Waluchow, Legal Positivism Inclusive Versus Exclusive
Waluchow, Legal Positivism Inclusive Versus Exclusive
Waluchow, Legal Positivism Inclusive Versus Exclusive
Modern legal positivism views law as a human creation; the existence and content of law
are, fundamentally, matters of social fact. This is usually termed ‘the social thesis’. A
second thesis integral to the positivist tradition is John Austin’s famous ‘separation
thesis’: ‘the existence of law is one thing, its merit or demerit another…’ – there is
nothing in the nature of law as a social institution that guarantees its moral worth. Despite
a shared commitment to the social and separation theses, defenders of legal positivism
differ significantly in their understanding of these fundamental tenets. In response to
Ronald Dworkin’s claim that moral principles partly determine the content of legal
systems, positivists have divided into two major camps. Inclusive positivists assert that it
is conceptually possible, but not necessary, that the legal validity of a norm should
depend on its consistency with moral principles or values. Exclusive positivists assert the
opposite: the legal validity of a norm can never be a function of its consistency with
moral principles or values. Recent debates among positivists have focused on whether
inclusive positivism is consistent with the law’s claim to authority and with the role it
purports to play in guiding our conduct. Where these debates will lead is, at this stage, an
open question.
In its traditional form, legal positivism asserts that human law is essentially a social
institution the existence and content of which is, fundamentally, a matter not of God’s
law but of human will and power. According to the English legal theorist John Austin,
any civil law is the command of a human sovereign. Law exists, that is, when someone
who alone enjoys the habitual obedience of the bulk of a population has expressed the
will that others act or forebear from acting in prescribed ways, and has coupled these
expressions of will with the threat of sanctions should their wishes be unfulfilled. Modern
positivists uniformly reject Austin’s command theory, substituting for the commands of
the sovereign a set of norms (e.g., rules, principles and variable standards) whose legal
validity depends on fundamental conventions or social rules. Nevertheless, though the
emphasis is now on convention rather than on command, every modern positivist views
law as a matter of human creation; its existence and content are, fundamentally, matters
of social fact. This basic commitment, shared by all proponents of legal positivism, is
usually termed ‘the social thesis’ (Raz 1979).
A second thesis integral to the positivist tradition is Austin’s famous ‘separation thesis’,
that ‘the existence of law is one thing, its merit or demerit another … ’ (Austin 1832,
Lecture 5). Any connections there might be between law and morality are contingent
only. There is nothing in the nature of law – as a social institution grounded in the will of
a sovereign or in fundamental social conventions – that guarantees its moral worth. It is
therefore possible to have profoundly immoral laws and wicked legal regimes. Aquinas’s
‘acts of violence’ may yet be valid law on the positivist model.
Despite a shared commitment to the social and separation theses, defenders of legal
positivism differ significantly in their understanding of these fundamental tenets. We
have already noted the split between Austin and his positivist descendants on the
tenability of the command theory as a proper account of the social thesis. Since H.L.A.
Hart’s devastating critique of Austin, most positivists assert that the true test of legal
validity lies not in the will of a sovereign but in something like Hart’s ‘rule of
recognition’, the fundamental social rule whose criteria determine the laws valid within a
particular legal system (Hart 1961) (see Hart, H.L.A.). The connection between Hart’s
rule of recognition and the social thesis is evident in the former’s conventional nature: its
existence and content are matters of pure social fact – matters of the criteria that are in
fact accepted by legal officials in their practice of establishing legal validity. The
connection with the separation thesis is likewise evident. There is nothing in the bare
notion of a social rule of recognition that guarantees the moral worthiness of the laws it
validates. More important, there is nothing which requires that moral worthiness be
included as a condition of legal validity. The accepted criteria can be as simple and
morally neutral as ‘Whatever the Queen in Parliament enacts is law’, or ‘Supreme Court
rulings constitute valid law’.
2 Dworkin’s challenge
Although contemporary positivists largely agree on how best to interpret the social thesis,
the same cannot be said of the separation thesis. In response to Ronald Dworkin’s
critique of Hart’s positivism (Dworkin 1978) (see Dworkin, Ronald), defenders of legal
positivism have divided into two major camps: ‘inclusive legal positivism’ (sometimes
referred to as ‘soft positivism’ or ‘incorporationism’) and ‘exclusive legal positivism’
(sometimes referred to as ‘hard positivism’ or ‘the sources thesis’). Defenders of the
latter include Joseph Raz, Andrei Marmor and Scott Shapiro; defenders of the former
include Hart, Jules Coleman, Matthew Kramer and Wil Waluchow. Among Dworkin’s
principal criticisms of Hart was that the latter’s ‘model of rules’ lacks the theoretical
capacity to account for the widespread use of principles within legal adjudication
(Dworkin 1978). Not only are such principles widely used, Dworkin maintained, but they
are treated by judges as binding law. Yet a principle is treated as binding law not because
it satisfies criteria of validity contained within a conventional rule of recognition, but
because (in the view of the judge who employs it), it expresses an ideal of justice,
fairness or due process – an ideal which clearly cannot be established independently of
substantive, and contestable, moral argument. So both the separation thesis and Hart’s
rendering of the social thesis in terms of his conventional rule of recognition are
incompatible with treating legal principles as binding norms. Hart must instead relegate
principles to the realm of non-legal standards to which judges may appeal, but need not
appeal, as they exercise their discretion to fill in gaps left by valid law (e.g., when a
relevant statute is indeterminate and no other legal source can be invoked to resolve the
indeterminacy). But this relegation, Dworkin urges, is something to be avoided. We must
therefore reject Hart’s model of rules – and the separation and social theses – in favour of
Dworkin’s interpretive theory of law, within which legality is partly determined by moral
principles which place legal practices in their best moral light (Dworkin 1978, 1986).
According to proposition A, which we might call the strong separation thesis, legality and
morality are necessarily separate from one another; moral argument can never be used to
determine what the law is, but only what it ought to be. According to proposition B,
which we might call the separability thesis, legality and morality are only separable, not
necessarily separate. The two can be brought together if the right conditions prevail, if,
e.g., a society’s rule of recognition includes conformity with a moral principle like
fairness as a condition of legal validity (see Justice §5). Inclusive positivists reject the
strong separation thesis but fully endorse the separability thesis. In response to Dworkin’s
claim that sometimes moral arguments figure in attempts to determine binding law,
defenders of inclusive positivism reply: Yes, but this is not necessarily so. Although there
is nothing in the nature of law (as characterized by the social and separability theses)
requiring the use of moral arguments to determine legality, there is nothing which
prohibits their use either – as Hart himself recognized. Hart was clear that the rule of
recognition can be as austere as ‘Whatever the Queen in Parliament enacts is law’, a rule
which separates legality from any and all moral conditions. But he was equally clear that
‘[i]n some systems [of law], as in the United States, the ultimate criteria of legal validity
explicitly incorporate principles of justice or substantive moral values’ (Hart [1961]
1994: 204). Such criteria form a rule of recognition in which the separable properties of
legality and morality are brought together, the one being a condition of the other.
So inclusive positivists reject the strong separation thesis and Dworkin’s assertion that it
reflects the theoretical commitments of legal positivism. Other legal positivists, most
notably Joseph Raz, agree with Dworkin on the theoretical commitments of legal
positivism and set out to defend their theory – exclusive positivism – both against
Dworkin’s critique and against the claim of their inclusive counterparts that the
separability thesis provides a sufficient account of positivism’s theoretical commitments.
In response to Dworkin, exclusive positivists accept that many principles do indeed
figure as binding law, but they maintain that the status of those principles is perfectly
explicable in terms of morally neutral criteria of validity which make no mention of
moral worth. Many legal principles come into existence through enactment in the
preambles to statutes and constitutions (Raz 1972). A principle can also ‘crystallize’ into
law when – and because – it is applied by judges in a sufficient number of cases. Yet
another exclusivist strategy is to distinguish between (a) moral criteria for legal validity,
and (b) legal rules which empower judges to appeal to non-legal norms – in this instance,
morality – to invalidate what are otherwise valid laws. The difference is as follows.
According to an inclusive positivist, it is perfectly possible that the due process clause
(fairness) of the USA’s constitution establishes a moral condition for legal validity. If it
does, then any statute or judicial decision that violates the principles of fairness is
actually not valid law in the USA. Since exclusive positivists are barred from allowing a
moral principle to serve in this way as a criterion of legal validity, they must construct an
alternative explanation of the due process clause. According to the exclusive positivist,
this clause does not incorporate the moral principle of fairness into the law as a condition
of validity; rather, it provides judges with a directed power to invalidate a statute or
precedent which, antecedent to the exercise of this power in a legal case, is perfectly
valid. The difference between the two accounts is analogous to the difference between a
void and a voidable contract. The former is of no force and effect; the latter can become
so, but only if the innocent party exercises their legal power to get the contract voided.
Until such time as the power to void is exercised, a voidable contract is binding.
According to inclusive positivism, conflict with the due process clause might mean that
the statute already is invalid (void). According to exclusive positivism, this conflict
means only that the statute is ‘subject to invalidation (voidable)’ – which is the only
interpretation consistent with the separation thesis (Raz 1980: appendix).
So defenders of exclusive positivism resolutely insist on the strong separation thesis, that
the moral worthiness of a normative standard can never figure among the conditions of its
status as a legally valid norm. In defence of this claim a number of arguments have been
put forward. The most powerful and influential is Joseph Raz’s argument that the
separability thesis undermines the law’s capacity to serve as a practical authority.
According to Raz, it is in the very nature of a legal system that it claims justified practical
authority over a population. If this claim is to be intelligible, a legal system must be the
kind of thing that is capable of exercising authority. But what is it to have authority? In
answer, Raz develops the ‘service conception of authority’ (Raz 1985a, 1985b). Consider
the following. Each of us is constantly faced with the question: What should I do? In
answering this question, we often appeal to reasons for doing one thing rather than
another. Typically, these reasons are either prudential or moral. Let us call these ‘first-
order reasons’. In many situations, we weigh the applicable first-order reasons against
each other and decide on the basis of the balance of reasons. In so deciding we view
ourselves as acting in accordance with ‘right reason’. Sometimes, however, reason
suggests that we should not ourselves attempt to act directly on the balance of first-order
reasons. This is where authority comes in. Sometimes right reason dictates that we act in
accordance with a second-order reason which reflects someone else’s judgement about
the relevant first-order reasons. Often the relevant second-order reason will be a directive
issued by an authority, say a financial counsellor if we are making investment decisions,
or a priest if we are attempting to answer a difficult moral question. On Raz’s account, Y
is justified (normally) in accepting X’s directives as authoritative when Y is more likely
to act in accordance with right reason by following them than by attempting to act
directly on the balance of first-order reasons. Authority provides a kind of second-order
‘exclusionary reason’ – a reason which excludes, and replaces, the relevant first-order
reasons. In Raz’s view, it is the normal role of an authority to provide this ‘service’ – to
issue such second-order reasons reflecting the proper balance of first-order reasons.
It follows from Raz’s account of authority that, if the law is to be capable of having the
authority it necessarily claims, its directives must be second-order reasons intended to
replace the first-order reasons which apply to its citizens. But why would one want a
legal system whose directives replace first-order reasons? There are two principal
reasons. First, sometimes we all have first-order reasons to contribute to the realization of
certain public goods – e.g., education, health care and security. Were each of us to act
independently in pursuit of such goods, the likely result would be total failure. Their
realization requires a kind of co-ordinated, collective action that is almost impossible to
achieve in the absence of public institutions created and governed by law. Thus we are
often better able to achieve our collective goals if we act on the basis of second-order
reasons established by publicly accessible and enforceable legal directives. When these
directives are well designed and administered, we are far more likely to achieve what
right reason demands of us – the provision of decent levels of education, health and
security – if we follow the law’s directives instead of trying to act individually on first-
order reasons without the law’s coordinating framework. In such circumstances we are
justified in submitting to the authority of law. A second reason for wanting a system of
legal directives is that the law sometimes does know best, or at the very least has the
time, energy and resources to determine what is best. Just as I might be justified in acting
on the advice of my financial advisor instead of attempting myself to divine the mysteries
of the market, I might be justified in being guided by the law’s attempt, on our behalf, to
discern the relevant requirements of right reason.
Now comes a crucial step in Raz’s authority argument. If authoritative legal directives are
to serve their role in guiding us on the path to right reason, the identification and
interpretation of them cannot depend upon our considering the first-order reasons they are
meant to replace. But this is precisely what happens, raz thinks, if we accept inclusive
legal positivism. A statute, as a second-order, authoritative directive is meant to replace
whatever reasons of prudence and morality apply to us in the situations over which the
statute governs. But if its identity as a valid legal directive depends on its conformity with
a moral principle like fairness – a moral reason – then its identity as an authoritative
directive will depend on the very reasons it was meant to exclude and replace. We will
have to deliberate on at least some of those first-order reasons to determine whether the
statute is valid and therefore authoritative. But if so, then we might just as well have
dispensed with the statute altogether and instead acted directly on the relevant first-order
reasons, including fairness. Hence, Raz concludes, inclusive legal positivism is
incompatible with the authority of law. Inclusive positivists have mounted vigorous and
multi-faceted replies to Raz’s authority argument. Coleman and Waluchow point out, for
example, that the moral reasons replaced by a legal directive need not be identical with
the moral factors invoked to challenge its validity (Coleman 1998; Waluchow 1994,
2000). In its landmark abortion case (R. v. Morgentaler [1985] Supreme Court Reports
30), the Canadian Supreme Court declared Canadian Criminal Code rules governing
abortions unconstitutional because they violated the moral rights to procedural fairness
guaranteed by the Charter. But whatever the first-order moral reasons underlying those
abortion rules might have been (e.g., a right to life versus rights to bodily integrity and
autonomy), they were not reasons of procedural fairness. Coleman further attempts to
meet Raz’s challenge by distinguishing between two functions of a rule of recognition: an
identification function and a validation function, suggesting that Raz’s authority
argument requires only that the rules for identifying valid law employed by ‘ordinary
folk’ make no reference to excluded moral reasons (Coleman 1996, 2001). Since a rule of
recognition is seldom used by any ordinary person to identify valid laws, the fact that it
might require appeal to excluded first-order reasons poses no threat to the law’s authority.
Waluchow, for his part, takes direct aim at Raz’s conception of authority, arguing that
there is nothing in the nature of an authoritative directive which precludes us from ever
consulting any of the relevant first-order reasons. It is possible that at least some of the
relevant first-order reasons might in various ways figure in attempts to determine the
identity and content of a directive which remains authoritative nevertheless (Waluchow
1994, 2000). Authoritative guidance can be partial guidance.
Although Raz’s authority argument remains the most powerful challenge to inclusive
legal positivism, a more recent challenge by Scott Shapiro has led to further divisions
within the inclusive camp. Shapiro argues that inclusive positivism is inconsistent with
the ‘practical difference thesis’, i.e., the claim that ‘in order to be law, authoritative
pronouncements must in principle be capable of making a practical difference: a
difference … in the structure or content of deliberation and action’ (Coleman 1998: 383).
Consider a rule of recognition which cites conformity with some moral principle as a
condition of the validity of a legal rule. We cannot, Shapiro argues, be guided by the
latter rule in the way in which authoritative directives are supposed to guide us. Owing to
the fact that we must consult moral principles, i.e., the first-order reasons the rule was
meant to replace, the rule cannot guide us. It can provide neither motivational nor
epistemic guidance and cannot, therefore, make a practical difference (Shapiro 1998a,
1998b). This is true, Shapiro argues, whether conformity with moral principle is taken to
be a necessary or a sufficient condition for legal validity. In response to Shapiro’s
argument, inclusive positivists have declared their allegiance to one or the other of the
two possibilities contemplated by Shapiro, thus introducing a further division among
defenders of legal positivism. According to Jules Coleman, only the ‘sufficiency’ version
of inclusive positivism is capable of parrying Dworkin’s original claim that principles are
sometimes treated as law if – and because – they express ideals of justice, fairness or due
process (Coleman 1998). Kramer and Waluchow disagree, arguing that the sufficiency
version in any sweeping form is untenable and that the ‘necessity’ version is more than
capable of meeting Dworkin’s challenge (Kramer 2000; Waluchow 2000). Where these
ongoing debates will ultimately lead is, at this stage, an open question. But wherever they
lead, the result is likely to be a far better understanding of the social and separability
theses than had been attained before Dworkin prompted positivists to consider more
deeply their own theoretical commitments.
Bibliography
Aquinas, Thomas (1266–73) Summa Theologica, various editions. (Classic natural law
text.)
Coleman, J. (1982) ‘Negative and Positive Positivism’, Journal of Legal Studies 11: 139.
(Coleman’s first reply to Dworkin’s critique of Hart’s legal positivism.)
Coleman, J. (1996) ‘Authority and Reason’, in R. George (ed.), The Autonomy of Law,
Oxford: Oxford University Press. (Coleman on Raz’s authority argument.)
Coleman, J. (2001) The Practice of Principle, Oxford: Oxford University Press. (An
elaboration of earlier arguments for inclusive positivism, with an attempt to ground them
in a version of philosophical pragmatism.)
Dworkin, R.M. (1978) Taking Rights Seriously, London: Duckworth. (A set of essays
including Dworkin’s original critique of Hart’s legal positivism together with a thumbnail
sketch of Dworkin’s alternative and a ‘Reply to Critics’.)
Dworkin, R.M. (1986) Law’s Empire, Cambridge, MA: Harvard University Press.
(Dworkin’s fully developed alternative to legal positivism, ‘law as integrity’.)
Green, L. (1988) The Authority of the State, Oxford: Oxford University Press. (An
excellent discussion of authority in the context of legal and political philosophy, with a
defence of exclusive positivism.)
Green, L. (1996) ‘The Concept of Law Revisited’, Michigan Law Review 94: 1687. (A
critical discussion of Hart’s Postscript to the second edition of The Concept of Law,
advancing the thesis that Hart’s theoretical commitments should have led him away from
inclusive and towards exclusive positivism.)
Hart, H.L.A. (1958) ‘Positivism and the Separation of Law and Morals’, Harvard Law
Review 71: 593. (Hart’s early thoughts on legal positivism and the separability thesis).
Hart, H.L.A. (1961) The Concept of Law, Oxford: Oxford University Press, 1994. (The
twentieth century’s most powerful and influential statement of legal positivism, with – in
the 1994 edition – a posthumous Postscript containing Hart’s final response to Dworkin’s
challenge and an expression of his commitment to inclusive positivism.)
Hart, H.L.A. (1982) Essays on Bentham: Jurisprudence and Political Theory, Oxford:
Oxford University Press. (Essays that include Hart’s reactions to Raz’s account of the
authority of law and to Dworkin’s early critique of legal positivism.)
Himma, K. (2000) ‘H.L.A. Hart and the Practical Difference Thesis’, Legal Theory 6: 1.
(Critique of Scott Shapiro’s argument that inclusive positivism violates legal positivism’s
commitment to the practical difference thesis.)
Kramer, M. (2000) ‘How Moral Principles Can Enter Into the Law’, Legal Theory 6: 83.
(A defence of inclusive positivism against Scott Shapiro’s argument that inclusive
positivism violates legal positivism’s commitment to the practical difference thesis.)
Leiter, B. (1998) ‘Realism, Hard Positivism and Conceptual Analysis’, Legal Theory 4:
533. (A defence of exclusive legal positivism and a vigorous critique of Coleman’s
arguments.)
Lyons, D. (1977) ‘Principles, Positivism and Legal Theory’, Yale Law Journal 87: 415.
(An early discussion of Dworkin’s claim that legal positivism lacks the theoretical
resources to account for legal principles; among the first to urge that Dworkin’s critique
was based on the unwarranted assumption that Hart is committed to what became known
as exclusive legal positivism.)
Mackie, J. (1977) ‘The Third Theory of Law’, Philosophy and Public Affairs 7: 3.
(Critical discussion of Dworkin’s early theory of law and of his critique of Hart; among
the first to urge that Dworkin’s critique was based on the unwarranted assumption that
Hart is committed to what became known as exclusive legal positivism.)
Marmor, A. (1992) Interpretation and Legal Theory, Oxford: Oxford University Press.
(A defence of exclusive positivism that includes an analysis of the theory of legal
interpretation upon which Dworkin builds his theory of law as integrity.)
Raz, J. (1972) ‘Legal Principles and the Limits of Law’, Yale Law Journal 81: 823. (An
early critique of Dworkin’s claim that Hart’s legal positivism fails to accommodate legal
principles.)
Raz, J. (1979) The Authority of Law, Oxford: Oxford University Press. (A collection that
includes a number of influential essays on the theoretical commitments of legal
positivism and on the nature of legal authority.)
Raz, J. (1980) The Concept of a Legal System, Oxford: Oxford University Press, 2nd
edn. (Raz’s first book with a later appendix that contains an explanation, from the point
of view of exclusive positivism, of the role of moral principles in arguments concerning
legal validity.)
Raz, J. (1985a) ‘Authority and Justification’, Philosophy and Public Affairs 14: 3. (A
further development of Raz’s theory of authority and his argument that only exclusive
positivism is compatible with legal authority.)
Raz, J. (1985b) ‘Authority, Law and Morality’, The Monist 68: 295. (A further
development of Raz’s theory of authority and his argument that only exclusive positivism
is compatible with legal authority.)
Shapiro, S. (1998a) ‘On Hart’s Way Out’, Legal Theory 4: 469. (A critical discussion of
Hart’s inclusive positivism and a defence of the claim that inclusive positivism robs law
of its capacity to make a practical difference in deliberations about our reasons for
action.)
Shapiro, S. (1998b) ‘The Difference that Rules Make’, in Brian Bix (ed.), Analyzing
Law, Oxford: Oxford University Press. (A further discussion of inclusive positivism and
the practical difference thesis.)
Shiner, R. (1992) Norm and Nature: The Movements of Legal Thought, Oxford: Oxford
University Press. (An argument that the conflicts between legal positivism and natural
law theory or ‘anti-positivism’ are irresolvable, with detailed discussions of exclusive and
inclusive positivism and Dworkin’s theory of law as integrity.)
Soper, E.P. (1977) ‘Legal Theory and the Obligation of a Judge: The Hart/Dworkin
Dispute’, Michigan Law Review 75: 473. (One of the earliest essays to urge that
Dworkin’s critique of Hart is based on the unwarranted assumption that Hart is
committed to what became known as exclusive legal positivism.)
Waluchow, W.J. (1985) ‘Herculean Positivism’, Oxford Journal of Legal Studies 5: 41.
(A defence of a Hartian version of inclusive legal positivism in response to Dworkin’s
challenge.)
Waluchow, W.J. (1989) ‘The Weak Social Thesis’, Oxford Journal of Legal Studies 9:
23. (A defence of inclusive positivism and a critique of Raz’s arguments for exclusive
positivism – particularly Raz’s argument that only exclusive positivism is compatible
with the law’s claim to authority.)
Waluchow, W.J. (1990) ‘Charter Challenges: A Test Case for Theories of Law’,
Osgoode Hall Law Journal 29: 183. (An argument that inclusive positivism provides the
best theoretical account to challenges to legal validity based on an appeal to moral
principles.)
Waluchow, W.J. (1994) Inclusive Legal Positivism, Oxford: Oxford University Press.
(The first book-length defence of inclusive positivism against Dworkin’s challenge and
Raz’s exclusive positivism.)
Waluchow, W.J. (2000) ‘Authority and the Practical Difference Thesis: A Defense of
Inclusive Legal Positivism’, Legal Theory 6: 45. (A reply to various critics of Inclusive
Legal Positivism and to Scott Shapiro’s argument that inclusive positivism violates legal
positivism’s commitment to the practical difference thesis.)