The Practical Difference Between Natural-Law Theory and Legal Positivism
The Practical Difference Between Natural-Law Theory and Legal Positivism
The Practical Difference Between Natural-Law Theory and Legal Positivism
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=oup. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to Oxford Journal
of Legal Studies.
http://www.jstor.org
THE PRACTICAL DIFFERENCE BETWEEN
NATURAL-LAW THEORY
AND LEGAL POSITIVISM
DERYCK BEYLEVELD AND ROGER BROWNSWORD*
*Deryck Beyleveld is Senior Lecturer in Criminology and Socio-Legal Studies at the University of
Sheffield. Roger Brownsword is Senior Lecturer in Law at the University of Sheffield. An earlier
draft of this paper was presented to a jurisprudence group at the University of Edinburgh. Our
thanks are due to those who participated in that group, especially Neil MacCormick, David Nelken
and Zen Bankowski. Our work on this paper has been facilitated by grants (Nos. 605 and 607) that
have been made available to us by the Research Fund Committee at the University of Sheffield. We
are grateful for the support of the Committee.
i In our article 'Law as a Moral Judgment vs Law as the Rules of the Powerful' 28 AmerJJuris 79
(I983), we outline our understanding of the dispute between natural-law theory and legal
positivism as follows:
The conflictbetweennatural-lawtheory and legal positivismis standardlypresentedas follows:natural
lawyersmaintainthat thereis a necessaryconnectionbetweenlaw andmoralitywhichpositivistsdeny.The
ideaof a necessaryconnectionis, however,opento interpretation, andnot all'necessary'connectionsbetween
law and moralityare incompatiblewith legalpositivism.Accordingto positivists,whetheror not thereare
objectivelyvalidmoralprinciplesby whichthe meritsof positivelaw canbe assessed,'The existenceof law is
one thing;its meritor demeritis another'[Austin,TheProvinceof Jurisprudence Determined (London,1955
184].Whateverconnections,necessaryor otherwise,mayexistbetweenlaw andmorality,thereis no necessary
connectionbetweenlaw andmoralitywithinthe conceptof law itself. Legalprovisionsareto be identifiedas
legalby theirempirically observablesourcesin legislation,decidedcases,custom,codesor whatever,the moral
status of these sourcesbeing irrelevantto their status as sourcesof legal provisions.Legalvalidityis not
relativeto morality.If naturallawyersandpositivistsgenuinelydisagree,and if the standardlocationof their
disputeis correct,then naturallawyersmust contest this particulardoctrineof the separationof law and
morals.They must hold that moralityentersinto the conceptof law itself; that to judge a provisionto be
legallyvalid is to judgeit to be in conformitywith specificmoralprinciples;that a provisionwhichviolates
theseprinciplesis defectivein its legality(79-80).
Some influential legal philosophers would argue against such a hard-nosed characterization of
natural-law theory. For example, in his 'The Matchmaker or Towards a Synthesis of Legal
Idealism and Positivism' 12 Journal of Legal Education, I, 4 (I959-60), Iredell Jenkins says of
'the insistence that Natural Law . .. stands as a transcendent body of principles and rules against
which positive law must be measured and declared invalid-not merely "bad" or "wrong" or
"mistaken" in case of conflict' that 'it is ... true that statements of this sort do occur in the
idealistic tradition: but they have to be searched for among the writings of third and fourth-class
minds, and even more among the writings of secondary sources that purport to expound this
doctrine'. Moreover, John Finnis in his Natural Law and Natural Rights (Oxford, I98o)
consistently and strenuously rejects any such hard-nosed view. According to Finnis, 'Far from
"denying legal validity to iniquitous rules", the tradition explicitly (by speaking of "unjust laws")
accords to iniquitous rules legal validity', op cit, 365.
continued on page 2
it must be because one is superior to the other in the way in which it will assist our
theoreticalinquiries or advance and clarify our moral deliberations,or both.2
Some positivists, for example Joseph Raz,3 employ the theoretical test; but others,
notably Neil MacCormick4 and perhaps Hart himself,5 appeal to the moral test,
treating this as a short-cut through the debate between positivism and natural-law
theory.
Our purpose in this article is to examine what difference, if any, it makes to our
moral deliberations whether we adopt a natural-law theory or a positivist concept
of law. To begin with we will consider what appears to be the standard positivist
reason for believing that natural-law theory provides an inferior concept of law
from a moral point of view. John Ladd expresses it neatly:
Legal positivists . . . are quick to point out that the practicaleffect of identifying law with
a part of morals is either to nullify existing law in favor of an ideal law, or to elevate all
existing law to the status of what is moral; in other words, the natural-lawtheorist, they
maintain,has to be either a radicalrevolutionaryor an unregeneratereactionary.6
In either event, it is alleged,the natural-lawtheorist is led to oversimplifycomplex
moral issues and cannot 'retain a critical moral stance in face of the law which is'.7
It is not always clear whether it is being asserted that natural-law theory
logically entails anarchismor conservatism and, in either event, dogmatism (we
can call this kind of thesis the 'positivist strong claim'), or merely that it is more
likely than positivism to generate such attitudes (this we can call the 'positivist
weaker claim')8.We contend that the positivist strong claim is false, resting either
continuedfrompage i
Notwithstanding such powerful denials, our position remains that if (i) natural lawyers and
positivists genuinely disagree, and (ii) the focus of the disagreement concerns a necessary
conceptual connection between law and morality, then natural lawyers must hold that a relevant
moral failure entails a legal failure. In our view, iniquitous rules cannot be legally valid; and, for
the record,we would not consciously speak of unjust laws.
2 Hart, The Conceptof Law (Oxford, I961) 204-205.
3 See especially Raz, The Authority of Law: Essays on Law and Morality (Oxford, I979) Chap 3.
4 See MacCormick,H.L.A. Hart (London, 1981) i6o. And see n 24 infra.
5 See Hart op cit, n 2 supra, 203-207.
6 Kant, The Metaphysical Elements of Justice (translated by John Ladd), (Indianapolis, 1965)
(translator's introduction) xxix. Similarly see Hart, 'Positivism and the Separation of Law and
Morals' 71 HLR 593, 598 (1958).
7 MacCormick,op cit, n 4 supra, 24.
8 The context should make it clear what kind of strong or weaker claim (or thesis) we are
considering but the scheme we have in mind is as follows:
strong claim - particular concepts of law entail particular (different) practical
attitudes;
weaker claim - particular concepts of law tend towards particular (different)
practical attitudes;
positivist strong claim - natural-lawtheory entails undesirablepractical attitudes;
positivist weaker claim - natural-lawtheory tends towards undesirablepractical attitudes;
natural-lawstrong claim - positivism entails undesirablepractical attitudes; and
natural-lawweaker claim - positivism tends towards undesirablepractical attitudes.
DERYCK BEYLEVELD AND ROGER BROWNSWORD 3
on too specific versions of natural-law theory or else involving logical fallacies:
most notably, the fallacy of equivocation (by employing positivist concepts of law
within the frameworkof what is supposed to be natural-lawreasoning), and the
fallacy of Denying the Antecedent (by arguing that because natural-law theory
holds that the legal validity of a rule entails a moral obligation to obey it therefore
the legal invalidity of a rule entails the absence of a moral obligation to obey it, or
even a moralobligationto disobey it).
A likely response to our analysis, however, is that this defence of natural-law
theory is bought only at the price of turning the 'dispute' between positivism and
natural-lawtheory into a verbal quibble. Positivists, it may be said, like natural
lawyers, are interested both in the law-as-it-is and in the law-as-it-ought-to-be.
They insist, however, that an objective science of law requiresthat the law-as-it-is
be defined and thereby identified in a morally neutral fashion. Natural-law
theorists, on the other hand, wish to define law in a morally loaded way: they
insist on withholding the title 'law' from rules promulgatedin the name of law
unless these meet specifiedmoral criteria.But if our defence of natural-lawtheory
is employed then this insistence is merely linguistic, for the defence, it seems,
implies that what may be apprehendedto be the social (legal) facts, and the
distance between these (if any) and what they ought to be, is unaffected by the
conceptualscheme which we choose to adopt.
We contend, however, that there is a significant (not merely linguistic)
conceptual and theoretical issue between natural-lawtheory and legal positivism.
The two positions may be seen to contest different views about the relationship
between practical reason in general and the moral point of view as a part of
practical reason. As a result the two positions have different views about the
natureof social facts.
Nevertheless, even if this is accepted, it may be maintained that our analysis
makes the dispute, although theoreticallysignificant,trivial from a moral point of
view. Provided that a positivist and a natural-lawtheorist share the same moral
principles, and there is no logical reason why they should not, positivism and
natural-lawtheory do not requiredifferentrules, proceduresand institutions for a
just system of legislation, adjudication,and government.
Although we agree that positivism and natural-lawtheory are not logically tied
to differentmoral points of view, it does not, however, follow that the adoption of
a concept of law has no moral implications. It may be the case, for example, that
positivism is regularlyassociated with different moral attitudes from natural-law
theory, or there may be reasons why the two positions are likely to encourage
different attitudes towards the institutions or agencies of social control, or
alternatively,positivism, whilst not requiringdifferent attitudes from natural-law
theory, may be compatible with attitudes with which natural-lawtheory is not
(and vice versa). In other words, we should seek the practicaldifferencebetween
natural-lawtheory and legal positivism, not as a thesis about logical entailment
(i.e. not as a strong claim), but as one about contingent conjunction, likely
tendency, or logical compatibility(i.e. as a weakerclaim).
4 NATURAL-LAW THEORY AND LEGAL POSITIVISM
We contend, however, that none of these weaker claim strategies can achieve
practical significance unless they can explain why positivism and natural-law
theory should tend to foster different attitudes. This requiresthe formulationof a
theory of ideology9which is missing in attempts to arguefor the moral superiority
of natural-law theory or positivism on the basis of conjunctions, tendencies, or
compatibilities.
We suggest, therefore,that if the second of Hart's criteriafor choosing between
natural-lawtheory and positivism (the advancementand clarificationof our moral
deliberations)is to be employed-whether as a means of demonstratingthat the
debate about the concept of law has some practicalsignificance,or as a polemical
strategy in the dispute between positivism and natural-law theory-then the
development of a theory of legal ideology must become a major preoccupationof
jurisprudence.Drawing together these strandswe conclude that Hart's moral test
is anythingbut the short-cutthat some positivists suppose.
II
9 In his Ideology (London, I971), John Plamenatz rightly notes that there is 'a tradition of careless
use' (20) of the concept of ideology. By a 'theory of ideology' we mean a theory exploring how
people come to have the ideas that they do, how ideas relate to one another, and how abstract
ideas bear on practical attitudes and actions.
Io Accordingly we would take issue with the implication in Harris, Legal Philosophies (London,
I980) i9-2i, that natural-law theory, strictly speaking, must be 'objectivistic' and restricted in
its ethical theory.
DERYCK BEYLEVELD AND ROGER BROWNSWORD 5
our version then they cannot be valid against natural-lawtheory per se-which
they are intendedto be.
In our view, the concept of law is the concept of morallylegitimate power and
the concept of a legally valid rule is the concept of a rule which there is a moral
right to enforce. Where the rule is positive (i.e. promulgatedand administeredin
fact) the existence of the moral right to enforce-equally the existence of a fully
legally valid rule-is subjectto three conditions:
(a) the rule itself must not be immoral;
(b) the rule must be administered in a morally legitimate fashion (according to
somethinglike the principlesthat compriseFuller'sinner moralityof law);" and
(c) the rule must be promulgatedby a morallylegitimate source.
When all three conditions hold there is not only a moralright to enforcethe rule
but also a moral obligation to obey it. However, our view of the relationship
between legal validity (LV), the moralright to enforce a rule (MR), and the moral
obligation to comply with a rule (MO), is not symmetrical and requires some
discussion.
We can start by stating baldly our view of the relationshipsbetween the critical
law concepts. They are as follows:
(i) (LV) entails (MO)
(ii) (MO) does not entail (LV)
(iii) (LV) entails (MR)
(iv) (MR) entails (LV)
It will be seen that there is an asymmetrybetween the (LV) (MO) relationshipand
the (LV) (MR) relationship.We shall speakbriefly to our four root propositions.
rules fail the test of legal validity simply because they are not moral per se.
Providedthat such rules are not immoral(that is, do not prohibitmorallyrequired
conduct, or require morally prohibited conduct, or regulate morally optional
conduct in some illegitimate way),12 and provided that the other conditions of
legal validity are met, then directly or indirectly'3such rules will generate moral
obligations.
I4 A caveat is in order here. The traditional categorization of ethical positions into deontological
and teleological theories has supported some serious misunderstandings. One is that ethical
theories either take account of consequences and nothing but consequences or else they take no
account whatsoever of consequences. Another is that all consequential theories are utilitarian.
These misunderstandings might foster the idea that we are utilitarians. This would be quite
wrong. We stand firmly on the deontological side of the traditionaldivide, drawing support from
contemporaryliberals such as Ronald Dworkin, John Rawls, and Alan Gewirth, but where basic
rights or duties conflict we look to the consequences of giving effect to one set of rights (or
duties) as against a conflicting set. This does not let utilitarianism in through the back-door but
development of this point is too much for even a long footnote.
15 See the discussion of controversialmoral cases in text, infra at 10o-3.
8 NATURAL-LAW THEORY AND LEGAL POSITIVISM
here.'6 For the moment we merely wish to point out that what is morally
permissiblein a non-ideal world may not be permissiblein an ideal world and that
judgements of legality should always be referred to the context of the social
arrangements in which 'the enterprise of subjecting human conduct to the
governanceof rules'"7takes place. What is law must be judged accordingto what
is practicallypossible, not accordingto what is merely ideallydesirable.
III
We are in a position now to consider the positivist strong claim that there is a
logical entailment between natural-law theory and reactionary or revolutionary
attitudes. Our contention is that this entailmentdoes not hold.
First, we can examine the supposed entailmentbetween natural-lawtheory and
a revolutionaryattitude. Why should a natural-lawtheorist have to be a radical
revolutionary or, to use Neil MacCormick's phrase, a revolutionary anarchist?
MacCormickpresents the reason very cryptically:
[S]ince[forthe natural-law
theorist]whateveris law mustbe moral,governments mustbe
disobeyedor evenoverthrownif whattheypropoundas 'law'is not morallyjustified.'8
Granted,this is presentedin parentheses,not as a developedargument;even so, it
is not merely elliptical, it is a non-sequitur. For, while the antecedent states
perfectlycorrectlythat natural-lawtheory holds that (LV) entails (MO), logic does
not commit us to MacCormick's conclusion that not-(LV) entails not-(MO)
(actually,that not-(LV) entails disobedience or revolution,but since (MO) entails
obedience and thereforedisobedience entails not-(MO), this entails that not-(LV)
entails not-(MO)).As we saw in the previous section, logic does not compcl one to
argue that because (LV) entails (MO) therefore (MC) entails (LV); this is the
error of Affirmingthe Consequent. Equally, it is an error to think that because
(LV) entails (MO) thereforenot-(LV) entails not-(MO); in this case it is the error
of Denying the Antecedent. And it is precisely this error that undermines
MacCormick's reasoning. To force MacCormick's conclusion we need the
proposition that (MO) entails (LV) for it will follow from this that not-(LV)
entails not-(MO). However, the snag is that (MO) entails (LV) is not entailed by
(LV) entails (MO), and so this appearsto be a dead-end.
MacCormick might try to derive the critical bridge proposition that (MO)
entails (LV) by an indirectroute. The three steps in the reasoningmight be thus:
I6 See e.g. Rawls, A Theory of Justice (London, 1972) for the contrast between ideal theory and
non-ideal theory, and Ackerman, Social Justice in the Liberal State (New Haven, I980) for the
distinction between ideal theory and second-best theory. On the latter generally see our
'Liberalism in Quest of Itself' 45 MALR104 (I982).
17 We take Fuller's famous phrase (see Fuller op cit 96) as the referentof the word 'law' or 'the legal
enterprise'. This seems to be a pretty general starting-point for the debate about the concept of
law and it has the virtue of eliminating the possibility that rival concepts of law can be attributed
to differences in the initial focus of interest.
I8 MacCormickop cit, n 4, supra, 24.
DERYCK BEYLEVELD AND ROGER BROWNSWORD 9
substitute something like 'the moral law' or 'the dictates of the true morality'for
'the Divine law'.
Firstly, Austin is perfectly entitled to charge our version of natural-lawtheory
with the view that an immoral rule cannot be a law. The mistake he makes is to
think that this commits the natural-lawtheorist to the view that there can never
be a moral obligationto comply with a legally invalid rule. As we have indicated at
length alreadythis is a striking non-sequiturin positivist thought, and we can be
excused yet furtherrehearsalsof the argument.
Austin's main point, however, seems to revolve around the inherent
controversiality of moral questions, his argument being that if we follow
natural-lawtheory by employing a controversialand uncertain(moral)test for law
then this will necessarilyinvite anarchy.Before we can comment on this aspect of
Austin's criticism of natural-lawtheory we have to explain our own thinking on
the status of rules the moralityof which is genuinelycontroversial.21
Suppose that A has been delegated the right to govern a community, the terms
of the delegation being such as to confineA's governmentwithin the bounds of a
particular(let us suppose, objectivelycorrect)22moral theory. The moral theory is
the constitution under which A has the moral right to govern. So long as A
endeavours to observe the moral groundrulesof his office it seems unlikely that
any obvious violations of the constitution will occur. Even so, and
notwithstandingA's best endeavours, there is the standing possibility that moral
violations will take place as soon as A's governmenthas to make decisions that go
beyond the straightforwardapplication of the cornerstones of the constitution.
The moral principles of the constitution provide the basic materials but, on any
number of issues, we must expect genuine disagreement between good faith
protagonistswho attempt to arguethrough the logic of the principles.Do we hold
that there is a moral right to enforce rules in such controversialcases and/or a
moralobligation to comply with such rules?
The short answer to this is as follows. If the rules are not in fact immoraland if
the other conditions of legal validity are met then there is a moral right to enforce
the rules; if not, then there is no moral right. But, correct though this is, it seems
to miss the thrust of the problem.For, given that we are not morally omniscient,
what is required is some practical guidance not some abstract theoretical
statement.
In our view there are two legitimate practical tacks on this tricky problem.
First, we can approachthe question in terms of controversiallegality. Given that
the moral arguments are inconclusive either way we are forced to suspend
judgment on the legality of the rule. With the legality of the rule bracketed off
21 The idea of genuine controversiality raises a complex issue about the distinction between a
question which is rationallyregardedas open and one which is not. Ronald Dworkin's distinction
between anthropological and discriminatory moral positions (see Dworkin, Taking Rights
Seriously (London, 1978 new impression) 248-253) provides some useful leads but there is
clearly a lot of work to be carried out here.
22 We are making this assumption to simplify our discussion. In principle, however, our discussion
is applicableequally to the case of a morally controversialconstitution.
DERYCK BEYLEVELD AND ROGER BROWNSWORD II
with controversial decisions provided that the decision has been made in good
faith and provided that the possibility of further debate on the question has not
been closed.24In this way one could plausibly argue that this is anothercase of a
moral obligation exisiting in respect of a rule which is known to be controversial
and which in fact (although,of course, this is not known) is legally invalid.
Now we can return directly to Austin. The first point is this. While there is no
denying that the governmentof moder complex societies is a minefieldof moral
controversiality,this is not to say that rational citizens treat themselves as under
no moral obligation to observe the requirements of decisions/rules which they
judge to be morallywrong. On the contrary,and as we have just explained, it is at
the very least plausible to argue that the terms of a rational delegation of
governing authoritywould provide for a collateralmoral obligation to conformto
the rules in the genuinely controversialcase (subject to moral good faith). Thus,
contrary to Austin's suggestion, this does not present any kind of threat to 'wise
and benign rule' if by that he means good faith moralgovernment.
Secondly, and building on the previous point, we have seen already that our
kind of natural-lawtheorist holds a complex view of legal validity. Legal validity is
not just a matter of the content of the rule; questions of procedureand title are
relevant as well. Moreover, there can be a failure to satisfy the criteria of legal
validity not only along the dimension of attainmentbut also along the dimension
of attempt. The idea that the natural-lawtheorist is a simple-mindeddemagogue
fits very uneasilywith our view of such a theorist.
Thirdly, it is worthwhile pausing to consider how positivism's guarantee of a
non-controversialtest for law eliminates the seeds of anarchyinevitably sown by
people's disagreementsconcerning what the law ought to be. Surely, holding the
positivist view that the law is to be determinedindependentlyof moraljudgement
is not going to insure us against anarchy, unless we adopt the extremely
paradoxicalview that what is law (independentof its morality)morallydetermines
our obedience to it. Indeed, Austin gets perilously close to this unsavoury view.
We are to be permitted to question and criticize the rules of government,but not
to disobey or attempt to overthrow government in any circumstances. We can
work for change within the constitutional structuresof the system with which we
are faced, but never outside them. This may be all very well if we are faced with a
25 In his celebrated debate with Lord Devlin, Hart puts this question forcefully (Hart, Law,
Liberty, and Morality (London, I963) I9): 'We might wish to argue that whether or not a
society is justified in taking steps to preserve itself must depend both on what sort of society it is
and what the steps to be taken are. If a society were mainly devoted to the cruel persecution of a
racial or religious minority, or if the steps to be taken included hideous tortures, it is arguable
that what Lord Devlin terms the 'disintegration' of such a society would be morally better than
its continued existence, and steps ought not to be taken to preserve it.'
26 See Hart, op cit, n 2 supra, I89-195.
27 MacCormick,op cit, n 4 supra, 24.
I4 NATURAL-LAW THEORY AND LEGAL POSITIVISM
30 Even MacCormick finds the argument unconvincing, see MacCormick, op cit, n 4 supra,
I59-t60. Also, for an effective refutation of Hart on this point see Hall, Foundations of
Jurisprudence(Indianapolis, 1973) i6.
31 Hart, op cit, n 2 supra, 206.
32 Ibid, 206-207.
I6 NATURAL-LAW THEORY AND LEGAL POSITIVISM
IV
Our general opinion is that positivist argumentsfor the moral superiorityof their
position are extremely poor if read as arguments of entailment. They invariably
beg the question, oversimplifytheir targets or involve simple logical fallacies.
There is a difficultywith our defence, however, and this is that it may lead to
the charge that it is bought only at the price of turning the 'dispute' between
natural-lawtheory and positivism into a verbal quibble. The defence at various
points, and most notably in connection with the allegationthat natural-lawtheory
33 For a variation on this theme see Jenkins, Social Order and the Limits of Law (Princeton, I980)
369, where the author embellishes Hart's argumentsthus:
To Hart'sarguments,I would add one of a practicalnature.If we thinkof iniquitouslaw as not really
law-and evenmore,if we thinkof an unjust'legal'systemas not actuallya legalsystem-this hasthe effect
of focusingour attentionsolelyon the legal aspectof the matter.And this in turn leads us to treat the
problemin strictlylegalterms:sincethe supposed'law'is not such,thereis no reallaw present,and all we
needto do is to enactlawsthatwillrightthe actualwrongs.Andthatis quitecertainlya self-defeating
policy.
And that, we should respond, is quite certainly a non-sequitur.
DERYCK BEYLEVELD AND ROGER BROWNSWORD I7
leads to reaction, in effect, forecloses the objection by employing a definitional
stop. The consequences alleged to follow from natural-lawtheory do not follow
because natural-lawtheory simply defineslaw in a way which does not allow them
to follow. But if this is so, it may be said, then there is no genuine debate between
natural-law theory and positivism. Apparent differences are merely a matter of
terminology. When the positivist identifies a rule as law, he does so because of
morally neutral features which the rule possesses. The natural-lawtheorist takes
the rules selected by the positivist accordingto these morallyneutralidentification
proceduresand divides them into two categories-rules which are legal, and rules
which are not legal accordingto whether or not they fulfil specifiedmoral criteria.
If, for the sake of convenience we call rules which fit the moral criteria 'moral
rules' and those which do not 'immoral rules' then what we have is a case of
positivists dividing laws into moral and immorallaws, whereas the naturallawyer
divides rules into moral and immoral rules and only allows the moral rules to be
called 'laws'. But the effect of this is that there is no differencebetween what the
natural-lawtheorist calls 'laws' and what the positivist calls 'moral laws' and no
difference between what the positivist calls 'immoral laws' and what the
natural-law theorist calls 'immoralrules claimed to be law'. The terminology of
positivism and natural-law theory is intertranslatable,and once translation is
effected then we see that what has happened is merely that different languages
have been used to classify the same social facts. Provided that we are clear about
which language we are using there can be no confusion. By the same token,
however, there cannot be anything significantto debate between the two ways of
speaking: it is merely a matter of which convention we choose to adopt.
Something like this is, in fact, alleged to be the case by GlanvilleWilliams in his
celebratedpaper'The ControversyConcerningthe Word "Law"'.34
When considering this position we must be careful to distinguish two types of
significance, (a) conceptual (or theoretical) significance and (b) practical
significance.
34 Williams, 'The Controversy Concerning the Word "Law"', in Laslett (Ed), Philosophy, Politics
and Society (Oxford, 1956) 134. Thus, for example, Williams says: 'Those who construe law as
command refuse to apply the word law to any system of rules that cannot be regarded as a
command. Those who construe law as a principle certified by reason refuse to apply the word law
to any system of rules that cannot be regarded as certified by reason. And so on. In each case
the controversy is simply as to the application of a word, not as to the characteristics of any
particularsystem of rules under dispute', (ibid, 155).
i8 NATURAL-LAW THEORY AND LEGAL POSITIVISM
36 See n I supra.
37 See Hart, op cit, n 2 supra, i98-199.
38 See Finnis, op cit, n i supra, especially Chap i. Finnis' argument is not, however, that practical
reason presupposes moral principles/reason, but that the viewpoint of moral reason is the
practically (morally) significant one within practical reason, and we do not believe that this will
secure natural-lawtheory.
20 NATURAL-LAW THEORY AND LEGAL POSITIVISM
will have to be seen as having an ideal content. This is to say that their existence
can only be realized when they meet ideal conditions. Whatever exists without
meeting these conditions will be a fact but one which represents itself as
somethingother than that which it really is.
This is still rather dark but perhaps we can illuminate a little if we disentangle
two crucial ideas at the bottom of this discussion. First, there is the idea that
social facts are concept-neutral,this being the idea propoundedby Williams-type
sceptics and disputed by ourselves. But what precisely is the nub of this dispute?
Suppose, for example, that an IRA bomb man blows up some English soldiers in
Northern Ireland. The physical facts, here, are clear enough but what are the
social facts? Some might say that this is a case of murder,others that it is a killing
for political purposes or a liberating act. The sceptic, it seems, wants to say that
the social facts are the brute facts of the killing and that such social facts remain
unaffectedby our ex post facto characterizationof the situation. Thus, argues the
sceptic, the social facts are not alteredby calling the killing a murder,or a political
killing, or anythingelse for that matter. We, on the other hand, hold that the social
facts are constituted by the concepts we employ over and above the raw physical
situation. In our view, the sceptic is in danger of reducing social facts to simple
physical facts. Hans Kelsen, it seems to us, makes the sound point that external
appearancesalone do not yield the distinctively legal facts.39Externalappearances
do not tell us whetherA's killingB is a murderor a State execution, whetherA is
making a valid contract, whether this group of people are legislating a new rule or
whatever. Similarly, external appearances alone do not yield the distinctively
cultural facts: a purely physical account of the situation will not tell us, for
instance, whether this activity is a cricket-match or a rain-making ceremony.
Thus, wherever we are seeking to identify the distinctively social facts we see no
plausible basis for asserting that such social facts are concept-neutral.Secondly,
there is the idea that even if social facts are not concept-neutral,neverthelessthey
are value-neutral.This is where, we suggest, positivists and natural-lawtheorists
join issue. Whereas positivists conceive of the social facts associated with law in
morally neutral terms, natural-law theorists conceive of such facts in morally
dependent terms. These rival conceptualframeworksyield different sets of social
facts, not (as the sceptic would have it) merely differentclassificationsof the same
social facts. In a nutshell, then, the core of the positivism/natural-lawdebate is
value-neutrality,the core of the debate with the sceptics concept-neutrality.
To round off our discussion we can return to the sceptic's claim that the social
facts gathered under one conceptual scheme can be translated without factual
alteration into the language of a different conceptual scheme. This follows, of
course, from the sceptic's epistemology being such that concepts do no
epistemological work: our direct experience of the social world yields the social
facts, which facts are then assembledinto whichever conceptualclassificationswe
choose to employ. If positivists and natural-lawtheorists are to have a significant
39 See Kelsen, Pure Theory of Law (Berkeley, 1967) 3-4.
DERYCK BEYLEVELD AND ROGER BROWNSWORD 21
40 Gewirth, Reason and Morality (Chicago, 1978). Gewirth argues that the supreme principle of
morality is the Principle of Generic Consistency (PGC), namely: 'Act in accord with the generic
rights of your recipients as well as of yourself (ibid, 135). This is very similar to the rather more
familiar Dworkinian ideal that one should treat others with equal concern and respect, but is
more specific in that the features which should be treated equally are indicated.
DERYCK BEYLEVELD AND ROGER BROWNSWORD 23
association or connection. The weaker claim may run in at least three different
forms, namely:
(a) The espousal of natural-lawtheory has been regularlyassociated with different
political and moral views or attitudes from the espousal of positivism (the
'conjunctionthesis').
(b) There are reasons, located in the structure and mechanics of society, why
natural-lawtheory should tend to promote different political and moral views or
attitudes from positivism (the 'tendencythesis').
(c) Natural-law theory is compatible with political views etc with which
positivism is not (and/or vice versa) (the 'compatibilitythesis').
44 The 1949 textbook on the Theory of the State and Law published by the Institute of Law of the
Academy of Sciences of the USSR defines law in a hard-nosed positivistic fashion as follows:
'Law is a combination of the rules of behaviour (norms), established or sanctioned by state
authority, reflecting the will of the ruling class-rules of behaviour, whose application is assured
by the coercive power of the state for the purpose of protecting, strengthening and developing
relationships and procedures suitable and beneficial to the ruling class' (cited by Hazard in his
Introduction to Soviet Legal Philosophy (Twentieth Century Legal Philosophy Series: Volume
V) (CambridgeMass., I951) xxxvi).
45 Suppose that it were conceded that the English natural-law theorists of the eighteenth-century
were dogmatic conservatives, that Bentham used positivism as a way of exposing such
dogmatism, and that English positivists from Bentham through to Hart have combined their
positivism with an open-minded reforming attitude. What implications would such concessions
have for the weaker thesis ?
So far as the conjunction thesis is concerned it would be of no significance that English
natural-law theorists at a particular time happened to hold conservative views. At other times
and in other places, as we have said in the text, natural-law theory has been associated with
radical political views. However, it might be suggested that the crucial aspect of eighteenth-
century English natural-law theory was not so much its conservatism as its dogmatism.
Moreover, it might be argued, this contrasts sharply with the reforming spirit of positivism.
Whatever truth there may be in this version of the conjunction thesis in the case of England, it is
impossible to generalize it. Positivism has not been invariably associated with a critical outlook
and natural-law theory has not been regularlypaired with a dogmatic attitude (on both points see
e.g. the German experience described in text infra, 26). So, whether one seizes on the
conservatism or the dogmatism of Blackstone and Co., there is no straightforwardconjunctionto
be identified.
Nevertheless, one might think that the English experience is evidence of some kind of tendency
to be found in positivism. Maybe there is no general conjunction between positivism and an
open-minded reforming attitude, but is there not at least a tendency in this direction? As we
argue in the text, this kind of speculation can only be taken seriously if it is grounded in a theory
of ideology. In this note suffice it to say that, in principle, natural-law theorists can be just as
open-minded (or dogmatic) as positivists, and if Bentham had found eighteenth-century thought
dominated by dogmatic positivists might he not have turned instead to a critical natural-law
position? If so, it has yet to be shown that English positivism has the intrinsic tendencies that
the weaker thesis requires.
26 NATURAL-LAW THEORY AND LEGAL POSITIVISM
Hart who, although he speculates that the tendency towards revolutionhas been
overrated,46neverthelessalleges a tendency towards reaction.Why, however, does
Hart think that this tendency should exist? He provides no empirical evidence
that the tendency exists. Furthermorethe only mechanism which he provides for
expecting the tendency to exist is the attributionof fallaciousreasoning to people
who will believe the natural-lawthesis (viz: the move from 'Only what is moral
can be law' to 'Anything called "law" by anyone must be moral'). But, if telling
people that only what is moral can be law will make them think that everything
called 'law' must be moral, why is this so? Why, in any case, should Hart think
that academicpronouncementswill have any effect at all on the attitudes of people
in society generally?Without answers to these questions what we have is merely
the drafting of a logical (in fact illogical) thesis to do service as an explanationof
why people will think in particularways.
Moreover, Hart himself is quick to point out weaknesses of just this kind when
natural-lawtheorists run a tendency thesis against positivism. For example, Hart
has no hesitation in finding 'an extraordinarynaivete in [Gustav Radbruch's]view
that insensitiveness to the demandsof morality and subservienceto state power in
a people like the Germansshouldhave arisenfrom the belief that law might be law
though it failed to conform with the minimum requirementsof morality'.47The
lesson Hart draws from this is that:
[T]his terriblehistorypromptsinquiryinto why emphasison the slogan'law is law',and
the distinctionbetweenlaw and morals,acquireda sinistercharacterin Germany,but
elsewhere,as with the Utilitariansthemselves,went along with the most enlightened
liberalattitudes.48
But this lesson is double-edged.If Radbruch'sspeculations are inadequatewhen
positivism stands accused of undesirable tendencies then natural-law theory is
entitled to the same due process.
An interesting version of the tendency thesis is presented by J. L. Mackie.
Whereas the traditional positivist focus has been on the attitudes of citizens
operating with natural-lawtheory, Mackie quite legitimately shifts the focus to
the attitudes of natural-lawjudges:
The argumentof this bookthereforehas, as a corollary,the rejectionof the doctrineof
natural-lawas a philosophicaltheory.Whetherit is, none the less, a usefulfictionis a
furtherquestion,andone to whichno generalanswercanbe given.Wherethe doctrineis
adopted,judgeswill treatenactmentswith somewhatless respect,andwill attendalsoto
their'conscience'or their'reason'.Whateffectsthis will havedependsuponthe current
state of these judicial organs, and on the characterof the legislatureson which they impose
a check. The doctrine may provide a barrier against excesses of governmental policy; it
continuedfrompage 27
4 An enculturationhypothesis (members of the legal profession tend to absorb the tradition they
are brought up in), or a related selection hypothesis (judges are selected from members of the
legal profession according to the degree of enculturation which they display) is capable of
explaining (3) on the basis of (2).
5 It is difficult to believe that the existence of the divergent legal traditions owes nothing to
differences in legal ideology (natural-lawtheory v legal positivism) which characterizethe two
jurisdictions.
6 Therefore, a differential impact of natural-law theory and legal positivism may be responsible
for different levels of judicial activism.
This suggestion certainly raises a number of interesting questions which require empirical
investigation, notably the enculturation and selection hypotheses. It is also in line with our
thinking in highlighting the importance of investigating connections between legal ideology and
institutional practice and style. However, as it stands the suggestion presents some conceptual
difficulties. In particular, how are we to interpret the assertion that the USA's legal system is
closer to natural-lawtheory than its English counterpart?
Does this allege
(a) In the USA judges are taught that to make legal judgments they must make moraljudgments,
whereas in Englandjudges are taught to avoid moraljudgments ?
or (b) The USA has a constitution entrenching moral considerations in the law whereas England
does not ?
If we take (a) then the link between (i) and (2) in the suggestion above becomes a tautology and
nothing more than the highly plausible enculturation or selection hypotheses is required to link
different levels of judicial activism to differences between natural-law theory and legal
positivism. If we take (b) then the suggestion is equally plausible since (b) will lead to (a) on
simple and plausible assumptions.
The difficulty, however, is that neither interpretation (a) nor interpretation(b) is entailed by
the way in which we have defined natural-law theory v legal positivism. The doctrine of the
separation of law and morals (characteristic of 'legal positivism' as we wish this position to be
understood) does not entail that judges must not make moral judgments when making legal
decisions, only that when making legal decisions they are not necessarily making moral
judgments. Furthermore,if the doctrine of the separation of law and morals were to entail that
judges must not make moraljudgments then this would make a critical positivism impossible by
definition. Also, it is certain that the doctrine of the separationof law and morals does not entail
that a legal constitution cannot entrench moral considerations (if it did then positivists would
have to say that the USA does not have a legal system). So, whatever the constitutional and
practical requirements of natural-law theory, these requirements are not ruled out by legal
positivism in our primary sense, and it seems that (i) is false in relation to the way in which we
wish 'natural-lawtheory' and 'legal positivism' to be understood.
This, of course, does not mean that adoption of the doctrine of the separation of law and
morals does not create (in certain circumstances) a tendency to treat legal decisions as
autonomous from moral judgments in practice. But if this is so, it first has to be established that
this is so (despite the lack of entailment) and then explained. (See our discussion of the
compatibility thesis for futher elaboration.) In the terms which we are considering the present
suggestion begs the question at this crucial point.
DERYCK BEYLEVELD AND ROGER BROWNSWORD 29
them. If one follows Gewirth and uses his argument to justify a natural-law
position then it will have to be argued that a legal system must enshrine moral
principleswithin its constitution; that judicial discretion must be limited by these
values (indeed that courts must be moral courts) and that agencies of government
must be accountable to citizens in terms of these values. There can be no
justifiable attempt to separate out from a substantive moral practice a special,
autonomous category of 'legal justice' which is purely formal and amounts to no
more than impartialityin terms of the dictum 'Treat like cases alike, unlike cases
unlike'. Such a 'justice' is no justice at all. Of course there is, in ordinarylanguage
usage, a concept of such justice, but this must be taken to show no more than an
ideological sedimentationof (false) positivistic attitudes. The legal enterprisemust
be viewed as one which is organized around values of co-operation and
accountabilityas between rulers and subjects.53
A positivist might, of course, regard such an arrangementas highly desirable,in
which case he would not run this particularargumentagainst natural-lawtheory;
but there is no logical requirementfor him to do so. The positivistic sources thesis
entails a view of practicalreason which is morally neutral (practicalreason does
not revolve around moral reason; it contains autonomous moral and non-moral
spheres of reason). It entails a view of human personal identity which is again
morally neutral (persons have no essences which are definedby their real (moral)
interests, inasmuch as human nature is to be defined by a capacity to engage in
practicalreason). It entails a doctrine of judicial discretion not necessarilylimited
by specifiedmoral values. The generallegal relationship,as Fullerhas pointed out,
is not a reciprocal one, but is instead hierarchical, 'a one-way projection of
authority'.54These premises make it possible to interpretthe 'impartialityof the
law' as 'freedom from the values of a particulargroup or class' (rather than as
accountability to all classes), to present 'the law' as autonomous from political
institutions and their requirements,and to see the 'purpose' of law as primarily
social control.55There is thus the possibility of an ideological gap between
positivism and natural-lawtheory. Positivism may be used to justify arrangements
which cannot be justified using tenable natural-lawtheory.
But if this ideologicalgap constitutes a practicaldifferencebetween natural-law
theory and positivism, it is only aformalpractical difference.The existence of the
53 Generally on this theme see Lon Fuller, op cit, n I i supra, Selznick, 'Sociology and Natural Law'
6 Natural Law Forum 84 (I96I), and Law, Society, and Industrial Justice (New York, 1969);
also, Duff, 'Legal Obligation and the Moral Nature of Law' The Juridical Review 6 (1980).
54 Fuller, op cit, 192.
55 Fuller (op cit, I91-242) wants to argue, it seems, that positivism consists of a number of
interrelated premises gravitating around the idea that law is 'a one-way projection of authority'
with no special role morality attaching to the law-giver. From here positivists conceive of the
law-giver/citizen relationship in terms of managerial direction and this, of course, yields
particular practical attitudes towards this relationship. If Fuller's position is that these linkages
are actually entailments then he is taking a stronger view than that presented in the text. In fact,
he would be arguing for the strong thesis.
30 NATURAL-LAW THEORY AND LEGAL POSITIVISM
gap does not entail that it will be exploited, and the thesis tells us nothing about
the material conditions which enable it to be exploited. Like the conjunction
thesis, the compatibility thesis needs to be supplemented by an explicated
tendency thesis if it is to do any real work in demonstrating that there is a
practical difference between natural-law theory and positivism which is of
practicalsignificancein the way in which society operates.
Thus, the tendency thesis is, clearly,the key to developingthe weakerclaim. To
explicate it requires, however, nothing less than the development of a full-scale
theory of legal ideology. Beforewe start saying that to use a natural-lawtheory (or
positivism) will lead people to think and act in particularways we need to know
how they do think, how this is related to how they act, how they come to think as
they do, and, in particular,how legal rhetoric arises and affects these things. An
appeal to Hart's second criterion for choosing between natural-law theory and
positivism is nothing but empty rhetoricwithout such a theory.
Assuming that we are correct in isolating the tendency thesis as the key to the
weaker claim can we be a little cleareras to the way in which a theory of ideology
could support this thesis? Our underlying point is that the tendency thesis as
stated in the abstractlacks both social and ideologicalcontext. Let us suppose that
the abstract statement of the tendency thesis runs as follows: if X holds this
particularconcept of law then this will tend towards X holding such and such a
practicalviewpoint. It will be noticed that this version of the thesis claims only a
relational tendency within X; it makes no claim about the possibility that X's
holding a particular concept of law may tend towards persons other than X
holding particularpracticalviewpoints. So we are dealing with the tendency thesis
in a very simple form. In what respects is this abstract statement of the thesis
theoreticallyundeveloped?
First, the tendency thesis presupposes that one's concept of law tends to
influenceone's practicalattitudes in a more or less direct cause and effect manner.
It is almost as though the formation of a particularidea about the concept of law
triggers off a set of practicalviewpoints. This raises a whole host of questions. Are
our ideas related in such a mechanicalpush-pullfashion? If so, is our law idea the
antecedent cause of our practicalattitudes? Are these relationshipsthe work of a
rational or an irrational mind? If they are the work of a rational mind is this
tantamount to resurrectingthe strong claim? If the tendency only operates in an
irrationalmind how drasticallydoes this limit the scope of the thesis?
Secondly, although the simple version of the tendency thesis does not claim
anything about the possible effects that X's ideas may have on other persons, it
cannot escape the possibility that other persons' ideas and rhetoric may interact
with X's ideas in a significant way. Such interactions could be highly important
and they badly need to be theorized. This also raises tricky questions about the
effective cause of X's practical attitudes. Suppose, for instance, that X is a
natural-lawyer,that he believes that he should defer to the moral judgments of
those in power, and that he believes that those in power are natural-lawyers,
whereas, in fact, they are disingenuous positivists. The net effect of this is that X
DERYCK BEYLEVELD AND ROGER BROWNSWORD 31
then Hart's test of practical moral superiority is not the short-cut that some
suppose. For, the positivist strong claim is false and no weaker claim can work
without a theory of ideology; the theory of ideology may reveal no practical
significance between rival concepts of law, or (even worse from a positivist
viewpoint) a practical significancethat shows natural-lawtheory to be superior;
and, anyway, none of this may be viable until the theoretical question has been
settled, in which case no one will be interested any longer in a short-cut.
continuedfrompage 31
though, two problems with this. One is that the construction of a general theory of ideology is a
social scientific endeavour and, thus, the inquirer should employ those concepts that are generally
superior for social scientific purposes. In other words, the general theory of ideology should build
on the theoretically superior concept of law. The other problem is that so long as the theory of
ideology remains general it cannot have much purchase on the particularthesis advanced under
the weaker claim. Sooner or later the general theory has to be applied to attitudes to law and we
are back to square one.
A more promising response, perhaps, is to set up a hypothesis for empirical validation, such
hypothesis eliminating all references to law. For instance, we should test out a tendency between
natural-law theory and particular attitudes towards those in power. Suppose, then that such
research revealed some connection between natural-law theory and revolutionary or submissive
attitudes/reactions to the rules of those in power. What claim would this evidence support? It
would not, we suggest, clinch the positivist weaker claim, for it would be unclear whether the
attitudes/actions were undesirable. What the positivist weaker claim needs to show is that
natural-law theory tends to generate submissive attitudes to the immoral powerful and/or
revolutionary attitudes towards the moral powerful. A simple connection between submission/
revolutio- and the rules of the powerful is not enough for the positivist weaker claim. Neither
would su-,h a connection establish the basic weaker claim that particular concepts of law tend
towards particular (different) practical attitudes. For, the evidence that natural-law theory
cultivates a distinctive attitude towards the rules of the powerful presupposes that 'the rules of
the powerful' represents a unitary social phenomenon towards which one can have a practical
attitude. Yet this clearly begs the theoretical question.
A third response, drawing on situation ethics, argues that we have missed the point about the
undesirability of revolutionary or submissive attitudes. The point is that these are patterned
attitudes involving the loss of moral autonomy. Moreover, natural-lawtheory suffers inescapably
from this defect for its practitioners are slaves to the ethics of their systems. The short answer to
this is that natural-law theory, in principle, can have any kind of ethics-even situation
ethics-and so this response is an ad hominemclaim. Admittedly, there would be a serious point
here if the only tenable versions of natural-law theory were non-situationist but that is a bridge
to be crossed at some future date.
No doubt a good deal more could be said about the involvement of the theoretical question in
attempts to cash the weaker claim. But, we suggest that the burden is now with the positivists
who wish to engage natural-lawtheory on this front.