Citation: 11 Oxford J. Legal Stud. 174 1991
Content downloaded/printed from
HeinOnline (http://heinonline.org)
Wed Mar 23 11:01:51 2016
-- Your use of this HeinOnline PDF indicates your acceptance
of HeinOnline's Terms and Conditions of the license
agreement available at http://heinonline.org/HOL/License
-- The search text of this PDF is generated from
uncorrected OCR text.
-- To obtain permission to use this article beyond the scope
of your HeinOnline license, please use:
https://www.copyright.com/ccc/basicSearch.do?
&operation=go&searchType=0
&lastSearch=simple&all=on&titleOrStdNo=0143-6503
Dworkin and Legal Pragmatism
ROBERT WESTMORELAND*
Law as integrity, Ronald Dworkin's theory ofjurisprudence, often requires judges
to make highly controversial decisions that have no roots either in popular morality
or in the understanding of the authors of the black-letter law which judges are
pledged to interpret. Yet Dworkin contrasts law as integrity sharply with pragmatism,
a rogue 'activist" jurisprudence that gives judges a free hand to make rather
than interpret law. Part I of this article sets out and endorses part of the general
understanding of legal theory that grounds law as integrity. Part II sketches
pragmatism and law as integrity. Integrity is said to be a political virtue that
grounds a commitment to equal treatment deeper even than that required by
justice. Part III tests and rejects the claim that only integrity condemns the sort of
unequal treatment exemplified by 'checkerboard' statutes intended to avert the
total defeat of justice by partially enacting conflicting principles of justice. The
failure of this claim threatens the distinction, at least in cases about constitutional
rights, between law as integrity and the sort of pragmatism that demands that
courts do justice unmediated by institutional history. The distinction collapses in
Part IV where it is shown that no institutional constraints impede Hercules, law as
integrity's mythical superjudge, in the pursuit of justice simpliciterin constitutional
rights cases. I concentrate on such cases because oftheir importance-for Dworkin
and many others, constitutional rights trump all other legal claims-and because
they most clearly reveal the problems of law as integrity, though I refer occasionally
to statutory law. Part V argues against the view that integrity best accounts for
prevalent ideas about political legitimacy and the general obligation to obey the law
within Anglo-American political culture. I defend the argument from fair play as
the better account of these features of political life, and suggest that Dworkin's
portrait of obligation within the community of integrity fits the pragmatism of
justice better than it fits any legal theory that takes fidelity to law seriously.
I Interpretation
One of Dworkin's important contributions to philosophy of law is the insight that
legal theory is 'interpretive' not semantic (chs 1-3). A philosophically significant
* Assistant Professor of Philosophy, University of Mississippi, USA. I am grateful to Gerald J. Postema and a
referee of this journal for their comments on earlier drafts of this article.
I Law's Empire (Cambridge: Harvard University Press, 1986), 378. All page and chapter references in parentheses
in the text are to this book.
DOxford University Press 1991 Oxford Journal of Legal Studies Vol. 11, No. 2
Dworkin andLegal Pragmatism
legal theory is not an empirical exercise that uncovers linguistic criteria governing
the use of legal terms. Those who think it is wrongly assume that unless we all
apply the same semantic rules to our legal terms we will talk past each other, as we
would in 'an argument about banks when one person has in mind savings banks
and the other riverbanks' (44). Lawyers perennially disagree about the nature of
law and legal interpretation: some for instance say that what the law requires is
always an empirical question; others insist that discerning (not creating) the law's
requirements engages moral principles as well as institutional history. Yet their
disagreement seems real.
It is real, argues Dworkin. Unlike the 'banks' case, the disputants give
competing interpretations of roughly 'the same objects and events'-the Constitution,
federal and state statutes, judicial decisions, etc (46). Furthermore,
disputes about what makes these law are ultimately substantive moral and political
disagreements about the point and function of law. The positivist view that
discovering legal rights and obligations is an empirical activity usually relies, not on
linguistic considerations, but rather on the conception of law as (in Joseph Raz's
words) 'a public measure by which one can measure one's own as well as other
people's behaviour'2
without reference to endlessly contested moral principles.
Dworkin's brand of 'natural law' theory rejects positivism out of a commitment to
individual rights rather than on linguistic or purely conceptual grounds. Now if
such normative judgments were only (eg) disguised statements of or expressions (in
the emotivist sense) of speakers' feelings about the law, their disagreement would
not be real. But Dworkin the moral realist thinks that these judgments concern
something that can be got right or wrong-namely, the question of which
normative theory puts the history of the institution in its best light. Legal
philosophy is a dialectical process aimed at approximating such a conception of
law.
Dworkin calls the attempt to frame the substantively best conception of law that
fits the institutional record interpretation.Two points need emphasis here. First,
sound interpretation is said to succeed along the dimension of fit as well as
substance, for it interprets an existing legal system. Second-I shall attack law as
integrity on this point in Part IV-for Dworkin interpretation is a 'constructive'
(52) not a 'conversational' (50) process. You do not discover the point or function
of law by polling participants-judges, lawmakers, or citizens. Beyond the most
primitive stages of 'runic' traditionalism participants themselves assume an
interpretive attitude toward law and disagree about law's purpose (47). Law no less
than art, though a product of its authors, is 'an entity distinct from them', so their
views about its point are not decisive. Though authors' intent 'provides theformal
structure for all interpretive claims' (58), the purposes an interpreter ascribes to the
law are those he or she considers sound, 'not (fundamentally) those of some author'
(52). But again, these purposes must fit the pre-interpretive data. Thus, presumably,
no interpretation that posited aesthetic advancement as the main purpose
2 The Authority ofLaw (New York: Oxford University Press, 1979), 51.
SUMMER 1991
OxfordJournalofLegalStudies
of US law could pass muster, because it could not make sense of enough of that
data.
For Dworkin the judge's task no less than the legal philosopher's is 'constructive'.
The question for law as integrity is whether this makes institutional
history so protean that fit and thus the distinction between legislation and
interpretation become meaningless. To answer that question I must first set out
Dworkin's account of pragmatism and law as integrity.
If TheoriesofLegal Interpretation
A. Pragmatism
Pragmatism is said to be law as integrity's fiercest competitor. It denies that 'past
political decisions in themselves provide any justification for either using or
withholding the state's coercive power' (151). With no duty of fidelity to law, the
judge pursues some political virtue like justice or utility 'liberated from the dead
hand of the past and the fetish of consistency for its own sake' (ibid).
Pragmatism ... denies that people ever have legal rights; it takes the bracing view that
they are never entitled to what would otherwise be worse for the community just because
some legislature said so or a long string of judges decided other people were. (152)
There are as many kinds of legal pragmatism as there are moral theories: eg, a
pragmatism ofjustice, various utilitarian pragmatisms, eudaimonistic pragmatism.
I am most interested in the pragmatism of justice because ofits appeal to aggressive
neo-Kantian rights liberals like Dworkin.
Pragmatism of any kind would seem to misdescribe our legal practice, since
judges labour to fit their decisions to the black-letter law. But the wise pragmatist
will recognize more than one political virtue, even if the others are only
instrumental to the pursuit of some ultimate value like justice. A pragmatist of
justice who enforced only those statutes he approved of would defeat himself by
undermining the social co-ordination and stability that are preconditions ofjustice.
So he will act as if people have legal rights flowing from past political decisionsperhaps
nearly as often as some non-pragmatist judges-if those decisions are clear
enough to raise definite expectations among citizens. But if the only good reason
for 'respecting' statutes and precedents is this instrumental one, there is no point in
as-if rights when the scope and meaning of these decisions is unclear. Here 'the
right rule is whichever rule is best for the future' and the judge could have only 'an
indirect, noble-lie reason' for pretending to divine the "'true" ground' of past
decisions (158).
Now judges typically show a 'constant and relentless concern... for explicating
the "true" force of a statute or precedent decision when that force is problematical'
(157-8). Even if they are pragmatists they have no noble-lie reasons for dissembling
in such cases, for 'the public will not be outraged if it is told that precedents
will be confined to their facts' (159). So pragmatism may not describe legal practice
VOL. 11
Dworkinand LegalPragmatism
very well after all, says Dworkin. It is a sound legal theory only if its political
attractiveness offsets its weakness along the dimension offit.
Dworkin finds it quite attractive. When the black-letter law loses its coordinative
value, why should justice yield to some notion of doctrinal consistency?
A persistent theme of Dworkin's is his aggressive advocacy of rights of justice even
when these seriously hinder the pursuit of collective goals, and even when there is
strong and principled popular opposition to the idea that the particular rights (eg to
abort even for trivial reasons) exist. Especially in constitutional cases, Dworkin
thinks the task of the courts is to enforce the rights of individuals against the
majority. Dworkin finds the claim that judges should simply do justice seductive
(see 153).
Nevertheless Dworkin holds that the descriptive failure of pragmatism suggests
a deep interpretive failure: disregard for a political virtue said to ground
'distinctively legal rights [understood] as trumps over what would otherwise be the
best future [even from the standpoint of justice] properly understood' (160).
Integrity is said to be the virtue that makes judges responsible to legal history as
well as justice. I turn now to Dworkin's account of this virtue and its attendant
legal theory.
B. Law as Integrity
Integrity is said to be a warmer, more communitarian virtue than justice that
secures 'a kind of equality among citizens that makes their community more
genuine and improves its moral justification for exercising the political power it
does' (96). Integrity secures this deep kind of equality by requiring that 'government
speak with one voice' and 'extend to everyone the substantive grounds of
justice or fairness it uses for some' even if the resulting legal standards are less than
ideal (165). 3
Thus it supposedly restrains the pursuit of justice by judges and
lawmakers. It forbids the application by courts of principles of justice (such as one
requiring the rich to share their wealth), however attractive, that are ungrounded
in the black-letter law, for this would manifest less than equal concern for plaintiffs
in similar cases to whom the principle is not applied. Integrity condemns as well
both 'different laws each of which is coherent in itself, but which cannot be
defended together as expressing a coherent ranking of different principles of justice
or fairness or procedural due process' (184), and individual statutes which express
conflicting principles of justice.
What if anything integrity keeps judges from doing in the way of justice is a
question for Part IV. First I want to explore the claim, important to Dworkin, that
integrity alone forbids legislation which in the name of justice speaks with two
voices.
3 Though integrity is said to regulate the application of principles of fairness and 'procedural' due process as well as
justice, I focus on its alleged impact on justice.
SUMMER 1991
OxfordJournalofLegal Studies
III CheckerboardStatutesand EqualTreatment
Dworkin calls the two-voiced laws which split the difference between competing
conceptions of justice 'checkerboard statutes'. If for instance the community were
evenly divided about abortion, a checkerboard statute might prohibit only
abortions for women born in odd-numbered years. Such a statute would be fair,
says Dworkin, because it accurately reflects the community's division of opinion,
and fairness recommends 'that each person or group in the community should have
a roughly equal share of control over the decisions made by Parliament or Congress
or the state legislature' (178). Dworkin thinks justice too would be served in that
each side thinks that fewer injustices would be done under the checkerboard
statute than if the opposition prevailed. A group has sound reasons ofjustice for
condemning all such compromises only if it is a perduring majority of opinion that
can count on winning most or all winner-take-all votes. Dworkin denies that
discrimination based on birthdate in the abortion case is arbitrary from the
standpoint of justice, for its point is precisely to minimize the number of injustices
committed (see 180-1).
Since most of us would rule out such compromises in principle even if we
thought them just and fair (in Dworkin's sense of those terms), it is supposedly
integritythat moves us to hold
that a state that adopts these internal compromises is acting in an unprincipled way, even
though no single official who voted for or enforces the compromise has done anything
which, judging his individual actions by the ordinary standards of personal morality, he
ought not to have done. The state lacks integrity because it must endorse principles to
justify part of what it has done that itmust reject to justify the rest. (183-4)
Integrity for Dworkin is an emergent political virtue that 'commands that no one
be left out, that we are all in politics together for better or worse, that no one may
be sacrificed, like wounded on the battlefield, to the crusade for justice overall'
(213).
The idea that justice does not condemn checkerboard statutes depends on a
curious maximizing view of justice which appears nowhere else in Dworkin's work,
is inimical to his Rawlsian conception of justice as bound up with a foundational
right to 'equal concern and respect',4 and shows up here I think only to give
integrity some work to do. This maximizing conception says that if (eg) abortion is
unjust, the number of abortions should be minimized, by checkerboard statute if
necessary. It is an example of what Robert Nozick calls a 'utilitarianismn of justice'
that replaces happiness with 'some condition about minimizing the total
"weighted" [according to seriousness] amount of violations of rights'5 as the
desirable end state. It would in some circumstances require the state to convict the
innocent when doing so would prevent more injustices overall than it would
produce. Even worse for Dworkin is the fact that integritywould condone and even
requiresuch a policy if the state embraces a maximizing conception of justice. For
4 See TakingRights Seriously(Cambridge: Harvard University Press, 1977), chs 6, 12.
5 Anarchy, State, andUtopia(New York: Basic Books, 1974), 28.
VOL. 11
DworkinandLegal Pragmatism
in that case the policy represents the application ofa single maximizing principle of
justice rather than a checkerboard compromise between two inconsistent principles
of justice. Assuming again a maximizing conception of justice, the checkerboard
abortion statute honours integrity for similar reasons. If justice demands the
maximization of just outcomes, and if political circumstances are such that the
alternative to the checkerboard statute is one that would maximize injustice where
abortion is concerned, then oneprinciple of justice endorses the different treatment
of women born in odd- and even-numbered years. Where a maximizing conception
fits the existing law best, the state would violate integrity by giving voice to a
non-maximizing principle ofjustice that forbids checkerboard statutes. Now ifthe
state embraces a non-maximizing conception of justice-as Hercules' neo-Kantian
state surely does-it will condemn the checkerboard statute on the ground that it
makes us use ourselves as well as others in ways that offend the inviolability of
persons. But whether or not the state accepts a maximizing conception of justice, it
is justice, not 'integrity', that determines whether or not such unequal treatment is
condoned. All this means that as yet we have no evidence that integrity restrains
Hercules' pursuit of justice, since justice by itself explains his rejection of
checkerboard statutes. But might not Hercules' bold rights-oriented ideal of
adjudication prevail even if it loses this particular battle? After all, the best
interpretation of the black-letter law engages the soundest conception of justice,
which for Hercules the neo-Kantian is a non-maximizing one. So does not law as
integrity in the end endorse the kind of equality that condemns checkerboard
compromises?
Maybe it does; but it in no way follows that law as integrity requires fidelity to
institutional history, which is supposed to be the crucial difference between law as
integrity and the pragmatism of justice. Justice demands consistency, but of a
different sort from that demanded by fidelity to law. It is tempting to conflate the
two, for practical reason, to which justice is basic, 'transcends tenses as well as
persons';6
an action required by justice yesterday is required today and tomorrow,
assuming sufficiently similar circumstances. But the similarity among these duties
is not due to fidelity to any history analogous to the black-letter law; justice today
requires an action similar to yesterday's simply because that sort ofaction is correct
from the perspective of justice, not because of any moral analogue of precedent.
This is respect for reason and for persons, not for history. It gives us no reason to
think that the respect for equality and consistency that inspires Hercules' rejection
of checkerboard statutes springs from respect for institutional history rather ihan
background justice.
IV MakingLegalHistory
A closer look at Hercules' attitude toward institutional restraints on the pursuit of
justice further undermines law as integrity's claims of fidelity to law. I consider his
6 J. R. Lucas, OnJustice(New York: Oxford University Press, 1980), 42.
SUMMER 1991
OxfordJournalofLegalStudies
approach to several traditional restraints, and then close Part IV with an
examination of first his general views about the role of legal history in constitutional
rights cases, and then what I take to be his odd notion of respect for
constitutional language.
A. OriginalIntent
The intent of the framers, ratifiers, and/or authors of constitutional precedents
does not constrain Hercules even when it is clear. (I focus on framers' intent.) The
main case for original intent is the relative predictability and certainty it affords.
These are important not just for the sake of efficiency, but also as part of the
classical liberal ideal of law as placing strict and well-defined limits on the power of
officials (including judges) as well as of private citizens and groups. Dworkin
thinks that certainty and fixity, though important in, eg, rules of the road, are
'relatively unimportant' in constitutional cases where fundamental rights are at
stake (367); in such cases 'Fitting what judges [and lawmakers] did is more
important than fitting what they said' (248). What they did and what they think
they did may have only the most tenuous relation, though the words of the
Constitution (and statutes) they frame are said to be canonical. (We shall see about
that in IV E.)
Original intent as usually understood concerns what they thought they did. This
can be taken several ways. The first is that judges may only make decisions shown
by the historical record to have been explicitly endorsed by the framers. A less
restrictive variation of this view allows decisions based on defensible counterfactual
statements about how the framers would have decided cases they did not actually
envisage. Dworkin dismisses these views; original intent cannot refer, he says, to
'some particular conscious thought wielding its baton in an author's mind when he
said or wrote or did what he did' (55). The framers ofthe equal protection clause of
the fourteenth amendment, for instance, could have had in mind only a few of the
sorts of cases falling within its ambit, so original intent in this sense cannot exhaust
the meaning of the clause.
Important proponents of original intent agree. Robert Bork thinks that 'In such
a narrow form the philosophy is useless ... no intentionalist of any sophistication
employs the narrow version just described'. 7
Bork summarizes his own conception
of original intent as follows:
all an intentionalist requires is that the text, structure, and history of the Constitution
provide him not with a conclusion but with a major premise. That premise states a core
value that the framers intended to protect. The intentionalist judge must then supply the
minor premise in order to protect the constitutional freedom in circumstances the
Framers could not foresee... we are usually able to understand the liberties that were
intended to be protected. We are able to apply the first amendment's Free Press Clause to
the electronic media and to the changing impact of libel litigation upon all the media; we
are able to apply the fourth amendment's prohibition on unreasonable searches and
7 'The Constitution, Original Intent, and Economic Rights', (1986) 23 San DiegoL Rev 823, 826.
VOL. 11
Dworkin and LegalPragmatism
seizures to electronic surveillance; we apply the Commerce Clause to state regulations of
interstate trucking.8
On this approach courts 'must not hesitate to apply old values to new
circumstances', and in doing so are not confined to probabilistic statements about
what the framers would have decided. But Bork's approach is said to place 'Entire
ranges ofproblems... off-limits to judges' by confining them to application of'the
principles the Framers put into the Constitution'. 9
Does it? The question arises
because this approach requires some constructive interpretation of the Constitution:
the problem before the court in the electronic media case is not how the
framers would have decided the case, but whether the best interpretation of the
free press clause covers those media. Though the major premise is said to be
provided by the framers, the rest of the syllogism is the court's. Where does
constructive interpretation stop, once some reasons and standards not derived from
the framers' intent are deemed constitutional?If the first version of original intent is
'so narrow' as to be 'useless', what non-constructive way is there to know whether
and how a premise put into the Constitution by the framers applies? And since the
rest of the syllogism is constructive, what becomes of original intent? Bork
acknowledges in addition that constitutional 'values and principles can be stated at
different levels ofabstraction'; 10
the equal protection clause, for instance, might be
understood to condemn far more than racial discrimination. It could be understood
to forbid any statute-say, one that prohibits private consensual homosexual
acts-taken by the court to violate something like a Rawlsian right to equal concern
and respect. Were it objected that the fourteenth amendment was framed in full
knowledge of the existence of sodomy statutes, the court could appeal to Bork's
determination to 'apply old values to new circumstances' and claim that those
circumstances include a new and more sympathetic understanding of homosexuality.
The court might even use Bork's appeal to sensitivity to the structure
and history of the Constitution to glean a privacy right from the rights explicit in
the text as well as from various constitutional precedents, as the Supreme Court in
fact did in Griswoldv Connecticut.11
These are just the sorts of judicial cadenzas Bork wants to avert. I think he
would do so as follows. He would block the ascent to the privacy right by restricting
courts to the application ofrights explicitly enacted in the Constitution. Extension
of provisions like the equal protection clause to overturn homosexual sodomy
statutes would be possible only if the historical record showed that the framers
understood the clause to apply to matters of sexual orientation. 12
If they did then
presumably the court could apply the clause in a controversial way only ifthe court
is genuinely unsure about what the framers would have said. But that seems to be
the only sort of case where Bork would allow constructive interpretation.
8 Ibid.
9 Ibid, 827.
10Ibid.
1 381 US 479 (1965).
12 Robert H. Bork, op cit, above, n 7, 828.
SUMMER 1991
OxfordJournalof LegalStudies
There is no question that law as integrity rejects Bork's as well as more
restrictive versions of original intent. For Hercules even the framers' beliefs about
the sorts of cases they didhave in mind are not decisive. Neither their expectations
nor their hopes about how the courts will interpret, say, the equal protection clause
determines its meaning: their expectations, because they might fearor regretwhat
they think judges will do; their hopes, because these might spring from illicit
concerns. (They might oppose extending the clause to discrimination based on
national origin because they are xenophobes.)
What about their legitimate hopes? Suppose Hercules thinks their concrete
convictions about what equal protection means in particular cases are respectable
though in his view less than the best? Dworkin makes two points here. First he
argues that if we are to respect the words of the clause we must assume that the
framers' dominantintention was that justice be done, ie, that courts apply the best
interpretation of equal protection, which (they must admit if they were moral
realists) is not necessarily the one they think best. Thus to honour the intent of the
framers Hercules must depart from their convictions about equal protection when
his are different. But what if it is clear that the framers thought these concrete
convictions decisive? To enforce them for that reason 'would beg the question'
whether constitutional interpretation is constructive or conversational (364).
Hercules' answer to this question is that interpretation is constructive, and that
he must enforce what he considers the best conception of equal protection even if
the framers would think he misunderstands what they did in drafting the clause.
Hercules regards framers' statements about what equal protection means as
'political events' that are no more decisive than the views of others, officials (eg
authors of constitutional precedents) or non-officials (316). Consider here
Hercules' understanding of statutes:
Hercules interprets not just the statute's text but its life, the process that begins before it
becomes law and extends far beyond that moment. He aims to make the best he can of this
continuing story, and his interpretation therefore changes as the story develops. He does
not identify particular people as the exclusive 'framers' of a statute and then attend only to
their hopes or expectations or concrete convictions... (348)
This goes doubly for constitutional provisions like the equal protection clause,
which are especially 'sensitive ... to time' (365) because they were framed 'long
ago, when popular morality ... and almost everything else was very different'
(364). Constructive interpretation in the service of justice is not restrained by
original intent.
B. ProtectedExpectations
Another way to restrain courts is where possible to bind them to established
interpretations of the law. Though the extent to which such settled interpretations
exist is subject to dispute (see ch 4), there is no doubt that this doctrine can curb
judicial power: it would stop the Supreme Court from finding the death penalty
VOL. 11
DworkinandLegalPragmatism
unconstitutional, for instance. There is also no doubt that law as integrity dismisses
such expectations in constitutional rights cases: it rejects this sort of 'conversational'
interpretation as much as it does the others, and, as seen above, it virtually
denies the importance of fixity and certainty where fundamental rights (constructively
defined) are in play.
C. PopularMorality
The view that popular morality should govern the interpretation of constitutional
provisions containing abstract moral language is influential and represents a
potentially substantial discipline on the courts. The strictest version of this view
would allow courts, eg, to overturn a death penalty statute on the ground that it is
cruel and unusual, if but only if that punishment utterly offended contemporary
standards. A less restrictive version, accepted by the Supreme Court in Gregg v
Georgia,13
directs judges to place great but not conclusive weight on contemporary
standards. Law as integrity entirely rejects even this second, qualified requirement
of deference to popular morality. Since it understands constitutional rights to be
rights against the majority, the case for fairness, ie, for deferring to majority
opinion, 'disappears when serious constitutional rights are in question' (257).
D. FitandJustice
Law as integrity edges toward the pragmatism of justice when it dismisses the
particular restraints on judicial power discussed above. But could there be others it
does recognize? Dworkin's general remarks about legal history emphatically
suggest not.
A decision judged superior from the standpoint of justice can hardly fail to meet
the demands of constructive interpretation in hard constitutional rights cases. How
does Hercules distinguish a decision that best justifies institutional history from
one that has 'no purchase in American history and culture' but reflects only 'the
views of a local and transient political majority'? (337). We know that he cannot
appeal to the idea that the latter decision clashes with the concrete convictions of
the authors of the black-letter law (which law includes constitutional provisions,
statutes, and precedents) the decision supposedly must fit. We know too that he
cannot appeal to its inconsistency with popular morality. The meaning of the
tradition Hercules interprets is determined by his background theory of justice.
Law as integrity
begins in the present and pursues the past only so far as and in the way its contemporary
focus dictates ... It does not aim to recapture, even for present law, the ideals or
practical purposes of the politicians who first created it. (227)
Hercules has a heads-I-win-tails-you-lose approach to history:
13 See 428 US 153 (1976), 179-87.
SUMMER 1991
OxfordJournalofLegalStudies
His arguments embrace popular conviction and national tradition whenever these are
pertinent to the sovereign question, which reading of constitutional history shows that
history overall in its best light. (398)
Tradition matters when it supports the interpretation deemed best by Hercules'
background conception of justice.
Constitutional history is especially malleable; since the Constitution is foundational,
its 'interpretation ... must be foundational as well... [It] must be drawn
from the most philosophical reaches of political theory' (380). And since the
Constitution 'must fit and justify the most basic arrangements ofpolitical power in
the community', institutional support for bold new interpretations of abstract
constitutional clauses can come from the far reaches of the law. The purposes
disclosed by legal history are Hercules' own, so any contemporary trend that fits
his conception of justice can claim institutional support. Fit is in thrall to justice. Justice
Stephen Field's dissent in Munn v Illinois, which anticipated the
laissez-faireera of the Supreme Court early this century, is instructive here. In
language strikingly like that of Justices Blackmun and (in a concurring opinion)
Douglas in Roe v Wade, Justice Field wrote that state regulation of freight and
grain monopolies violated principles expressed-constructively-in the equal protection
clause, as well as in every statute and court decision protecting property
rights and 'freedom to go where one may choose ... as his judgment may dictate
for the promotion of his happiness'. 14
In Lochner v New York, 15
an equally broad
array of law was taken to support the principle that struck down regulations of the
working hours ofbakery employees. In his Lochner dissent Justice Holmes charged
that the Court ignored much established law that seemed perfectly consistent with
such regulations. But law as integrity's rejection of any 'conversational' interpretation
of legal history, and its reconstruction of that history to meet the present
demands of justice, guaranteed for the Lochner principles that purchase in constitutional
tradition needed to pass the threshold of fit; their substantive attractiveness
did the rest.
E. CanonicalLanguageandJustice
Even if Hercules' reading of a given constitutional clause is not responsible to a
substantial part of institutional history, it might at least be responsible to the
language of the clause itself. Perhaps the language of, eg, the due process clause of
the fourteenth amendment forces Hercules to restrict the application of the clause
to cases about procedural issues even though its meaning is not influenced even in
part by any other aspects of the institutional record. If law as integrity does
consider constitutional language canonical in this way it salvages a rump conception
of fit and so a semblance of a distinction between itself and pragmatism.
Whether or not it does respect constitutional language is debatable. Justicenon-maximizing
justice-is the foundational value of Hercules' background
14 94 US 113 (1877), 142.
15 198 US 45 (1905).
VOL. 11
Dworkin andLegal Pragmatism
political theory. Since that theory already recognizes the various rights specified by
the Constitution, Hercules is never tempted not to enforce them. And we have seen
that law as integrity's protean conception of fit lets Hercules discover in the
Constitution whatever background rights his theory of justice endorses. If constitutional
rights language requires nothing that justice forbids, and if that language
forbids the discovery in the Constitution of nothing that justice requires, what
duties does it really impose? What in the way of justice does it force Hercules to
sacrifice in the name of fidelity to law?
Hercules would protest that this argument rests on the flawed view that a duty
exists only if it requires actual sacrifices, only ifyou are seriously tempted to violate
it. If this notion of duty were sound, one whose sentiments and sense of duty both
recommended, say, truth-telling would have no duty16
to do so. That is a bizarre
conception ofduty. Hercules would argue that constitutional language iscanonical,
since it demands enforcement of explicit constitutional rights because they are in
the constitution, not (solely) because they are background moral rights. Thus on
his view even a judge whose background morality placed little weight on the rights
enshrined in the words of the Constitution would be duty-bound to enforce those
rights.
But how potentially confining is this duty for such a judge? Suppose Max is an
act-utilitarian judge who accepts law as integrity. (His reasons for doing so will not
be Hercules' neo-Kantian ones, but it is open to him to 'interpret' integrity
differently.) How would the individual rights explicit in the Constitution slow his
pursuit of utility maximization?
Law as integrity might force Max to deploy some conception of rights in his
interpretive theory, since the Constitution contains explicit rights and since law as
integrity allegedly respects constitutional language. But law as integrity is constructive
and does not demand an interpretation of those rights as foundational1 7
in
the Kantian sense endorsed by Hercules' background conception of justice. Max
would consider the rights explicit in the Constitution as rules of thumb of
considerable weight because of their conduciveness to utility maximization.18
But
when he is convinced in a particular case that the enforcement of the free speech
right would hinder the pursuit of utility then that collective goal will prevail.
If the duty to respect constitutional rights language burdens not even the
act-utilitarian, is it a duty at all? Maybe. Utilitarianism is, after all, within the
mainstream of the Anglo-American political tradition; perhaps then it should come
as no surprise that it could be constructively incorporated into the foundatiotis of
the constitution. So long as law as integrity excludes eg Nazi and Marxist
interpretations of the Constitution, perhaps it can claim some measure of fidelity to
constitutional language.
16 1ignore here the distinction sometimes made between duty and obligation according to which obligations but not
duties involve an element of consent.
17 They are foundational for a Kantian in that they are trumps against collective goals like utility maximization
rather than instruments for the pursuit of such goals.
IS In Taking RightsSeriously Dworkin suggests an understanding of the concept of rights latitudinarian enough to
embrace 'an act-utilitarian theory that holds that everyone has a duty to act on every occasion, so as to produce the best
continuedon page186
SUMMER 1991
OxfordJournalofLegalStudies
It is a wan conception of canonical language that allows constitutional rights to
be interpreted in accordance with Hercules' and Max's background political
principles. But even this skeletal conception is threatened by Hercules' apparent
reading of the due process clause. His creator Dworkin shares with many lawyers a
commitment to 'substantive' due process, 19
which directs courts to judge the
reasonableness of the burdens placed on citizens by the law rather than on the
fairness of the procedures by which those burdens are created and imposed. Now
the clause does admit interpretation. A narrow reading says a person has a right to
whatever legal processes the black-letter law elsewhere specifies; a broader
interpretation has it that in depriving citizens of liberty or other basic goods the
state must meet standards of proceduralfairness not necessarily spelled out by
law. 20
Both these interpretations are variations on the theme of due process.
Substantive due process is not; John Hart Ely is correct that 'There is simply no
avoiding the fact that the word that follows "due" is "process ..... "substantive
due process" is a contradiction in terms-sort of like "green pastel redness"'. 21 If
law as integrity allows for the 'substantive' reading of the due process clause it
seems to respect constitutional language no more than pragmatism does.
One way around this problem might be to say that Dworkin's impersonation of
Hercules interpreting the due process clause is a bad one, and so irrelevant to law
as integrity's commitment to honour constitutional language. But it is not so easy
to dismiss this attempt to apply law as integrity as an aberration. Substantive due
process has had important advocates, some on the Supreme Court, for decades,
and the criticism that it is flatly untrue to the language of the clause is a standard
one. Since Roe v Wade substantive due process has made a comeback and is once
again an important part of constitutional debate. So Dworkin's endorsement of it
cannot be dismissed as an unfortunate but irrelevant detail. There are other
instances where Dworkin among others has in the pursuit of what they see as
fundamental rights countenanced what on any traditional understanding is the
violation of canonical language. Consider Dworkin's (and many judges') treatment
of the Civil Rights Act of 1964 which, though not a constitutional provision,
addresses important individual rights. Dworkin apparently endorses many of the
famous busing decisions (see 221, 392), all of which came after the Civil Rights Act
became law, even though the Act defines desegregation as the assignment of public
school pupils 'without regard to their race. . . "desegregation" shall not mean the
assignment of students to public schools in order to overcome racial imbalance'. 22
It would be difficult to find a more flagrant example of what most people would
consider disregard for statutory language.23
continued fr'om page 185
results, as a utilitarian would define these, and that those who would benefit from such acts from time to time have a
"right" to them'. (Ronald Dworkin, op cit, above, n 4, 313.)
19 Dworkin's frequent references to 'procedural' due process suggest that he accepts 'substantive' due process.
20 See Christopher Wolfe, The Rise ofModernJudicial Review (New York: Basic Books, 1986), 131-2.
21 Democracyand Distrust: A Theory ofJudiralReview (Cambridge: Harvard University Press, 1985), 178.
2, Act of 2 July 1964, 78 Star 241, Title IV Sec 401(b), 246.
23 It might be objected that the busing cases interpret the Constitution, not the Civil Rights Act. But this would
mean that the Act's proscription ofquotas in pupil assignment is in effect unconstitutional, a position no one I know of
has taken.
VOL. 11
Dworkin andLegalPragmatism
If the examples above evidence disregard for canonical language, and if law as
integrity is wedded to them, then the contrast between law as integrity and the
pragmatism of justice would seem lost where fundamental fights are concerned.
Furthermore, law as integrity would have to forfeit its claim to fit legal practice,
assuming the canonical status of constitutional and statutory language is among the
indisputable pre-interpretive data of law if anything is. But law as integrity is
nothing if not resourceful, and it might be able to reinterpret the idea of canonical
language in a way that shows that the substantive due process and busing decisions
really respect canonical language.
Here is how it might do so. The question of what respect for statutory or
constitutional language means is, like all questions within law as integrity,
normative as well as descriptive; it is as much a matter of substance as fit. The
conception of canonical language embraced by law as integrity is thus determined
by the most morally attractive account of law that adequately fits legal history.
That account will be sensitive to the different functions of different kinds of law.
There are several examples of this: eg, certainty is important in interpreting rules
of the road and rules governing negotiable instruments, but not in interpreting
constitutional rights; popular morality can play a substantial part in cases at
common law, but not in constitutional rights cases. It is the importance of the value
of respect for persons expressed by constitutional rights that makes these other
values pale by comparison in such cases.
The supreme importance of these rights will also influence what it means to
respect constitutional language. We know that Hercules operates on the general
principle that, where such rights are concerned, what the framers (and later
interpreters) did is more important than what they saidor what they thought they
did. The framing of a constitutional provision is but the beginning of a long story
in which the framers have little say; the authors of the eighth amendment might
have prohibited capital punishment even if they thought the death penalty was
outside the extension of 'cruel punishment', and those who wrote the laws and
precedents cited in Griswoldmight have effected a privacy right even if they had no
inkling they had done so, and were quite sure they had not. Especially in the
privacy cases Hercules ascends far above the words and concrete convictions of the
authors of the black-letter law to construct a highly abstract principle that
supposedly justifies bold new interpretations of that law. All it takes to justify the
substantive reading ofthe due process clause is to ascend a bit farther and take the
framers' wordsas well as their concrete convictions about those words to mean"that
justice simpliciterbe done. Understood this way, the clause could mean inter alia
that anti-abortion and anti-contraception statutes are unconstitutional.
This interpretation of the notion of canonical language is well within the spirit of
law as integrity, and ifaccepted saves Hercules, albeit in an eccentric way, from the
charge that he, like the pragmatist of justice, has no principled commitment to
constitutional language. But salvation is expensive. As is the case with its doctrine
of fit with institutional history, law as integrity's conception of respect for
constitutional rights language leaves it extensionally equivalent to the pragmatism
SUMMER 1991
OxfordJournalofLegalStudies
of justice. Ifthat is so, the citizens of the community ofintegrity have reason to lose
interest in the arcane intentional distinctions between the two theories of law, and
to conclude that where fundamental rights are concerned they are ruled by
pragmatist judges.
V Legitimacy and Obligation
I have denied that integrity explains our commitment to a certain kind of equality
before the law that puts constraints on Hercules' pursuit of justice in constitutional
rights cases. It is time now to explore the claim that integrity explains better than
its rivals fundamental intuitions about political legitimacy and the general obligation
to obey the law. I contend that Dworkin's portrait of the community of
integrity, and especially of the conception of legal obligation that prevails there,
supports the view that law as integrity demands the open-ended enforcement of
justice. I close by suggesting that Dworkin rejects the argument from fair play as
the best account ofpolitical legitimacy and obligation because it does not entail that
kind of commitment.
A. Legitimacy, Obligation,andIntegrity
Dworkin thinks that
the best defense of political legitimacy-the right of a political community to treat its
members as having obligations in virtue ofcollective community decisions-is to be found
not in the hard terrain ofcontracts orduties of justice or obligations offair play that might
hold among strangers ... but in the more fertile ground of fraternity, community, and
their attendant obligations. (206)
The argument for this claim is serpentine, but it goes like this. A striking feature
of political obligation is its non-contractual or non-consensual nature. In a
reasonably just state citizens cannot escape the general obligation to obey the law
by claiming that it was not voluntarily assumed. Dworkin's explanation of this is
that political obligation is a species of 'associative' obligation (see ch 6), a type of
obligation said to arise only in communities that exhibit integrity.
Dworkin says that the obligations a community generates are associative under
the following conditions: (1) the obligations its practices create are special, 'holding
distinctly within the group, rather than as general duties its members 6we equally
to persons outside it'; (2) these obligations are personal,running 'directly from each
member to each other member, not just to the group as a whole in some collective
sense' (199); and (3) the practices of the group exhibit a general concern members
have for one another which is (4) equal concern, such that even if the group is
hierarchical its 'structure and hierarchy must reflect the group's assumption that its
roles and rules are equally in the interests of all' (200). Members of a group that
meets these conditions do not assumeobligations; rather, the practices of the group
'attract obligations, and we are rarely even aware that we are entering upon any
VOL. I I
Dworkin andLegalPragmatism
special status as the story unfolds' (197). Such is said to be true of family and
friendship, whose obligations are strong and 'sustained among people who share a
general and diffuse sense of members' special rights and responsibilities from or
toward one another' (199) even when (as in the case of family) they 'are matters of
the least choice' (198).
A state whose laws individually and collectively 'speak with one voice' is the only
state that generates associative obligations, says Dworkin. That state exhibits unity
in its principles of justice, fairness, and 'procedural' due process, and so 'expresses
a concern by each for all that is sufficiently special, personal, pervasive, and
egalitarian to ground communal obligations according to standards for communal
obligation we elsewhere accept' (216). The citizen of a state thus committed to
equal concern has a general though not absolute obligation to obey its laws even
though he has never consented to do so, and even though those laws are less than
perfectly just. The other side ofthe coin is supposedly that judges must for similar
reasons interpret existing law rather than simply do justice.
I have already rejected the claim that integrity restrains rather than demands the
pursuit of justice in constitutional rights cases. Dworkin's attempt to ground
political legitimacy and obligation on integrity undermines that claim in two
additional ways. The first concerns the previously sketched argument that the state
whose practices constructively manifest less than equal concern for its citizens
relative to whatever conception of justice it deploys is illegitimate. It follows that,
even if the state displays some level of concern for targets of invidious discrimination
sufficient to pass some minimum threshold ofjustice, it is illegitimate unless
it passes the equal concern test. Now this test of legitimacy does not seem as
stringent as it could be: it demands that the state consistently enforce principles of
justice that are adequate, not sub specie aeternitatis,but rather from the perspective
of social morality, though a state that grossly violates ideal justice is said to be
illegitimate no matter what (203). If,however, the arguments in Part IV above are
sound, this test is more stringent than it sounds, for in constitutional rights cases
the deference ofcourts to institutional history is illusory. But even if this test is less
than absolutely stringent, in denying legitimacy to systems which regard some
groups as less worthy than others even ifsome important rights of those groups are
respected, law as integrity underestimates (in Alexander Bickel's words) 'how
much is human activity a random confusion', 24
and the implications of this for
political theory. Integrity fails to recognize that knowing and doing what justice
requires-whether we speak of ideal justice or a society's conception of justice--is
an arduous achievement, and that the overzealous pursuit of justice can promote
the summum malum of civil strife or totalist government. And if I am correct so far
integrity does restrain the pursuit of background justice.
The second way that law as integrity's idea of legitimacy undermines the claim
that it imposes restraints is this. Dworkin says that the conditions of associative
obligation sketched above are conditions of the communitarian virtue of fraternity.
24 The Morality of Consent (New Haven: Yale University Press, 1975), 77.
SUMMER 1991
OxfordJournalof LegalStudies
He hastens to add that these conditions 'do not themselves demand' that members
of the community 'actually feel some emotional bond with one another'; they are
not "'psychological" conditions' (201). This means that 'associative communities
can be larger and more anonymous than they could be if it were a necessary
condition that each member love all others, or even that they know them or know
who they are' (ibid). In other words, though political society lacks the 'psychological
conditions' of intimacy, such as the affection born of personal ties characteristic
of face-to-face communities like families, law as integrity 'constructively' ascribes
these properties to political society. Affective properties are on this view mere
epiphenomena that have no bearing on what kinds of obligations are generated by
the practices of a group.
That is a recipe for the smothering pursuit of justice which I have said law as
integrity requires, as Dworkin's own portrait of associative obligations attests.
Dworkin depicts such obligations as typically informal and non-legalistic. They are
not in the central family-type case legislated or governed by formal standards of
identification and application. The history of an associative community is said to
generate a quite abstract obligation of care and concern; as mentioned earlier,
members have 'a general and diffuse sense of their rights and responsibilities',
which suggests that those rights and responsibilities are pervasive rather than
carefully delimited. All this makes sense in the context of a family; the obligations
of their members can be open-ended and informally applied because they grow out
of a real mutual concern possessed only by those who share a history and know
each other as flesh-and-blood individuals. Formal, arm's-length methods of
determining such obligations are inappropriate and offend fraternity; a couple
whose pre-nuptial agreement goes into effect is finished as a community in
Dworkin's sense.
Such formal methods find their home in political society. It is useful briefly to
'interpret' certain fundamental features of law to point up the contrast between
political society and real associations. Law's often elaborate and explicit lawdetermining
procedures, its demand for fully articulated grounds for judicial
decisions, and its coercive apparatus spring from the peculiar needs ofgroups most
unlike what Dworkin calls associations. This structure reflects the concern that, if
the larger society were treated as an association, if rights and responsibilities of
citizens were pervasive, we might well face what public choice theorists call 'a legal
warof allagainstall'25
in which individuals and groups organize to get whatever
they can, not necessarily from selfishness, but to pre-empt the open-ended claims
of anonymous others driven to do the same for similar reasons. That is the likely
result of the attempt to graft onto political society the sort of open-ended concern
for others typical ofintimate associations. Integrity, which was supposed to take us
off the hard, adversarial terrain of justice, really beckons us there.
25 John Gray, Hayek on Liberty (New York: Basil Blackwell, 1984), 120.
VOL. 11
Dworkin andLegalPragmatism
B. Legitimacy, Obligation,andFairPlay
I have said that law as integrity's views about political legitimacy and obligation fit
the pragmatism of justice better than they fit a theory that really demands fidelity
to law. I conclude this article by defending the thesis that the argument from fair
play gives a better account of these fundamental political concepts, and by
suggesting that law as integrity rejects the argument because it is less conducive to
pragmatism.
The argument from fair play says that the benefits a political community confers
on its members create obligations on their part to obey the laws of that particular
community. Dworkin concedes that the argument has one attractive feature: it
explains the 'special' character of political obligation; it ties 'political obligation
sufficiently tightly to the particular community to which those who have the
obligation belong . . .' (193). But he thinks the argument fails in two respects. It is
said to assume that we incur obligations just by receiving unsolicited benefits which
we might reject had we the choice. And it is supposedly ambiguous: it assumes
either that citizens are better off under their present government than they would
be under any system that might replace it-an impossibly stringent test-or else
that they are better off than they would be with no social or political organizationan
absurdly permissive test (see 193-5).
Each of these weaknesses is only apparent. Apropos the first, Dworkin accepts
Nozick's view that the fair play argument no more grounds political legitimacy
than it justifies an obligation to pay a philosopher who broadcasts an edifying but
unsolicited lecture from a sound truck outside your house (see 194). I think this
argument is inseparable from Nozick's radical libertarianism, which denies any
'positive' right-a right correlative to a duty of others actively to aid its bearer in
some way-that is not the product of agreement. From this libertarian perspective
an obligation to obey the law in virtue of the benefits derived from political society
can be lumped together with an obligation to pay for an unwanted lecture. But the
two cases are different. Philosophy is a significant good, but even good philosophy
lectures cannot be assumed to be requirements of a minimally decent life. In
contrast, the benefits provided by law and political society, at least in complex
industrial nations, arerequirements of such a life. Liberals warn us that what some
see as needs which government should satisfy-say, religious instruction or universal
day care-are seen by others as nuisances or even threats. But even in a
'pluralistic'-fragmented-society like our own, we can assume that each person
needs certain basic goods that only political society is likely to provide, and we have
reason to doubt severely the protests of those who deny the need for these things.
Now to be effective political authority requires general compliance by those subject
to it; even those who accept it for non-selfish reasons need assurance that most of
those who do not accept it will be forced to comply. It is reasonable, then, to
demand compliance as the price of living in political society even when its benefits
are unrequested, and even if expecting payment for a splendid but unsolicited
lecture on meaning and reference is unreasonable.
Dworkin's second argument poses a false dilemma. There is a middle way
SUMMER 1991
OxfordJournalofLegalStudies
between the extreme that citizens have a general obligation to obey the law because
they are better off than they would be under any other system, on the one hand, or
because they are better off than they would be without a legal system, on the other.
Dworkin himself shows that way in TakingRights Seriously, in a discussion of how
a judge might frame an interpretive theory that justifies the practices of his legal
system:
The constitution sets out a general political scheme that is sufficiently just to be taken as
settled for reasons of fairness. Citizens take the benefit of living in a society whose
institutions are arranged and governed in accordance with that scheme, and they must
take the burdens as well, at least until a new scheme is put into force either by discrete
amendment or general revolution.2 6
This is of course an argument from fair play, and it is a more moderate basis for
political obligation than the integrity argument. It demands only that the law
achieve some minimum level of justice; it does not assimilate political society to the
category of 'associations' such that the conditions ofpolitical legitimacy include the
recognition of open-ended associative obligations implicit in the black-letter law.
Fair play is thus consistent with a significantly 'conversational' conception of law
on which even constitutional law should 'make it possible to foresee with fair
certainty how the [legal] authority will use its coercive powers in given circumstances
and to plan one's individual affairs on the basis of this knowledge'. 27
Though this conception of the rule of law lacks certain inspirational qualities, it at
least accommodates the insight that
The small society, as the milieu in which man is first found, retains for him an infinite
attraction; he undoubtedly goes to it to renew his strength; but.., any attempt to graft
the same features on a large society is utopian and leads to tyranny. With that admitted, it
is clear that as social relations become wider and more various, the common good
conceived as reciprocal trustfulness cannot be sought in methods which the model of the
small, closed society inspires; such a model is, on the contrary, entirely misleading.28
No wonder law as integrity abandons fair play.
26 Ronald Dworkin, op cit, above, n 4, 106.
27 FriedrichA. Hayek, The Road to Serfdom (Chicago: University of Chicago Press, 1972), 72.
28 Bertrand de Jouvenel, Sovereignty (Chicago: University of Chicago Press, 1957), 136, quoted in Friedrich A.
Hayek, Law, LegislationandLiberty, Vol 2, The Mirage ofSocialJustice, 182 n 38.
VOL. 11