, films should increase our
awareness of the range of the new world and the magnitude
of change. Through such awareness, Grierson assumes, we
will come to understand our world the better. Like Housing
Problems and Workers and Jobs, film should venture directly
into social problems and social criticism. They should show
' the common man, not in the romance of his calling as
in the earlier documentaries, but in the more complex and
 intimate drama of his citizenship.'® And at the most general
level of all, the cinema should shift its focus from .. the
drama of personal habits and personal achievements' to the
collective and cooperative element in modern life. Within
societies and across the world man must be unified. Film, says
Grierson, '... is the medium of all media born to express the
living nature of interdependency."
This, then, is the argument of a reformer. Thirty years later
it sounds at the very least a little romantic. What looked like a
great and powerful medium then, now looks to be a relatively
minor instrument of persuasion. 'Democracy' seems rather less
of a worry, at least in Grierson's sense, and it is clear that
simply to show the rest of the world in their normal pursuits
does not have the automatic benefits which Grierson tended to
assume. In many respects his argument has dated, as would any
case derived from specific social conditions, and Grierson was
above all a social pragmatist. Unlike the 'philosophical' realists
who try to justify realism in less socially sensitive terms, he is
rarely to be found considering the 'medium'perse.Rarely does
he invoke photography, and its affinity to reality, as the fount
of cinema, and then only in passing. For him the intention was
clear, the raison d'etre practical, the method realistic. If film
was expensive and the moguls unsympathetic (he frequently
raises this as the obstacle to a good cinema), then turn to the
Government, turn to the sponsors. If the 'big films' are inacces
sible, then the little documentaries will do. If aesthetic choices
must be made then Grierson's path is clear: realism.
The penalty of realism is that it is about reality and has to botherfor
ever not about being 'beautiful' but about being right... In our
world it is especially necessary these days to guard against the
aesthetic argument. It is plausible and apt to get under the defences of
any maker in any medium. But, of course, it is the dear bright-eyed
old enemy and by this time we knowit very well. Documentary was
6. Forsyth Hardy (ed.), Grierson on Documentary, pp. 148—9.
7. ibid., p. 247.

Ermler: Fragment of an Empire ^

. , «-

< J-^- a -- A

I
 h
'kJ- p- •

.
A o ♦

' -

• *

»•-w ?A

* /

l!>4*
from the beginning - when we first separated our public purpose
theories from those of Flaherty - an 'anti-aesthetic' movement.®
Grierson asks that the cinema come to earth. Although his
argument for realism over and against aestheticism is made in
the context of the documentary movement, it must clearly
apply to the whole range of cinema. If he seriously wished to
achieve his social aim then the whole cinematic castle must be
captured, not just the outposts. Heasks that the cinema should
keep in touch with the 'common people', and is saddened by
this failure in the later films of Eisenstein, Pudovkin and
Dovzhenko. The General Line, Deserter and Earth, he says,
suffer in that their directors are alien to the material with which
they are working. Interestingly it is the neglected Ermler —
occasionally remembered for the montage ingenuity of
Fragment of an Empire —whom Grierson singles out from the
Russians. The truth is that Grierson's populism derives not a
little of its power from a deep-seated mistrust of 'highbrows' in
particular, and the middle-class in general. As we have already
seen, Hitchcock was allegedly undone by the highbrows. In a
more evangelical strain Grierson is concerned for Cagney:
It [the cinema] began in the gutter and still trails the clouds of glory
with which its vulgar origin was invested. But if we ask it to go deep,
be sure we are not just asking it to go middle-class. And be sure that
the next phase of cinema may not be to eliminate the Cagneys in
favour of the Colmans and indeed to Colmanise Cagney himself.'
The implied siding with 'popular' culture as against 'high' is
a brave sally in advance of its time and a reflection of
Grierson's frequent expression of his cinematic aim as
sociological. This last, presumably, in the traditional English
sense: what has been aptly called 'a kind of Marxified
Fabianism'!
Whatever sympathy one might feel for Grierson's position,
8. Forsyth Hardy (ed.), Grierson on Documentary, p. 179.
9. ibid., p. 72.
and it would be difficult not to feel any, his aesthetic argument
is clearly problematic. Even allowing the particular social
analysis it is not at all clear why realism should follow of
necessity. His intention, after all, is not simply to show one half
of the world how the other half lives. Although the early years
of the movement saw the production of many such films,
Grierson came to conceive of them as the weakest weapon in
the armoury. More important to him was to enter the arena of
social debate, and, ultimately, change people's ways of thought,
their cultures. What he is really asking for is a purposive
cinema; if you like, a 'moral' cinema. Obviously not in the
limited sense, but an attempt to encompass the concept of
social responsibility in the cinema, to include the 'moral'
dimension in criticism as the overriding concern. But,of course,
there is no good reason to suppose that Grierson's realism, its
material shot 'in the raw', is any more powerful in its effect
than is staged cinema. Propaganda is a far more subtle business
than it seemed some thirty years ago. As a justification for
realism Grierson's argumentdoes not hold water; his is reallya
case for 'responsible propaganda'. Propaganda that is 'right'.
At heart Grierson sees film aesthetics as a by-product of
historical 'destiny'. Society is far too complex to be unified and
stabilized in the old familiar ways. The close-knit community
has gone, and industrial democracy is unworkable on the old
bases. A new and 'Democratic' culture is to be forged and film
is the medium to do it. Not merely through spreading infor
mation, though this is part, but by creating a cultural context in
which social interdependency is a practical possibility. Once
Grierson expands his case beyond the simple spread of know
ledge then we find two separate strands to the argument. Both
begin with a sociological assumption about the nature of the
changes experienced in modem society. In the terms developed
by the French sociologist, Emile Durkheim, the passage from
mechanical to organic solidarity. The form of solidarity
characteristic of modern societies is organic: resting on the
complex interdependence of the members of that society. For
Grierson —although he did not employ these terms —the
problem lies in providing a cultural basis for this organic
solidarity; a system of beliefs which will make modern, demo
cratic, industrialized societies work. It is not possible to live by
total consensus on all points (Durkheim's mechanical solidar
ity) any longer, and, as I have suggested, Grierson puts forward
two interlinked but separable arguments. The one, which leads
to documentary in particular and realism in general, sees the
spread of information about the range of differences between
man and man as a basis for social solidarity. The weakness of
this argument, at least as an exclusivecase, is that it approaches
the problem of organic solidarity with the weapons of
mechanical solidarity. It assumes that a consensus achieved by
increasing knowledge of the 'real' world will bind society
together. Knowing how the other half lives will enable us to
recognize and hold on to our common humanity. This assump
tion does not seem entirely realistic, unless we also assume that
the simple spread of knowledge automatically promotes mutual
understanding. The evidence hardly favours such an optimistic
Grierson himself was obviously not happy with this as it
stood. His later essays implicitly develop the view that the
problem is rather more complicated than can be dealt with in
these terms. So we find a further case made for 'propaganda'.
Information and realism are not sufficient. Our films must
involve the individual in the process of creating solidarity, in
the interdependence of his society, in the 'drama of his citizen
ship'; in Eisenstein's terms, the need is for 'pathos'. Although
we no longer live in a simple society depending on common
agreement, there is a level at which we can be 'persuaded' to
agree: the level of recognizing our interdependence as
transcending our differences. And this, of course, is not a brief
for any particular aesthetic limitation such as realism. It is an
argument that, whatever the particular form, the crucial factor
is its social role vis-d-vis this fundamental need for solidarity.
In the last analysis Grierson's position leads in the direction of
the end (providing a culture to underwrite the workings of
modern democratic societies) justifying the means (any form of
propaganda directed toward this morally accepted end).

Context Domination
Grierson's theory, then, is context-dominant. It elevates one
area of contextual concern above all other factors. Pushed to
the limit it can have no implications for our aesthetic judge
ments of film separate from the social function performed by
the film. The measures of aesthetic taste are limited to two: the
social responsibility of the film, and its effectiveness in achiev
ing this socially responsible aim. As it stands the Grierson
argument is no basis for anything more than this and, as such,
it exemplifies the problems which have been attendant upon the
few attempts to specifically involve context factors in theories of
film. The problem is more acute in aesthetic argument; in
developing models of film there is at least a further arbiter, how
well the models fit the reality. If they do not, then we may be
forced to revise the particular form of our contextual assump
tions. But aesthetic argument is directed toward the justifi
cation of a set of general evaluations on the basis of certain
axioms, and some of these latter are invariably contextual
assumptions. In Grierson's case the position is extreme in that
aestheticjudgements about film are not simply conditioned by
contextual assumptions about the social and psychological role
of cinema, they are dominated. Aesthetics is reduced to
morally prescribed social theory. 'Purposive cinema' em
phasizes the 'purposive' at the expense of the cinema.
It would be entirely wrong, however, to take this as a reason
for avoiding the context issue in aesthetic discussion. The
attempt to develop a context-free aesthetic - the effective aim
of the 'philosophical realists' — is just as problematic as
Grierson's context-domination. The problem is to find a
balance between evasion and subjugation. Aesthetic argument
can be reduced neither to 'essence' nor to social (or psy
chological) discussion. It may be that it ought to combine both.
And equally film models are neither created in a vacuum (which
implies unstated assumptions) nor can they be reduced to a
sociology or psychology of film. Whatever our theories they
must in some ways be abstractions and simplifications. To
attack the issue of context is to begin to specify the precise
ways in which they are so. Grierson's theoretical importance
lies in the impetus he gives to this endeavour.
4: Aesthetics of Realism: Bazin and
Kracauer

As we have seen, the spectre of realism has haunted film

aesthetics from the beginning. From the moment Lumiere
singled out the 'rippling of the leaves' as fundamental to film,
realism was marked out for immortality. Eisenstein frequently
found his wings clipped by its Stalinist variant, while John
Grierson bluntly asserted the aesthetic in the context of his
moral and political analysis. The standard texts are never
without their statutory reference to the realistic propensities
of the cinema. Predictably, however, it is outside the Anglo-
Saxon context that the tradition finds its full and 'philosophic'
flowering. Disparate in style and opinion, the French critic
Andre Bazin and the emigre German academic Siegfried
Kracauer share at least their starting point. Often differing
on the direction in which their thought carries them, they
are both crucially involved in the attempt to create a non-
social aesthetic of the real. Theirs is the pursuit of the Holy
Grail.
Basically they share two assumptions. The first of these is
that a medium possesses certain distinct characteristics which
have specific implications for the nature of communication in
the particular medium and for any related system of aesthetics.
Hence Kracauer's study, '... rests upon the assumption that
each medium has a specific nature which invites certain kinds
of communications while obstructing others'.' The special
characteristics of film have to do with photography: for Bazin,
'those categories of resemblance which determine the species
photographic image likewise, then, determine the character of
its aesthetic as distinct from that of painting.'^ In other words,
to create a valid system of aesthetic standards we must ground
it in the central distinguishing characteristics of the medium
in question. For those bothered by relativism in aesthetic
standards this offers some relief from the flux of subjective
judgements; such a criterion seems independent of particular
individuals and yet deeply implicated in the nature of the
medium. Aesthetic dispute ceases to be in terms of 'what I like'
versus 'what you like'. Now the questions to be asked are: 'Is
this film genuinely cinematici Does it fulfil the proper poten
tialities of the medium?' Of course, like all solutions to this
problem - if indeed it is a problem - the argument is really
only pushed back one stage. Instead of making directly subjec
tive judgements about the quality of the film(s), we now make
them about the 'essence' of the medium. From this general
evaluation, particular judgements follow. In this sense, then, the
attempt is misdirected. A body of aesthetic propositions is no
more 'objective' for being based on the central characteristics
of the medium, and such a method does not necessarily offer
any extra justification for the judgements involved. What is the
'essence' of film to one man may well be the residue to another.
The basic assumption of these arguments, then, is not
entirely without its difficulties. A second is even more prob
lematic. Here both Bazin and Kracauer fall into what Pauline
Kael calls 'the great, lunatic tradition'. The essence of film, they
would argue, lies 'in its power to lay bare the realities',^ in 'the
1. Siegfried Kracauer, Theory of Film: The Redemption of Physical
Reality, Oxford University Press, New York, 1965, p. 3.
2. Andre Bazin, What is Cinema?, Vol. 1, University of California Press,
Berkeley and Los Angeles, 1967, p. 15.
3. ibid., p. 15.
redemption of physical reality'. There is a natural affinity
between the cinema and the recording and revealing of reality,
an affinity which becomes the central axiom of the realist
aesthetic. The problems raised by this assumption are very
extensive and their discussion forms a large proportion of this
chapter.
Both writers wish to found all aesthetic judgements on the
central characteristics of the medium as theyperceive them. If,
for the sake of making any progress at all, we bracket discus
sion of this questionable procedure, we can then explore the
divergent cases put by Bazin and Kracauer. It is perhaps worth
noting at the outset that Bazin seems the more sympathetic of
the two, if only because his work is far less deliberately
systematic. The reality aesthetic is something that informs most
of his work, and if he does not seem to sink into the quicksands
in which Kracauer flounders, it may only be because he does
not push his position to its limits. This has been left to some of
his more recent followers. Kracauer, by way of contrast, offers
a teutonic epic to match any of the silent German films on
which he has so extensively written. One is irresistibly
reminded of the well-known sociologist who writes, it has been
said, in Heidelberg English. But even if his assumptions are
faulty, if some of the argument is blurred, and if the build-up is
laboured, Kracauer at least tries to formulate a consistent
aesthetic system. The execution may falter but the intention
seems legitimate. And because he tries to be systematic,
Kracauer, above all, shows us some of the strange places to
which the realist aesthetic can lead. What may be implicitly
involved in Bazin is desperately explicit in Kracauer. For this
reason, even if for no other, it is useful to turn first to Theory of
Film.

Kracauer: Systematics of the Real

Theory of Film attempts to cover a wide area. It involves two
separate arguments in support of the realist aesthetic, and a

Musical 'realism'? Fred Astaire and Ginger Rogers in Top Haf, and (right) Gene •
Kelly in Singin' in the Rain
Si

•7 ^

wfw
 ?T».'
Nr.''
I,

f
tMAI) f. ••!
sr '
^1^

I i I A* itiit.

j I'^'®^W-?-1^
iir.^
' • TT?
•'• "•m
j*

t' If
r"*
21 •'•-•> If-:!!
^ §*Jt .^T—••^

• -^'fL' ® '

V-« w
r t • . %

—• -^.iSr- -« ^ C- w.
sometimes laborious drawing out of the implications of the
theory. I do not intend to much concern myself with the
inconsistencies and oddities of this last. Pauline Kael's scathing
review (most reviews were highly complimentary), 'Is There a
Cure for Film Criticism?"' provides ample evidence of his
specific peculiarities. Perhaps one or two will serve as signposts
for the rest. Witness the way in which he squeezes the Musical
into the 'goody' category by dint of arguing that Astaire's
dancing emerges from the real-life events of his films, though
Top Hat hardly seems a paragon of realism. And, evidence of
the peculiar limits of his discussion, he fails entirely to mention
the Gene Kelly of AnchorsAway and Singin' in the Rain, who
is probably a far better example in support of his case. Or
again, the way in which he rescues Song of Ceylon from hell-
fire and perdition by arguing that its montage sequence (a sin
against realism, though since he likes early Eisenstein perhaps
only sometimes a sin) serves to make the 'real' sequences
around it that much more real. This ignores the fact that the
sequence in question is a way of developing forcefully one of
the main themes of the film. Or, lastly, his failure to recognize
that the romanticism of Louisiana Story is more than 'simply'
a moral issue, but can also impinge on the 'realism' of the
movie. And so it goes on, with Kracauer turning this way and
that apparently in an effort to 'fit in' all his favourite movies.
As a practical guard to the gates of cinematic heaven his
judgements are odd. To understand why we must turn to the
chapter and verse.
Kracauer makes two central arguments in support of his
position. One is fundamental to the whole theory; a second is
added almost as an afterthought. I shall briefly suggest the
second, and then explore the first in some detail. Kracauer's
4. Pauline Kael, 'Is there a Cure for Film Criticism? or: Some Unhappy
Thoughts on Siegfried Kracauer's Nature of Film: The Redemption of
Physical Reality', Sight and Sound, 31, 2, 1962. Reprinted in Pauline
Kael, I Lost it at the Movies, Jonathan Cape, London, 1966.
Romantic realism: Flaherty's Louisiana Story

justificatory addendum is really simply a question ofhis fitting

his view into the intellectual ethos of the time at which he
wrote. It is a pessimistic variant of the 'end-of-ideology' thesis
of the late 1950s (usefully compared with Grierson's earlier
optimism) and is founded on the view that contemporary
society is best characterized by the absence ofa set ofgenerally
agreed beliefs and values. Man is without the ideological shelter
once provided by religion; he is aware of the world only
through the abstractions of thetechnological ethos. 'He touches
reality only with the fingertips.'^ He has a need to once more
grasp reality, a need which can be met through the cinema
(again, cf. Grierson). A realist aesthetic is therefore in tune
with the requirements of the age. The obvious problem with
this argument is, ofcourse, that this diagnosis ofcontemporary
5. Siegfried Kracauer, op. cit., p. 294.
society is not quite as convincing as it might have been a
decade ago. It seems less likely now that the 'golden age' beliefs
are giving way to a vacuum, and more likely that the beliefs
themselves are changing in some fairly drastic ways. One
wonders how Kracauer would have reacted to Easy Rider.
Happily, though, diagnosing contemporary society is not at
present necessary. It only finds a mention here for the sake of
completeness. Rather, I shall look at the core arguments of
Kracauer's theory, which fall into five linked parts.
(a) It is the essence of photography to incline toward the
straightforward recording and revealing of reality.
(b) Film involves photography.
(c) Although film involves other elements (such as editing
and sound), photography '... has a legitimate claim to
top priority among these elements, for it undeniably is
and remains the decisive factor in establishing film con
tent .. .'^
(d) Therefore, film shares with photography the inclination
toward capturing unaltered reality. This is its major
characteristic.
(e) Hence, given the assumption about basing aesthetic
standards on the fundamental nature of the medium, it
follows that realism is the principal criterion of aesthetic
value in the cinema. The realistic tendency is at one with
the very essence of film.
Of these five propositions only (b) seems completely un
questionable. Film, we must agree, does involve photography,
although whether the continuous frame photography of the
movie camera is that close to the single shots of the still camera
is arguable. For present purposes, however, there is no need
to become involved in the finer points of the argument. The
remaining four propositions provide ample material.
First (a): the nature of photography argument. When it
6. SiegfriedKracauer, op. cit., p. 27, my italics.
actually comes to the point, Kracauer is not so absolutely on
the side of the realist angels as this bald formulation suggests.
Although he speaks of following the realistic tendency as a
minimum requirement of the photographic approach, he is also
willing to admit that.. the formative tendency [meaning the
interference of man], then, does not have to conflict with the
realist tendency. Quite the contrary it may help substantiate
and fulfil it —an interaction of which the nineteenth-century
[photographic] realists could not possibly be aware.'^ One's
heart must go out to the nineteenth-century realists. For them
the self-evident distinguishing characteristic of the new medium
was that it revealed the world before the lens in considerable
detail and with absolute veracity. Their point of comparison
was the realism offered by painting. Rather like Thomas in
Blow Up they believed in the power of the camera as revealer of
the world; unlike Thomas, they were never really required to
face the consequences of their position. But surely they were
correct on the very issues of which Kracauer says they could
not be aware, precisely because they were not aware. If we once
claim that the revealing function of the camera vis-d-vis some
given reality 'out there' is the crucial essence of photography,
as they did and as does Kracauer (and, for that matter, Bazin),
then any formative effect which the photographer may have
must be a denial of this essence. Ultimately this leads to the
situation wherein the human agency must be totally removed —
angle, exposure, shutter speed, all randomly determined. Any
human agency has a formative effect. But once this last is
admitted it must also be admitted that the 'reality' with which
we are dealing is never independent of human agency. The
'reality' revealed by the camera must also depend on the photo
grapher. And that admission denies the absolute basis of the
realist position.
Because Kracauer is conscious of this problem (presumably)
he introduces a number of caveats concerning the relation
7. Siegfried Kracauer, op. cit., p. 16.
between realistic and formative tendencies. The one quoted
above is typical. But it is not possible to hold both to the
position that there is an independent 'reality' which it is the
essence of photography to reveal, and to the position that the
'reality' revealed by the camera must also depend on the photo
grapher and, for that matter, the audience. If the latter is the
case - as it must be for Kracauer since he is willing to admit
the place of the formative tendency as appropriate in certain
circumstances (an admission which the nineteenth-century
realists would not be foolish enough to make) - then the first
position is absurd. This is so because to admit the influence of
the photographer assumes that 'reality' varies according to who
perceives it and how, while the 'essence' position- characteris
tically nineteenth century - assumes there is one fixed 'reality'
independent of the observer. In consequence, once we allow
the photographer to creep in, he necessarily undermines the
'essence'. It then becomes necessary to detail the varying con
ceptions of 'reality' held by photographer and audience, and
to distinguish between photography as realistic in essence and
photography which is thought of as realistic. We learn to think
of photography as realistic in the sense that we accept a photo
graph as showing whatever was before the lens at that par
ticular moment. This has nothing to do with some mystical
essence of the medium; it is a response we have developed until
it has become second nature, which, incidentally, is not found
in those cultures hitherto lacking photography.
It is basically because Kracauer believes that it is possible to
isolate the essence of the medium apart from the way in which
people respond to it that he finds himself in this mess. Certainly
the classic realists were wrong in that they failed to see that
some 'formative tendency' must always be present. If the
photographic spirit does lie in the neutral revelation of reality,
there can hardly be many truly photographic photographs. If
this is Kracauer's 'reality' then there are no films in the world
which meet his aesthetic requirements. Since he fills over three
hundred pages with discussion of films ranging from Forty-
Second Street through Hamlet to Paisd we must charitably
assume that his conception is rather less strict than it appears.
But then whatnotion of reality does he employ? There appears
to be no single answer. But it is clear, whatever variations he
may develop, that he cannot acceptably claim to have isolated
The Essence of Photography. Instead he is implicitly asserting
the commonplace that a photographer can attempt to render
the illusion of reality with the minimum of interference, or
he can interfere to different extents, in different ways. The
dichotomous tendencies epitomized in Melies and Lumiere now
become a question of degree; the opening shot of Kracauer's
argument reduces to: 'Photographers can variously interfere
with the process of photography.' Proposition (a) and the
essence of photography are no more.
Proposition (c) asserts that photography has top priority
among the elements of film; it is decisive in establishing film
content. What Kracauer means by this is not entirely clear. The
limiting case - the content of a film must be photographable —
is uninteresting in this context. But Kracauer offers no real
support for, or specification of, this position apart from his
previous argumentsabout reality. That is, having established (!)
that the essence of photography is to record and reveal reality,
to propose the centrality of photography in film is to arrive at
the essence of the film. But why should we accept any element
as having top priority? Film is a combination of many diverse
characteristics. To sink the lot beneath the iceberg-tip of photo
graphy seems unnecessary and misleading. To fail to justify
such a procedure is critically criminal. It is clearlyan arguable
proposition that the differences between movie and still photo
graphy far outweigh the similarities, though perhaps when one
is dealing with essences ... But Kracauer does not give the
problems a decent airing; the possibility is cursorily dismissed
within the first paragraph of Chapter Two. It is not that
proposition (c) is necessarily wrong —it is an unsupported

Forty-Second Street ^
 ; ' . «ll!l»i| -

4»aj'

"o'pfj
4 .C.*
r Zf '
"m

ttr* -f'
i
r
assumption for which no grounds are presented-for accepting it
as right.
Proposition (d) falls for a number of very obvious reasons.
Even accepting that photography is the top priority in film, the
dilution of the 'nature of photography' arguments leaves us
only with the assertion that 'film-makers can variously interfere
with the cinematic process'. But even by-passing this objection,
other of Kracauer's specific assertions can bear further
scrutiny. Some of the emphases in his discussion of the basic
nature of the medium have changed in the move from photo
graphy to cinema. 'The basic properties [of film] are identical
with the properties of photography. Film, in other words, is
uniquely equipped to record and reveal physical reality and,
hence, gravitates toward it.'^ Because a medium is well
equipped to fulfil some task it naturally 'gravitates' in that
direction. The general assumption is that media tend to operate
in the field for which they are best equipped. If we are to take
this at its face value all the many films from Melies to Planet of
the Apes which are not in the puritan tradition of realism
(including many that Kracauer favours) are not just bad films,
they are contrary to the natural tendency. They should never
have gravitated in this direction at all. Some new laws of nature
must be in operation! Since Kracauer can hardly mean this, he
must, presumably, mean that films are only really films in so
far as they record and reveal physical reality. Anything else is
simply not a film.
But what exactly is this 'physical reality' to the revelation of
which the cinema should be devoted? The answer is fascinat-

Now there are different visible worlds. Take a stage performance or a

painting; they too are real and can be perceived. But the only reality
we are concerned with is actually existing physical reality - the
transitory world we live in. (Physical reality will also be called
'Material reality' or 'physical existence', or 'actuality', or loosely just
8. Siegfried Kracauer, op. cit., p. 28, my italics.
'nature'. Another fitting term might be 'camera-reality'. Finally the
term 'life' suggests itself as an alternative expression ...)'
Resisting the obvious temptation to infer that anything which is
not a stage performance or a painting constitutes 'reality' (or
vice-versa), we are left in the midst of a welter of misnomers.
Just savour 'actually existing physical reality'! Anything
stuck in front of a camera and photographed must surely be
actual, must exist, and must be physically real. Whether we
believe it to be a man in a costume or the title figure of The
Creature from the Black Lagoon matters little in respect of
its 'actually existing physical reality'. What Kracauer must
be concerned with is not whether it is real, but whether it is
really real!
And in this respect, one of his synonyms is quite instructive.
The expression 'camera-reality', although Kracauer's usage
varies, is suggestive of the reality status lent to a subject by the
camera. In other words, the tendency for our belief in the
truthfulness of the camera to lend credence to the subject
matter of the film. Now this clearly gets into complex psy
chological and sociological questions which are not at all
involved in the formal notion of reality with which Kracauer is
operating. However, his actual usage does get close to this
meaning. He is prepared, for instance, to speak of Dreyer's Joan
of Areas attempting to transform '... the whole of past reality
into camera-reality'.'" Since in this case it is not possible to film
real reality, the concept of camera-reality must include the
possibility of recreating reality in front of the camera. It is a
tiny step from here to the creation of familiar realities (as in
most narrative films), and then to creating unfamiliar realities
(as in many 'fantasy' films).
Whatever Kracauer may say in principle, like Bazin, he is, in
practice, willing to include films which attempt to create the
illusion of reality. It is now a question of appearances rather
9. Siegfried Kracauer, op. cit., pp. 28-9.
10. ibid., p. 80.

The Creaturefrom the Black Lagoon •

 ir f ^ " : •.

• j,
^•' . • • . '• ;••••':.• -..-V^-fi^-' :.5« •» • .
 l|!«a»(*l^

:,. 'iW'!,?.
v^: •••

f
'-SSfe"'' -

♦ ' "'i'^ '!' 'f' -i'-?.'

-'" f "(»tA , f-

'-s-£

» '
 M

Neo-realism: Umberto D

than one of essences. Indeed, the reader begins to feel that

Kracauer is altering the meaning of 'camera-reality' according
to the films he likes, instead of judging the films on the basis of
the reality criterion. There are no problems with neo-realism,
Umberto D or Paisd. But he also admires Potemkin and Wild
Strawberries, which are hardly easy. And finally he opens
nearly all the flood-gates by accepting the presentation of
'special modes of reality' as legitimate. 'Films,' he says, 'may
expose physical reality as it appears to individuals in extreme
states of mind.'" Any individuals? This is the licence to print
anything, and what started life as a strict aesthetic criterion is
now whittled down to an eclectic hold-all.
Which leads to proposition (e): the aesthetically desirable
film is the one which lives up to the basic nature of the medium.
11. Siegfried Kracauer, op. cit., p. 58.
This approaches grand tautology. If we once accept the
proposition that the fundamental character of film lies in its
ability to represent and record reality, then the argument that a
film which does not fulfil this fundamental character is bad, is
one of two things. Either the reality proposition is a definition
of film and in consequence all films are therefore aesthetically
desirable, or the reality proposition is an empirical statement
about the nature of film. Since the first is uninteresting the
second must be the case. But if it is an empirical statement it
must be based on the evidence provided by the body of films,
that is, if we look at the body of existent films we will discover
that their basic common characteristic is their 'realism'. This is
evidently not the case. The third 'alternative' is that Kracauer
believes 'realistic' films to be good, and chooses an extraordin
arily circuitous way of 'justifying' his position. Ultimately such
subjectivism seems to me inevitable and not at all disturbing.
To Kracauer it is obviously anathema; hence this elaborate and
dubious general argument.
As if in recognition of these problems, Kracauer is per
petually hedging his bets.
Imagine a film which, in keeping with the basic properties, records
interesting aspects of physical reality but does so in a technically
imperfect manner; perhaps the lighting is awkward or the editing
uninspired. Nevertheless, such a film is more specifically a film than
one which utilises brilliantly all the cinematic devices and tricks to
produce a statement disregarding camera-reality. Yet this should not
lead one to underestimate the influence of the technical properties. It
will be seen that in certain cases the knowing use of a variety of
techniques may endow otherwise non-realistic films with a cinematic
flavour.'^

The reader alert to the style of Kracauer's passing amendments

will note that we are now concerned with 'interesting' aspects
of reality. To pursue this leads back into the maze from which
12. Siegfried Kracauer, op. cit., p. 30.
we have just departed, so let us look elsewhere. Although
Kracauer repeats that the dominant aesthetic standard is the
film's reality, he is now prepared to admit that unspecified
techniques can add just a soupQon of cinema to an otherwise
unacceptable mixture! Presumably, then, techniques can
enhance films which are already acceptable on the basis of their
realism. In other words, films of reality in which the lighting
is not awkward and the editing is inspired must be best of all.
Thus we are not concerned with some 'pure' relation with reality,
but basically with the efficiency with which a film can create
the illusion of reality. To commit the heresy of reducing
Kracauer's argument to a phrase: the more convincing a film,
the better it is.
The only conclusion to be drawn from all this is that
Kracauer's main argument ends up going in every-which-
way. The only systematic elaboration of a realist aesthetic of
the cinema founders at the start. And in Kracauer's case this is
immensely important since he does not have the sure touch of
Bazin in his discussion of the films themselves. His main
quality is in his argument, and his argument is lacking con
siderably. One of the many reasons for this failure may be that
the attempt itself is misdirected. To found an aesthetic on the
'nature of film' is misconceived whatever the problems arising
from identifying 'reality' as the crucial element. Ultimately
Kracauer derives from the purest of romantic aesthetics. One
of his synonyms for 'reality' is 'nature', and it is unspoiled
nature which he wishes film to preserve untouched. Clearly he
approves of Rossellini, about whom it is said that he
reprimanded his cameraman for removing a white rock from a
field of dark rocks in which they were shooting. What right had
the cameraman, he asked, to 'improve' on nature? Kracauer
responds to film in this way — this is what he likes. It is
unfortunate that he felt the need to erect such an immense
edifice in justification. He really only wants films to be nearer
to what he considers 'real life'.
Types of realism:Farrebique (top); La Terra Trema
Bazin: Two Types of Realism
One of the many oddities of Theory ofFilm is its failure to refer
to Bazin. Given Kracauer's concern with realist cinema,
especially as epitomized in Italian neo-realism, he mightreason
ably have been expected to draw upon the work of Europe's
finest critical authority on the subject. In Esprit and Cahiers du
Cinema of the late 1940s and 1950s, Bazin wrote extensively
on Rossellini, De Sica, Fellini, and the rest. When Theory of
Film first appeared (1960), two volumes of Bazin's collected
essays, Qu'est-ce que le cinema? had already been published.'^
They included in their pages a number of essays relevant to
Kracauer's themes, notably 'Ontologie de I'image photogra-
phique', 'Montage interdit', and 'L'evolution du langage cine-
matographique'. This material seems to have escaped
Kracauer's attention, a failing for which the theoretical price
was rather high. It is not that Bazin escapes the maze in which
Kracauer is lost; far from it! It is rather that he displays a
sufficiently impressive sensitivity to cinema as to encouragethe
irrational hope that some of it might have rubbed off on to
Kracauer. The combination of Bazin's 'feel' and Kracauer's
systematization might just have produced the first (and only)
full-scale aesthetic of film. As it is, we have only the two
surprisingly independent contributions.
Even so, Bazin does seem to sidle round many of the
obstacles into which Kracauer unseeingly blunders. Both
authors share a considerable admiration for the achievements
of Italian neo-realism; in particular, for the films of Rossellini.
And yet Bazin rarely falls into the trap of seeming to formulate
a puritan aesthetic which will include neo-realism at the ex
pense of all else. Unlike Kracauer (formally, at least) he admits
to different forms of realism. Thus, for example, the distinction
13. Andre Bazin, Qu'est-ce que le Cinema? I. Ontologie et Langage,
Editions du Cerf, Paris, 1958.
Andre Bazin, Qu'est-ce que le Cinema? II. Le Cinema et les autres
Arts, Editions du Cerf, Paris, 1959,
he draws between the 'documentary' realism of Farrebique and
the 'aesthetic' realism of Citizen Kane, both forms allegedly
finding their unification in La Terra Trema}* This willingness
to speak of different types of realism can lead to problems in
interpreting his position. In Signs and Meaning in the Cinema,
Wollen takes to task two contemporary inheritors of Bazin's
views (Barr and Metz) over their opposition of Rossellini and
Eisenstein. The villain for Bazin, he points out, was not
Eisenstein, but German Expressionism. But the real problem is
that at different times, and in different ways, Bazin occupies
both positions. He starts life invoking a case similar to
Kracauer's in favour of a 'purist' realism. But this proves too
limiting for his much more catholic tastes, and so he also
develops a second case for what I shall refer to as spatial
realism. Unfortunately, he never really brings the two concep
tions face to face; never really resolves the strains between
them. It seems useful here to take a closer look at these basics
of his argument, if only to fully understand how he dodges
some of Kracauer's more extravagant disasters.
The 'purist' case finds its main development in the 1945
essay, 'The Ontology of the Photographic Image'. Here Bazin
is concerned to isolate 'realism' as the fundamental character of
photography and hence of film. He makes a dubious distinction
between the two sorts of realism.
... true realism, the need that is to give significant expression to the
world both concretely and its essence, and the pseudorealism of a
deception aimed at fooling the eye (or for that matter the mind); a
pseudorealism content in other words with illusory appearances."
Needless to say, it is 'true realism' of which Bazin approves,
although he does not enlarge specifically on what constitutes
a 'significant expression'. The distinction is reminiscent of
Kracauer's attempt to distinguish what I called the 'really real',
14. Andre Bazin, Qu'est-ce que le Cinema? IV. Une esthetique de la
Realite: Le neo-realisme. Editions du Cerf, Paris, 1962, pp. 39-40.
15. Andre Bazin, tVhat is Cinema? op. cit., p. 12.

2001: A Space Odyssey ^

 . S9r*rr*..

•^T.

,
%
n
and it meets with the same sort of problems. If the 'pseudoreal'
fools both eye and mind, then who is to say that it is not truly
real? To found an aesthetic on a distinction of this type is
rather like founding a morality on the view that murder is only
wrong if you get caught. Contrast such a position with that
expressed six years later in 'Theatre and Cinema'.
The realism of the cinema follows directly from its photographic
nature. Not only does some marvel or some fantastic thing on the
screen not undermine the reality of the image, on the contrary, it is
its most validjustification. Illusion in the cinema is not based as it is
in the theatre on convention tacitly accepted by the general public;
rather, contrariwise, it is based on the inalienable realism of that
which is shown. All trick work must be perfect in all material
respects on the screen. The 'invisible man' must wear pyjamas and
smoke a cigarette.'®
Which implies, surely, that the only reality is the reality of
which the audience is convinced; quite conceivably, the
pseudorealism which fools the eye and mind. There is a clear
contrast between the 'purist' position wherein the cinema taps a
fixed 'true' reality and this latter case in which the cinema, by
its nature, lends realism to something which is illusory. The
most obvious conclusion from the second argument is that the
cinema is dedicated to representing '... a plausible reality of
which the spectator admits the identity with nature as he knows
it.' Presumably 'as he knows it' is flexible and depends on a
series of social and psychological conditions. Few of us in
1970 'know' the lunar landscape of 2001: A Space Odyssey in
any but the most indirect sense, yet we are willing to accept its
realism. But an audience in 1930 would probably have found it
much more difficult. In other words, our acceptance of illusion
is a sort of convention having nothing to do with our metaphy
sical conceptions about the 'inalienable realism' of the camera.
Clearly there is some strain between the components of the
argument. Bazin, an incurable metaphysician, wants to talk of
16. Andre Bazin, What is Cinema? op. cit., p. 108.
Frau im mond

the nature of the medium, its inalienable realism. But his wish
to allow 'some fantastic thing' to appear on the screen leads
him to the position that something is real if we are fooled into
thinking it so. And this depends on a range of factors which
have nothing to do with the 'nature' of the medium. Predic
tably, Bazin expresses unhappiness with the 'plausible reality'
explanation. It is over-simplified; insufficiently 'subtle'. Al
though the failure (to Bazin) of the stagey and implausible
decors of The Cabinet of Dr Caligari and the rest of German
Expressionism lend support to the thesis, there are, he feels,
more basic explanations. Like Kracauer, he is unhappy faced with
relativism. To make aesthetic judgements on the basis of'plausi
bility' raises the familiar bogey: plausible to whom? We are
returned (and why not?) to 'I think this is plausiblethough you do
not'. And faced with this, theorists ofthe persuasion of Bazin and
Kracauer produce the same old response. Back to the funda-
mental-nature-of-the-medium argument. Again the comparison
between Bazin's 'pure' and 'impure' positions is instructive.
Thephotographic image is the object itself, the object freed from the
conditions of time and space that govern it... The aesthetic qualities
of photography are to be sought in its power to lay bare the realities.
It is not for me to separate off, in the complex fabric of the objective
world, here a reflection on a damp sidewalk, there the gesture of a
child. Only the impassive lens, stripping its object of all those ways
of seeing it, those piled up preconceptions, that spiritual dust and
grime with which my eyes have covered it, is able to present it inall
its virginal purity to my attention
Apart from its extravagant romanticism the most interesting
characteristic of this passage is its stress on the identity be
tween image and object. Photography impassively reveals the
realities of the 'objective' world; it is this reality by its very
nature. Compare this once more with 'Theatre and Cinema'.
We are prepared to admit that the screen opens on an artificial world
provided there exists a common denominator between the cinemato
graphic image and the world we live in ... We may say, in fact,...
that 'the cinematographic image can be emptiedof all reality saveone
- the reality of space'.'*
'Identity' has become 'common denominator'; 'objective world'
has become 'artificial world'. Initially, the fundamental nature
of the medium revolves round the absolutes of 'identity' and the
'objective' world. But by 1951 the absolutes have been some
what mitigated. What Bazin is then concerned to discover are
the characteristics shared by the 'artificial' world of the film
and the world around us. And the most basic of these, he
argues, is our normal conception of space. The natural distribu
tion of objects in their spatial context. It is this factor which is
then crucial in determining cinematic reality.
17. Andre Bazin, What is Cinema? op. cit., pp. 14-15, my italics.
18. p. 108, my italics.
Rossellini: Paisd

These are the basic components of Bazin's two conceptions

of realism. Purist realism, at its limit, is the total documentary.
It is the neutral, impassive, non-human revealing of an objec
tive world. There is an absolute minimum of interference in the
sacrosanct identity of the cinematic image and the object it
presents. It is the 'honest' world of Vertov's kino-pravda, of
Nanook of the North, and, above all, the neo-realism of Paisd.
The dominant aesthetic criterion lies in a film's faithfulness to
outside natural reality. Spatial realism, however, derives from
a rather different set of assumptions. The objective world, the
identity of image and object, no longer figure. Here re^ism
stems from faithfulness to the basic common denominator, the
spatially natural distribution of objects. In 'William Wyler ou
lejanseniste de la mise en scene'" we find that Wyler - hardly
19. Andre Bazin, Qu'est-ce que le Cinema? I. Ontologie et Langage, op.
cit., pp. 149-73.

Spatial realism: Wyler's The Best Years of Our Lives ^

i g
 IpKW^
Vfr'-'/iV^'Mv ':.' i '
<'yi'ff?':xlM ' •

&ili!^liiM
The wedding in The Best Years of Our Lives

a 'pure' realist - qualifies on the basis of the spatial realism of

deep-focus photography. And in his famous essay on realism
and the Italian post-war film Bazin develops the contrasting
realisms of Welles' Citizen Kane and Rossellini's Paisd. His
hope —partly realized in La Terra Trema —was that the two
would join.
The problem of these two criteria of realism, in so far as we
are concerned with Bazin's thought as a systematic aesthetic, is
that they are not easily reconciled with one another. Certainly
we can say that a good film will be a film which reveals natural
reality in a spatially realistic manner. Someone else may then
disagree. But the whole tenor of Bazin's approach is to Justify
his aesthetic judgements by reference to the fundamental nature
20. Andre Bazin, Qu'est-ce que le Cinema? IV. Une esthetique de la
Realite: Le neo-realisme, op. cit., pp. 9-37.
 of the medium. The difficulty is that he has two fundamental
natures, two different views of the same thing. One says image
and object are identical; the other that they share a common
denominator. One says that film lays bare the objective world;
the other an artificial world. If we are really dealing with
fundamentals then only one of them can hold. If we are not
dealing with fundamentals then they cannot justify any par
ticular type of films as fulfilling the natural potentialities of the
medium. Thus, given the misdirected form of argument charac
teristic of this approach to aesthetics, pure realism and spatial
realism cannot both be justified. Like Kracauer, Bazin seems to
be unavoidably pushed, against his will, to the extremes. In con
sequence his practical aesthetic decisions display peculiarities.
Were it not for the subtlety of his critical writings, they would,
no doubt, look as obviously strange as do Kracauer's.
This can be seen in Bazin's rendering of the realist tradition.
Fairly naturally we find him defending Lumiere against the
incursions of Melies, though never in crude terms: 'Melies et
son Foyage dans la lune n'est pas venu contredire Lumiere et
son Entree du train en gare de la Qotat. L'un est inconcevable
sans I'autre.'^' In particular, one suspects, Melies is incon
ceivable without Lumiere. Continuing on the basis of purist
realism the peak of the silent era sees a contrast between
the expressionist 'heresy' and the realism of Potemkin. The
expressionists are so obviously staged, while the Russians are
drawing on the natural world. For this, at least, Eisenstein
can be hero; Dr Caligari is ultimate villain.
But if we now draw in the spatial realism criterion the
picture begins to alter. Bazin makes the point in contrasting
three products of the German silent era. On the one side
Caligari, and Lang's Die Nibelungen; on the other, Murnau's
Nosferatu. Presumably, all three must fail on pure realism
grounds - the vampire of Nosferatu as much as the somnam-
21. Andre Bazin, Qu'est-ce que le Cinema? I. Ontologie et Langage, op.
cit., p. 27.
bulist of Caligari and the legendary figures of Die Nibelungen.
But Nosferatu, Bazin argues, survives the others (personally, I
find them all pretty bad) because it is played out against natural
settings, against a world with a common spatial denominator to
share with our own. The distorted sets of Caligari and the
artificial forest of Die Nibelungen disobey this basic aesthetic
rule. (One wonders how much of this is a spin-off from the
'politique des auteurs'. Bazin, as a great admirer of the later
Murnau of The Last Laugh and Sunrise, may transfer this
admiration across the board. Nosferatu and the slightly later
Faust - also played against natural settings if memory serves -
seem in the end to be just as much failures as the rest of the
German films of the early period.) But in any comparison of
the Russian and German silent cinema Nosferatu, like the rest
of German 'expressionism', would surely come out the loser on
grounds of pure realism. There is clearly more than one
criterion in play when Bazin lists Stroheim, Flaherty, Murnau
and Dreyer as the great carriers of the realist tradition in the
silent era.
A consequence of Bazin's switch to spatial realism is that
montage becomes the arch-criminal. In the 1930s, Renoir is
the sole representative of realism in the cinema:
He alone in his searchings as a director prior to La regie du jeu
forced himself to look back beyond the resources provided by mon
tage and so uncovered the secret of a film form that would permit
everything to be said without chopping the world up into little
fragments, that would reveal the hidden meanings in people and
things without disturbing the unity natural to them.^^
Montage, Bazin now argues, is the 'anticinematic process par
excellence''-, proper cinema is to be seen in 'straightforward
photographic respect for the unity of space.'Spatial realism
thus encompasses his view of both the plastic and language
elements of the medium. Expressionism destroys natural spatial
unity by bending it; montage by fragmenting it. Unnatural
form becomes the cinematic original sin. What now of
Battleship Potemkini
Finally, Bazin sees both forms of realism in the develop
ments of the 1940s. The pure realism of Rossellini and the
spatial realism of Welles., But to choose Welles is to enter deep
water. Certainly Citizen Kane preserves the unity of space
through Toland's deep-focus photography. Certainly the cuts
are minimized by use of dissolves and joins across the sound
track. But Welles is, nevertheless, the true inheritor of expres
sionism (as The Trial, Touch of Evil,Macbeth, and by surmise.
The Third Man, bear witness), the specialist in the distortion
by camera angle, the mysterious shadows once painted but now
created through lighting, the grotesque, and the baroque.
22. Andre Bazin, What is Cinema?, op. cit., p. 38.
23. ibid., p. 46.

Inheritor of expressionism:Orson Welles in Touch ofEvil •

ii
The Third Man

Montage and decor are not the only ways of destroying the
visual unity of space; almost anything does. Why should they, in
particular, be singled out? The immense camera movements of
Touch ofEvil break up the world just as much as montage, and,
indeed, serve much the same function in terms of the tempo of
the film. If Bazin is really serious about spatial realism a lot
more than just montage will have to go. In the end, given that
he champions the neo-realists, he must be returned to the
argument for pure realism, to the romantic naturalist aesthetic.
Peter Wollen puts it well: 'Of Bicycle Thieves Bazin wrote that
it was the first example of pure cinema. No more actors, no
more plot, no more mise en scene-, the perfect aesthetic illusion
of reality. In fact, no more cinema.'^'*
24. Peter Wollen, Signs and Meaning in the Cinema, Seeker and
Warburg/British Film Institute, London, 1973, p. 131.
No More Cinema
If I have spent these many pages attacking the realist aesthetic
it is not —as it might have been —because I wish to support a
contradictory aestheticism. That debate should be certified
dead. Too much time has already been wasted in its by-ways.
Both those who take Melies and those who take Lumiere as
exemplifying the 'great tradition' are guilty of the same error;
an error which Kracauer and Bazin have in common. Refusing
the fence of relativism they gallop blindly into the ditch of
essentialism. Unwilling to admit the rightful subjectivity of our
aesthetic judgements, they dodge the consequent hail of non-
problems by recourse to a non-argument. The fundamental
character of film, they say, is naturally such-and-such, so
any film that fulfils this nature is therefore good. This, hope
fully but wrongly, is independent of subjective judgement.
Ultimately, both Bazin and Kracauer want a cinema and
an aesthetic from which human interference is absent. An
immaculate conception!
The tradition is a well-developed one. It reflects above all a
romantic faith in nature. Art must passively reveal the natural
world. And this faith is very much connected with the equally
nineteenth-century conviction that there is some great objective
world out there at which we may point our 'artistic' sensi
bilities. But romantic aesthetics and positivism are not quite the
powers that they once were. If we are tempted to reinstate them
we would do well to look at Bazin and Kracauer to see where
they lead. They lead, as Wollen suggests, to 'no more cinema'.
To the paradox that the fundamental essence of film is to
destroy everything which distinguishes it as film. And yet
Bazin and Kracauer obviously dearly loved a number of films
well outside the ambit of their respective pure aesthetics. One
can imagine a fitting epitaph: 'Here lie two theories. Their
cinematic eyes were bigger than their aesthetic stomachs, and
they perished from indigestion.'
5: Critical Method: Auteur and
Genre

As Italian neo-realism declined so also did the importance of

the realist aesthetic. In 1960 Kracauer's book saw the last
heavy weaponry in the 'philosophical' branch of the tradition.
In the related but less complex world of pragmatic realism the
moral impetus initiated by John Grierson worked itself out in
the 'Commitment Debate' of the late 1950s and early 1960s.
This is not to suggest that there are not still skirmishes on
either front; the realism issue is far too well installed to simply
disappear. But there has been a definite shift away. In some
contexts this has led to an increased interest in form, a move
toward Eisenstein's area of interest if not toward his termin
ology. In other contexts the aesthetic question has simply been
dropped and the preponderance of effort devoted to an attempt
to develop methods of interpretative analysis. Not unnaturally
these changes have been related to developments in the cinema
itself. In the former case the rise of a 'new' Continental cinema
at the end of the 1950s; in the latter an increasing interest in
the hitherto neglected body of the American sound film.
It is impossible to assess the magnitude of these changes, but
it is very clear that the rise in critical estimation of the
American film is of vital importance. Not that all critics are
now obsessed with American movies. Far from it. The next
Sight and Sound decade poll will no doubt show the 'intellec
tual' Continental cinema happily ensconced. But many more
 Budd Boetticher's The Tall T

people are now prepared to look seriously at a Don Siegel or a

Budd Boetticher, with directors like Howard Hawks becoming
a good bet for majority respectability where previously only
the names of Hitchcock and Ford were whispered —and the
latter pretty softly at that! In Sarris' suggestive terms the
American cinema has ceased to be the sole domain of the
'forest' critics; there is more inspection of the individual trees.'
However, it is not the shift itself which is the centre of my
attention. The focus here is on what such a shift implies in the
realm of film theory. If the realist aesthetic systems and the
text-book grammars have gone, whathas filled the vacuum they
have left? There being no single monumental work on which to
base analysis, what can we infer about the theoretical assump
tions of much contemporary practice? In short, where, if any
where, are we headed?
1. Andrew Sarris, The American Cinema, Button, New York, 1968.

John Ford: TheSearchers ^

I
Preminger: Laura

The easiest point of entry to such a discussion, lacking a

systematized perspective, is through the critical 'language
thrown up by contemporary practice. In particular, through
two terms: auteur and genre. The former has been associated
with an amazing variety of positions ranging from the obtusely
fascinating to the just plain silly. Some lead to fairly clear
aesthetic Judgements; others have been more concerned with
sympathetic descriptive analysis. Either way the historical root
of the term and certain of the assumptions involved are shared.
The second term, genre, has had a fairly extensive history. For
some years it was the principal guideline bywhich theAmerican
'forest' was mapped. Where Europe had an allegedly free and
creative director's cinema, the American, in contrast, was at
best seen as a cinema of popular genre. Now, however, genre
has become one of the hinges on which a revived interest in
 'i. : ' -...aaa
Walsh: White Heat

cinematic language hangs, although it also raises more general

questions. In what follows I shall try to partially demonstrate
where auteur and genre may be leading.

Auteur
The most misleading development in contemporary English
language writings on film lies in the joining of the two terms
'auteur' and 'theory'. It is something called the auteur theory
which has provided the touchstone of many a violent dispute.
Yet the direct ancestor of the auteur usage is to be found in the
'politique des auteurs' of Cahiers du Cinema, in the now
famous criticism of Truffaut and his colleagues. How 'policy',
the most obvious translation of 'politique', became 'theory' is a
tributary in the history of ideas which need not be dealt with
here. Sufficient to note that when the Cahiers group said
'policy' they meant 'policy'. Their use of auteur was exactly
that: a polemical position marking their views off from the
orthodox tradition in French criticism and, ultimately, when
they started making films, from the rest of the French cinema.
In the England of this period (the early and mid-1950s) op
position to the traditional approach was marshalled under the
banners of realism and commitment. Like Grierson before
them, these critics mistrusted the commercial American
cinema. But in France, fathered though not controlled by Bazin,
Cahiers used a partisan support of certain American directors
against the 'serious' Continental cinema. Which directors
exactly varied from critic to critic and group to group, but
names like Hitchcock, Ford, Hawks, Ray, Losey, Preminger,
and Walsh recurred. It was this polemical and exclusive sup
port for these American figures that was characteristic of the
'politique des auteurs'.
There were thus two notions central to the use of auteur
from the very beginning. First, the old idea that the director
was the true creator of the film. However controversial this
may have been in the past (Spottiswoode dedicated his 1935
book to 'the future of the director's cinema')^ it has surely now
passed forever into the realms of acceptability. The old
arguments, often tied to the attempt to 'prove' that film was
indeed the seventh art, have by and large been successfully
arbitrated. While no one would deny the collective nature of
film production, the crucial importance of the director's con
ception is part of the orthodox canon. The Antonionis, the
Bergmans, the Godards, and the Fellinis, are accepted as, at the
very least, the creative integrators of the disparate elements of
film. If auteur were simply this then it would have ceased to be
controversial years ago. But the second notion involved in the
Cahiers usage led in a breakaway direction. Applying the
notion of director as auteur to the maligned commercial
2. Raymond Spottiswoode, A Grammar of Film, Faber and Faber,
London, 1955.
Hollywood cinema raised a whole new series of bogeymen. In
Europe the director was thought to be relatively free from the
commercial pressures of Hollywood. It was this 'freedom'
which allowed him to be an auteur. But for years past conven
tional wisdom had seen Hollywood as a collection of variously
qualified craftsmen turning out variously competent films. Of
course, it was never entirely clear that the contrast between the
American and European situation was so great; folklore, how
ever, could always fall back on the great artists ruined by the
commercial citadel: Stroheim, Murnau, even Eisenstein. What
the Cahiers critics did was to find auteurs where none had been
dreamt of before. Directors like Welles, Hitchcock, and to a
lesser extent Ford, had always been accorded some admira
tion. The fate of the emigres had been bemoaned: Lang, the
paradigm case, was frequently and wrongly claimed to have
declined in Hollywood. But until Cahiers the rest were simply
a part of the commercial forest. ART was to be found else
where, by definition.
It was this singling out of American directors which was
first taken up in the English and American critical context. In
America Andrew Sarris provided a focus in the pages of Film
Culture-, in Britain the group of critics writing in Movie for
mulated their own set of American auteurs. In both contexts, as
indeed in France, there were inevitable extravagances. In em
ploying the notion of auteur as a basis for evaluating films there
was always the open invitation to elevate the worst films of an
auteur over the best films of another director as a matter of
course. Because an auteur made the film it must be good. The
reductio ad absurdum of this position is that it is not necessary
to actually see the films, sufficient only to know who directed
them. It becomes self-evident that bad Hitchock {Topaz) is
better than good Rossen {The Hustler or Lilith); bad Hawks
{Hatari) is better than good Zinnemann {High Noon)-, bad
Preminger {Exodus) is better than good Lumet {TheHill). And
all of them are better than the 'respectable' European directors.
Ironically enough, it is probably a good measure of the dilet
tantism of traditional criticism that the Anglo-American auteur
critics veered so close to these absurdities. For such extreme
positions are only really intelligible in terms of the polemical
needs of the situations in which they arose. This was guerrilla
warfare against an apparently safely established enemy. It was
no time for sweet reason.
Hence 'politique des auteurs' led to the formulation of lists,
of slogans. The 'Ten Best' ideas were no longer parlour games
but declarations of position. The guilty party who added the
notion 'theory' to this loosely formulated policy appears to
have been Andrew Sarris. In 'Notes on the Auteur Theory in
1962'^ he provided a rather highly coloured suit of clothes for
the polemically naked emperor, and, in addition, an easy open
ing for attack from the 'traditionalists'. It is only necessary to
demonstrate the peculiarities of Sarris' article - hardly difficult
- and we can forget about auteur forever! The invitation was
avidly taken up by that '... lady critic with a lively sense of
outrage', Pauline Kael, who, carried away on a rising wave of
sarcasm, pushed herself into some equally strange corners."*
She made the still common mistake of throwing out all the
auteur babies with Sarris' admittedly murky bathwater.
Between his theoretical premises and her vituperation most of
the point was lost. Sarris, no doubt moved by the salutary
experience, made some amends next time round. In a much less
extravagant piece he admitted that'... the auteur theory is not
so much a theory as an attitude,' ^ which is, of course, exactly
what it had been up to the Sarris-Kael exchange. Were it not
that the rumpus in question is still invoked as the nail in the
coffin of the 'auteur theory' (Roy Armes' almost hysterical
3. Andrew Sarris, 'Notes on the Auteur Theory in 1962', Film Culture,
27, 1962-3.
4. Pauline Kael, 'Circles and Squares, Joys and Sarris', I Lost it at the
Movies, Jonathan Cape, London, 1966.
5. Andrew Sarris, The American Cinema, op. cit., p. 30

DonSiegel: Invasion ofthe Body Snatchers ^

m

•*%£

X ";.'• ^ SK -• <•.. ;-V

^r - ,t-- ^

j' '.^. ^i- ' j*''. ^!§1;-i!!A*,'^.''3

review of Wollen's Signs and Meaning in the Cinema is a case
in point),® it would barely be worth a mention. The fact that it
lingers on is a hangover from earlier and more poleniical days.
With this debate, if such it can be called, passed the 'auteur
theory'. In its origins the idea of auteur served to isolate the
preferences of various groups of critics. It served as a sub
stitute for an aesthetic, a newly developed criterion for making
judgements of value. In its pristine form it was obviously
unsatisfactory and, pressured by the critical situation in which
they found themselves, the auteur groups rapidly displayed a
tendency to descend into 'aesthetic cults of personality'. Not
that this was entirely fruitless. As I have suggested, their great
contribution lay in forcing critical opinion to sit up and take
notice of the American cinema, hitherto concealed in the fogs
of aesthetic obscurity. It is very notable that such established
Journals as Sight and Sound, a decade ago only tossing the
occasional snide brickbat in the direction of the auteur critics
(the 'Cahiers menagerie' is a phrase that sticks in the mind),
now employ the rationale almost as a matter of course.
American movies are here to stay, and the genuine polemical
days of the auteur aesthetic have probably run their course.
There still remains, however, a considerable 'spin-off', much
of which awaits development. The auteur position can also be
employed as a principle of analysis instead of as a source of
critical evaluation. It is the critical method to which auteur
leads that is potentially most rewarding, a point rather missed
by the attackers. Pauline Kael:
It's obvious that a director like Don Siegel or Phil Karlson does a
better job with what he's got to work with than Peter Glenville, but
that doesn't mean that there's any pressing need to go to see every
tawdry little gangster picture Siegel or Karlson directs; and perhaps
if they tackled more difficult subjects they wouldn't do a better job
than Glenville.^
6. Roy Armes, 'A Polemic', Screen, 10, 6, 1969.
7. Pauline Kael, op. ait., p. 313.
 it employs the
the snobbery
rst place, and,
en it shows up
;h Karlson but
e would never
tchers. Riot in
an are hardly
3 draw people
?hers and also
tion. As Siegel
s becoming an
and Madi2an

8. Interview with Don Siegel, Positif, no. 71, Sept. 1965.

offer rather 'difficult' explorations of certain characters within
their 'gangster' framework. How does Miss Kael know they are
tawdry anyway, if she feels no pressing need to
see them? This is the mirror image of the most absurd auteur
cases: these films must be tawdry because Siegel or Karlson
made them. There is some lack in humility here: before dis
missing a director's work it is at least reasonable to look at it.
Auteur used as a method enshrines this from the start. The
point is that when we have seen every 'tawdry little gangster
picture' they might really turn out to be quite good, precisely
because of the insights afforded by looking at the range of a
director's work. Whatever else may be said in its favour,
through such a process a director as good as Siegel can be
rescued from Miss Kael's dustbin.
Alan Lovell puts it well:
By the 'auteur' principle I understand a descriptive method which
seeks to establish, not whether a director is a great director, but what
the basic structure of a director's work is. The assumption behind
this principle is that any director creates his films on the basis of a
central structure and that all his films can be seen as variations or
developments of it.'
The assumption that 'any' director creates on the basis of such
a structure is not, however, essential to the working of the
principle. It is only necessary to operate 'as if' such an assump
tion held. If no structure is forthcoming from the analysis then
it may be that a director fails to impose his personality on his
materials. But we can only discover if this is the case through
the operation of the principle in the first place. Hence the
'assumption' is a working hypothesis through which we can
make detailed analyses of a director's work. The real assump
tion of the process is that we can leam more about a director's
films by considering them in relation to one another.
9. Alan Lovell, 'Robin Wood —A Dissenting View', Screen, 10, 2, 1969,
pp. 47-8.
 There is an obvious way in which this method could be used
as a basis for aesthetic judgement. In so far as a director's work
does display a central structure then it is good. Equally
obviously it would be as well to avoid such a simplistic
application. It is a hazardous trail leading back to the
extravagances of the earlier auteur aesthetics. The more in
teresting consequence is methodological rather than aesthetic. If
the aim is to elaborate the central structure of meaning in a
director's work certain anal3d;ical facilities are necessary to
achieve this aim. The auteur principle directs our attention to
groups of films having in common one thing - the director. It
asks us to isolate his conception of the world as presented in
the films, and to do so in some considerable detail. But it does
not provide us with the tools necessary for such an analysis.
We have a prescription for the direction of our critical action
but we still face detailed problems of interpretation. The auteur
principle is thus a sort of 'pre-theory', a methodological
instruction. Whether it is invoked in the analysis of themes, of
common stylistic characteristics, or, as Sarris would wish it, in
our accounts of film history, it pulls our attention back to film
models. For ultimately the sorts of questions suggested by the
auteur principle can only "be answered through detailed and
systematic knowledge of the workings of film. To look at films
as the work of an auteur involves close textual analysis rather
than brief critical comment. Unfortunately, we are still not
entirely sure of the language in which the text is written. Auteur
directs our attention back to these concerns.

Genre
Auteur at least originated in film criticism in the recent past;
genre had a lengthy pedigree in literary criticism long before
the advent of the cinema. Hence the meaning and uses of the
10. A longerand slightly different versionof this sectionwas publishedas
'Genre: Theory and Mispractice in Film Criticism', Screen, 11, 6,
1970.
The Man Who Shot Liberty Valance

term vary considerably and it is very difficult to identify even a

tenuous school of thought on the subject. For years it provided
a crudely useful way of delineating the American cinema. The
literature abounds with references to the 'Western', the
'Gangster' movie, or the 'Horror' film, all of which are loosely
thought of as genre. On occasions it becomes almost the end
point of the critical process to fit a film into such a category,
much as it once made a film 'intelligible' to fit it into, say, the
French 'nouvelle vague'. To call a film a 'Western' is thought of
as somehow saying something interesting or important about it.
To fit it into a class of films about which we presumably have
some general knowledge. To say a film is a 'Western' is im
mediately to say that it shares some indefinable 'X' with other
films we call 'Westerns'. In addition, it provides us with a body
of films to which our film can be usefully compared;
The Last Hurrah

sometimes, the only body of films. The most extreme, and

clearly ridiculous, application might be to argue that it is
necessarily more illuminating to compare, say. The Man Who
Shot Liberty Valance with a Roy Rogers short than with The
Last Hurrah. Not that the first comparison might not be
instructive; merely that it is not necessarily the case. Extreme
genre imperialism leads in this direction.
Now almost everyone uses terms like 'Western'; the neurotic
critic as much as the undisturbed cinemagoer. The difference,
and the source of difficulty, lies in the way the critic seeks to
use the term. What is normally a thumb-nail classification for
everyday purposes is now being asked to carry rather more
weight. The fact that there is a special term, genre, for these
categories suggests that the critic's conception of 'Western' is
more complex than is the case in everyday discourse; if not.
why the special term? But in quite what way critical usage is
more complex is not entirely clear. In some cases it involves
the idea that if a film is a 'Western' it somehow draws on a
tradition, in particular, on a set of conventions. That is,
'Westerns' have in common certain themes, certain typical
actions, certain characteristic mannerisms; to experience a
'Western' is to operate within this previously defined world.
Jim Kitses tries to isolate characteristics in this way, by
defining genre in terms of such attributes; '... a varied and
flexible structure, a thematically fertile and ambiguous world
of historical material shot through with archetypal elements
which are themselves even in flux.'" But other usages, such
as 'Horror' films, might also mean films displaying certain
themes, actions, and so on, or, just as often, films that have
in common the intention to horrify. Instead of defining the
genre by attributes it is defined by intentions. Likewise
with the distinction between 'Gangster' movies and
'Thrillers'.
Both these uses display serious problems. The second, and
for all practical purposes least important, suffers from the
notorious difficulties of isolating intentions. In the first and
more common case the special genre term is frequently entirely
redundant. Imagine a definition of a 'Western' as a film set in
Western America between 1860 and 1900 and involving as its
central theme the contrast between Garden and Desert. Any
film fulfilling these requirements is a Western, and a Western is
only a film fulfilling these requirements. By multiplying such
categories it is possible to divide all films into groups, though
not necessarily mutually exclusive groups. The usefulness of
this (and classification can only be justified by its use) depends
on what it is meant to achieve. But what is certain is that just as
the critic determines the criteria on which the classification is
based, so he also determines the name given to the resultant
11. Jim Kitses, Horizons West, Thames and Hudson/British Film
Institute, 1970, p. 19.
groups of films. Our group might just as well be called 'type
1482/9a' as 'Westerns'.
Evidently there are areas in which such individually defined
categories might be of some use. A sort of bibliographic classifi
cation of the history of film, for instance, or even an abstract
exploration of the cyclical recurrence of certain themes. The
films would be simply defined in terms of the presence or
absence of the themes in question. But this is not the way in
which the term is usually employed. On the contrary, most
writers tend to assume that there is some body of films we can
safely call the 'Western' and then move on to the real work —
the analysis of the crucial characteristics of the already recog
nized genre. Hence Kitses' set of thematic antinomies and four
sorts of genre conventions. Or Bazin's distinction between
classic and 'sur-western' assuming, as it does, that there is some
independently established essence of the Western which is
distilled into Stagecoach.^^ These writers, and almost all
writers using the term genre, are caught in a dilemma. They are
defining a 'Western' on the basis of analysing a body of films
which cannot possibly be said to be 'Westerns' until after the
analysis. If Kitses' themes and conventions are the defining
characteristic of the 'Western' then this is the previously
discussed case of arbitrary definition - the category becomes
redundant. But these themes and conventions are arrived at by
analysing films already distinguishedfrom otherfilms by virtue
of being 'Westerns'. To take a genre such as a 'Western',
analyse it, and list its principle characteristics, is to beg the
question that we must first isolate the body of films which are
'Westerns'. But they can only be isolated on the basis of the
'principal characteristics' which can only be discovered from
the films themselves after they have been isolated. That is, we
are caught in a circle which first requires that the films are
12. Andre Bazin, 'Evolution du Western', Cahiers du Cinema, December,
1955, reprinted in Qu'est-ce que le Cinema? III. Cinema et
Sociologie, Editions du Cerf, Paris, 1961.
f

yi
isolated, for which purposes a criterion is necessary, but the
criterion is, in turn, meant to emerge from the empirically
established common characteristics of the films. This 'empiricist
dilemma' has two solutions. One is to classify films ac
cording to a priori chosen criteria depending on the critical
purpose. This leads back to the earlier position in which the
special genre term is redundant. The second is to lean on a
common cultural consensus as to what constitutes a 'Western',
and then go on to analyse it in detail.
This latter is clearly the root of most uses of genre. It is this
usage that leads to, for example, the notion of conventions in a
genre. The 'Western', it is said, has certain crucial established
conventions - ritualistic gun-fights, black/white clothing corre
sponding to good/bad distinctions, revenge themes, certain
patterns of clothing, typed villains, and many, many more. The
best evidence for the widespread recognition of these conven
tions is to be found in those films which pointedly set out to
invoke them. Shane, for example, plays very much on the stereo
typed imagery contrasting the stooping, black-clad, sallow,
be-gloved Palance with the tall (by dint of careful camera
angles), straight, white buckskinned, fair, white-horsed Ladd.
The power of this imagery is such that the sequence in which
Shane rides to the showdown elevates him to a classically
heroic posture. The point is reinforced by comparing Stevens'
visualization of his characters with the very different descrip
tions offered in Schaefer's novel. The film 'converts' the images
to its own conventional language. Other obvious examples are
provided by the series of Italian Westerns. The use of Lee Van
Cleef in leading roles depends very much on the image he has
come to occupy over two decades of bit-part villains. Actors in
the series - Van Cleef, Eastwood, Wallach, Jack Elam, Woody
Strode, Henry Fonda, Charles Bronson - perpetually verge on
self-parody. The most peculiar of the films —Once Upon a Time
in the West —is a fairy-tale collection of Western conventions,
verging on self-parody, and culminating in what must be the
most extended face-off ever filmed. Indeed, the most telling
suggestions as to the importance of conventions are to be found
in the gentle parodies of Cat Ballon, Support Your Local
Sherriff, and The Good Guys and the Bad Guys. Without clear,
shared conceptions of what is to be expected from a 'Western'
such humour is not possible. One of the best sequences in Cat
Ballon encapsulates the importance of the imagery, the
sequence in which Lee Marvin is changed from drunken wreck
to classic gunfighter. Starting very humorously with Marvin
struggling into a corset, the transformation not only alters him
but brings out a response in us as piece by piece the stereo
typed image appears.
In short, to talk about the 'Western' is (arbitrary definitions
apart) to appeal to a common set of meanings in our culture.
From a very early age most of us have built up a picture of a
'Western'. We feel that we know a 'Western' when we see one,
though the edges may be rather blurred. Thus in calling a film a
'Western' the critic is implying more than the simple statement,
'This film is a member of a class of films ("Westerns") having
in common x, y, z'. He is also suggesting that such a film would
be universally recognized as such in our culture. In other
words, the crucial factors which distinguished a genre are not
only characteristics inherent to the films themselves; they also
depend on the particular culture within which we are operating.
And unless there is world consensus on the subject (which is an
empirical question) there is no basis for assuming that a
'Western' will be conceived in the same way in every culture.
The way in which the genre term is applied can quite con
ceivably vary from case to case. Genre notions - except the
special case of arbitrary definition - are not critic's classifi
cations made for special purposes; they are sets of cultural
conventions. Genre is what we collectively believe it to be.
It is for precisely this reason that genre notions are so
potentially interesting. But more for the exploration of the
psychological and sociological interplay between film-maker.

Sergio Leone: Once Upon A Time In the West ^

. ^ I- .a

r
 m/1

fe
' 'Jd
Cat Ballou

film, and audience, than for the immediate purposes of film

criticism. (Given that it is not entirely possible to draw a clear
line between the two, this is really an argument for using a
concept in one area rather than another.) Until we have a clear,
if speculative, notion of the connotations of a genre class, it is
difficult to see how the critic, already besieged by imponder
ables, could usefully use the term, certainly not as a special
term at the root of his analysis. To use the concept in any
stronger sense it becomes necessary to establish clearly what
film-makers mean when they conceive themselves as making a
'Western'; what limits such a choice may impose on them; in
effect, what relationship exists between auteur and genre. But
specific answers to such questions must needs tap the concep
tions held by particular film-makers and industries. To
methodically analyse the way in which a film-maker utilizes
 mmm

a genre for his own purposes (at present a popular critical

pursuit) requires that we clearly establish the principal com
ponents of his conception of the genre. But this is not all. The
notion that someone utilizes a genre suggests something
about audience response. It implies that any given film works in
such-and-such a way because the audience has certain expec
tations of the genre. We can only meaningfully talk of, for
instance, an auteur breaking the rules of a genre if we know
what these rules are. And, of course, such rule-breaking has no
consequence unless the audience knows as well. Now as I have
suggested, Shane may well take on its almost 'epic' quality
because Stevens for the most part sticks to the rules. In a
similar way. Two Rode Together and Cheyenne Autumn are
slightly disconcerting because they break the rules, particularly
vis-a-vis the Indian/White Man relation. And, most obviously
 v;'./

Rule-breaking: the car in The Wild Bunch

in recent years, Peckinpah's 'Westerns' use such elements to

disturb the conventional Western universe. The much remarked
opening scene of Ride the High Country with its policeman and
motor cars; the cavalry charging the French Army in Major
Dundee-, the car in The Wild Bunch. Now you, the reader, may
agree that these are cases of deliberate rule-breaking, and such
agreement reflects that there is, in America and much of
Europe, some considerable consensus on what constitutes the
characteristic 'language' of a Western. But this could well be a
special case. To infer from it that all genreterms are thus easily
employed is hardly justified.
This is not to suggest that genre terms are totally useless. It
is merely that to employ them requires a much more
methodical understanding of the workings of film. And this in
turn requires that we specify a set of sociological and psycho-
The 'art-movie' genre: Antonioni's L'Avventura

logical context assumptions and construct explicit genre

models within them. If we imagine a general model of the work
ings of film language, genre directs our attention to sub
languages within it. Less centrally, however, the genre concept
is indispensable in more strictly social and psychological terms
as a way of formulating the interplay between culture,
audience, films, and film-makers. For example, there is a class
of films thought of by a relatively highly-educated, middle-class,
group of filmgoers as 'art-movies'. Now for present purposes
genre is a conception existing in the culture of any particular
group or society; it is not a way in which a critic classifies films
for methodological purposes, but the much looser way in which
an audience classifies its films. On this meaning of the term 'art-
movies' is a genre. If a culture includes such genre notions
then, over a period of time, and in a complicated way, certain
conventions become established as to what can be expected
from an 'art-movie' as compared to some other category. The
critics (the 'posh' critics in this case) are mediating factors in
such developments. But once such conventions develop they
can in turn affect a film-maker's conception of what he is doing.
Hence we get a commercial playing up of the 'art-movie'
category.
Let me take an impressionistic example bearing in mind that
much more extensive work would be needed to establish this in
anything more than an intuitive way. At the beginning of the
1960s in this country the general conception of an 'art-movie'
revolved around the films of a group of European directors.
Bergman was already established with, in particular, The
Seventh Seal and Wild Strawberries. The first year of the new
decade had seen Antonioni's L'Avventura, Resnais' Hiroshima
man Amour, and Fellini's La Dolce Vita. These four —though
perhaps Resnais less than the others - served to define the
'conventions' of the developing 'art-movies' genre. Deliberately
and obviously intellectual (there is nothing more deliberate
than the final scene of La Dolce Vita), with extremely visible
individual stylistic characteristics. Bergman's silhouettes,
puritan obsessiveness, and grunting Dark Age meals;
Antonioni's minimal dialogue, grey photography, and carefully
bleak compositions; and Fellini's self-indulgent surrealistic
imagery (partly in La Dolce Vita but much more clearly in Sy)
circumscribed what wasexpected of an 'art-movie'. Increasingly,
European films, whether 'deliberate' copies (a sub-Antonioni
example is Patroni Griffi's II Mare) or later films made by the
original directors, met the conventions which the earlier films
had established. Antonioni's II Deserto Rosso, Fellini's
Giulietta of the Spirits, Bergman's Winter Light and The
Silence, are almost stylistic parodies of their director's earlier
films. Giulietta of the Spirits becomes the ultimate in colour
supplement 'art-movies'; a combination of the earlier films and
the newly established conventions of the genre.
 iW/
Fellini's

This should serve to illustrate the way in which genre

notions might constructively be used in tapping the socio-
psychological dynamics of film, although it is not designed to
convince anyone of the particular case of 'art-movies'. To
properly establish such an argument would require detailed
research on the changing expectations of 'art-movie' audiences
(perhaps via analysis of the 'posh' critics), on the genreconcep
tions (and self-conceptions) held by individuals and groups in
various film industries, and on the films themselves. Now there
does not seem to me to be any crucial differences between the
most commonly employed genre term —the 'Western' —and the
'art-movie' category which I have been dealing with. They are
both conceptions held by certain groups about certain films.
Many of the theoretical problems about using genre terms
have, however, been overlooked in the case of the 'Western'. It
 c
J \ ' ^

From the 40s {My Darling Clementine)...

has become so much a part of our cultural patterning that film

criticism has tended to use it as if it were possible to assume
common agreement in all the respects on which research would
be necessary in the 'art-movie' case. It may be that there is such
common agreement on the 'Western'; but it does not follow
that this would be true of all genre categories. Anyway, it is not
at all clear that there is that much consensus on the 'Western'.
It seems likely that for many people the most Western of
Westerns (certainly the most popular if revivals are any
indicator) is John Sturges' The Magnificent Seven. On the other
hand, in the 1940s the same position might be filled by My
Darling Clementine, in the 1950s by High Noon. Conventions
change often for reasons entirely out of the control of film
makers and film critics.
In sum, then, genre terms seem best immediately employed
 I
18
-3i»

... to the 60s {The Magnificent Seven)

in the analysis of the relation between groups of films, the

cultures in which they are made, and the cultures in which they
are exhibited. That is, it is a term which can be usefully
employed in relation to a body of knowledge and theory about
the social and psychological context of film. Any assertion we
might make about the use a director makes of genre conven
tions —Peckinpah uses the contrast between our expectations
and actual images to reinforce the 'end of an era' element in
Ride the High Country and The Wild Bunch - assumes,
wrongly, the existence of this body of knowledge. To labour the
point, it assumes (1) we know what Peckinpah thinks; (2) we
know what the audience thinks (a) about the films in question,
and (b) about 'Westerns'; (3) Peckinpah knows the answer to
(2(b)) and it is the same as our answer, etc. Most uses of genre
effectively invent answers to such questions by implicitly claim-
ing to tap some archetypal characteristic of the genre, some
universal human response. This, as we have seen with
Eisenstein, depends on the particular context assumptions em
ployed, and on a more general notion of film language. To leap
in with genre immediately is to put the cart before the horse.

Critical Methodology
Both auteur and genre started life deeply implicated in sets of
aesthetic judgements. Auteur as part of a glorification of the
American cinema, a way of looking at trees; genre as a con
demnation of the American cinema, a way of merging trees into
forests. In the course of the 1960s, however, they have becom.e
increasingly divorced from this context. Much of their
evaluative content has been drained off, leading to their use in
an increasingly descriptive sense. Whatever the difficulties and
assumptions in their use, and I have tried to show that they are
not inconsiderable, they reflect a growing interest in detailed
and responsible critical interpretation. The stress on thematic
structure in the use of the auteur principle has led to an interest
in the application to film of methodologies from other
disciplines, in particular techniques borrowed from structural
linguists and anthropology. The attempt to find a construc
tive use for genre terms has led to an interest in techniques for
analysing the recurrent motifs in groups of films. Often this has
been more concerned with the visual iconography of the genre.
Either way, thcwight on film has shifted away from theoretical
interests toward discussion of the methodology of analysis.
There is now much more self-conscious interest in the
processes involved in analysing, comprehending, and evaluat
ing films.
It is much too early to say whether the methodologies under
consideration will prove fruitful. Although it seems obviously
potentially rewarding to look at the thematic structures implicit
in a director's work, the particular structuralist techniques
invoked are rarely satisfactory. To reduce the thematic con-
cerns of a group of films to a set of polarities may be adequate
in some contexts - consider, for instance, Peter Wollen on
Hawks or Alan Lovell on Siegel - but not in others. The
assumptions necessary in order to analyse all content in terms
of polar opposites are far from safely established. Structures do
not 'leap out' from the subject matter as one notable structur
alist has suggested; they are at least partly imposed by the
consciousness of the observer. Any schematizations produced in
this way must, therefore, be treated as hypotheses to be tested
against the material, not conclusively established truths. The
problems of detailed analysis still remain. The 'structural'
method is hardly a magic formula.
Still, such methods have shown some pay-off. The search for
otherwise unnoticed patterns, encouraged by the auteur prin
ciple and the crude application of structural techniques, is not
an inconsiderable achievement. It breaks down the regal
isolation of the film as basic unit of analysis. Similarly with the
use ofgenre, demanding, as it does, thata film or group offilms
be considered in a larger context. And in the end, auteur and
genre do not retreat into methodology as totally as they
initially seem to do. Although they serve to focus our attention
predominantly on problems of descriptive analysis, such
analysis in turn leads back to theoretical questions. Aesthetic
issues, becausethe greater our analysis the greater the invitation
to judge or criticize. Models of film, because there are no final
methodological answers to our problems of understanding. If
the auteur principle is to be pursued it is not sufficient to pick
out the clearest thematic oppositions; to do so risks losing that
which makes film specifically film. We also need to know how
film works, the means of expression at levels other than the
narrative surface, in a word, what language a film is speaking.
It remains to be seen whether semiology (the general science of
13. Peter Wollen, Signs and Meaning in the Cinema, Seeker and
Warburg/British Film Institute, London, 1969, pp. 81-94; Alan
Lovell, Don Siegel —American Cinema, op. cit.
signs) will provide us with a starter in this respect. It has not
started too well. But certainly such demands return us to the
interests initially developed by Eisenstein over thirty years ago.
If genre notions are also to be developed we will inevitably be
led to sociological and psychological theories of film, to
questions about the context within which the cinema is operat
ing. The form of such questions remains unclear but there is no
doubt that they will be asked. Perhaps Eisenstein's self-imposed
task of creating a unified theory of film, of understanding the
medium in which he was so involved, may yet be completed. If
we can escape the aesthetic disputes of the past, and some of the
more anti-intellectual prejudices of the present, film theory
might yet receive the attention that it merits.
6: Epilogue

In the cause of symmetry a book beginning with an introduc

tion should perhaps end with a conclusion. Rationalist aes
thetics, at least, might legitimate such a pattern. But in point of
fact it is very difficult to draw any clear and positive con
clusions here short of embarking on a further and rather more
extensive essay. The only genuine next step would seem to be
the development of a body of theories which would deal
adequately with the sorts of problems I have been discussing.
As a conclusion such a sentiment is merely another introduc
tion. This does not mean that my discussion here has been
entirely aimless; future trends in film theory will hardly emerge
from an intellectual vacuum. Indeed, the ultimate justification
for looking at works of the past must surely be in terms of
future utilities - in the widest sense of that much misused term.
Anything else trips dangerously close to scholasticism.
In the light of these diffuse reservations, then, my last few
pages are by way of epilogue. A concluding chapter but with
out conclusions. Almost every theme thus far considered has
been located specifically in relation to some particular theorist
or to some fairly easily recognized tradition. Here, then, I shall
try to isolate one or two of the crucial areas which in some
sense transcend the disputes and positions with which we have
been occupied. As I implied in Chapter One the theorists
represented in this book have not a little in common, a fact
somewhat obscured in their conventional counter-positioning.
Quite obviously the critical concept in this 'common heritage'
—though not always explicitly employed - is that oflanguage'.
This is clearest in Eisenstein and the so-called auteur theory
where notions variously associated with the idea of a language
are absolutely central. Somewhat more subtle, but equally
important, is the involvement of language concepts in the
classic discussions of realism as both movement and aesthetic.
Metz's development of the Bazin tradition in semiological
terms is a case in point. And in recent years there has been an
increasing willingness to borrow from the expanding discipline
of structural linguistics. I do not think that these various usages
have been quite as disparate as they might at first appear.
Substantively they have differed; but in their formal focus we
find a strikingly similar set of problems.
To round out my picture of this development, then, I shall
take a sketchy look at the growing crystallization of ideas
about film language. But in so doing I am conscious of by
passing a further cluster of themes which seem implicitly
involved in much contemporary discussion of film. The precise
form taken by these developments remains unclear, but the
questions with which they are concerned are fairly abstract.
What is the nature of theorizing about film? What is its pur
pose, its relation to other disciplines, and its relevance to
problems of evaluation? Interestingly enough, the problems
generated in creating a more disciplined analysis of film lan
guage increasingly draw attention to these more general issues.
We can no doubt expect more and more problems to be thrown
up by this area of cross-over. However, a number of these
questions seem to me better left until we have mastered less
intractable problems. Concern with this philosophic hinterland
—what is the nature of film, its art-ness, its film-ness? —seems
to me to be presently misplaced. In the long run, of course, it
is bound to be important. But in short-run practice these
questions are littered with pitfalls which are by no means easy
to avoid. Kracauer's failings are a blatantly obvious case of
such magnitude that it may be worth curtailing our ambition to
avoid such entanglements. For such reasons, among others, it
seems more acceptable to focus on 'language' and exclude some
of the more traditional areas of aesthetic discourse. Specific
strategy depends on what alternatives we have.

Language and Structure

As I have suggested, a loose idea of film language has almost
always been part and parcel of the apparatus of film criticism.
Anyone whose youthful interests went beyond simply reading
film reviews inevitably came across traditional ideas about the
'grammar of film'. The analogy was rarely precise. The 'gram
mar' of film was usually only a way of organizing discussion in
terms of the obvious components of the film process: shot,
sequence, cut, and the rest. Although these discussions were
not unhelpful they were only superficially theories about the
'language' of film. Only Eisenstein consistently delved deeper.
He was interested in the language of film in that he wished to
know how and why films communicated 'pathos'. What formal
characteristics of the medium produced the required response?
In fact he began to develop a poetics based largely on the idea
of rhythm, encapsulated in his well-known theory of montage.
But he did not approach the study of film language in general,
although he occasionally made the mistake of confusing his
partial rendering with the whole story. Although he was
undoubtedly conscious of the need for a general 'linguistics' of
film, and although he had pretensions to theoretical generality,
his analysis remained sadly incomplete.
This is hardly a damning criticism. What else was he to do?
Even now we have barely delved deeper than Eisenstein with all
the armoury of terms borrowed (and stolen) from semiology,
structuralism and linguistics. It is even arguable that this very
borrowing is only serving to conceal the true nature of our
problems beneath a bushel of ambiguities. Certainly these more
recent techniques must be used with extreme care lest they fool
us into considering the obvious an insight. It also seems helpful
—temporarily at least —to try to divorce this structural and
linguistic impetus from the critical context within which it has
mainly developed. Basically, from the so-called auteur theory.
Historically, of course, some such structural approach was
inevitable in the auteur tradition. If one takes as axiomatic the
notion that the individual author is the highest common factor
in a group of films, then the invitation is to discover the overall
structure (of sentiments and styles) which characterize the
whole group. Separated from its polemical context, the
methods involved in auteur analysis lead inevitably to
questions of structure, however weakly they may be put. In
effect, the patterned relations between themes, styles, and
motifs in a director's work are conceived as some kind of
reflection of his personal conception of the world, of the struc
ture of his sentiments. Usually an assumed homology of struc
ture between art and artist. And shorn of the term structure
there is obviously nothing very new here.
Given that the focus of such weakly 'structural' approaches
is on large groups of films, there is no outstanding reason why
the principal locus should always be the director. Provided the
analysis comes up with a common set of characteristics, any
further question as to the source of these characteristics
remains formally open. For example, it has been customary to
ascribe the shared characteristics of a group of 1940s' horror
films to their producer Val Lewton, their directors varying.
The Cat People, I Walked with a Zombie, The Leopard Man
by Jacques Tourneur; The Seventh Victim, Isle of the Dead hy
Mark Robson; The Curse of the Cat People, The Body
Snatcher by Robert Wise. The fact that there intuitively seems
to be a series of common concerns in these films has led to
some analysis of their common structure, and an additional
ascription of this communality to their producer. It never-

The Haunting (Robert Wise)

 ~ i PM
' im If
y -J--.

A'

l!i% »

; -»>
theless remains that Tourneur's later —and excellent - Night of
the Demon (also known as Curse of the Demon) and Wise's The
Haunting have not uncommon similarities with the earlier
Lewton films, besides a cinematic reference group in the notion
of a horror genre. Clearly the horror film, the Western, the
thriller —given the difficulties raised in Chapter Five - are all
in principle amenable to analysis of structure. The individual
artistic persona is not the only possible hypothetical 'explan
ation' of the existence of a communality of structures across a
range of films. A set of genre conventions may be equally
apposite.
But this almost purely decoratively terminological use of the
notion of structure says little or nothing about the problems of
application. This has always been the limitation of straight
forward genre or auteur applications. What are the categories
of thematic and/or stylistic structure? What class of units are
involved? How are they related? Any fully-fledged discussion
of film must of necessity deal with such questions, and as a
priority. But in the main the 'structuralist' impetus has not yet
led to refinement at this level. It has instead given rise to a
methodology of analysis based largely on dichotomization; to
the idea that culture, thought, or whatever, is innately struc
tured in terms of a series of dichotomies. This has been mainly
derived from a vulgarization of the work of Claude Levi-
Strauss, an anthropologist who has proved something of a
critical messiah. His analysis of the structure of myth in similar
terms has led the unwary to claim 'myth' status for the film
genre in modern society. And where application of 'structur
alism' has not directly involved such sociological absurdity, it
has invariably been simplified in terms of a descending series
of dichotomies. Through such techniques it is hoped to render
explicit the 'deep' structures of the films in question: what used
to be termed in more mundane and facetious manner, the
'hidden meanings'. But of course there is no a priori reason for
employing dichotomization, particularly when it is assumed —
as it frequently is - that these dichotomically related themes
are inherent in the films themselves independent of the obser
ver. Dichotomization can be at best only an epistemological
\veapon. Its claim to ontological universality is deeply prob
lematic.
In practice a great deal of criticism labelled 'structural' has
little to do with this reductio ad Levi-Strauss. It simply reflects
some degree of interest in the structure of relationships be
tween thematic clusters in, say, a director's work. And the
development of that focus was, as we have seen, largely
independent of contemporary ideas about 'structuralism'. The
act of analysing structure and structural/s/n may usefully be
kept separate. Apparently structuralist work, such as Peter
Wollen's analysis of Sam Fuller, is only superficially so; a
focus on auteur and a desire for method would be sufficient to
generate the approach.' Indeed, what is valuable about such
work is not that it is sufficiently similar to other disciplines to
claim some part of their intellectual legitimacy, but that it
persists in directing attention toward the notion of structure. It
demands that we explicitly look at the film(s) as a total config
uration rather than a discrete set of elements. But basically it
tells us little more. It does not, as it stands, seriously address
itself to the problem of film language, of how films in fact
communicate these various allegedly structured clusters. In this
context other associated inputs have proved more interesting, if
not yet more lucrative.
The model for this much more specified interest in film
language has been linguistics. Thus, there has been a suggestion
that it might be possible to create a cinematic equivalent to
Chomsky's transformational grammar. Such a theory would
enable us to generate all possible grammatical statements in
the 'language'. But this hope leans heavily on the possibility
of analogous development, a faith which may be misplaced.
1. See Peter Wollen, 'Notes towards a Structural Analysis of the Films of
Samuel Fuller', Cinema, 1, Dec. 1968.
Perhaps the non-cognitive element is so important in film
(along with most other arts) that linguistic strategies are too
cognitively oriented for constructive application. It is a little too
early to know, but even Bernstein's now familiar distinction
between restricted and elaborated codes has a suspiciously
cognitive bias. Applications, as opposed to programmatic
suggestions, have not been common. More often than not
'linguistic' inclinations have manifested themselves in relation
to semiology, the general science of signs. But again, we have
yet to see a developed semiological mode in relation to film.
Writers on the subject have often been basically more con
cerned to illumine the ontological nature of film as a stage in
developing an aesthetic position. Metz, in one guise or another,
develops and reiterates versions of Bazin's classic position.^
And while Wollen makes illuminating general use of Peirce's
distinction between iconic, indexical, and symbolic signs, the
traditional semiological focus on the symbolic or arbitrary sign
prevents us from adequately filling in the detail.^ And reference
to Barthes' works is demonstration of the stringent limitations
on semiological analysis of more complex communication
systems."
We are then relatively well provided with sources of'linguis
tic' sensitization, but little more. Excellent cases have been
made for focusing on the persistent structures running through
one or a series of films. By and large, however, this interest has
developed in relation to theme at the expense of style; needless
to say there is no reason to think of this as a necessary con-

See, among many others, Christian Metz, 'Le dire et le dit au cinema:
vers le declin d'un vraisemblable'. Communications, 11, 1968;
'Propositions methodologiques pour I'analyse du film'. Social Science
Information, VII, 4, 1968.
Peter Wollen, Signs and Meaning in the Cinema, Seeker & Warburg/
BFI, London, 1969, pp. 120fT.
Roland Barthes, Elements of Semiology and Writing Degree Zero, both
published by Jonathan Cape, London, 1967.
junction, though style is undoubtedly more difficult to handle.
But a clear prerequisite, if'structural' analysis is to be anything
more than a dressed up list of dichotomous distinctions, must
be the development of a detailed theory of film language.
Whether this will derive from a general science of signs or will
later be slotted into it remains open. Most likely both strategies
will be involved in some part. But at the moment all we really
have is an a priori faith derived partly from the respectable
overtones of terms like 'structuralism' and 'semiology'. As far
as film criticism is concerned both labels are still getting by on
credit. They constitute very general pointers in the direction of
'progress' but have thus far shown little real analytical power.
Naive structuralism especially might be argued to have had
predominantly negative consequences in reducing analysis to a
formula simplification of a few themes. And none of the major
contemporary approaches have come to terms with the par
ticular structured relations which were of so much interest to
Eisenstein: rhythm, tempo, and formal compositional contrast.
Undoubtedly there is promise but it remains largely unfulfilled.
We have only a few fragmentary clues as to our best directions
of development.

Theories of Film
It is a commonplace to suggest that all processes of comprehen
sion involve a framework of assumptions, propositions, and the
like. Attempts to evaluate the aesthetic importance of an object
as much as attempts to empirically investigate it are basically
processes of 'mapping' the object into some previously consti
tuted framework. Neither endeavour exists in a vacuum. Of
course, nothing follows as such from this observation: several
further premises are needed for that. But it does seem that
experience in other areas of inquiry suggests that there are
important payoffs to be derived from explicit discussion of
such frameworks. Their nature, their internal consistency, and
the manner of their relation to their subject. For facts do not
L'Annee Derniere a Marienbad

speak for themselves and aesthetic judgements are hardly self-

evident. If this is true, then a concern with theories of film is
not simply a decorative addendum to the real thing. On the
contrary, we must know about the theories lest we confuse a
necessarily partial theoretical account with 'absolute truth', or,
more familiarly, absolute canons of beauty. Explicit theorizing
is essential. Without it we simply dissolve our studies into a
series of variously recognized prejudices.
Given this general point there still remain many alternative
strategies within the rubric 'theory'. We have looked at some
of the more exemplary cases: Bazin, Kracauer, Grierson,
Eisenstein. But is it possible to single out any particular line of
theoretical development as presently most potentially reward
ing? It will be clear to the reader who has got this far that I am
not without belief in such a thread. Loosely conceived, film
'language' seems as good a label as any. The basic reasoning in
favour of such a focus is straightforward. Though our studies
may be directed at many things - the relation of film to social
change; the psychological appeal of genre; the bases for a set of
aesthetic criteria; and endless others - whatever the aim we
come up against one persistent problem: how to tap the mean
ing of the film(s). Not in the very specific sense of that term; not
in the 'What does L'Annie Derniere d Marienbad mean?'
idiom. Meaning in the sense of raising questions about the ways
in which language is used, the form of recurring patterns, the
incidence of unlikely absences, the contextual significance of
the film, and so on. Trying to make sense of the narrative
thread is only one element in the range of levels of meaning
involved. Whatever our specific interests - and as I have said
they may be many —the core problem lies in analysis of
'meaning'.
So, improving our understanding of the ways in which film
may be meaningful is a priority; any further analysis must
needs build on this foundation. This is patently obvious in areas
of contextual interest. Most commonly in sociology and psy
chology. Whatever linkage the social scientist is exploring
inevitably derives in some respect from an understanding of the
meaning of the films in question. Even rampant behaviourism
has some such covert point of reference. The intermediary
connecting human action and aesthetic object is always some
generalized notion of meaning. And elsewhere the importance
of meaning is not unclear. Aesthetic evaluation, for instance,
invariably has reference to particular patterns which are mean
ingful in certain aesthetically prescribed ways. It is quite
apparent - even from the limited materials discussed in this
book - that assumptions about the manner in which films are
meaningful (about 'language') are very basic indeed. Not a few
of the problems we have been discussing derive from question
able assumptions at this level. In this area we really need
renewed theoretical and analytical energies.
It is partly for these reasons that Eisenstein might be
reconsidered. In some areas at least he made a good start. Un
fortunately much of his theoretical contribution to the solution
of such problems has been sunk without trace in the back
waters of Realism. The image of Eisenstein as formalist has
dominated. In some ways this is deeply ironical. It does seem
that a number of the problems found in the theoretical develop
ments of the realism position derive from misunderstandings in
just the areas toward which Eisenstein points. Take the contor
tions discussed in Chapter Four. The core of the tangle, for
both Bazin and Kracauer, lies in the relation between particular
elements of cinematic language and conventions about the
'real'. Bazin appears to have been aware of this, though he fails
to push it through to any single conclusion. Instead he remains
ambivalent, in the end postulating two contradictory realist
'essences'. If Eisenstein's proper impetus had not been sub
sequently sidetracked the realist aestheticians might not have
found themselves in such a mess. They have much to answer
for.
Three epilogic conclusions then. First and foremost we
must theorize. Otherwise our knowledge will simply stultify at
its present primitive state. Secondly, in theorizing we would be
better advised to try to explore what is for a while instead of
arguing about what ought to be: models rather than aesthetics.
And lastly, models of film language must be a first priority.
With resources scarce this seems to be not just the best invest
ment, but ultimately the only available possibility. As we have
seen earlier in this chapter recent years have brought a number
of shifts in this direction. The combination of the classical
theories I have been considering here along with these later
inputs seems at least a promising beginning. Let us hope it will
lead to a promising end.
Bibliography

This bibliography includes all materials referred to directly in the text as

well as further relevant references.

Anderson, Lindsay, 'British Cinema: the Descending Spiral', Sequence, 7,

1949.
Anderson, Lindsay, 'Stand Up! Stand Up!', Sight and Sound, 1956.
Armes, Roy, 'A Polemic', Screen, 10, 6, 1969.
Amheim, Rudolf, Art and Visual Pereeption, Faber and Faber, London,
1956.
Arnheim, Rudolf, Film as Art, Faber and Faber, London, 1958.
Arnheim, Rudolf, Toward a Psychology ofArt, Faber and Faber, London,
1956. 2nd ed. 1966; Berkeley, University of California Press.
Balazs, Bela, Theory of the Film, Dennis Dobson, London, 1952.
Barthes, Roland, Elements of Semiology, Jonathan Cape, London, 1967.
Barthes, Roland, Mythologies, Jonathan Cape, London, 1972.
Barthes, Roland, Writing Degree Zero, Jonathan Cape, London, 1967.
Bazin, Andre, 'Evolution du Western', Cahiers du Cinema, December,
1955.
Bazin, Andre, Qu'est-ce que le Cinema? I. Ontologie et Langage, Editions
du Cerf, Paris, 1958.
Bazin, Andre, Qu'est-ce que le Cinema? II. Le Cinema et les autres Arts,
Editions du Cerf, Paris, 1959.
Bazin, Andre, Qu'est-ce que le Cinema? III. Cinema et Sociologie, Editions
du Cerf, Paris, 1961.
Bazin, Andre, Qu'est-ce que le Cinema? IV. line Esthetique de la Realite,
Editions du Cerf, Paris, 1962.
Bazin, Andre, What is Cinema?, University of California Press, Berkeley
and Los Angeles, 1967.
Bazin, Andre, What is Cinema?, Vol. 2, University of California Press,
Berkeley and Los Angeles, 1971.
Eisenstein, Sergei, Film Essays, Dennis Dobson, London, 1968.
Eisenstein, Sergei, Film Form, New York, Harcourt, Brace, 1949; Dennis
Dobson, London, 1951.
Eisenstein, Sergei, Notes of a Film Director, Lawrence & Wishart, London,
1959; rev. ed. New York, Dover Publications, 1970.
Eisenstein, Sergei, The Film Sense, Eaber & Faber, London, 1943; New
York, Meridian Books, 1957.
Hall, Stuart ; Whannel, Paddy, The Popular Arts, Hutchinson Educational,
London, 1964.
Hardy, Forsyth (ed.), Grierson on Documentary, Collins, London,
1946.
Hardy, Forsyth (ed.), Grierson on Documentary, Faber & Faber, London,
1966; New York, Praeger, 1971.
Huaco, George A., The Sociology of Film Art, Basic Books, New York,
1965.
Kael, Pauline, 'Is there a cure for Film Criticism? or: Some Unhappy
Thoughts on Siegfried Kracauer's Nature of Film: the Redemption of
Physical Reality', Sight and Sound, 31, 2, 1962.
Kael, Pauline, I Lost it at the Movies, Boston; Toronto; Little, Brown &
Co., 1965; Jonathan Cape, London, 1966.
Kitses, Jim, Horizons West, Thames & Hudson/British Film Institute,
London, 1969. (Cinema One).
Kracauer, Siegfried, 'National Types as Hollywood Presents Them', Public
Opinion Quarterly, 13, 1949.
Kracauer, Siegfried, From Caligari to Hitler: A Psychological History of
the German Film, Princeton University Press, Princeton, 1947.
Kracauer, Siegfried, Theory of Film: the Redemption of Physical Reality,
Oxford University Press, New York, 1965.
Lawson, John Howard, Film: the Creative Process, Hill & Wang, New
York, 1964.
Lindgren, Ernest, The Art of the Film, Allen & Unwin, London, 1948.
Lovell, Alan, 'Robin Wood —A dissenting View', Screen, 10, 2, 1969.
Lovell, Alan, "The Common Pursuit of True Judgment', Screen, 11, 4/5,
1970.
Lovell, Alan, Don Siegel: American Cinema, British Film Institute, London,
rev. ed. 1975.
MacCann, Richard Dyer (ed.). Film: a Montage of Theories, F. P. Dutton,
New York, 1966.
Metz, Christian, 'Le Cinema: Langue ou Langage', Communications, 4,
1964.
Metz, Christian, 'Le dire et le dit au cinema: vers le declin d'un vraisembl-
able', Communications, 11, 1968.
Metz, Christian, 'La Grande Syntagmatique du Film Narratif,
Communications, 8, 1966.
Metz, Christian, 'Propositions Methodologiques pour I'analyse du film'.
Social Science Information, VII, 4, 1968. In English in Screen, 114, 1/2,
1973.
Metz, Christian, Essais sur la signification au cinema. Editions Klincksieck,
Paris, Vol. 1, 1968 (English Trans. Film Language, OUP, 1974); Vol. 2.
Metz, Christian, Language and Cinema, Mouton, The Hague, 1974.
Nizhny, Vladimir, Lessons with Eisenstein, Allen & Unwin, London, 1962.
Pryluck, C., 'Structural analysis of Motion Pictures as a Symbol System',
Journalism Quarterly, 16, 4, 1968.
Pryluck, C., and Snow, R. E., 'Towards a Psycholinguistics of Cinema', A V
Communications Review, yi\, 1, 1967.
Pudovkin, Vsevolod, Film Technique and Film Acting, Vision Press,
London, 1953; Mayflower, 1958.
Sarris, Andrew, 'Notes on the Auteur Theory in 1962', Film Culture, 27,
1962-3.
Sarris, Andrew, The American Cinema, E. P. Button, New York, 1968.
Seton, Marie, Sergei M. Eisenstein, London, Bodley Head, 1952; Grove
Press, New York, 1960.
Spottiswoode, Raymond, A Grammar ofFilm, Faber & Faber, London, 2nd
ed. 1955; Berkeley, University of California Press, 1950.
Talbot, Daniel (ed.). Film: an Anthology, New York, Simon and Schuster,
1959; pbk. and shortened ed. California University Press, Berkeley,
1966.
Tudor, Andrew, 'Genre: Theory and Mispractice in Film Criticism',
Screen, 11, 6, 1970.
Wollen, Peter, 'Notes Towards a Structural Analysis of the Films of Samuel
Fuller', Cinema, 1, 1968.
Wollen, Peter, Signs and Meaning in the Cinema, Seeker & Warburg/BFI,
London, 1969. (Cinema One); 2nd ed. (revised), 1973.
Wollen, Peter (ed.). Working Papers on the Cinema: Sociology and
Semiology, British Film Institute, London, 1969.
Wood, Robin, 'Ghostly Paradigm and H.C.F.: An Answer to Alan Lovell',
Screen, 10, 3, 1969.
Wood, Robin, Hitchcock's Films, Zwemmer, London; New York, A. S.
Barnes 1965; 2nd ed. 1969.
Wood, Robin, Howard Hawks, Seeker & Warburg/BFI, London, 1968.
(Cinema One).
Acknowledgements

I would like to express my gratitude to a number of people. To

Peter Wollen with whose help the original idea of the book was
developed. To Alan Lovell, Terry Lovell, and Christopher
Williams, who were all kind enough to read the manuscript and
comment on it in detail. They were more helpful than they
realize. To Marion Haberhauer whose secretarial aid was im
peccable. To Min Tudor whose general help was invaluable.
This book is in memory of Phaedra and Myshkin.
Stills by courtesy of British Lion, CIC, Columbia,
Connoisseur, Contemporary, Gala, M-G-M, Paramount, Rank,
RKO, 20th Century-Fox, United Artists, Universal, Warner
Bros., and the Stills Library of the National Film Archive,
London.
 cinema Onelill

'This clear and 1 Jean-Luc Godard 16 Meiviiie on Meiviiie

straightforward account of by Richard Roud by Rui Nogueira
film theory Is Intended to
provide a concise and 2 Losey on Losey 17 Straub
critical guide to the edited by Tom Miine by Richard Roud
principal approaches of
film study. Andrew Tudor 4 How it Happened Here ISSirk onSirk
analyses the six schools or by Kevin Browniow by Jon Haiiiday
practices which have
contributed most towards the SBiiiy Wiider 19 Orson Weiies
still necessary production by Axei Madsen by Joseph McBride
of a systematic method of
thinking about the cinema. 9 Signs and Meaning 20 Underworid USA
He also tests the two in the Cinema by Coiin McArthur
theories which dominate by Peter Woiien
much of current film 21 Studies in Documentary
10 Buster Keaton by Aian Loveii and
criticism: the Idea of
by David Robinson Jim Hiiiier
Individual authorship of
films, and studies In terms
of genre and convention.
11 Pasoiini on Pasoiini 22 Vai Lewton: The Reaiity
by Oswaid Stack of Terror
This Is a valuable handbook
which will be much used by by Joei E. Siegei
12 Horizons West
serious students of the by Jim Kitses 23 Theories of Fiim
cinema.
by Andrew Tudor
13 Rouben Mamoulian
by Tom Miine 24 Fran9ois Truffaut
by Don Alien
14 Hoilywood Cameramen
by Charies Higham 25 Westerns
by Philip French
15 Samuei Fuiier
by Nichoias Garnham

36 09937 3 Seeker & Warburg/ £2.50 net