Maryland Law Review

Volume 49 | Issue 2 Article 6

Should a Good Lawyer Do the Right Thing? David

Luban on the Morality of Adversary
Representation
David Wasserman

Follow this and additional works at: http://digitalcommons.law.umaryland.edu/mlr

Part of the Legal Profession Commons

Recommended Citation
David Wasserman, Should a Good Lawyer Do the Right Thing? David Luban on the Morality of Adversary Representation, 49 Md. L. Rev.
392 (1990)
Available at: http://digitalcommons.law.umaryland.edu/mlr/vol49/iss2/6

This Article is brought to you for free and open access by the Academic Journals at DigitalCommons@UM Carey Law. It has been accepted for
inclusion in Maryland Law Review by an authorized administrator of DigitalCommons@UM Carey Law. For more information, please contact
[email protected].
 Review Essays
SHOULD A GOOD LAWYER DO THE RIGHT THING? DAVID
LUBAN ON THE MORALITY OF ADVERSARY
REPRESENTATION

A REVIEW ESSAY BY DAVID WASSERMAN*

INTRODUCTION

Lawyers andJustice' has three audiences: ordinary citizens, who

denounce lawyers as vicious predators when not threatening to un-
leash them on their neighbors; lawyers themselves, who reserve
their piety for an adversary system that allows them to take a cyni-
cally instrumental view of everything else; and moral philosophers,
who find little interest in role morality unless the professions are
paying for moral audits. As a member of all three audiences, I was
inspired and chastened. Luban's book offers a devastating critique
of the standard rationales for the adversary system, an expos6 of
existing practice, and an impassioned defense of an alternative
model of legal representation.
Luban's goal is not merely to bury existing norms of legal prac-
tice, but also to praise the moral and practical opportunities in the
law. He attempts to steer a middle course between the' "standard
conception" of adversary representation, which sees the law as a
source of special privileges and immunities for its practitioners, and
the legal realism popular in academic circles, for which the law is
simply a highly-manipulable instrument of social control. Luban's
general thesis is that legal practice has moral responsibilities which
constrain it more than its apologists concede and legitimize it more
than its critics acknowledge. 2 In this Review, I will consider his ar-
guments for the moral value of adversary representation.
Luban's defense of conscientious legal practice takes several
forms. In chapter three, he argues that citizens have a duty to obey

* Research Scholar, Institute for Philosophy and Public Policy, University of Mary-
land; B.A., Yale University, 1975;J.D., University of Michigan, 1978; M.A., University of
North Carolina, 1984 (Psychology). This Review Essay was written while the author was
a Fellow in Ethics and the Professions at Harvard University. It has benefitted greatly
from criticism and suggestions by Douglas MacLean, Judith Andre, Greg Dees, Amy
Gutmann, Dennis Thompson, Alan Strudler, Henry Richardson, and David Luban.
1. D. LUBAN, LAWYERS ANDJUSTICE (1988).
2. See id. at xix-xxvi.
1990] REVIEW EsSAY-Lawyers andJustice 393

generally beneficial laws 3 and lawyers a concomitant duty to respect

the laws' generality.4 In chapter seven, he argues that although law-
yers are morally accountable for their actions in a way that the stan-
dard conception denies, 5 a strongly justified adversary criminal
defense can license acts that conflict with common morality. 6 Fi-
nally, in chapters eleven and twelve, he argues that the right to legal
services has a different, and less controversial, source than the right
to more urgent goods like health care,7 and that lawyers have a spe-
cial obligation to make their services available to the poor.' I find
these arguments provocative but largely unconvincing. In reviewing
them I come to the bleak conclusion that the moral challenges of
legal practice are not as singular, nor as interesting, as Luban makes
them appear.

I. JUSTIFYING LEGAL INSTITUTIONS AND JUSTIFYING

LAWYERS' ACTIONS

I begin with Luban's discussion of role morality because of it's

importance to his book and its originality as a contribution to legal
ethics. Part I of Lawyers andJustice is a sustained attack on the law-
yer's use of the "adversary system excuse" to cover a multitude of
professional sins. Luban argues that this "excuse" fails to justify
most objectionable aspects of adversary practice, and concludes that
there are only a few settings in which a lawyer justifiably may engage
in conduct which ordinary morality condemns. 9 Unlike many of
Luban's critics, I feel he is too generous.
In chapters four and five, Luban challenges virtually all the cur-
rent rationales for the "standard conception" of the lawyer's role, as
defined by two principles: (1) partisanship, which requires the law-
yer to do everything within the bounds of the law to maximize her
client's likelihood of success,i ° and (2) nonaccountability, which
frees her of any moral responsibility in so doing." Luban's critique
is directed largely against the partisanship principle; in particular,
against claims that it maximizes the discovery of truth 12 or the pro-

3. Id. at 35-43.
4. Id. at 47-49.
5. Id. at 128-47.
6. Id. at 145-47.
7. Id. at 265-66.
8. Id. at 240-66.
9. Id. at 148-49.
10. Id. at 50-92.
11. Id.
12. Id. at 68-74.
394 MARYLAND LAW REVIEW [VOL. 49:392

tection of individual rights,' 3 that it is an intrinsic good like friend-

ship or a condition of human dignity,14 or that it is essential to
maintain the existing social fabric.' 5 The nonaccountability princi-
ple is an indirect casualty of this critique because there is no reason
to immunize the lawyer for the excesses required by partisanship,
and no need to immunize her against the acts required by a more
qualified principle.
Luban finds full partisanship justified solely in the defense of
accused criminals and, possibly, in the representation of other poor,
oppressed, or vulnerable clients against powerful state-like institu-
tions. 6 In these areas, the single-minded zeal of the attorney helps
to maintain the balance in an otherwise lopsided contest by "over-
protecting" the rights of the weaker party. 7 Although the repre-
sentation of guilty criminals often is presented as the hardest case
for adversary representation, it is actually one of the easiest (the eas-
iest being the representation of an innocent criminal defendant). The
partisanship principle derives spurious force from the criminal para-
digm when it is extended to other settings in which the exigencies of
criminal defense are absent. In ordinary civil settings, the adversary
system is justified only "pragmatically" because there is no better
way of resolving disputes in a society with our history and values.'
I find this line of argument compelling, but I think that Luban
misconstrues its relevance to the issue he addresses in his next two
chapters: the dilemma of conscientious practitioners trying to
choose between their professional obligations, and the apparent
conflicting demands of common morality.'" Luban believes that the
strength of the policies underlying the role will be reflected in the
range of "morally dissonant role acts" it licenses the agent to per-
form." ° Luban attempts to steer a middle course between a "policy
over acts" approach, which "eschew[s] the moral assessment of indi-
vidual acts"'" and exempts the agent from ordinary moral consider-
ations, and an "acts over policies" approach, which fails to take
"general rules, policies, and institutions very seriously" and holds

13. Id. at 74-78.

14. Id. at 81-86.
15. Id. at 87-92.
16. Id. at 58-66.
17. Id.
18. Id. at 92-103.
19. Id. at 104-47.
20. Id. at 145.
21. Id. at 138.
1990] REVIEW ESSAY-Lawyers andJustice 395

the agent fully accountable to common morality.2 2

Luban, however, does not argue that the role-agent must give
the policies underlying the role greater weight in his decision than
common morality would assign them-the position Alan Goldman
calls "strong role differentiation.- 23 Nor does he argue that the
agent should conform to the role except in circumstances in which
the assumptions justifying his deference do not appear to hold-the
rule-of-thumb approach. Rather, Luban insists that the agent is re-
quired, in every case, to weigh the marginal contribution of his role-
act to the "moral goodness" of the institution against the harm and
offense it may cause.2 4
Despite his claim to steer a middle course, Luban's approach
ultimately favors acts over policies. It is really a form of "sophisti-
cated" act consequentialism, taking account of roles, policies, and
acts only to the extent that they bear on the consequences of specific
acts. Because common morality also takes cognizance of the full
range of effects produced by individual acts, Luban's approach ap-
pears to be more of a refinement than an alternative. And if his
approach does depart from common morality, it is only in the rare
case in which a morally-dissonant role-act makes an important con-
tribution to the moral goodness of the institution.
Luban begins his discussion of role-morality by rejecting the
prevailing approach, which, following F.H. Bradley, he calls "my
station and its duties." '25 In this approach, which is little more than
a philosophical version of the nonaccountability principle, an agent
bears no moral responsibility for the acts dictated by a desirable
role. 26 This conception of role-morality gives obvious and decisive
relevance to Luban's justification of adversary criminal defense: the
role of zealous criminal advocate is sufficiently desirable to immu-
nize the agent for otherwise immoral acts, while the role of zealous
civil advocate is not.2 7
Luban's critique of "my station and its duties" focuses on the
claim that the role justification gives the agent an "exclusionary rea-
son"-a reason for refusing to consider any moral objections to in-
dividual role-acts. 28 He does not argue that an unqualified

22. Id.
23. See A. GOLDMAN, THE MORAL FOUNDATIONS OF PROFESSIONAL ETHICS 2-6 (1980).
24. D. LUBAN, supra note 1, at 139-40.
25. Id. at 116-27.
26. Id. at 117-20.
27. Id. at 58-66.
28. Id. at 120-23.
396 MARYLAND LAW REVIEW [VOL., 49:392

exclusionary rule could never be justified, but rather, that only a

"moral prig" or moral monster could adhere to it. 29 Decent moral
agents cannot always blind themselves to the consequences of their
acts, however justifiable the rule that so requires them.3 0
Having denied that the duties attached to a morally desirable
station always are decisive in justifying individual role-acts, Luban
also wishes to deny that they are irrelevant. In balancing "the moral
reasons incorporated in [a] role . . . against the moral reasons for
breaking the role expressed in common morality,"'" he constructs a
form of argument in which the justification for the role-act derives
from the good achieved by the institution: "[T]he agent (1) justifies
the institution by demonstrating its moral goodness; (2) justifies the
role by appealing to the structure of the institution; (3) justifies the
role obligations by showing that they are essential to the role; and
'
(4) justifies the role-act by showing that the obligations require it. "32
Luban illustrates this "four-fold root" by applying it to the di-
lemma of an "Oxfam" logistics officer working for a philanthropic
institution "whose sole function is to distribute food to famine-
stricken people in impoverished areas of the world," and whose role
within that institution imposes the role-obligation of "procuring the
means of transporting food[,]" which in turn requires various role-
acts to carry out.3 3 Luban presents a situation in which one of these
role-acts remains silent about an impending murder:
Let us suppose that to get food to a remote village in
an underdeveloped country stricken by famine, the logis-
tics officer must obtain several trucks from a powerful local
boss named P .... Imagine further that the logistics officer
overhears P dispatching a murderer to kill [a] man that very
night; that P discovers that the logistics officer has over-
heard him; and that P tells her that if the man is warned
and escapes, P will withhold the trucks.3 4
If the logistics officer feels obliged to keep silent, she will justify, or
excuse, her act in the following way:
[S]he points out that the role act of complying with P is
required by her role obligation, which in turn is necessary
to perform the role's institutional task, which (finally) is

29. Id. at 121.

30. Id. at 120-23.
31. Id. at 125.
32. Id. at 131.
33. Id. at 129-30.
34. Id. at 130.
 1990] REVIEW ESSAY-Lawyers andJustice 397

justified by the positive moral good of the institution: the

saving of many innocent lives. Taken together, these justi-
fications for the role act outweigh the obligation to P's un-
35
fortunate victim.
Her silence is "required by" her role obligation instrumentally, be-
cause it is necessary to discharge it: no silence, no trucks. This illus-
tration makes clear that the four-fold root is intended to provide a
consequentialist justification for morally dissonant role-acts, by
showing how the disputed act contributes to the moral work of the
institution. 36
An institution that only is justified pragmatically will not, on
this approach, provide the agent with a very strong reason for en-
gaging in immoral conduct. The justification is only as strong as its
weakest link, and the first link consists of nothing more than the
claim that the institution is no worse, or perhaps slightly better, than
the alternatives. 3 7 If a pragmatic justification, however, does not li-
cense departures from common morality, it is not clear whether a
stronger justification will either. Because "the moral goodness of
the institution "-the ultimate source ofjustification in the four-fold
root3--rarely is affected by a single role-act, it is hard to see how it
can underwrite that act. 39 The problem becomes apparent in the
setting in which Luban is most anxious to license departures from
common morality-adversary criminal defense.40

35. Id.
36. The consequentialist character of the approach is reinforced by Luban's con-
trasting example of an anthropological research team in the same village. Although the
researcher's silence is necessary to advance anthropological knowledge, that advance's
importance does not outweigh the loss of an innocent life. Id. at 133.
37. The moral value in sustaining the institution, however, would still be considera-
ble, if there were high "replacement costs" in distrust, disruption, and public resistance
in substituting an equally adequate institution.
38. D. LUBAN, supra note 1, at 131.
39. David Ross has argued generally that those who make utilitarian arguments for
such practices as promise-keeping, typically overestimate the harm done to the practice
by a single breach. D. Ross, THE RIGrr AND THE GOOD 39 (1930).
40. D. LUBAN, supra note 1, at 58-66. Luban confuses the issue somewhat by sug-
gesting that the lawyer who helps win the acquittal of a client he believes to be guilty,
commits a morally-dissonant role act. Id. at 59. If this is true, so does the jury which
believes the defendant guilty but acquits him because of a reasonable doubt about his
guilt. If the lawyer's belief in his client's guilt arises solely from the evidence and im-
pressions available to the jurors, and not from his client's confidences, he does not, any
more than they, need to justify his actions in terms of the distal goal of overprotecting
defendants' rights. The claim that it is better to acquit a hundred guilty defendants than
convict one innocent person, id. at 58, does not prescribe a ratio for "overprotection;" it
expresses a sense of comparative evil. We believe that it is inherently wrong to deprive a
person of his liberty unless his guilt is nearly certain, and a lawyer's efforts to assure that
398 MARYLAND LAW REVIEW [VOL. 49:392

In applying the four-fold root to legal representation, Luban

claims that "since in the criminal defense paradigm the adversary
system has weighty justifications, it can support broad institutional
excuses-provided, of course, that the acts being excused are genu-
ine requirements of role obligations that are themselves important
to the defense attorney's job."' This again suggests an instrumen-
tal justification for morally offensive role-acts-they are necessary
for the institution to accomplish the goals that justify it. The plausi-
bility of this claim, however, seems to rest on a confusion between
acts prescribedby a role or role-obligation with acts necessary to achieve
their objectives. While justifiable roles sometimes prescribe morally
dissonant acts, those acts may not achieve or maintain the moral
work of the role or institution. This equivocation is seen by examin-
ing Luban's discussion of the Lake Pleasant Bodies Case.4 2
In the Lake Pleasant case, a defendant charged with one murder
told his lawyers about two other teenage murder victims, still classi-
fied as missing persons.4" His lawyers felt obliged to keep this
knowledge secret, although their silence prolonged the anguish of
the victims' families, still hoping to find their children alive. 44 In
justifying the lawyers' decision, Luban invokes the value of confi-
dentiality in protecting criminal defendants against a powerful,
overreaching state:
The lawyers' role acts (preserving the defendant's confi-
dences, photographing the bodies but telling nobody) were
required by the general duty of confidentiality-the role
obligation. This is justified by arguments that confidential-
ity is required in order to guarantee an adequate criminal
defense-the institutional task . . . . The next step is to
show that zealous criminal defense is required by the ad-
versary system, and this in turn . . . serves the positive
moral good- of overprotecting individual rights against the
encroachments of the state.4 5
It is instructive to note how the structure of this excuse differs
from that offered by the Oxfam logistics officer. The officer's role-
act of keeping silent was required instrumentally by her role-obliga-

his client is not convicted without the requisite level of certainty need no extrinsic justifi-
cation. It may be that this high standard of proof has external, consequentialist origins,
but it certainly has become internalized in our culture.
41. Id. at 148.
42. Id. at 53-54.
43. Id. at 53.
44. Id.
45. Id. at 149.
1990] REVIEW EsSAY-Lawyers andJustice 399

tion of procuring transport-her silence was necessary to obtain the

trucks. In contrast, the lawyers' role-act of preserving the defend-
ant's confidences is prescribed by their duty of confidentiality, "re-
quired" only in the trivial sense that keeping silent is required to
fulfill the duty to keep silent. The fact that the role-act is required in
this sense simply is not relevant to ajustification based on the moral
good of the institution.
This is apparent in moving up the four-fold root: while procur-
ing the trucks was necessary for the institutional task of transporting
food to starving people, and thus for the moral good of saving lives,
fulfilling the duty of confidentiality in the Lake Pleasant case is not
necessary for the institutional task of "guarantee[ing] an adequate
criminal defense," 4 6 or for the moral good of "overprotecting indi-
vidual rights against the encroachments of the state." 4 7 While it
may seriously compromise the adequacy of criminal defense if every
criminal attorney were to reveal her client's confidences whenever
such action appeared likely to produce more benefit than harm, the
impact of this decision on the "institutional task" of adversary crimi-
nal defense most likely would be negligible. The Lake Pleasant law-
yers might have tipped off authorities about the bodies without
arousing their clients' own suspicions or causing wider erosion of
attorney confidentiality or client trust.
If "an adequate criminal defense" is read to mean an adequate
defense in this case, and assuming that a breach of confidentiality
prevents an adequate defense, then the lawyers' silence is required
by their role-obligation. This, however, merely shifts the justifica-
tory gap: ensuring an adequate defense is not necessary for the in-
stitutional goal of "overprotecting individual rights." A single
inadequate defense hardly impairs the vigor of the adversary system
or upsets the "balance of advantages" in criminal cases.
What is necessary to guarantee an adequate criminal defense
(or so we may assume) is the practice of maintaining client confi-
dences, for without it, the trust essential to effective representation
does not exist. It makes no sense, however, to claim that the Lake
Pleasantlawyers' silence is required by this practice. The practice is
a social or legal fact, not a moral duty, and it clearly is a mistake to
claim that their silence is necessary to maintain that practice. Luban
equivocates between the duty of these lawyers to keep silent and the
practice of confidentiality by speaking of a "general duty of confi-

46. Id.
47. Id.
400 MARYLAND LAW REVIEW [VOL. 49:392

dentiality. ' ' 48 If this refers to the duty that binds the lawyers, it is
not necessary for an adequate criminal defense. If it refers to the
practice that is necessary for that institutional task, it requires noth-
ing of these lawyers."9
If the Lake Pleasant lawyers had reasoned like the Oxfam logis-
tics officer, they might have felt far less constrained by their role and
its duties. Even taking into account not only the indirect conse-
quences of their actions, but their own tendency to underestimate
those consequences and overestimate the immediate harm in con-
forming to their role, they might well have concluded that the harm
in satisfying their role obligation far outweighed the harm in violat-
ing it. It is unclear how the four-fold root would hinder them in
reaching that conclusion. If the concealment in the Lake Pleasant
case contributed significantly to the larger good of criminal defense,
Luban certainly has not shown it. Accordingly, it is difficult to see
how the concealment is justified by that good.
The same difficulty occurs in Luban's explanation of the crimi-
nal lawyer's duty of confidentiality in terms of the client's "morally
based legal right" to compel his attorney's silence. 50 Luban argues
that "[a] morally based right against self-incrimination will serve
just as well as a moral right to establish the lawyer's duty of confi-
dentiality."'" To the contrary, it does not serve as well. While a
moral right would "trump" the considerable gains in utility that fre-

48. Id.
49. John Rawls introduced the distinction between "justification of a rule or prac-
tice" as a system of rules to be applied and enforced and the "justification of a particular
action falling under [these rules]." Rawls, Two Concepts of Rules, 64 PHIL. REV. 3, 4
(1955). Rawls illustrates the distinction with the examples of punishment and promise-
keeping, justified as practices by their long-term social benefits but unjustified in partic-
ular applications by moral duties for which consequences are largely irrelevant. Id. at 4-
18. He does not explain why practices justified by their consequences should generate
or correspond with independent nonconsequential moral duties, and that remains one
of the unsolved mysteries of role morality.
Unless there is a fortuitous or providential correspondence between justified prac-
tices and moral duties, then either: (1) we will not always have a duty to adhere to
practices that justifiably permit no exceptions; or (2) the justification for the practice will
generate an unconditional duty of adherence. The former is not acceptable to those
who think that the demands of morality must be consistent, but the latter is a difficult
claim to defend. Why should we have a duty to conform to a generally beneficial prac-
tice when our conformity confers no marginal benefit? It is hard to see why we should
be constrained by the possibility that the benefits of the practice would be lost if every-
one reasoned as we did, when we know that everyone will not. We can hardly be ac-
cused of free-riding if we violate the practice to achieve what is, all things considered,
the greater good.
50. D. LUBAN, supra note 1, at 192-97.
51. Id. at 196.
1990] REVIEW EsSAY-Lawyers andJustice 401

quently arise from breaking confidences, a morally-based legal right

would not, for the reasons discussed above. The moral basis for the
legal right arises from the need to resist the threat to individual lib-
erty posed by a powerful state. Unless that threat is increased mate-
rially by the contemplated breach, however, it is difficult to see how
the breach would be a moral, as opposed to a legal, wrong.5 2 The
practice of confidentiality may be necessary to restrain the state, but
this act of concealment clearly is not.
The difficulty of invoking a morally-based legal right to license
a morally dissonant role-act reflects the broader tension between
policies and acts, usually discussed under the rubric of rule-conse-
quentialism.5" There are several reasons why a sound rule or policy
may require acts that individually are not necessary to achieve its
objectives. Because case-by-case judgments may be fallible, biased,
distracting, time-consuming, or impervious to collective effects, the
greater good often is achieved by per se rules, which require greater
conformity, obedience, or zeal than actually may be desired. The
role agent faces a moral conflict whenever she believes that the
objectives of the rule are not served by compliance. These, how-
ever, are simply the circumstances in which the four-fold root does
not license a "morally-dissonant role act."
Thus, consider the duty of unqualified zeal in adversary crimi-
nal defense, which Luban endorses as a bulwark against state
power.5 4 While the goal of "overprotecting" defendants' rights 55
against the state does not require unqualified zeal in every case,
the official requirement helps the attorney resist the pervasive temp-
tation to distrust her client and accommodate the state.5 6 If defense
attorneys were permitted explicitly to make exceptions, those excep-
tions quickly would swallow the rule. Role-breaking by attorneys,

52. This is somewhat analogous to the claim that the admission of highly probative
but unconstitutionally obtained evidence against a defendant is justifiable when there
would be no marginal deterrence in excluding it, if the sole purpose for its exclusion is
to deter police misconduct. A defendant suffers no injustice, because she did not have a
right to its exclusion, which is only an incidental benefit derived from a deterrent policy.
Of course, the claim that its exclusion has no marginal deterrent value is highly ques-
tionable: unlike a single act of role-breaking by an individual attorney, a court's admis-
sion of illegally obtained evidence may indeed have a subversive effect on the policy
underlying the rule.
53. See, e.g., D. LYONS, FORMS AND LIMITS OF UTILITARIANISM 27-29 (1965).
54. D. LUBAN, supra note 1, at 62-63.
55. Id. at 58 n. 17.
56. Luban discusses the interest of the experienced defense attorney "in playing ball
with the prosecution and encouraging the client to plea bargain." Id. at 60. Thus, Luban
implicitly recognizes the temptation to acquiesce to state power.
 MARYLAND LAW REVIEW [VOL. 49:392

like nullification by juries, may be desirable only when it is carried

out in defiance of the law.
If this is the justification for unqualified zeal, the attorney
hardly can invoke it to decide whether to check her zeal in an excep-
tional case. After she reminds herself that the law prohibits excep-
tions by recognizing a danger of bias or bad judgment from which
she is not exempt, she must still decide how to act.5 8 If the contem-
plated breach would threaten significantly the protection of the de-
fendant's rights, her decision to comply will be easy. If, more
typically, the indirect and long-term effects of her breach appear
negligible, the four-fold root will not license her performance of a
59
morally dissonant role-act.
Although Luban insists that the four-fold root may be applied
equally well by the individual role agent and the legislator,60 it is far
more appropriate for the legislator. 6 ' A strong institutional justifi-
cation supports a role requiring morally-dissonant acts more readily
than it licenses those acts themselves, because the design of the role
usually has a much greater impact on the goals of the institution
than does the performance of any single act. By recognizing a duty
of confidentiality (for example, by exempting the lawyer from civil

57. This justification for unqualified zeal has much the same structure, and the same
tension, as the utilitarian justification for per se prohibitions against lying and promise
breaking. As Philip Devine explains,
[t]he gist of the argument... is that only if the obligation to tell the truth and
keep one's promises is given force in excess of that provided by the utility of
specific acts of truth telling and promise keeping can the utilities resulting from
honesty as a social practice be realized.
Devine, The Conscious Acceptance of Guilt in the Necessary Murder, 89 ETHIcs 221, 223 (1979).
While the reasons for excessive force may be quite different in the case of zealous advo-
cacy, the resulting tension is similar: the conscientious moral agent inevitably will find
herself in the position of endorsing the policy while having good, and sometimes com-
pelling, moral reasons for refusing to conform to it. Devine argues that this position
ultimately is indefensible, or immoral, id. at 224, but recognizes that some variation on it
underlies a large number of policies, from nuclear deterrence, id. at 228, to civil disobe-
dience. Id. at 229-33.
58. See D. LUBAN, supra note 1, at 138-39.
59. Luban himself recognizes that a single act of role-breaking rarely has more than
a negligible effect on the role and the institution. He comments that the effect of a role
violation "cannot amount to much" unless the violation is widespread. Id. at 138 n.7.
60. Id. at 138-39.
61. This is not surprising because as I have argued, the four-fold root provides a
strictly consequentialist justification. As Rawls pointed out 35 years ago, "utilitarian
arguments are appropriate with regard to questions about practices .... " Rawls, supra
note 49, at 5. Using the example of punishment, Rawls argued that while sentences
should be assigned by the judge on retributive or "just-deserts" grounds, the institution
of punishment is established by the legislature on the basis of deterrent and other utili-
tarian considerations. Id. at 6-8.
1990] REVIEW EsSAY-Lawyers and Justice 403

or criminal liability for failure to disclose confidences, or by making

their disclosures an offense), the legislature may have a substantial
impact on the protection of individual rights, an impact it must
weigh against the harm that occurs by the practice of confidentiality.
The institutional justification also plays a more constructive
role for the legislator than the individual role-agent. As Luban
notes, the agent often confronts a stark choice between conformity 62
and disobedience, while the legislator's options are much broader.
Luban argues very effectively in chapter five that some of the less
attractive features of the adversary role are not, or have not been
shown to be, necessary for the goals of the adversary system.6 3 As
Luban suggests in his discussion of modified adversary advocacy, a
legislator might be better able to redesign those roles to achieve the
same good at less moral cost. 6' Once the legislator has done her
best, her job is complete. In contrast, the role agent still must de-
cide whether to perform the morally dissonant acts required of
her.65 The fact that her role was designed to keep such acts to a
minimum will not help her decide; all that matters is the good or evil
she now achieves by conforming with or breaking that role.6 6
If the Lake Pleasant lawyers were justified in keeping silent, it is
because their duties were only partially determined by their role and
its justification. If they had induced their client to confide in them
by promising confidentiality, they would be obliged to keep silent
whether or not the duty of confidentiality was strongly justified.6 7
Our ordinary moral duties to keep promises, fulfill reasonable ex-
pectations, and protect the vulnerable, often licenses morally disso-
nant role-acts that contribute very little to the good of a strongly
justified institution, or which onl.y are prescribed by a weakly justi-
fied institution.6 s

62. D. LUBAN, supra note 1, at 138. While the role-agent "takes the institution, the
role, and the role obligations as givens[,]" the legislator looks upon these as vehicles for
regulation. Id.
63. Id. at 72-73.
64. Id. at 138-39.
65. Id.
66. Id.
67. The difference between the Lake Pleasant case and its civil counterpart may be
that the strong policy underlying adversary, criminal defense will justify placing role-
agents in moral dilemmas that it cannot resolve, while the weak policy underlying adver-
sary civil defense will not. Perhaps civil lawyers should not elicit incriminating confi-
dences; even if having done so, they are obliged to honor them, while criminal lawyers
are justified in eliciting such confidences, although the obligation to honor them has the
same nonconsequentialist source as for civil lawyers.
68. A general critique of the four-fold root has been developed by Judith Andre,
using the issue of obligations to one's children and the institution of the family:
 MARYLAND LAW REVIEW [VOL. 49:392

In sum, I believe that the "moral goodness" of the adversary

system in criminal cases contributes far less, and less directly, to li-
censing morally dissonant acts by criminal lawyers than Luban
claims. Because the duties of unqualified zeal and strict confidenti-
ality are by their nature overbroad, they cannot justify every act of
zeal and concealment. In essence that policy requires greater zeal
and concealment than actually is desired, because that is the only
way to attain the desired level. On the other hand, many acts of zeal
and concealment not licensed by the institution nevertheless may be
justified by such ordinary moral considerations as the duties to obey
promises, to protect the vulnerable, and to defend against improper
tactics by an adversary. Luban's discussion fails to show how law-
yers, even those engaged in the most strongly justified forms of ad-
versary practice, are ever exempt from the demands of common
morality.

II. THE DUTY TO OBEY THE LAW

Another source of special constraints and prerogatives for law-

yers may be the law itself: Even if lawyers rarely are licensed to en-
gage in morally dissonant acts by their role in the adversary system,
their obligation to uphold the moral authority of the law signifi-
cantly may affect their range of permissible conduct.6 9 Ironically,
while lawyers usually invoke the adversary system to expand their
range of conduct, their obligation towards the law may in fact con-
tract it, requiring them to honor its spirit as well as its letter in serv-
ing individual clients.70
In chapter three, Luban defends the notion that citizens have a

[O]ne can and should ask about obligations to one's children, say, without also
asking whether the institution of the family is morally good. This is a separate
and important question; but however one decides, it remains a fact that these
children are as a matter of fact-and as a result of a social role which might not
be justified-particularly dependent on their parents.
J. Andre, Role Morality as a Complex Instance of Ordinary Morality (June 17, 1989) (to
be published in AM. PHIL. Q.). There is, however, no reason to believe that such ordi-
nary moral obligations always are available to support a role-act with a consequentialist
justification. See supra note 49.
A related point is that a wide range of deceptive, dilatory, and coercive legal tactics
are excused, if not justified, by the fact that they are employed by one's adversary. As
Professor Dees has pointed out, the case against "unilateral disarmament" is often pow-
erful, albeit treacherous, and those who rely on that defense may have a moral obliga-
tion to reduce the need to use it by working for multilateral disarmament. J. Dees, I'd
Be a Fool Not To! (May 5, 1989) (unpublished manuscript) (available from author at
Harvard University, Graduate School of Business Administration).
69. See D. LUBAN, supra note 1, at 48-49.
70. See id.
1990] REVIEW ESSAY-Lawyers andJustice 405

duty to obey generally beneficial laws, 7 ' and lawyers a concomitant

duty to maintain the generality of those laws. 72 He regards the rec-
ognition of these duties as integral to both his critique of the adver-
sary system and his proposals for reforming legal practice.7 3 If laws
simply are the self-serving ukases of a power elite, there is little op-
probrium in violating or manipulating them to selfish ends, but also
little potential in using them as a vehicle for redressing social injus-
tice.7 4 In claiming that a substantial subset of laws represents the
efforts of the community to distribute equitably its benefits and bur-
dens, and as such place a moral claim on the members of that com-
munity, Luban argues that lawyers have a duty to uphold the
fairness or generality of the law,75 thus preserving the feature that
gives the law its moral authority.
Luban's argument is not as central to his project as he suggests:
even if citizens have no duty to obey the law, lawyers would never-
theless do important moral work in promoting and enforcing equi-
table laws. Still, the claim that lawyers are responsible for
preserving the moral authority of the law gives legal practice a
unique moral urgency, and its rejection reinforces the realist posi-
tion that the moral force of the law lies exclusively in its social value.
Luban attempts to steer a middle course between those who
argue that there is a presumptive, or "prima facie," duty to obey all
laws and those who believe that the moral force of a law can be no
greater than that of the conduct or restraint it requires.7 6 The for-
mer cannot account for our readiness to ignore manifestly unjust
and irrational laws, nor the latter explain our resentment towards
those who regard themselves as "above the law." Luban's strategy
is to extract from our seemingly conflicting intuitions iri this area the
recognition of a duty to obey laws that establish generally beneficial
cooperative schemes, as a matter of respect for our fellow citizens. 7 7
Luban notes that those who defend a duty to obey the law usu-

71. Id. at 37.

72. Id. at 48-49. Luban defines the "generality requirement" to -mean that similar
cases should be treated similarly, and that laws should not confer benefits on one group
at the expense of another. Id. at 43-44.
73. Id. at 31-32.
74. See id. at 48-49.
75. Id. Luban is not concerned directly with the duty of lawyers to obey the law,
which might be both stricter and easier to justify than the duty of ordinary citizens.
Rather, he is interested in discovering the source of the citizen's duty and to assess its
implications for legal practice. Id. at 37.
76. Id. at 44-45.
77. Id. at 38. Luban might have adopted a very different strategy, arguing for a duty
to obey democratically enacted laws, however fair their provisions. This "purely proce-
406 MARYLAND LAW REVIEW [VOL. 49:392

ally argue that fair play conditions obligation on consent: it is wrong

to accept the benefits of a social scheme without sharing its bur-
dens.7" He, however, argues following Simmons,7 9 that we only can
be regarded as accepting benefits if we could deny those benefits
without significant cost.8 0 Most benefits conferred by the law fail to
satisfy that criterion because we cannot decline them without up-
rooting ourselves from our communities.
Luban accepts the idea that our duty of fair play rests on the
benefits we receive from cooperative schemes, but he seeks to show
that these benefits may be binding even if they have not been volun-
tarily accepted. He argues for a duty of fair play based on reciproc-
ity but not consent: we are bound, out of respect for our fellow
citizens, to assume the burdens of a generally beneficial cooperative
scheme only if we receive or expect to receive important benefits
from it. 8 Thus, a scheme conferring such benefits creates a duty
that would otherwise not exist. This approach may appeal to Luban
because it gives the law a role in creating duties, rather than merely
enforcing duties that already exist.
An approach based on reciprocity, however, construes the duty
of fair play too narrowly: we may be obliged to comply with fair
cooperative schemes regardless of whether we have received or ex-
pect to receive any benefit from them. The ex ante balance of bur-
dens and benefits that makes the schemes binding on us, not the
benefits they actually confer. In arguing for a reciprocity-based duty
of fair play, Luban fails to distinguish free-riding-the unearned en-
joyment of nonexcludable collective goods-from unfair dealing,
which need not involve unreciprocated benefits. We may act un-
fairly without free-riding, if we refuse to comply with a fair coopera-
tive scheme that has not benefitted us. Conversely, we may free-ride
without acting unfairly, if we have no prior duty to act as the scheme
requires and we have not actively sought or accepted the scheme's
benefits.

dural" approach would have been consonant with his later emphasis on political legiti-
macy in a democratic state. Id. at 239-40.
78. Id. at 37-38. Consistent with Luban, I use "consent" to include acceptance and
"acceptance" to include what contract law calls "ratification," for example, drawing
water from the well constructed by our neighbors. Id. at 36-38. Simmons uses the term
"participation" to cover these and other forms of voluntary involvement in a coopera-
tive scheme. A. SIMMONS, MORAL PRINCIPLES AND POLITICAL OBLIGATIONS 127-36
(1979).
79. A. SIMMONS, supra note 78, at 127-29.
80. D. LUBAN, supra note 1, at 39.
81. Id. at 41-42.
 1990] REVIEW ESSAY-Lawyers and Justice 407

Luban offers three examples to show how the unaccepted bene-

fits of fair cooperative schemes can bind their recipients. 82 These
examples prove both more and less than he intends: they show that
we can be bound without consent or benefit, but do not show how
unaccepted benefits alone can bind us.
Case 1: Drunken teen-agers smash fourteen six-packs
of Heineken's empties on your street, making it impassible
for autos. Your neighbors grab rakes, shovels, and
brooms, and clean up the glass; you sit on your front porch
(meditatively downing a nice frosty Heineken's) and watch
them. As soon as the glass is gone, you climb into your car
and drive off into the sunset....
Case 2: On a trip to London, you observe people
queuing up at a bus stop. How nice! you think. How un-
like Manhattan! Forthwith, you cut into the front of the
line....
Case 3: You are driving on a highway, and two lanes
must squeeze into one.... You see that the cars in the two
lanes are taking turns. You know (let us suppose) that this
method advances the line of traffic most rapidly.... When
you come to the head of the line, you ...skip ahead out of
83
turn.
Luban contends that in all three examples, the scheme confers a sig-
nificant benefit-enhanced mobility-on individuals who have had
no opportunity to consent.8 4 He argues that in each case we are
obliged to comply, because the failure to do so, as evidenced by cut-
ting in front, skipping ahead of turn, or declining to help out, shows
disrespect toward our fellow citizens.8 5
Luban's examples do not establish a duty to participate in coop-
erative schemes based on the benefits they confer. 8 6 Luban con-

82. Id. at 39-40.

83. Id.
84. Id. at 40.
85. Id. at 41. iuban recognizes that not all schemes which confer significant benefits
also impose an obligation to comply: he supports the libertarians in denying that we
have a duty to join our neighbors in planting flowers on the median strip in front of our
homes. Id. at 41-42. He distinguishes his examples by the importance of the benefit
they confer: "hassle free transportation" is "more basic to daily life" in our society than
neighborhood beautification. Id. This difference is in part a matter of convention: "in
most urban neighborhoods it is considered presumptuous to demand civic commitment
from residents when it goes beyond necessities." Id. at 42.
86. While Luban does not claim that cooperative schemes are binding simply because
of the benefits they confer, he insists that when they are sufficiently important or reason-
able, our duty of respect requires us 'to reciprocate: "The obligation of fair play arises
out of an obligation not to exhibit disrespect for ones' moral equals by free-riding." Id.
408 MARYLAND LAW REVIEW [VOL. 49:392

flates two distinct types of situations in which we feel bound by a

duty of fair play. In his first two cases, we have strong antecedent
beliefs about how the burden or inconvenience should be distrib-
uted and confront a scheme that makes distribution feasible. In the
third, we lack such strong antecedent beliefs, and any duty we feel
arises from the operation of the scheme. It is only in this third case
that noncompliance looks like free-riding, and even here, the ap-
pearance is deceptive.
If our sense of fairness in the first two cases depends on a bal-
ance of benefits and burdens, it is an abstract, hypothetical balance:
we may believe that first come, first served, or turn taking, is the
fairest arrangement because it tends to equalize the burden of wait-
ing, or because on average, it will yield the shortest waiting time for
a passenger or driver. This weighing of benefits and burdens is
strictly ex ante: the actual or prospective benefits to an individual
waiting for the bus or driving towards the roadblock are irrelevant;
he is obliged to wait in line, or take his turn, simply because that is
the fair arrangement. If he fails to comply with that scheme, he is
acting unfairly, but he is not necessarily free-riding; that will depend
on the benefits he actually has received or is likely to receive from
the scheme.
In analyzing the first two cases, I argue: (1) that the indignation

at 45. Respect is a far more imperfect duty than fair play, rarely requiring us to shoulder
burdens proportionate to unaccepted benefits.
Luban is unclear about the relationship between the duty not to disrespect our
fellow citizens and the duty to play fairly with them. Towards the end of his discussion,
he states:
[T]he obligation to join in the clean-up is an obligation of fair play. The obliga-
tion of fair play arises out of an obligation not to exhibit disrespect for one's
moral equals by free riding. But noncooperation in extenuating circumstances
(such as minding the baby) exhibits no disrespect for one's moral equals.
When I am minding the baby, I am playing fair, even though I do not join in the
clean-up. So there is no obligation to join in the clean-up.
Id. While the logic of this argument is obscure, it seems clear that disrespect, and not
fair play, must wear the pants. While we satisfy the duty to avoid disrespect by playing
fairly, we need not satisfy the duty of fair play to avoid disrespect. Luban's claim that we
play fairly by minding the baby distorts the usual understanding of fair play, which re-
quires burdens proportionate to benefits within a scheme of cooperation. Our perform-
ance of tasks external to that scheme, like child-care, may exempt us from the duty of
fair play, but it does not satisfy that duty. Had we given our prior consent to the scheme,
we might be obliged to compensate the others if child care prevented us from taking
part. We are free-riding when we mind the baby instead of cleaning up the glass, but we
show no disrespect in so doing because of the personal importance of our alternative
chore. If Luban wants to define the duty of fair play so that it always is satisfied when we
display respect, it will not require us to assume burdens proportionate to the benefits we
receive.
1990] REVIEW ESSAY-Lawyers and Justice 409

we feel toward the person who fails to comply does not depend on
his having received, or being likely to receive, any benefits from the
scheme, (2) that the role of the cooperative scheme is not to create a
duty of fair play, but to make it feasible to satisfy that duty, thereby
removing an excuse for its violation, and (3) that we would not even
acknowledge a duty to comply unless we thought that the underly-
ing distribution of inconvenience was fair, independent of any
scheme for effecting it.
In the bus-queue case, Luban may be correct in claiming that
queuing benefits even those who arrive last, by expediting the
boarding process and by reducing their average standing-time over
the long run.8 7 I, however, do not think that our indignation at the
person who cuts in front is aroused by her ingratitude at these fairly
slight advantages. We would be equally indignant even if the cir-
cumstances denied her any short- or long-term advantage from the
queuing arrangement, for example, if she cut ahead of a relatively
short queue to board the Heathrow-Gatwick bus on her only pas-
sage through England.
Our sense of obligation arises from the fact that the queue al-
lows us to honor our pre-existing duty to defer to those who have
arrived first. A rule of "first come, first served" equalizes the bur-
den of waiting in settings in which variations in need are relatively
slight and impractical to ascertain: it is appropriate for a bus stop,
not for an organ transplant. Most of us would acknowledge a duty
to yield to those who have arrived before us, but would feel com-
pelled to honor that duty only if we could do so at reasonable cost.8 8
In the absence of a cooperative scheme, the cost often will be
excessive. In the free-for-all of a Manhattan bus stop, we may not
even know who came before us, and if we did, our attempt to defer
to them might allow dozens of late arrivals to squeeze ahead. There
is no practical way of giving our predecessors priority without losing
our place to our successors. The London queue provides the an-
swer to the dilemma. We are obliged to respect it because it does
not create the obligation and removes a strong defense for breach-

87. Id. at 41.

88. Some evidence of the cost we are willing to bear is provided by Dick Larson, a
queuing expert interviewed in The Boston Phoenix. Gower, Life in the Slow Line, Boston
Phoenix, Aug. 11, 1989, § 2, at 1. He observes that customer satisfaction is reported to
be higher at fast food outlets with single rather than multiple queues, although the ac-
tual waiting time is twice as great. Id. at 5. Larson attributes this effect to "the assurance
of equity." Id. Admittedly, the preference for single queues also may reflect risk-
aversion.
410 MARYLAND LAW REVIEW [VOL.. 49:392

ing it. 89
The lane-closing case 9° is somewhat intermediate between the
bus queue and the glass clean-up, because a turn-taking rule for a
lane obstruction may seem less obviously fair than a first-come, first
served rule at the bus-stop: cars side-by-side at the obstruction may
have suffered very disparate delays in reaching that point. If, how-
ever, we think that a turn-taking rule is the fairest way to distribute
the delay from a closed lane, the arrangement Luban describes al-
lows us to do so without unreasonable cost. It would ask too much
to require us to let in the car on our right unless we could count on
the cars behind it to wait their turn, but the arrangement we observe
gives us the assurance that they will.
Luban may assert correctly that a turn-taking arrangement at
the blocked lane confers a significant benefit by expediting the flow
of traffic. 9 ' Again, however, I do not think that it is this benefit that
obliges us to comply. As in the bus-queue case,9 2 our obligation is
not conditioned on having benefited from the scheme, but on the
assurance it provides that we can act fairly without too great a bur-
den. If we felt that turn-taking was equitable, we would be obliged
to wait our turn without a pre-existing arrangement, e.g., if we were
among the first to arrive at the obstruction, if our predecessors in
the left lane simply had barreled ahead, or if the lead driver in the
right lane had been too timid to take his turn. Assuming we could
safely yield, it would be inconsiderate not to permit the car to our
right to enter, even if we were the first driver to do so.
In both the bus-queue and blocked-lane cases,9 3 the obligation
we feel to comply with the cooperative scheme is contingent upon
our acceptance of the priority rule it enforces. 94 If we reject that
rule altogether, we feel little or no obligation to comply. Thus, if we
regard highway driving as a lottery in which drivers took the luck of
their lane, like customers in separate lines at a bank or a McDon-

89. This implies that if circumstances permit us to defer to earlier arrivals without
risk, we are obliged to do so even in the absence of a queuing arrangement. I fully
accept this implication. Thus, if we are the last to arrive at a Manhattan bus stop, it
would be rude of us to board before the earlier arrivals.
90. D. LUBAN, supra note 1, at 40.
91. Id. at 40-41.
92. Id. at 40.
93. Id.
94. Alternatively, we may feel obliged just because we come to accept the consensus
expressed in the cooperative scheme about the appropriate priority rule. If our own
beliefs about the correct rule are uncertain or inconsistent, it may be reasonable to defer
in this way to the "decent opinion" of our fellow citizens. If we do, however, their con-
duct has caused us to recognize an obligation; it has not created one.
1990] REVIEW EsSAY-Lawyers andJustice

aid's, we would feel no duty to take turns with drivers in an ob-

structed right lane, and would resent the fact that the drivers ahead
of us in the left lane felt so obliged. 95 While we might regard their
deference as part of our bad luck, we certainly would feel no later
obligation to defer when our turn came up.9 6
Luban's first case, the glass clean-up, 9 7 differs from the other
two in that it does not involve an antecedent duty of fair play. There
are an indefinite number of ways the effected neighbors could divide
fairly the burden of cleaning up, and in the absence of a cooperative
scheme, fairness would not require them to pick up a single shard of
glass. Here, the very existence of a cooperative scheme appears to
create a duty where none previously existed; the question is whether
it can do so without consent. The actual or prospective benefits
conferred by the scheme appear to contribute to the recipient's duty
even if he does not accept them voluntarily. The person who re-
ceives the benefits of such a scheme without assuming the burdens
can be seen as a free-rider. This is the strongest case for Luban's
claim that we are sometimes required to take part in cooperative
schemes, because our offense in opting out appears to consist in
nothing but the fact that we are free-riding on the efforts of our
neighbors.9 8 The question, however, is what distinguishes this
scheme from others, like the flower-planting,99 in which we also
seem to be free-riding if we opt out. The answer does not lay in the
size of the benefits conferred, nor the ratio of benefits to burdens,
because the planting might yield enormous pleasure at very slight
cost. 10o

95. Luban might respond that, given our "moral baseline," we would receive no
benefit on this occasion from the turn-taking arrangement-it certainly would not get us
through as fast as a "tough luck" approach. This, however, would relativize the benefits
of a scheme to the recipient's preferred alternative: people who were delayed equally by
the obstruction would be benefited unequally by the turn-taking scheme, if they favored
different alternatives which would have resulted in different delays. It is absurd to claim
that the more self-serving the preferred alternative, the less the benefit conferred and
the corresponding burden imposed by the turn-taking scheme.
96. An obligation to take turns also might arise from the fact that in so doing, we
contribute to the solution of a coordination problem, while we impair the solution in
refusing to do so. But, as Regan has noted, this is a strictly consequentialist concern,
having nothing to do with a duty of fair play. Regan, Law's Halo, 4 Soc. PHIL. & POL'Y
15, 17 (1986).
97. See D. LUBAN, supra note 1, at 39-40.
98. See id. at 39.
99. See id. at 41-42.
100. Luban argues that the obligation to participate in whichever type of cooperative
scheme hinges "on the importance, or at any rate the reasonableness, of the cooperative
scheme in question." Id. at 41.
 MARYLAND LAW REVIEW [VOL. 49:392

Rather, I believe that the critical factor only is related contin-

gently to the objective size or social importance of the benefits-it is
our willingness to burden ourselves to obtain those benefits in the
absence of a cooperative scheme. We exploit our neighbors if, by
opting out, we allow their efforts to relieve us of a burden that we
would otherwise have shouldered. 10
Our hypothetical willingness to burden ourselves differs signifi-
cantly in the flower-planting and the glass clean-up: the description
of those cases suggests that we would not have personally lowered a
finger to till the soil, while we would have tried to clear a path for
our car in the absence of a group clean-up.10 2 By opting out of the
clean-up, we exploit our neighbors to avoid labor that we would
have undertaken independently.'0 3 If we had been willing to plant
flowers on our own, our opportunistic withdrawal would be equally

101. This is true even if we would not in fact have been willing to do anything without
group assistance, because a solo effort would have been futile. What matters is whether
we would have been willing to do as much for ourselves as our "share" of the group
effort required, if that would have gotten the job done. I think we take advantage of our
neighbors by opting out in circumstances in which we would be willing to put "x" effort
into a project which would require greater than "x" to do by ourselves but less than "x"
to do with others. While we would not have done the work if the group had not, because
the effort would have been excessive, we would have done as much for ourselves as the
group scheme required of us if that lesser effort sufficed to get the job done. In this
sense, we are allowing others do our work for us if we do not participate. This suggests
that the strongest case of exploitation will not be the paradigm case of free-riding, in
which the individual is always better off doing nothing. This sense of exploitation re-
quires that there be at least some hypothetical circumstances in which the individual
would be better off by exerting herself.
102. See D. LUBAN, supra note 1, at 39-40, 41-42.
103. I think this sense of exploitation explains the duty to participate that Christie
claims to find in a neighborhood campaign to restore basic services cut off by civil war or
natural calamity. G. Christie, On the Moral Obligation to Obey the Law 15 (Oct. 28,
1988) (unpublished manuscript) (paper given at Columbia University Law School Legal
Theory Workshop) (copy on file with Maryland Law Review). We can safely assume that
virtually anyone would work to restore his own services, and thus that the holdouts are
exploiting their neighbors' efforts. Simmons relies on a similar inference from objective
importance to individual disposition in an exchange with Klosko. Klosko, Presumptive
Benefit, Fairness, and Political Obligation, 16 PHIL. & PUB. AFF. 241 (1987); Simmons, The
Anarchist Position: A Reply to Klosko & Senor, 16 PHIL. & PUB. AFF. 269 (1987). Klosko
argues for a duty, not conditioned on consent, to contribute to schemes providing goods
that are (1) "presumptively beneficial"-those which, like Rawls' "primary social
goods," everyone is presumed to want, Klosko, supra, at 246 (citingJ. RAWLS, THEORY OF
JUSTICE 62 (1971)); and (2) worth the cost to the typical recipient. Id. at 248. Simmons
contends that to the extent that such a duty rests on fairness rather than general benefi-
cence, presumptive benefits will be binding only because they can be presumed to elicit
participation; in other words, because it can be presumed that almost anyone would
want them at the price. Simmons, supra, at 273-74. Simmons further contends that
schemes providing only "discretionary" goods do not bind just because participation in
obtaining them cannot be presumed. Id. at 274.
1990] REVIEW EsSAY-Lawyers and Justice 413

offensive. 01 4 Conversely, if we were so lazy that, left to our own de-

vices, we would have relied on mass transit until the next good rain,
0 5
we would not exploit our neighbors by sitting out the clean-up.1
Luban might respond that I simply argue for the principle of
respect for one's fellow citizens; exploitation, after all, is an aggra-
vated form of disrespect.'0 6 If, however, we have a duty to respect
our fellow citizens, it does not require us to assume burdens propor-
tionate to the unaccepted benefits we receive. For example, if we
clean up less than "our share" of the broken glass, but as much as
we would have independently, we show no disrespect. The individ-
ual who only does as much as he would have without the coopera-
tive scheme may lack public spirit, but he does not lack respect.
Luban recognizes that respect does not compel us to join our fellow
citizens if they "stand immersed up to their necks in the outhouse
tank,"' 0 7 but fails to see that it may not compel us to join them in far
more beneficent ventures. Thus, I do not think that Luban's exam-
ples demonstrate that our duty of respect obliges us to reciprocate
the benefits of a cooperative scheme to which we have not
consented.
Even if Luban's examples established that some cooperative
schemes imposed a duty to participate independent of consent, the
law itself would play only an incidental role in creating that duty. In
all three of his examples, noncompliance shows disrespect only on
the assumption that most other citizens are complying. 0 8 If not,
our compliance may be impractical or supererogatory: impractical if
the lack of compliance makes it difficult for us to ascertain how to
comply, supererogatory if the general lack of compliance means that
others will take costly advantage of our self-restraint. 0 9 Thus, in

104. But not if we only had decided after seeing the results of the planting that it
would have been worth the effort to us to take part. In these circumstances, we would
not have been exploiting our neighbors at the time of the planting when we sincerely
believed that we would not have made the effort independently.
105. The case would be even stronger, but quite different, if we were unwilling to join
in because we did not own a car. In that case, we would receive objectively less benefit
from the clean-up, and it might be unfair to impose the same burden. I believe that a
notion of fair play requires an objective measure of benefits and burdens, although I
recognize the distinction may be hard to draw. I also think certain reasons for opting out
may be inherently disrespectful, e.g., the belief that street-cleaning is "beneath our
dignity."
106. See D. LUBAN, supra note 1, at 42-43.
107. Id. at 44.
108. See id. at 49.
109. For someone who regarded duties of fair play as antecedent to any cooperative
scheme for enforcing them, the absence of such a scheme would at most excuse non-
compliance. For Luban, however, the duty to comply would not even arise, since fair
 MARYLAND LAW REVIEW [VOL. 49:392

the free-for-all of a Manhattan bus-stop or traffic-jam, we may have

no way of assigning priority, and we risk being delayed by those with
lower priority if we defer to those with higher priority.
On the other hand, if most citizens comply with a cooperative
scheme, it is their compliance, not the fact that it is undergirded by
law, that obliges us to do likewise. For example, it might seem fair
to require slow drivers to pull over for the cars behind them,
although few slow drivers do so, and few states require them to. If
pulling-over became a general practice, Luban would regard those
who failed to do so as free-riding, since they would benefit from the
compliance of slower drivers without burdening themselves on be-
half of faster ones. But the driver who failed to pull over would be
no more disrespectful, no more of a free-rider, in states in which the
practice was enforced by powerful social norms than in states in
which it was established by law.
In those cases in which general compliance with the scheme
would not be possible without the coercive force of law, our duty to
cooperate is analogous to our duty to stop at red lights at busy inter-
sections. As Donald Regan has argued, what obliges us to stop is
not the law requiring us to do so, but the fact that others rely on our
compliance, and that we risk considerable harm in violating their
expectations.'1 0 Similarly, what obliges us to comply with coopera-
tive schemes is not the laws creating them, but the beneficial con-
duct those laws elicit from others.
Even if citizens do not have a moral duty to obey the law, how-
ever, Luban's objection to the instrumentalist lawyer"' has consid-
erable force. The lawyer who twists the letter of the law to serve the
narrow interests of a client "undermine[s] the generality of law," 1 2
and this is a serious offense. Luban, however, cannot claim that the
lawyer thereby "abrogate[s] the moral authority of the law . .. re-
ducing it once again to coercion," because he has not 13
demonstrated
that the law has any independent moral authority.'

play is based on reciprocity, and in the absence of an effective scheme there would be no
benefits to reciprocate.
110. Regan, supra note 96, at 16.
111. Luban implicitly acknowledges this in accepting that the compliance of most citi-
zens is a condition of the obligation to obey the law. D. LUBAN, supra note 1, at 38. He,
however, fails to recognize that this condition significantly attenuates the connection
between fair play and the obligation to obey even a small subset of laws.
112. Id.
113. Id.
1990] REVIEW EsSAY-Lawyers andJustice 415

III. THE RIGHT TO LEGAL SERVICES AND THE OBLIGATION TO

PROVIDE THEM

Luban finds a third source of moral constraints on legal practice

in its relationship to the state. He argues that a right to legal serv-
ices is implicit in our basic form of government, because profession-
ally-assisted access to the courts is necessary for government's
legitimacy.' 14 He contends that lawyers have a special obligation to
make their services available to the poor because the legal profes-
sion is a creature of the state and subject to its ends.1 5 These argu-
ments are best considered in the context of the contemporary
debate on the distribution, of legal services.
Two distinct claims recently have been made concerning the
distribution of legal services: (1) that the case for redistributing
legal services is stronger than, or at least distinct from, the case for
redistributing other professional services, and (2) that lawyers have
a special obligation to make their services available to those who
cannot afford to pay for them. These claims may be independent:
arguments for redistribution may say nothing about who should
bear its cost, while arguments for pro bono service may be based on
the obligation of lawyers to the system at large, not on the rights of
the poor to legal services.
The independence of the two claims has been recognized by
writers with strong positions on both sides. Charles Fried claims
that forcing lawyers to represent the poor is incompatible with the
lawyers basic role; however, he does not object to subsidizing the
representation of poor people through the general revenues." 16
Alan Wertheimer has argued that justice requires the equalization of
legal resources in specific conflicts, but he takes no position as to
whether equality is best achieved by providing legal services for the
"have nots" or by limiting the legal resources of the "haves.""'17
Luban, in contrast, argues for both the right of the poor to free legal
services 18 and the obligation of lawyers to provide them." 19

114. Id. at 255.

115. Id. at 256.
116. See Fried, The Lawyer as Friend: The Moral Foundationsof the Lawyer-Client Relation,
85 YALE L.J. 1060, 1077-80 (1976).
117. See Wertheimer, The Equalization of Legal Resources, 17 PHIL. & PUB. AFF. 303, 304-
05 (1988).
118. D. LUBAN, supra note 1, at 237-64.
119. Id. at 282-88.
416 MARYLAND LAW REVIEW [VOL. 49:392

A. The Right to Legal Services

Although Luban and Wertheimer propose different standards
for an acceptable distribution of legal services, their arguments for
redistribution have interesting parallels. Luban argues for universal
access to minimally competent legal services,1 21 Wertheimer for
parity in the legal resources devoted to specific disputes.' 2 ' Luban's
notion is wider, but thinner than Wertheimer's: wider because the
right to access is not triggered by a specific dispute and thinner be-
cause minimally competent services may leave indigents severely
22
disadvantaged in disputes with wealthier adversaries.
While their proposed distributional standards differ, Luban and
Wertheimer both attempt to make a case for redistribution that is
independent of the seemingly intractable controversy over such
"welfare rights" as secure housing and adequate health care. Both
argue that the need for redistribution arises not from a moral right
to legal services, but from the constraints imposed by the larger so-
cial system in which those services are embedded-for Wertheimer,
the adversary system, t 21 and for Luban, the American form of gov-
ernment. 124 Wertheimer contends that equal resources are neces-
sary to achieve the adversary system's basic goal of maximizing
correct results in legal disputes.' 2 5 Luban argues that universal ac-
cess is necessary to legitimate a government based on equality

120. Id. at 240-41. Luban concedes that access to equal legal services would be a
"better interpretation of the meaning of equal access," but feels that it would be a pro-
hibitively expensive, politically impossible goal. Id.
121. Wertheimer, supra note 117, at 320-2 1. Wertheimer believes that universal ac-
cess would be harder to defend against a libertarian objection than would "negative"
equalization, because it is less restrictive of liberty to prohibit a rich litigant from over-
spending on legal services than to require a third party to pay for a poor litigant's access
to those services. Id. I doubt that many libertarians would find it more objectionable to
take money from someone by taxation than to prohibit that person from spending it on
protecting what she believed to be her entitlements. To appeal to libertarians, the dis-
tinction would have to rely on the acceptance of the judicial system as a legitimate func-
tion of the minimal state, allowed to impose spending limits on disputing parties. But if
people can be taxed to set up the court, why can they not be taxed to subsidize access to
it?
122. These two forms of redistribution have complementary deficiencies. Equaliza-
tion is impracticable if it is limited to specific disputes, while access is ineffective unless it
is more than minimal. Because private attorneys structure their clients' transactions to
avoid disputes and to ensure success when they occur, equalization in specific disputes is
difficult to achieve without massive handicapping of the more privileged party. Because
the legal problems of the poor often require considerable skill and perseverance, mini-
mal access barely will scratch the surface.
123. Wertheimer, supra note 117, at 304.
124. D. LUBAN, supra note 1, at 243-44.
125. Wertheimer, supra note 117, at 304.
1990] REVIEW EsSAY-Lawyers andJustice 417

before the law.' 26 In permitting an inadequate distribution of legal

services, the system undermines its own justification.
Both arguments have an appealing parsimony, as they derive
distributional claims from the system's self-justification. Because
these distributional claims arise from within the system, however,
their moral force depends on the justice of that system and on the
very issues concerning distributive justice that Wertheimer and
Luban attempt to sidestep. These strengths and limitations become
apparent in considering their arguments in greater detail.
Wertheimer argues that because an adversary system requires
evenly-matched disputants for accurate results, equality of resources
is not merely desirable as a matter of distributive justice but neces-
sary to achieve the system's own ends:
Let us assume that the adversary system is the best (or at
least a satisfactory) mechanism for achieving just re-
sults .... Yet once the argument for the adversary system
is put in these terms, it seems reasonable to suppose that
the equalization of legal resources, broadly construed, will
result in a higher probability ofjust results than the present
laissez-faire system . . . . It is arguably inconsistent and
self-defeating to allow the use of grossly disparate
resources.... 27
Thus, a system which permits great disparities in legal resources vi-
olates the conditions necessary to maximize just results.
Luban argues that (1) access to minimally competent legal serv-
ices is necessary for access to the legal system,i 28 (2) access to the
legal system is necessary for equality under the law,' 22 and (3)
equality before the law is a "legitimation principle" of American
government. 13 0 Luban argues quite persuasively, in light of Gideon
v. Wainwright 1' and the complexities of modern procedural law,
that meaningful access to the legal system is impossible without a
lawyer."t 2 The further claim that equality under law requires access
to the legal system is necessarily true, because Luban defines equal-
ity under law to mean access to the courts to enforce one's legal
rights. 133

126. D. LUBAN, supra note 1, at 244.

127. Wertheimer, supra note 117, at 310-11.
128. D. LUBAN, supra note 1, at 264.
129. Id.
130. Id.
131. 372 U.S. 335 (1963) (establishing right to counsel in felony cases).
132. D. LUBAN, supra note 1, at 243-48.
133. Id. at 252-54.
418 MARYLAND LAW REVIEW [VOL. 49:392

The critical step in Luban's argument is the third, in which he

claims that equality before the law, or equality of legal rights, is a
"legitimation principle" for our form of government. 13 4 Luban de-
fines a legitimation principle as "an account of legitimacy offered by
a government to its citizens," under which its citizens consent to be
governed.'35 A government which defies a legitimation principle
undermines its own authority: "[It] would be like the pope declar-
ing ex cathedra that God does not exist! If there is no God, the pope
1 36
cannot make ex cathedra utterances."'
Luban also makes an historical argument for equality under law
as a legitimation principle. He notes that it has been embraced by
most Americans over the past two centuries, despite their sharp dis-
agreements about distributive justice. 3 7 He further observes that
access to the courts traditionally has been the first incident of mem-
bership in our political community: the right to sue was granted to
women, blacks, and now to aliens, before the right to vote.' 3 '
The strength of Luban's and Wertheimer's arguments is also a
limitation. Both offer what Luban calls an "immanent critique,"
which does not depend on the system's satisfaction of independent
criteria of justice."5 9 The moral force of the critique, however, de-
pends heavily on the extent to which those criteria are satisfied. If
the laws enforced by the system essentially are unjust, greater jus-
tice may well be achieved by disparate legal resources or unequal
access.
This limitation is more obvious for Wertheimer. If the goal of
140
adjudication is to maximize just results, as Wertheimer assumes,
and equal legal resources maximize legally correct results, which is all
that defenders of the adversary system can plausibly claim, then
equalization maximizes the former only to the extent that they cor-
respond with legally correct results.' 4 ' To achieve justice in a basi-
cally unjust system, it often is necessary to produce legally incorrect

134. Id. at 244.

135. Id. at 251.
136. Id. at 252.
137. Id. at 252-53.
138. Id. at 253.
139. Id. at 265.
140. Wertheimer, supra note 117, at 304.
141. Wertheimer implicitly concedes this point in placing his equalization proposal
under the rubric of "partial compliance theory," which he understands to deal with un-
just departures from "the criteria of a just society." Id. at 314-15. He obscures the gap
between just and legally correct outcomes by defining equalization of legal resources as
the allocation of resources which maximizes just outcomes. Id. at 304-05.
1990] REVIEW EsSAY-Lawyers andJustice 419

results, which may require that the weaker party receive far greater
legal resources than a powerful adversary.
A similar qualification applies to Luban's argument. Even if
equal access to the courts is required for the legitimacy of the Amer-
ican government, differential access may serve as a corrective to the
injustices produced by that government. Thus, it has been sug-
gested that a moral equilibrium has been achieved in tort law be-
tween access barriers favoring defendants and rules of liability
favoring plaintiffs.14 2 Removing access barriers might well destroy
this equilibrium and iricrease substantive injustice. To achieve jus-
tice in an unjust political order, it may sometimes be necessary to
violate its legitimation principles.
Wertheimer and Luban also suggest another reason why the
state is obliged to provide legal services to those who cannot afford
them. If the government declines to do so, it does not merely fail to
help the poor, but harms them in a way different than denying them
adequate health care or housing. 43 Thus, Luban argues:
[T]he selective exclusion of the poor from the legal system
does not simply fail to confer an advantage on them-it ac-
tively injures them. For a legal system does more than pro-
tect people from each other: it enormously expands our
field of action .... [A] network of practices-of power and
privilege-is set up from which those who have no access
to the system are excluded; and this exclusion itself intensi-
fies the pariah status of the poor. . . . The state has not
been an innocent bystander observing the regrettable spec-
tacle of economic inequality and poverty: it shares primary
responsibility with the legal profession . . . for the fact that
the poor have no meaningful
144
access to justice and are made
worse off by that fact.
Wertheimer draws a specific contrast to medicine:
Although the distribution of medical resources may be
somewhat competitive, it is not directly adversarial. By con-
trast, an improvement in the legal resources available to
one party has a direct negative effect on the interests of
another. For these reasons, the distribution of legal re-
sources-as opposed to the absolute level-is arguably

142. Cf Krier & Gillette, The Un-easy Casefor Technological Optimism, 84 MICH. L. REV.
405 (1985).
143. D. LUBAN, supra note 1, at 265-66; Wertheimer, supra note 117, at 322. Only
Wertheimer takes his argument to imply the priority of the right to legal services over
the right to health care. Id. at 305.
144. D. LUBAN, supra note 1, at 247-48.
420 MARYLAND LAW REVIEW [VOL. 49:392

more important than the distribution of medical resources

145

Although both arguments reinforce the case for a more equal

distribution of legal services, neither establishes the priority of
those services over more urgent, and controversial, welfare rights-
1 46
a priority that Wertheimer, but not Luban, appears to endorse.
Norman Daniel's argument that health care is necessary for the fair
equality of opportunity 4 7 may be precisely cast in the same terms:
health care greatly expands our opportunities for economic success
and personal fulfillment, and its uneven distribution puts those who
cannot afford it at a competitive disadvantage to those who can. Ac-
cordingly, because of this disadvantage, an improvement in the
medical resources available to one person may have "a direct nega-
48
tive effect on the interests" of others.'
The government may play a more extensive and visible role in
excluding the poor from the legal rather than the medical system,
but it certainly is complicit in the latter. The state enforces the med-
ical monopoly over various forms of health care, severely limits the
availability of drugs and other treatments, and imposes high stan-
dards for licensing hospitals and other health-care facilities. While
health care may be a less integral and exclusive function of the state
than providing "a locus for the accurate and impartial adjudication
of civil disputes," as Wertheimer argues,1 49 our own state is in-
volved extensively in regulating both services, and its involvement
gives it a responsibility for their distribution.
Even if the case could be made that the state more actively in-

145. Wertheimer, supra note 117, at 314 (emphasis in original).

146. Id. at 313-16.
147. N. DANIELS, JUST HEALTH CARE 45-48 (1985).
148. Wertheimer, supra note 117, at 314. Wertheimer fails to explain how a competi-
tion among several claimants for a single organ is less adversarial than a lawsuit, i.e., how
the donation of the organ to one patient has a less directly negative effect on the other
than the loss of the lawsuit. He might argue that starkly noncorrespondent outcomes
are more common in law than health care, but that is simply a matter of circumstance
and degree. Alternatively, he might contend that the end of medicine is to cure, while
the end of law is to adjudicate justly rival claims. Thus, the medical good of healing
would be accomplished whomever got the organ, while the legal good of justice only
would be accomplished if the party with the stronger entitlement prevailed. This, how-
ever, merely would restate the justice-maximizing argument discussed above without
giving an independent explanation of why the rejected transplant patients are harmed
less directly in losing the organ than the disputant in losing the lawsuit. The harm to the
other needy patients is more direct, because an under-represented litigant may still win
the lawsuit. In either case, the distribution of scarce resources may deny claimants what
they are entitled to on the correct distributional theory.
149. Id.
1990] REVIEW EssAY-Lawyers and Justice 421

jures the poor in denying them legal than medical services, that
would not settle the issue of priority. While it may be worse, ceteris
paribis, for a government or an individual to harm than to fail to
help, the magnitude of the injury also must be considered. The
stakes in health care may be, in general, much higher than in law: it
may be far worse for the government passively to allow illness and
debility to go untreated than actively to prevent the vindication of
most legal rights. If we believe in the rights to adequate health care,
secure housing or meaningful employment, we may well regard
those rights as more urgent than the right to legal services.

B. The Obligation to Provide Legal Services

As long as people have a right to more legal services than they
can afford, those services will have to be provided out of general
state revenues or lawyers' pockets. While Luban strongly supports
government funding for legal services, he also argues that lawyers
are obliged to provide free services. 150 He contrasts lawyers with
grocers, who do not have a special duty to feed the poor:
[T]he lawyer's lucrative monopoly would not exist without
the community and its state; the monopoly and indeed the
product it monopolizes is an artifact of the community ....
The community, as a consequence, has the right to condi-
tion its handiwork on the recipients of the5 monopoly fulfil-
ling the monopoly's legitimate purpose. 1
There are two distinctions in this comparison. The first is simply
that lawyers, unlike grocers, owe their substantial earnings to gov-
ernment intervention (but what about agricultural and commercial
policies that enable grocers to charge higher than the market
price?). This may distinguish lawyers from grocers, but not from
doctors, engineers, teachers, and others whose earnings receive
some boost from state certification or licensing requirements.
The second distinction is more helpful: law, unlike grocery or
medicine, is a creature of the state, a vocation that only can exist in a
civil society and whose activities play an integral role in the opera-
tion of the state. ("Integral role" alone might not distinguish law-
yers from engineers or teachers without some debatable notion of
"essential" government functions.) Whereas the state created the
legal system for a specific end (which is, remember, equality under
law), it can regulate the profession in any way that serves that end.

150. D. LUBAN, supra note 1, at 282-88.

151. Id. at 286.
422 MARYLAND LAW REVIEW [VOL. 49:392

Requiring lawyers to volunteer a few hours a week to the poor or

the public interest is hardly a major imposition; the state would be
entitled to demand much more if it would advance equal justice. I
find this distinction persuasive in explaining why we do not need an
argument from distributive justice, or even from monopoly status,
to impose substantial pro bono requirements on the practice of
52
law. '

CONCLUSION

Luban makes a powerful but incomplete case against the moral

prerogatives claimed by the legal profession. Although he rejects
the standard conception of adversary representation, he exagger-
ates the moral importance of legal practice in several areas: in serv-
ing long-term institutional goals, in upholding the moral authority
of the law, and in maintaining the legitimacy of our government.
Even the most strongly justified forms of legal practice rarely will, if
ever, license conduct that is not justified by ordinary moral consid-
erations. While lawyers may enhance or impair the community's ef-
forts to distribute equitably its benefits and burdens, they do not
thereby uphold or subvert the moral authority of the law. Finally,
although lawyers may have a greater obligation than other profes-
sionals to make their services available to the poor, those services
are no more important to a just social order.
In challenging Luban on these points, I intend to praise him
with faint damnation. Even if his claims for the moral significance of
legal practice are wrong or overstated, he has raised the level of

152. Less persuasive, and also unnecessary, is Luban's fallback argument, that law,
unlike grocery retailing, is not "a victimless pastime." Id. Luban suggests that the ad-
versarial nature of legal practice creates some sort of compensatory obligation on the
part of those who practice it. Id. at 286-87. Initially, this argument from direct harming
might seem to have more force against lawyers than the government, because lawyers
harm the disadvantaged more directly in representing the advantaged than the govern-
ment does in allowing them to do so. The individual lawyer, however, hardly can re-
dress this harm by representing the actual victims-to the extent they were identifiable,
it would present a conflict of interest for her to represent them. Moreover, it seems odd
to demand redress for an injury while not requiring the lawyer to avoid it in the first
place, by assuming a less adversarial role.
If the lawyer owes no specific compensatory duty to those actually injured by her
actions, it is unclear why she should have a more general obligation to make her services
available to those belonging to the same class as her victims, though something like this
appears to have been the moral logic of robber-barons turned philanthropists. While
the harm of adversary representation may require subsidies for the injured class, it does
not require that those subsidies come out of lawyers' pockets. If lawyers should pay, it is
only because the state's role in creating the legal profession permits it to structure that
profession as it sees fit.
1990] REVIEw ESSAY-Lawyers and Justice 423

debate tremendously by making them. Lawyers and Justice forces

legal ethics to concern itself with difficult moral issues about obedi-
ence to the law, role-morality, and distributive justice. In placing
professional ethics within the jurisdiction of moral philosophy, it
upgrades the former and enriches the latter.