A Skeptical Answer to Edmundson's Contextualism: What We Know We Lawyers Know

Florida State University Law Review Florida State University Law Review

Article 3

A Skeptical Answer to Edmundson's Contextualism: What We A Skeptical Answer to Edmundson's Contextualism: What We

Know We Lawyers Know Know We Lawyers Know

Rob Atkinson [email protected]

Part of the Law Commons

A SKEPTICAL ANSWER TO EDMUNDSON'S CONTEXTUALISM: WHAT WE KNOW WE LAWYERS KNOW

Rob Atkinson

FALL 2002

A SKEPTICAL ANSWER TO EDMUNDSON’S CONTEXTUALISM: WHAT WE KNOW WE LAWYERS KNOW

How can lawyers help criminal defendants they know to be guilty avoid conviction and punishment? In Contextualist Answers to Skepticism, and What A Lawyer Cannot Know,2 Professor William Edmundson3 answers that nagging question, a question asked not only at cocktail parties, but also in law faculty lounges and legal ethics classes. Prof. Edmundson’s answer is clever, even brilliant, and appealing, almost seductive. It is also, I am afraid, fundamentally flawed. But it is flawed in deeply interesting and revealing ways. In Part I of this Comment, I try to show why both Prof. Edmundson’s question and his answer are more important than he himself suggests. In Part II, I try, at greater length, to isolate what I believe to be his answer’s fundamental flaw. Finally, in Part III, I suggest why Prof. Edmundson’s answer, flawed though I think it is, may mark an

important realization in legal ethics scholarship, underscoring the need for significant new work in several related directions.

The question Prof. Edmundson addresses in his paper, big though it is in its own right, is actually a particularization of an even bigger question: “Can a good person be a good lawyer?” If Prof. Edmundson’s question is the most basic in criminal defense lawyering,4 this latter is, by general consensus, central to legal ethics as a whole.5 To fully appreciate the importance of Prof. Edmundson’s question, we need to see how closely it has come to be related to the other, broader question. Not surprisingly, all legal ethicists and practicing lawyers answer the “can a good person be a good lawyer” question in the affirmative; no more surprisingly, as a legal ethicist and former lawyer, I am convinced we are right. Where we differ is not on whether a good person can be a good lawyer, but on how. Here we divide logically, but by no means numerically, in half. On the one hand are those who believe that the good person, as a good lawyer, may (more strongly, should) do all that the letter of the law allows for any client. Call them the neutral partisans.6 On the other hand are those, myself and Edmundson both included, who believe that other normative limits, sometimes narrower than the letter of the law, govern what the good person, as good lawyer, may do for at least some clients, at least

some of the time.7 Call us the critics of neutral partisanship or, more positively, the reformers.8 The central question of legal ethics arises as a practical matter because lawyers are sometimes asked or required, in their role as lawyers, to do things that strike all conscientious people, lawyers and non-lawyers alike, as morally suspect. These apparent conflicts between what theorists call role morality and ordinary morality9 include, to put it in classical terms, making the true look false and the false, true. Discrediting truthful witnesses is an example of the former; arguing for factual or legal positions in which you don’t believe nicely illustrates the latter.10 For all their disagreement on these matters, legal ethicists have tended to agree on one context where what we have called neutral partisanship, or something very close to it, should be practiced.11 That context is criminal defense. In what has come to be called the criminal defense paradigm,12 the way to be a good person and a good lawyer is to do all that the law allows to acquit your clients, even clients whom you know to be guilty, dangerous, and unrepentant. Some have argued that this should include assisting the client in presenting perjured testimony.13 Although neither the law nor many legal scholars go quite this far, both the law and virtually all commentators agree that the criminal defense lawyer has, to quote the oft-

quoted language of Justice Byron White, “a different mission.”14 Just how different that mission is, however, is a point of considerable disagreement between neutral partisans and their critics. Curiously, each school tends to minimize the “difference” of the criminal defense paradigm, but in opposite directions. Reformers try to move the criminal defense paradigm toward all other lawyering, which, for them, involves checking lawyerly advocacy with supplemental norms;15 neutral partisans try to move all lawyering in the direction of the criminal defense paradigm, which, for them, involves pressing any client’s advantage to the fullest extent of the letter of the law.16 All lawyers, according to neutral partisans, should be as unalloyedly loyal to all clients, in all causes, as criminal defense lawyers are to those charged with the severest of crimes. Reformers have responded to this position in two related ways. First, and with greatest accord, they have tried to distinguish the need for extraordinary zeal in criminal defense from the kind of client loyalty appropriate in other contexts.17 This has been a careful, even judicious, process. In the initial phase, critics of neutral partisanship showed how routine, non-adversarial, in-office counseling differs from all litigation, not just criminal defense, in ways that make the injection of publicspirited advice desirable, even essential, to the proper lawyerly role.18 Next, they pointed that, in some civil litigation contexts, the criminal defense tables are effectively turned, from individual Davids defending against the Goliath state to corporate Goliaths like Kerr-McGee going after individual Davids like Karen Silkwood.19 Finally, they argued that, even when private clients are civil defendants, the similarities to criminal defense may be more diagramatic than real. This is particularly true, they point out, where large, well-heeled corporate defendants can spend their opponents—employees, consumers,

and even government agencies—into unfavorable settlement or outright defeat, despite the merits of their cases.20

This series of arguments cleared the field for the reformers’ second line of assault on the criminal defense paradigm.21 Having countered every effort of neutral partisans to expand the criminal defense paradigm into other areas of law, critics then turned their analysis on criminal defense itself. Several accepted elements of criminal defense lawyering, they have argued, are not always required, and hence not universally justified, by the social policies that support the criminal defense lawyer’s “different mission.” The policies typically invoked are preventing the conviction of the innocent and protecting the dignity interests—rights, if you prefer—of even the guilty.22 But neither of these policies necessarily entails helping the known guilty avoid conviction whenever they are prosecuted.23 At least in principle, these policies could be served as well by a “friendly inquisitor” as by a “zealous advocate,” to use Edmundson’s useful distinction. In his words,

That justification, as Edmundson points out, has traditionally been instrumentalist. Unless the criminal defense lawyer acts as a zealous advocate, not just a friendly inquisitor, the criminal justice system cannot effectively guarantee the acquittal of the innocent and the dignity-based rights of the guilty. What the reformers argue is possible in principle, the neutral partisans deny is possible in fact. But this, as Edmundson demonstrates, is not where the argument over the central issue in criminal defense work begins. That beginning, back to which he effectively takes us, is the following syllogism:

As Edmundson points out, both neutral partisans and their critics accept the validity of this argument. They agree that assisting the known guilty in avoiding conviction is a prima facie wrong; what they disagree on is whether that prima facie wrong is justifiable. Neutral partisans maintain that it is justifiable as a necessary means to deeply shared values; their critics argue that it isn’t. Thus all legal ethicists, neutral partisans as well as their critics, have traditionally taken criminal defense as a special case of this general principle. There has never been much question that knowingly assisting a criminal defendant in avoiding a just penalty for his or her crime is prima facie a harm to the criminal’s victims, or an injustice to society, or a vice on your own part, or all of the above.25 Defenders of the criminal defense paradigm have, accordingly, almost invariably proceeded to justify this prima facie wrong that lies at the root of the prevailing conception of the criminal defense lawyer’s role. As Edmundson points out, these defenses follow modern moral philosophy more generally along one of three routes. Either the consequences of the prima facie harm promise a preponderance of some social good, or the moral dignity of the criminal defendant trumps any such harm, or any such harms are part of social arrangements that self-interested individuals would assent to ab initio.

As we have seen, these defenses are, as it were, on the defensive, under sustained assault from reformers. What Edmundson offers is a radically new defense, in the nature of a counterattack. Edmundson reminds us that the conclusion of our syllogism need not be taken for granted; it can, in fact, be attacked at either of two obvious points, the major premise or the minor. The major premise, an obvious corollary of the “no harm” principle,26 may seem compelling and commonsensical to most of us, but it is hardly unassailable. Some of Socrates’s more outré interlocutors denied it, or came extremely close, in identifying justice with the will of the strong: Thrasymachus in Book I of the Republic;27 Callicles in the final round of the Gorgias.28 Nietzsche, in his more exultantly blond-beastly, proto-Nazi moods, notoriously denounced it as close to the root of all evil (or, as he put it, bad);29 Ayn Rand is a somewhat less extreme, if much less eloquent, fellow-traveller.30 Theirs, not surprisingly, is not the road Prof. Edmundson recommends. He, instead, attacks the minor premise, the assumption that criminal defense lawyers knowingly assist the guilty in avoiding conviction. Rather than concede that criminal defense lawyers need to justify the prima facie wrong of knowingly assisting the guilty in avoiding punishment, he denies that their assistance is “knowing,” (the minor premise in our syllogism). Here lies the ingeniousness of his approach; it opens a new front in the defense of the criminal defense paradigm. And it opens that front by marshalling a sophisticated philosophical argument that draws less from ethics, the standard resort of reformers, than from epistemology, a discipline they invoke much less often. With respect to criminal defense lawyers, Edmundson’s position is not a mere “forgive them, for they know not what they do.” He argues that, if they don’t know what they’re doing, then they do no moral wrong, prima facie or otherwise, even if what they are doing works very grievous individual or social harm. Thus, unlike earlier defenders of the criminal defense paradigm, he does not merely tell us to lower our accusing fingers, now that a prima facie wrong has been explained away. Beyond that, he chides us, at least implicitly, for ever having raised an accusation in the first place. What is more, Edmundson sweeps within his defense some of the very practices that reformers have argued cannot be justified by the

standard instrumentalist defenses of the criminal defense lawyer’s role. He explicitly defends presenting false defenses,31 “making the false look true,” and his defense could easily be extended to “making the true look false,” not only the general practice of discrediting truthful witnesses, but also the particular practice of “brutal crossexamination” of complaining witnesses in rape cases.32 Perhaps most significantly, Edmundson’s argument could be extrapolated beyond the criminal defense paradigm to all other modes of lawyering—not just civil contexts analogous to criminal trials, like child custody cases, where the state threatens a single individual with calamitous personal loss, but also the entire spectrum of corporate representation, from defending litigation brought by the government with respect to past actions to the most insulated in-office counseling about proposed future courses of conduct. The exchange between Boswell and Johnson on which he relies is certainly susceptible, on its face, to just such an expansive reading. What Johnson defends, it is worth remembering, is not just criminal defense but, more generally, “supporting a cause which you know to be bad.”33 To see how this extrapolation from Edmundson’s argument would go, consider another syllogism, which starts where the first leaves off. For neutral partisans and for Edmundson, though for different reasons, criminal defense lawyers are not morally culpable for the harms or injustice they assist in when they work to acquit the guilty. This is where the second syllogism begins:

As we have seen, neutral partisanship’s critics have long, and I think persuasively, attacked this extrapolation in terms of the traditional, instrumentalist defenses of the criminal defense paradigm. Whatever social goods the criminal defense paradigm may be plausibly said to be necessary to advance, it has been shown that the same zealous representation is not necessary on the part of civil lawyers to advance those goals. These critics have, in effect, shown that the minor premise in our second syllogism is false, or at least subject to deep doubts, insofar as it relies on A, the traditional reasons for the criminal defense paradigm, because those reasons do not apply to other lawyers. Notice, however, that Edmundson’s argument offers an alternative version of the minor premise, in terms of B, his epistemological defense of the criminal defense paradigm. It is not too much to say that, if Edmundson’s argument is sound, the entire decades-long debate between neutral partisans and their critics has been shifted radically in favor of the former. Edmundson himself, I hasten to underscore, does not attempt these extrapolations; in all likelihood, he would lament them.34 His case is, in its own terms, limited to the criminal defense paradigm. But Prof. Edmunson’s defense of neutral partisanship in the context of criminal defense, if successful, would, as our second syllogism shows, apply to other contexts as well. If the history of that paradigm is any indication, those extrapolations will be made. This, then, is the context in which Edmundson makes his contextualist case; these are the stakes in the game he invites us to play. In the next part, I will try to show why Edmundson’s argument is fundamentally unsound, why the reformers’ criticisms of neutral partisanship have lost none of their purchase, in either the criminal defense paradigm or elsewhere.

Edmundson candidly admits, at the outset, the apparent oddity of denying that criminal defense lawyers know that their clients are guilty. To account for that oddity, he reminds us of an important epistemological point about the relationship between ordinary and special claims of knowledge. We can, he points out, comfortably affirm that we know things that we can’t prove absolutely, where “prove absolutely” means ruling out all imaginable possibilities of er-

ror. We (well, most of us) believe that the universe is billions of years old; yet, as Bertrand Russell famously pointed out, “there is no logical impossibility in the view that the world was created five minutes ago, complete with memories and records.”35 Even more mundanely, we know that the sun will come up tomorrow, as we (again, most of us) know it has for billions of years. Yet, as David Hume pointed out, the causal nexus between past and future is surprisingly difficult to establish;36 as your stockbroker will tell you (more readily now than a year ago), past performance is no guarantee of future earnings. As Descartes pointed out, further back and more radically still, we can doubt pretty much everything we experience—except, ironically, our most immediate doubts.37 Major philosophical reputations have been made, in very different times and places,38 by showing us how well we can get along without absolute knowledge. Reasonable certainty serves us well enough, day in and day out, as an entirely adequate substitute for absolute knowledge; if we can’t rule out all imaginable sources of error, we can readily enough rule out all relevant sources. Similarly, ordinary language about knowledge serves us quite adequately for ordinary life—or, more precisely, for most situations in life. What we ordinarily mean when we say we know something is that we have eliminated not all possible doubts, but all contextually relevant doubts. Thus, when someone tells us that we don’t “really” know what we think we know, we know that the asserted “really” is in inverted commas and that the person making the assertion is a philosopher, professional or amateur. If we begin to suspect that the “really” really isn’t in inverted commas, then we also begin to suspect that we are talking, not to a philosopher, but to a nut (assuming we know our interlocutor to be older than nineteen and not one of our own offspring). The insight that reasonable belief is almost always enough carries a corollary point, on which Edmundson relies more directly: In epistemological language, as in the law of torts and elsewhere, reasonableness varies with the circumstances. Sitting in a restaurant and

asking the waiter for something to cut your meat with, you know you’ll get a steak knife; if what you want is a microtome, you know you’ll have to ask for it specifically. If you are in the cytology lab, on the other hand, these expectations about cutlery will be reversed (and, following linguistic conventions as old as the Norman Conquest, the mammalian muscle before you won’t be beef, pork, or mutton, but cattle, swine, or sheep). So it is with language more generally, including the language of knowledge. We know that what we know—more precisely, what we need to know to claim we know— varies with the situation. As the philosophers of linguist analysis are wont to say, in their typically flippant way, we play lots of games39 or do lots of things40 with words, and we typically know the rules and the tools. But some tasks require special tools, and, in life and in language, we have to fashion what we don’t find in the literal or metaphorical toolbox. Microbiologists need microtomes, not steak knives; as Katharine Ross told Paul Newman and Robert Redford in Butch Cassidy and the Sundance Kid, “your line of work requires a specialized vocabulary.”41 In that respect, at least, our epistemological enquiry is like their bank robbing. To show how well we get along without absolute knowledge, we need to be able to distinguish our everyday sense of “knowing” from more restrictive and demanding senses. This is not to say that our ordinary use of “know” is sloppy, much less wrong, any more than we are wrong or sloppy to say, in a postCopernican world, that the sun rises in the east and sets in the west, or to cut our meat at the table with a steak knife rather than a scalpel, or a microtome. What’s right—more precisely, what’s appropriate or reasonable—depends on the task at hand, be it verbal or culinary. The meanings of words, then, like the functions of tools, are taskspecific.42 If we come up with a new task, we may need to find or create new tools. Where the task is a language-task, the new tools will be new terms, or new meanings for old terms. From Edmundson’s perspective, defending bank robbers, like bank robbery itself, can be seen as just such a special task, requiring just such a specialized vocabulary. For that task, we may need, as we do in epistemology, a special sense of the word “know.” In particular, he maintains that we

need a sense of “know” for criminal defense lawyers’ knowledge of their clients’ crimes that is almost as restrictive as the deeply skeptical, gold-standard, “absolute knowledge” that epistemologists sometimes invoke. Edmundson calls this “film noir” skepticism. Before the criminal defense lawyer can be said to know a client is guilty under that standard, “the lawyer must rule out the possibility that by some unfathomable whim of fate, a kaleidoscope of otherwise disjointed appearances has frozen itself into a damning configuration.”43 This standard, as Edmundson indicates, is analogous to, but even stricter than, our criminal justice system’s high standard of proof for conviction, proof “beyond a reasonable doubt.” Under that latter, proof beyond a reasonable doubt standard, we could say that, for purposes of deciding whether a criminal defendant is guilty, jurors do not “know” the defendant did what he or she had done even though, in other contexts, they might “know” it well enough. Thus the jurors in the O.J. Simpson trial could be said not to “know,” for purposes of convicting him of murder, that O.J. killed his ex-wife. But they probably “know” it well enough not to recommend him for a blind date. Notice two things about this example. First, we as a society have discussed the Simpson trial, and many another criminal trial, long and well without invoking any special sense of the word “know.” Why didn’t the jury convict someone who seemed so obviously guilty? The widely believed, but less than charitable answer, of course, was that Johnny Cochran played the “race card.” The less cynical, if less plausible, answer was that, because of reasonable doubts about the evidence traceable to a suspiciously racist police department, the jury found that the government had not proved its case “beyond a reasonable doubt.” But what, many lay-folk wondered aloud, about the second trial, in which a different jury found Mr. Simpson liable for the wrongful death of his ex-wife? We lawyers explained (the more telegenic among us on TV): In the later civil trial, with its lower, preponderance of the evidence standard, the jury was able to find Mr. Simpson civilly liable for the very same alleged wrong. To make it an epistemological sound-bite, with “scare quotes” filling in for voiceinflections:

We, lawyers and layfolk alike, can explain the different outcomes in the civil and criminal Simpson cases, compare them, and criticize them, all without invoking any special sense of the term “know.” The second thing to note about the Simpson case is that we all understand (or, if you will, know) why the jurors were required to use a higher standard of proof, to be more skeptical, in the criminal case. They had a very serious task before them: determining a fellow citizen’s guilt or innocence of a heinous act, to which we attach the most severe of possible penalties, loss of liberty and even life. We as a society insist that, before we make that determination and attach those penalties, we are very, very sure we are correct. To be as sure as we reasonably can, we require our jurors to resolve all plausible doubts in favor of the accused. Another part of that assurance is to require that every criminal defendant be represented by a lawyer, whose job it is to make sure that the government meets that high standard of proof. How the lawyer goes about that, of course, is a matter of much dispute, to which we shall return shortly. But notice, for now, that, whatever the particular requirements of that role, it has two general features in common with the two we have just seen with respect to jurors. First, we all understand (again, if you will, know) that it is defined as a means to, and justified in terms of, our deep-seated insistence on protecting criminal defendants. Whatever special things the criminal defense lawyer has to do, we know the reason for their having to be done. Second, we can discuss and explain those things, and the reasons for them, without invoking special standards of knowledge. In particular, we don’t typically say, with Edmundson, that the criminal defense lawyers treat clients differently from the way we treat them, or from the way lawyers treat other clients, because they know less about their clients than we do. Indeed, we typically suspect that, unless criminal defense lawyers take special steps to preserve their ignorance,44 they know a good deal more about their clients than we do. That suspicion, of course, is at the root of our cocktail party question about criminal defense lawyers: How can they try to acquit those they know to be guilty? And that, as we saw at the outset, is the very question with which Edmundson wrestles. He would have criminal defense lawyers give

an answer that is, to them and to us, very odd: “Because we criminal defense lawyers don’t know our clients are guilty.” In response, it makes perfect sense for us to ask, “What do you mean, you don’t know?” We know either that they are being defensive to the point of disingenuousness or that they are invoking a very special sense of the word “know.” Edmundson would have us believe it is the latter; more precisely, he wants us to substitute the latter for the former. To appreciate fully the specialness of this criminal defense notion of knowledge, consider, again, the two Simpson cases (the first criminal, the second civil). In the first, under our standard conception of the criminal defense function, Simpson’s lawyers were permitted, even expected, to put the state on its proof as to each aspect of its case, even if those lawyers knew him to be guilty as charged. In the parallel civil damages case, by contrast, matters stand quite differently. Under the ABA Model Rules of Professional Conduct45 and the Federal Rules of Civil Procedure,46 lawyers representing civil defendants are forbidden to contest what they know to be true. If they know there is no defense of their clients’ conduct other than denying facts they know to be true, they must concede liability. They may assert affirmative defenses (the statute of limitations, for example), and they may contest the amount of damages. But, in contrast to the criminal case, they may not deny what they know their client did, in an ordinary language sense of the word “know.” Nor are criminal defense lawyers themselves wholly absolved of ordinary standards of knowledge. For precisely this reason, as Edmundson himself concedes, the law of perjury is a particular puzzle for his theory.47 As he notes, the ABA Model Rules forbid a lawyer, even a criminal defense lawyer to “‘knowingly . . . fail to disclose a material fact to a tribunal when disclosure is necessary to avoid assisting a criminal or fraudulent act by the client . . . [or] offer evidence that the lawyer knows to be false.’”48 If a lawyer knows that a client or witness has offered perjured testimony, the lawyer must inform the tribunal; if the lawyer knows the client is about to testify falsely, the lawyer must forbid it. How can these requirements apply

to criminal defense lawyers, if, on Edmundson’s hypothesis, they cannot know when their clients are guilty of perjury? Here is Edmundson’s answer: “Knowledge of the underlying offense simply does not enter into the lawyer’s duty to report client or witness perjury. What the lawyer ‘knows’ is the existence of an irreconcilably inconsistent prior account; it is unnecessary to suppose in addition that the lawyer knows which (if either) is the true account.”49 But that knowledge is precisely what the law of perjury does suppose, as the pre-condition of the lawyer’s duty to report. Otherwise, Edmundson’s argument proves far too much, and the lawyer would have to report far more than current law requires. Consider the facts in Nix v. Whiteside,50 the case in which the U.S. Supreme Court considered the constitutionality of a state’s requirement that criminal defense lawyers report client perjury. In that case, the criminal defense lawyer suspected perjury when, a week before trial, his client added a new and exculpatory detail to his selfdefense story: On the fateful night when he admittedly committed homicide, he had seen something “metallic” in his victim’s hand. Here we have Edmundson’s two inconsistent stories, the earlier one with “something metallic” in the victim’s hand, the later one without. What’s the lawyer to do? Virtually all authorities, even the severest critics of a requirement to report perjury,51 agree that the first step is to confront the client with the inconsistency. If the two stories really can’t be reconciled, then the lawyer must do what Edmundson says the lawyer needn’t do: decide which story is really true. This is precisely what the lawyer in Nix v. Whiteside did. He asked his client why, at so late a date and on so important a point, he suddenly referred to having seen “something metallic.” Here is his client’s fatefully candid reply: “[I]n Howard Cook’s case there was a gun. If I don’t say I saw a gun, I’m dead.”52 Having heard this attempted reconciliation, the lawyer found the “something metallic” story incredible. He concluded that his client had added the detail, not because he remembered it belatedly, but because he thought he needed to improve his story with false, self-serving elaboration. The lawyer further concluded that, because he knew his client was about to offer perjured testimony, he could not let him take the stand to present that testimony—a conclusion

which the United States Supreme Court held not to violate the client’s constitutional right to effective assistance of counsel.53 Had the client convinced his lawyer that he really had remembered, rather than invented, the “glint of metal”—a possibility that Justice Stevens emphasized in a concurring opinion54—the lawyer would have had no duty to report the inconsistency in the stories. Indeed, under current law, the lawyer would be not only forbidden to reveal the inconsistency, but also required to let the client take the stand to present the story, including the supplemental detail about seeing the glint of metal.55 To know which set of obligations bound him—to refuse to allow his client to put in the false testimony, or to allow him to put in the true—the lawyer must, contrary to Edmundson’s claim, know which story is true and which, false. It cannot be enough simply to know that the two are irreconcilable. Otherwise, criminal defense lawyers would have to report not just the changed stories they find incredible, but all inconsistent stories, which is clearly more than current law requires or even permits. There is, I should point out, a way out of this perjury puzzle that is entirely consistent with Edmundson’s position. We can simply say that, for purposes of determining whether a criminal defendant is committing perjury, as for determining whether a civil defendant has a contestable case, ordinary standards of knowledge apply. What this underscores is precisely my point: Edmundson’s claim that criminal defense lawyers cannot know of their clients’ guilt is a very special, and peculiar, sense of the word “know.” This is Edmundson’s critical move; we must look at this peculiar sense of “know” very closely, even (if you will) skeptically. The standard answer to “how can you assist those you know to be guilty in avoiding conviction” is closely parallel to the answer we saw with respect to jurors, in both form and content. In form, the answer invokes the systemic need to protect criminal defendants; criminal defense lawyers work to acquit the guilty because that is what it takes to protect criminal defendants. In substance, the lawyer’s standard answer is closely related to the jury’s “guilt beyond a reasonable doubt” standard: In order to insure that we convict only under that stan-

dard, lawyers have to do what, in some cases, involves helping those they know to be guilty to avoid conviction. One way of describing what lawyers do for criminal defendants would be to say that they treat them as if they do not know that they are in fact guilty, even when they do know. That is close to the answer Edmundson would have criminal defense lawyers give: “We don’t know our clients are guilty.” Seen against that background, and understood that way, Edmundson’s answer looks a lot less peculiar. In response to it, we might say, “Oh, we see, you’re not saying that you don’t know that they are guilty, in the ordinary sense of ‘know’; what you’re saying is that you’re just acting as if you didn’t know it, the way jurors do, though even more aggressively than they.” But that raises an immediate and obvious question: Since you can explain what criminal defense lawyers do for their clients without invoking a special sense of “know,” why invoke it? Here, I believe, is where Edmundson’s argument gets into trouble. He gives two reasons, neither of which works as well as he wants. The first, and most basic, reason is to avoid having to justify the criminal defense lawyer’s role in the usual ways (consequentialist, dignity-based, or contractarian). And the second is like unto it: to give criminal defense lawyers “breathing space” in which to formulate those very justifications. Neither works, I’m afraid, for pretty much the same reason, as we shall see. With respect to the first, Edmundson’s own odd way of defending criminal defense in terms of a special sense of knowledge ultimately brings us back to the very standard defenses, in terms of consequences, rights, or hypothetical contracts, that he hopes to avoid. As we have seen, special senses of knowledge-terms, like special senses of other terms and special uses for other tools, are quite common, and the special contexts in which they occur are readily identifiable. But sometimes those contexts need more than identification; sometimes they need justification as well. As Edmundson himself admits, “[a] justification has to be given for the social practices that support the creation of special epistemic contexts, especially ones that attach to socially created and sanctioned roles, such as lawyer and juror.”56 In other words, if he is to have criminal defense lawyers claim not to know that their clients are guilty, he must justify that strikingly narrow standard of knowledge. Edmundson, again, admits as much: “[A] justification is demanded of the social institution—call it the adversary system—which assigns an especially rigorous epistemic standard to criminal-defense attorneys.”57

The demand for that justification, curiously, carries us back to ground familiar to all scholars of legal ethics generally and criminal defense particularly. As Edmundson puts it:

And our own adversarial system, as he points out, has a distinctive element to justify in one of those three ways. We must explain why lawyers operate under not just a high, beyond a reasonable doubt standard, but, beyond that, under a “super-exacting,” film noir standard. That, as we have seen, is simply Edmundson’s way of rephrasing an old and familiar aspect of our system: the criminal defense lawyer’s knowingly assisting the guilty in avoiding conviction. But, if we have to defend that aspect of the criminal defense lawyer’s role in the same terms in which it has always been defended, where has Edmundson’s argument gotten us? His answer at this point is doubly surprising, and, I have to say, doubly disappointing—surprising and disappointing both in what he admits he hasn’t given us, and in what he claims he has given us. He does not claim to have added anything to the standard defenses of the more troubling aspects of the criminal defense lawyer’s role; indeed, after reviewing them, he finds what he takes to be the best of them wanting.59 What, then, does he claim to have given us? In a word, help, in the form of the best available experts to assist us in examining the normative (as opposed to linguistic) justifications of the criminal defense lawyer’s role. These experts, he tells us, are none other than criminal defense lawyers themselves. “It is they who are the learned professionals in most frequent sympathetic contact with people accused of crime.”60 “In the absence of systematic empirical study, . . . their anecdotal wisdom may be the best we can get” as the basis for evaluating our current version of the criminal defense lawyer’s role.61 Presumably we are to settle for this admittedly second best source because “systematic empirical study” is not to be forthcoming. But there is still a puzzle here. What exactly has Edmundson done to

enlist these supposed experts in aid of our inquiry? One would think that they would be eagerly engaged in that inquiry already; it is, after all, their role that is being questioned. That, Edmundson says, is precisely the problem: “They are . . . not inclined toward dispassionate evaluation, largely, I would hazard to say, because they have been too willing to accept the popular view of themselves as persons regularly engaged in knowingly helping the guilty to avoid just punishment.”62 Locked in a siege mentality by questions about the moral justifiability of their role, particularly the aspect that involves knowingly assisting the guilty in avoiding conviction, criminal defense lawyers have hunkered down behind what Edmundson sees as a patently unsound justification of their position, what he calls “an arid anti-state ideology.”63 Edmundson hopes that they will divorce that “bad ideology” now that he has absolved them of the charge of prima facie wrong-doing that drove them into its arms in the first place.64 But how has he removed that charge, if what criminal defense lawyers need to answer that charge, a justification of the criminal defense system, is precisely what he is liberating them to help us find? The key that is to release them from their bad ideology seems to be exactly the key to the “good ideology” that everyone, we as well as they, are looking for. To put it most pointedly, hasn’t Edmundson assumed his conclusion, leading us in a circular argument, and a very wide one at that? He recognizes this implicit charge, and offers an ingenious, but ultimately unsatisfying answer—an answer, I’m afraid, that turns out to be another iteration of the same circle, a wheel within a wheel. At two critical points in his paper, Edmundson distinguishes the need to justify the criminal justice system, on the one hand, from the need to explain apparently prima facie wrong-doing on the part of criminal defense lawyers within that system, on the other.65 This is not, to be sure, a distinction without a difference. Unfortunately for this critical aspect of Edmundson’s argument, however, it is a difference that compounds, rather than corrects, the circularity problem. Let us concede, quite willingly, the distinction Edmundson draws between justifying individual, role-required acts, on the one hand, and, on the other hand, the systems that require those roles and acts. As David Luban and others have nicely demonstrated, the need for

defenses of role-specific moralities typically begin with the perception that someone is doing, as part of a socially defined role, what would appear to be improper outside that role.66 Thus, to take the example at hand, criminal defense lawyers are making heroic efforts to acquit those whom they know to be guilty; anyone else, working toward that end, would be morally culpable of aiding injustice and perhaps legally culpable of obstruction of justice. Having identified this prima facie conflict between role morality and ordinary morality, analysis then turns to reducing the two to a common denominator. The prima facie wrongful act (working to acquit the known guilty) is shown to be required by, and essential to, a social role, such as that of criminal defense lawyer. That role is then shown to be required by a particular social institution; in our example, the adversarial system of justice. And that system, finally, is shown to be critical to advancing certain values that are, themselves, shared by ordinary morality. Thus, as we have seen, the criminal justice system is said to protect either the innocent from wrongful conviction, or the guilty themselves from dehumanizing indignities. What seems to violate ordinary morality in the short run and at the level of individual defense lawyers, their helping acquit the known guilty, is thus justified, if it is justifiable at all, in terms of its systematically advancing the shared values of ordinary morality in the longer run. Broken down this way, the justification of particular prima facie bad acts is clearly distinguishable from the justification of the system in which those bad acts occur. This is just what Edmundson says. But notice how, in the very mode of analysis Edmundson seems to accept, the justification of the suspicious act stands or falls with justification of the system. If the prima facie bad act is justified, it is only justified because it is required by a system that is itself justified. It is this justificatory linkage that Edmundson attempts, I think without success, to break. He would have the criminal defense lawyers deny moral responsibility for assisting those whom they know to be guilty to avoid punishment, pending proof that the system in which they render this assistance is itself morally justified (and, beyond that, also requires this particular sort of act). But, unfortunately for Edmundson’s analysis, the two defenses stand or fall together. Think of it this way: You know that the foundation of the building you’re now reading in is distinct from the room in which you’re sitting; the foundation’s down there, and you’re up

here. But you also know that, if the foundation is not structurally sound, the position you’re in is, to precisely that extent, precarious. So it is with the morally dubious acts criminal defense lawyers do without grounding them in a well-justified adversarial system. Without the justification of the system, the act itself remains unjustified, and prima facie moral wrongs remain, prima facie, morally wrong. At this point Edmundson wants to interpose his special knowledge definition; he wants to tell the criminal defense lawyer that he or she doesn’t “really” know that he or she is helping the guilty escape conviction. But, as we have seen, this linguistic more itself requires justification by reference to the adversarial system in a way that, we can now see, is precisely parallel to the standard analysis of role morality. Edmundson’s highly restrictive use of the word “know,” like the defense lawyer’s conduct to which it applies, requires justification. And the justification for creating that particular social context is, precisely, its function in either protecting the dignity of the guilty or preventing the conviction of the innocent. But that is none other than the standard defense of the adversarial system.

It seems we are back, quite simply, where we began. We know that assisting the guilty in avoiding conviction is prima facie wrong, and we don’t know anything new about the adversarial system that would justify that apparent wrong. I will suggest in Part III that we are not quite back where we were, if only because the difficulty of Edmundson’s heroically creative effort shows us the more general difficulty of the task at hand. Before turning to that point, though, I want to take a look at what we might call the psychological, as opposed to the logical, side of Edmundson’s argument. As we have seen, Edmundson tells us, quite candidly and I think quite appropriately and helpfully,67 why he offers us his epistemological defense of criminal defense lawyers: to free them to assist us in undertaking a policy-based re-evaluation of the criminal defense paradigm. Laudable though this purpose is, I fear that in undertaking it Edmundson is overly optimistic in two related ways. These are two sides of the same psychological coin. He deals explicitly, but I think too sanguinely, with the psychological state of conscientious criminal defense lawyers; he omits entirely their opposite number, criminal defense lawyers who are hopelessly unscrupulous. To take the conscientious first, Edmundson believes that he can relieve their anxiety about prima facie wrong-doing by showing them that, seen in the proper light or described in the proper way, what they are doing is not “really” wrong. He shows them and us how

criminal defense lawyers could, in their off-duty, ordinary language hours, know that they are working to acquit the guilty even though, in their day-jobs, they don’t “really” know it, in the sense that they are role-required to work as if they didn’t know it.68 As a matter of logic and linguistics, Edmundson may well be right. But I suspect that, as a matter of psychology, he is missing something quite important, a distinction quite well supported by both theory and at least anecdotal evidence. What he is missing is the distinction between knowing that a prima facie wrong is morally justified and being able to commit that sort of wrong over and over again without deep psychological damage. As Bernard Williams has nicely demonstrated,69 one can be intellectually convinced that role-required acts are justified and yet still retain a strong moral disposition against doing them. Soldiers can find killing, even in what they believe to be just wars, deeply distasteful, even unbearable; lawyers may find making the false look true, and the true false, psychologically painful, 70 even if morally justified.71

It will be helpful here to refer you to two real-life examples. In the first, Randy Bellows, formerly a staff attorney at the Washington, D.C. Public Defender’s Office, describes how, over the course of several years, he found his defense of people he knew to be guilty of heinous crimes personally unbearable. In the second, Fred Armani, a middle-aged, small-firm general practitioner in Syracuse, New York, recounts in a series of videotaped interviews with journalist Fred Graham his anguish in representing serial murderer Robert Garrow.72 For both cases, my summary is woefully inadequate. To fully appreciate the lawyers’ pain, you really need to read Bellow’s graphic depiction of his client’s crimes and the heart-rending decline of his personal life, particularly his marriage;73 you need literally to see the anguish in Armani’s face as he recounts to Graham how he refused to tell the parents of his client’s victims where their bodies lay—one in

an abandoned mineshaft, the other in a shallow grave outside a cemetery.74 For our present purposes, however, the point is clear enough: Both Bellows and Armani explicitly believed that what they had been doing—in both cases, trying to acquit those they knew to be guilty— was completely justified, even required, by the Constitution itself. Even so, it was almost more than they could bear. It is hard to see how the comfort Edmundson offers lawyers like these, the prospect of someday finding a full justification for what they are doing now, could work any better than the full justification they firmly believed they already have. There is another side to this psychological coin. If the examples of Bellows and Armani are any indication, at least some criminal defense lawyers will find no solace even in a more thorough defense of their prima facie wrong-doing than Edmundson purports to offer. But, if Williams’s dispositional theory is any indication, others may need no such solace. They will have inured themselves to the very real harm they know they do; they will have adapted in either of two ways. On the one hand, they will have specifically adapted; they will have come to feel no compunction about the harms they do in their professional capacities. They will feel no more compunction about putting an ax murderer back on the street than a surgeon feels about making an incision in a chest. On the other hand—and, for Williams, more worrisomely—they may change their general dispositions, what he calls general adaptation. They won’t feel any compunction about making the false look true and the true, false in their lawyerly lives, because they don’t feel any compunction about such things anywhere else in life. This raises a possibility—for my money, a very safe bet—that Edmundson, perhaps in an excess of charity, overlooks: Some criminal defense lawyers may not be very admirable people. They may not need the moral comfort that Edmundson proffers, much less engage with us in the moral inquiry he anticipates, because they may have no compunction about what they do.75 And they may have no compunction about what they do, not because they believe it to be morally justified, but because they are wholly unconcerned with moral justification. I hope you won’t think it overly cynical of me to suspect that, just as some soldiers are sadists, so some lawyers are mercenaries. They may mouth the platitude that every criminal defendant, however empty his purse or heinous his crime, deserves a lawyer.

But don’t expect to learn that, though many of their clients have committed heinous crimes, few can’t afford heavy fees. What distinguishes these lawyers from their criminal clients is not the social benefit of their practices, much less the loftiness of their motives, but the barest legality of their deprepations.

To say, as I have said at some length, that Edmundson’s theory doesn’t work as intended is not to say that his effort has been fruitless. At very least, his theory nicely illustrates the wisdom of prefering a fertile error over a sterile truth.76 To shift to a more mundane metaphor, Edmundson’s effort to sweep around the flanks of the present debate between neutral partisans and their critics may confirm that the contest between them is joined at more or less the right point. If that is so, then the burden of moral proof remains where the current generation of legal ethics scholarship has placed it: squarely on the defenders of neutral partisanship, to show why the prima facie wrongs lawyers do are not ultimately wrong, as much in zealously assisting in the acquittal of guilty criminal defendants as in socially harmful civil representations. But, if I am right, Edmundson’s theory fails, not only at these edges, but also in the center. His argument works no better for the core of the criminal defense paradigm, merely getting the guilty off by putting the state to its proof, than it does for the periphery, getting the guilty off by presenting false defenses and discrediting truthful witnesses. That, too, is significant. On the need for the former kind of criminal defense, everyone, reformers and neutral partisans alike, is in agreement. But, as Edmundson rightly points out, the best contemporary defenses, in terms of preventing the conviction of the innocent and protecting the rights of the guilty, leave us deeply dissatisfied. As he says, they “excuse[] the lawyer’s conduct, but need not relieve her guilty conscience.”77 When a necessary social role proves deeply disquieting, even damaging, to conscientious people who fill that role, something is radically wrong. Edmundson’s piece poignantly reminds us that this is true of criminal defense work, and thus pointedly raises a basic question: What are we to do? Problems with Edmundson’s argument, despite its cleverness, lead me to suspect that one avenue is closed: If there were a way to solve the problem in theory, Edmundson or his

predecessors would have found it. I am convinced that, in effect, they have proven a negative: There is no way to have our cake and eat it, too, no way both to warrant the current model of criminal defense work and to guarantee that those who do that kind of work will, at the end of their long day, sleep perfectly well through the night. Seen from this perspective, Edmundson’s epistemological argument is not so much a circle, assuming the justification he is seeking, as a tightening of the screws, forcing us to face deep problems with the position we all share. Edmundson reminds us that all civilized societies face the same basic dilemma. On the one hand, some citizens do extraordinarily anti-social things, including killing their fellow citizens, and some cover their tracks and their transgressions in extremely artful ways. On the other hand, wholly innocent people will occasionally find themselves in deeply incriminating circumstances, often through no fault of their own, and sometimes through the machinations of their most anti-social compatriots.78 Society must protect itself from the former,79 always at risk of confusing them with the latter. And errors in either direction are costly: the danger of erroneous exculpation on the one hand, the horror of erroneous conviction on the other. We know, beyond a peradventure of a doubt, that innocent people have been convicted, even executed; we know, equally well, that killers (not to mention rapists and child molesters) notoriously kill (and rape and molest) again. Dilemma dissolves into paradox: We desperately need to know precisely what, quite often, we cannot know; we must act decisively, but we dare not. Our resolution is a motto—better a hundred guilty go free than one innocent suffer80—a motto both Edmundson and I wholeheartedly endorse. But that motto all too often degenerates into a state of mind akin to both the just world self-delusion and the ostrich syndrome.81 It is unthinkable that the innocent should be convicted, so we don’t think about it; innocents shall not suffer, therefore we cannot see that they do suffer. We worry over the excesses of zeal on the part of criminal defense lawyers, while barely 10% of our crimi-

nal cases actually go to trial.82 Plea bargains, the routine resolution of the vast bulk of our criminal cases, are routinely forbidden in virtually all of our sibling democracies,83 which look with even deeper dismay at our death penalty.84 And our prison population, particularly our death row population, is predominantly poor and overwhelmingly Black.85 All societies face an ultimately irreducible dilemma: separating the wolves from the sheep, in a world where wolves not only kill sheep, but also wear their clothing. It is hard to imagine how that problem can ever be made fully to disappear. But our society has long confronted a very different set of problems, which compound and complicate that basic dilemma. We have good reason to suspect that the hundred guilty whom we free are disproportionately wealthy and white, while the innocent ones whom we send to prison and to death are predominantly poor and Black.86

We, with Edmundson, understandably want to give aid and comfort to the lawyers for the latter.87 We would do well to remember, however, that lawyers for the former may deserve something very different. What’s fair for the goose may be fair for the gander, but it doesn’t follow that justice requires identical treatment of shepherds and serpents, watchdogs and wolverines. We know that zeal for the good and zeal for the bad are both zeal; we also have reason to believe that one is good and the other, bad. We all know that criminal defense lawyers sometimes work very grievous social harms. In terms as old as Plato’s dialogues, they sometimes make the false look true and the true, false; in phrases as fresh as today’s headlines, they help put dangerous sociopaths back on the street. Yet some criminal defense lawyers, I’m happy to say, rank with the holiest of secular saints. They live and work among society’s morally worst, and economically worst off, assisting them out of the sincerest conviction in ways that place the maximum imaginable stress on their own social, psychological, and moral well-being, all for a fraction of the money they might make in other employment. Their work cannot await either perfect justifications for what they have to do, or progressive reforms of the system in which they have to do it; their hesitation would literally mean others’ deaths, even innocent others’ deaths, at hands that are ultimately our own. Other criminal defense lawyers, I’m only somewhat less comfortable in saying, are among the crassest of mercenaries, moral if not legal criminals themselves. They represent the dangerous and depraved—drug kingpins and corporate defrauders, millionaire murderers and well-connected con-artists, socially secure date rapists and drunk drivers—all with apparent indifference to both public injustice and private suffering. Some of them, no doubt, are conveniently self-deceived; they really believe, in some distorted way, in the demonstrably dubious defenses of what they do. But most of them, I suspect, simply don’t care. What we mustn’t forget is what they all have in common: they make a lot of money helping the evil and harming the innocent. Even if love of money were not the root of all evil, large incomes would seldom be the stamp of disinterest as to their source. Even if it’s not the solution to all mysteries, moral and

political, Deep Throat’s admonition is helpful, here as elsewhere: Follow the money.

In philosophy, both moral and linguistic, we have all the refinements we need to know the difference between the saints and the sinners, the shepherds and the wolves. But in law and in legal ethics, our world is less tidy, and our task more challenging. We need a great deal more work to square the demands of our criminal justice system with those of our ordinary moral commitments. And we need a great deal more work to make criminal defense itself less demanding of moral heroism on the one hand and less rewarding to moral mercenarism on the other. Most of all, we need a great deal more work toward a world where injustice does not breed poverty, and poverty, crime. To believe that Prof. Edmundson’s ambitious and excellent article does not finish these tasks is certainly not to doubt that he has worked at them wisely and well. Thanks again, Bill.88

--- FOOTNOTES ---

ROB ATKINSON*

I. INTRODUCTION: PROF. EDMUNDSON’S CONTEXTUALISM IN CONTEXT, OR THE STRANGE CAREER OF THE CRIMINAL DEFENSE PARADIGM................................. II. ANALYSIS: WHY PROF. EDMUNDSON’S EPISTEMOLOGICAL APPROACH CANNOT ANSWER THE BASIC QUESTION ABOUT THE CRIMINAL DEFENSE PARADIGM..... III. CONCLUSION: TOWARD A SOCIAL DEMOCRATIC PERSPECTIVE ON THE CRIMINAL DEFENSE PROBLEM............................................................................

BOSWELL. “You say, Dr. Johnson, that Garrick exhibits himself for a shilling. In this respect he is only on a footing with a lawyer who exhibits himself for his fee, and even will maintain any nonsense or absurdity, if the case requires it. Garrick refuses a play or a part which he does not like; a lawyer never refuses.” JOHNSON. “Why, Sir, what does this prove? [O]nly that a lawyer is worse.”1

* Ruden, McClosky, Smith, Schuster & Russell Professor of Law, Florida State University. I am grateful to William Edmundson, Stephanie Gamble, Mark Seidenfeld, and Donald Weidner for their comments and encouragement and to Chenell Garrido and Román Ortega-Cowan for their diligent and capable research assistance. Special thanks to Mary McCormick, Senior Research Librarian, Florida State University College of Law— our ultimate source for all obscure citations. I thank the Florida State University College of Law for providing a research grant and leave for the summer of 2002. 1. JAMES BOSWELL, THE LIFE OF SAMUEL JOHNSON 213 (1952). 2. 30 FLA. ST. U. L. REV. 1 (2002). 3. Following prevailing convention, I will refer to my interlocutor as “Prof. Edmundson” or, for brevity’s sake, simply “Edmundson.” Those usages belie, however, the fact that he and I are on a friendly, familiar, and first-name basis; he was my most gracious host on a visit several years ago to his home institution, Georgia State University, and I his on his visit to Tallahassee last fall for the presentation of his Contextualist Answers paper to the law faculty at Florida State University. Thank you, Bill, for making that presentation and for publishing your piece in the pages of our Law Review.

I. INTRODUCTION: PROF. EDMUNDSON’S CONTEXTUALISM IN CONTEXT, OR THE STRANGE CAREER OF THE CRIMINAL DEFENSE PARADIGM

4. See JAMES S. KUNEN, HOW CAN YOU DEFEND THOSE PEOPLE? THE MAKING OF A CRIMINAL LAWYER (1983); Barbara A. Babcock, Defending the Guilty, 32 CLEV. ST. L. REV. 175 (1983-84).

5. See Charles Fried, The Lawyer as Friend: The Moral Foundations of the LawyerClient Relation, 85 YALE L.J. 1060, 1060 (1976); Stephen L. Pepper, The Lawyer’s Amoral Ethical Role: A Defense, A Problem, and Some Possibilities, 1986 AM. B. FOUND. RES. J. 613, 614; Gerald J. Postema, Self-Image, Integrity, and Professional Responsibility, in THE GOOD LAWYER: LAWYERS’ ROLES AND LAWYERS’ ETHICS 286, 287 (David Luban ed., 1984) [hereinafter THE GOOD LAWYER]; see also DAVID LUBAN, LAWYERS AND JUSTICE 109 (1988) [hereinafter LAWYERS & JUSTICE]; Stephen Gillers, Can a Good Lawyer Be a Bad Person?, 84 MICH. L. REV. 1011 (1986) (reviewing THE GOOD LAWYER, supra, and STEPHAN LANDSMAN, THE ADVERSARY SYSTEM: A DESCRIPTION AND DEFENSE (1984 ; cf. THOMAS L. SHAFFER, ON BEING A CHRISTIAN AND A LAWYER 32 (1981) (“Is it possible to be a Christian and a lawyer?”); Harry T. Edwards, A Lawyer’s Duty to Serve the Public Good, 65 N.Y.U. L. REV. 1148 (1990). But cf. Serena Stier, Legal Ethics: The Integrity Thesis, 52 OHIO ST. L.J. 551, 560 (1991) (“To ask whether one can be both a good person and a good lawyer is to ask the wrong question.”).

6. Deborah L. Rhode, Ethical Perspectives on Legal Practice, 37 STAN. L. REV. 589, 605 (1985); William H. Simon, The Ideology of Advocacy: Procedural Justice and Professional Ethics, 1978 WIS. L. REV. 29, 36-38.

7. This second group is actually and theoretically divisible into two subgroups, those who believe that the extra-legal, supplemental norms are public, accessible to and binding upon all lawyers, and those who believe otherwise. The latter tend to invoke norms from particular religious traditions, though non-public norms can be, and occasionally are, imported from secular systems as well. For present purposes, however, we can safely ignore these subclasses; for more (much, much more) on why they matter, see Rob Atkinson, A Dissenter’s Commentary on the Professionalism Crusade, 74 TEX. L. REV. 259, 269 (1995) and Rob Atkinson, Beyond the New Role Morality, 51 MD. L. REV. 853, 883 (1992).

8. There is no generally accepted, universally used, term for neutral partisanship’s opponents. I choose “reformers” because it implies, almost literally, that neutral partisanship is a departure from an earlier, more desirable position, contrary to the suggestion that neutral partisanship has any logical or historical, much less normative, priority. Furthermore, for those versed in theology, the term reformer has additional resonances, for me quite strongly positive.

9. For perhaps the clearest elaboration of this distinction in the context of legal eth-

ics, see LAWYERS & JUSTICE, supra note 5, at 104-27.

10. See WILLIAM H. SIMON, THE PRACTICE OF JUSTICE: A THEORY OF LAWYERS’ ETHICS

171 (1998) (giving these examples in the context of criminal defense).

11. See id. at 170, 237 n.1 (noting that “[s]ome people who join me in rejecting the Dominant View [neutral partisanship] make an exception for criminal defense,” and citing the work of fellow reformers David Luban, Richard Wasserstrom, and Deborah Rhode).

12. LAWYERS & JUSTICE, supra note 5, at 65. 13. See MONROE H. FREEDMAN, UNDERSTANDING LAWYERS’ ETHICS 119-20 (1990) (arguing that criminal defense counsel should be allowed to present perjury testimony just as they present other testimony).

14. United States v. Wade, 388 U.S. 218, 256 (1967) (White, J., dissenting in part and concurring in part); see Harry I. Subin, The Criminal Defense Lawyer’s “Different Mission”: Reflections on the “Right” to Present a False Case, 1 GEO. J. LEGAL ETHICS 125 (1987).

15. SIMON, supra note 10, at 170-94; Subin, supra note 14; Harry I. Subin, Is This Lie Necessary? Further Reflections on the Right to Present a False Defense, 1 GEO. J. LEGAL ETHICS 689 (1988).

16. Pepper, supra note 5, at 622 (“In the criminal context the moral value of full ac-

cess to all that the law allows is simply clearer and more dramatic.”).

17. See id. at 621 (“The critics [of neutral partisanship] suggest that a role justified by the rather unusual context of the criminal justice system simply is not justified in the far more common lawyer roles.”).

18. Lon Fuller & John D. Randall, Professional Responsibility: Report of the Joint Conference, 44 A.B.A. J. 1159, 1160-61 (1958); Murray L. Schwartz, The Professionalism and Accountability of Lawyers, 66 CAL. L. REV. 669 (1978).

19. LAWYERS & JUSTICE, supra note 5, at 64. See Murray L. Schwartz, The Zeal of the

Civil Advocate, 1983 AM. B. FOUND. RES. J. 543.

[b]oth roles are designed to assure that no one is punished unless on evidence that excludes reasonable possibilities of factual innocence. But the friendly inquisitor’s job is to assemble facts, and if the facts exclude reasonable doubts about guilt, her remaining job is to bring the offender to accept responsibility, not to make efforts to help him avoid it. The zealous advocate, on the other hand, brings a far stricter epistemic standard to the facts, and she operates throughout on the presumption (if not the belief) that responsibility cannot justly be assigned to the accused. It is this distinc-

20. Id. at 58-66; see also SIMON, supra note 10, at 53-76 (arguing against instrumen-

talist defenses of neutral partisanship).

21. We should note that not all the reformers have gone along. See David Luban, Are Criminal Defenders Different?, 91 MICH. L. REV. 1729 (1993) (criticizing William Simon’s critique of neutral partisanship in the context of criminal defense).

22. See SIMON, supra note 10, at 177 (“We can all agree on a system that provides strong opportunities for the establishment of innocence and for the assertion of some intrinsic procedural rights.”).

23. Id. at 190; see also Subin, supra note 14; Murray Schwartz, On Making the True Look False and the False Look True, 41 SW. L.J. 1135 (1988). The official position of the ABA came close to that of the reformers on the point of discrediting truthful witnesses, but only for a while. As Murray Schwartz pointed out:

The ABA Project on Minimum Standards for Criminal Justice suggests that a lawyer’s knowledge that the witness is testifying truthfully “may affect the method and scope of cross-examination or impeachment,” STANDARDS RELATING TO THE ADMINISTRATION OF CRIMINAL JUSTICE, Standard 7.6(b) at 132 (1974), and the Commentary implies that the lawyer should not impeach a truthful witness (Approved Draft, at 272).

Schwartz, supra note 18, at 673 n.12. But that was 1974; the current version makes clear that “[d]efense counsel’s belief or knowledge that the witness is telling the truth does not preclude cross-examination.” STANDARDS FOR CRIMINAL JUSTICE § 4-7.6(b) (1992).

tive difference between the roles of the friendly inquisitor and the zealous advocate that is in need of justification.24

Major premise: Knowingly to help another do harm or injustice is, prima facie, to do harm or injustice oneself. Minor premise: When criminal defense lawyers knowingly assist the guilty in avoiding conviction, they knowingly assist in harm or injustice. Conclusion: Therefore, criminal defense lawyers, prima facie, do harm or injustice themselves.

24. Edmundson, supra note 2, at 17. 25. Prof. Edmundson himself concedes that a harm occurs when the guilty are acquitted; as we shall see, he is at pains, not to deny the harm, but to absolve the guilty defendants’ lawyers of any culpability for that harm, even prima facie. Id. at 19.

26. On the “no harm” principle, generally, see 1 JOEL FEINBERG, THE MORAL LIMITS

OF THE CRIMINAL LAW: HARM TO OTHERS (1984).

27. PLATO, THE REPUBLIC 15 (Allan Bloom trans., 1968) 28. PLATO, THE GORGIAS 77-80 (Walter Hamilton trans., 1960). 29. FRIEDRICH NIETZSCHE, ON THE GENEALOGY OF MORALS 28-34 (Walter Kaufmann

& R.J. Hollingdale trans., 1969).

30. AYN RAND, THE VIRTUE OF SELFISHNESS (1964).

Major premise: Criminal defense lawyers are not morally culpable for the harms or injustice they assist in when they represent their clients as zealously as the law allows, because of A (the traditional, instrumentalist reasons) or B (Edmundson’s epistemological reason). Minor premise: Other lawyers (some or all) are like criminal defense lawyers in the representation of their clients, because of A (the traditional reasons) or B (Edmundson’s reason). Conclusion: Other lawyers (some or all) are not morally culpable for the harms they assist in when they represent their clients as zealously as the law allows.

31. Edmundson, supra note 2, at 18-19. 32. See David Luban, Partisanship, Betrayal and Autonomy in the Lawyer-Client Relationship: A Reply to Stephen Ellman, 90 COLUM. L. REV. 1004, 1027-28 (1990) (detailing elements of “brutal cross-examination”).

33. David Luban has convincingly shown why, as a general defense of neutral partisanship, the Johnsonian position cannot hold. LAWYERS & JUSTICE, supra note 5, at 26-29. The crux of his long and careful argument is nicely captured in this parodic reductio ad absurdum: “‘Imagine that!’ Dr. Johnson exclaims. ‘I was wrong about the weakness of our case! My arguments turned out to be sound after all—for the judge relied on them in the opinion!’” Id. at 29. In important ways, Luban’s argument against Johnson’s position anticipates Edmundson’s reliance on that position, though it cuts much more strongly as to issues of law than issues of fact.

II. ANALYSIS: WHY PROF. EDMUNDSON’S EPISTEMOLOGICAL APPROACH CANNOT ANSWER THE BASIC QUESTION ABOUT THE CRIMINAL DEFENSE PARADIGM

34. So, I can now report, he has confirmed to me in our email correspondence. E-mail from Professor William Edmundson to Professor Rob Atkinson (Aug. 14, 2002, 11:47 EST) (on file with author).

35. BERTRAND RUSSELL, AN OUTLINE OF PHILOSOPHY 7 (1960). 36. DAVID HUME, AN INQUIRY CONCERNING HUMAN UNDERSTANDING 72-89 (Charles

W. Hendel ed., The Bobbs-Merrill Co. 1955) (1748).

37. RENE DESCARTES, DISCOURSE ON METHOD 20-21 (Laurence J. LaFleur trans., Lib-

eral Arts Press 1950) (1637).

38. I’m thinking of Kant’s effort to answer Hume’s skepticism, see IMMANUEL KANT, CRITIQUE OF PURE REASON (J.M.D. Meiklejohn trans., P.F. Collier 1901) (1787), and the Ordinary Language Philosophers’ response to reductionists like Bertrand Russell. Illustrative of a very wide field are J.L. AUSTIN, HOW TO DO THINGS WITH WORDS (1975), and LUDWIG WITTGENSTEIN, PHILOSOPHICAL INVESTIGATIONS (G.E.M. Anscombe trans., 1953). Prof. Edmundson has reminded me that the American Pragmatists, particularly John Dewey, are another case-in-point. See JOHN DEWEY, THE QUEST FOR CERTAINTY (1930).

39. WITTGENSTEIN, supra note 38, § 23, at 11e (“Here the term ‘language-game’ is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life.”).

40. AUSTIN, supra note 38. 41. BUTCH CASSIDY AND THE SUNDANCE KID (20th Century Fox 1969). 42. WITTGENSTEIN, supra note 38, § 11, at 6e (“Think of the tools in a tool-box: there is a hammer, pliers, a saw, a screw-driver, a rule, a glue-pot, glue, nails, and screws.—The functions of words are as diverse as the functions of these objects.”).

It’s a double standard. Even if the jurors in both cases “knew” Mr. Simpson killed his ex-wife, the jurors in the first trial could find the evidence insufficient for a criminal conviction, even though the jurors in the second trial could find essentially the same evidence adequate for civil liability.

43. Edmundson, supra note 2, at 7.

44. The most famous, or infamous, is “the Lecture,” from ROBERT TRAVER, ANATOMY OF A MURDER 35-36 (1958). The lawyer silences the client, tells the client a range of possible defenses, and lets the client fill in the necessary facts; if the sequence is properly followed, the lawyer supposedly can’t “know” when the client is lying. Id. See also KENNETH MANN, DEFENDING WHITE-COLLAR CRIME: A PORTRAIT OF ATTORNEYS AT WORK (1985) (describing “avoidance techniques,” ways developed by white collar criminal defense lawyers “to avoid inquiry that protect them from potential accusations of unethical misconduct by others and from a personal feeling of ethical impropriety that would make it difficult to perform their work”).

45. MODEL RULES OF PROF’L CONDUCT R. 3.1 (1999). California’s parallel provisions, which would have governed the Simpson case, are admittedly less clear on this point. See CALIFORNIA RULES OF PROF’L CONDUCT R. 5-200(A), (B) (1988) (requiring lawyers to employ “such means only as are consistent with truth” and forbidding them to “seek to mislead the judge, judicial officer, or jury by an artifice or false statement of fact or law”).

46. FED. R. CIV. P. 11. 47. Edmundson, supra note 2, at 7-8. 48. Id. at 7 (quoting MODEL RULES OF PROF’L CONDUCT R. 3.3 (1999 . See also ABA Comm. on Ethics and Prof’l Responsibility, Formal Op. 87-353 (1987) (clarifying, in the wake of the Supreme Court’s decision in Nix v. Whiteside, that this rule applies with full force to criminal defense counsel).

49. Edmundson, supra note 2, at 8. 50. 475 U.S. 157 (1986). 51. FREEDMAN, supra note 13, at 120. 52. Nix, 475 U.S. at 161.

53. But, as Prof. Edmundson’s patron sage pointed out, “when any man knows he is to be hanged in a fortnight, it concentrates his mind wonderfully.” 2 JAMES BOSWELL, THE LIFE OF SAMUEL JOHNSON 393 (1963).

54. Id. at 190-91 (Stevens, J., concurring):

A lawyer’s certainty that a change in his client’s recollection is a harbinger of intended perjury—as well as judicial review of such apparent certainty—should be tempered by the realization that, after reflection, the most honest witness may recall (or sincerely believe he recalls) details that he previously overlooked.

55. See Rock v. Arkansas, 483 U.S. 44 (1987) (affirming criminal defendants’ right to

testify).

56. Edmundson, supra note 2, at 6. 57. Id. at 16.

Any type of criminal justice system has to be justified, and the candidate types of justification emphasize one or more of the following: the social consequences of adopting one or another type of system; the extent to which one or another system respects human dignity; and the extent to which one, rather than another, system would be eligible as an object of reasonable agreement among selfinterested individuals.58

58. Id. 59. Id. at 18. Of the argument that he takes as the linch-pin of the adversarial system defense, Edmundson says “it has a certain plausibility, but may be less than compelling.” Id. at 21.

60. Id. at 21. 61. Id.

62. Id. See also id. at 16:

It is a regrettable fact that many criminal-defense attorneys bring the same zeal they exercise on behalf of their clients to the defense of the adversary system. Were this misplaced zeal relaxed, criminal-defense attorneys might be able to join the rest of us in cooly examining the justification of the adversary system.

63. Id. at 22. See also SIMON, supra note 10, at 173-79 (dismissing the “bogey of the

state” as “driven by what might be called the Libertarian dogma”).

64. Edmundson, supra note 2, at 22. 65. Id. at 15-16, 21.

66. My account of the structure of role morality follows LAWYERS & JUSTICE, supra note 5, at 128-47 (Chapter 7, “The Structure of Role Morality”). See also MICHAEL D. BAYLES, PROFESSIONAL ETHICS 18-19 (2d ed. 1989) (describing a similarly-tiered justification system).

67. See SIMON, supra note 10, at 18 (explaining that he grounds his critique of neutral partisanship in jurisprudence rather than ethics, the spirit of the law instead of ordinary morality, because reliance on the latter places lawyers at a psychological disadvantage).

68. Edmundson, supra note 2, at 8-9. 69. Bernard Williams, Professional Morality and Its Dispositions, in THE GOOD

LAWYER, supra note 5, at 259.

70. Randy Bellows, Notes of a Public Defender, in THE SOCIAL RESPONSIBILITIES OF

LAWYERS: CASE STUDIES 69 (Philip B. Heymann & Lance Liebman eds., 1991).

71. As Randy Bellows put it, in describing his cross examination of a police officer in a

criminal trial:

I did what I had to do. I did not regret it then and I do not regret it today. But I had tried to make an honorable man appear dishonorable. And that is a sad thing to have to do, even if you are a public defender and even if that is your job.

Id. at 71.

72. Videotape: Ethics on Trial (Jim Wesley 1987) (on file with the Florida State University Law Library) [hereinafter Ethics on Trial]. See also TOM ALIBRANDI & FRANK H. ARMANI, PRIVILEGED INFORMATION (1984). 73. Bellows, supra note 70, at 69-70.

74. Ethics on Trial, supra note 72. 75. See Paul R. Tremblay, Shared Norms, Bad Lawyers, and the Virtues of Casuistry, 36 U.S.F. L. REV. 659, 672, 704-08 (2002) (noting and trying to account for and respond to the presence of “felons, whores, and jerks” within the legal profession).

III. CONCLUSION: TOWARD A SOCIAL DEMOCRATIC PERSPECTIVE ON THE CRIMINAL DEFENSE PROBLEM

76. See Hugh Trevor-Roper, History: Professional and Lay,

in HISTORY & IMAGINATION: ESSAYS IN HONOUR OF H.R. TREVOR-ROPER 1, 13 (Hugh Lloyd-Jones et al. eds., 1981).

77. Edmundson, supra note 2, at 5.

78. See Frank H. Easterbrook, Plea Bargaining as Compromise, 101 YALE L.J. 1969, 1970 (1992) (“Innocent persons are accused not because prosecutors are wicked but because these innocents appear to be guilty.”).

79. See John H. Langbein, Torture and Plea Bargaining, 46 U. CHI. L. REV. 3, 5 (1978) (“[S]ociety cannot long tolerate a legal system that lacks the capacity to convict unrepentant persons who commit clandestine crimes . . . .”).

80. See LAWYERS & JUSTICE, supra note 5, at 58 (“Better, we say, that a hundred

criminals go free than that one person be wrongly convicted.”).

81. See id. at 59 n.17 (“The real objection to ‘Better a hundred criminals go free’ is not that it is a middle class bromide, but rather that the middle class is so willing to abandon it the moment that a suspected mugger or burglar enters the docket.”).

82. According to recent statistics, fewer than 10% of state felony cases and fewer than 15% of all federal cases. U.S. DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS, BULLETIN NCJ-173939, FELONY SENTENCES IN STATE COURTS 1996 7-8 (1999); ADMIN. OFFICE OF THE U.S. COURTS, OFFICE OF HUMAN RES. & STATISTICS, STATISTICS DIV., STATISTICAL TABLES at Table FOR http://www.uscourts.gov/judiciary2001/dectables/d04dec01.pdf (on file with author).

JUDICIARY

available

(2001),

FED.

D-4

THE

83. See Langbein, supra note 79, at 21 (“The contemporary nonadversarial criminal justice systems of countries like West Germany have long demonstrated that advanced industrial societies can institute efficient criminal procedures that nevertheless provide for lay participation and for full adjudication in every case of serious crime.”); see also id. at 3 (“My thesis is that there are remarkable parallels in origin, in function, and even in specific points of doctrine, between the law of torture and the law of plea bargaining.”); Rudolf B. Schlesinger, Comparative Criminal Procedure: A Plea for Utilizing Foreign Experience, 26 BUFF. L. REV. 361, 382 (1977) (“[A]lthough the continental systems do not recognize a plea of guilty, so that no defendant can be convicted or sentenced without an actual trial, most criminal cases are handled with surprising dispatch.”). Not everyone, it should be noted, opposes plea-bargaining. See, e.g., Easterbrook, supra note 78, at 1975 (“Plea bargains are preferable to mandatory litigation . . . because compromise is better than conflict.”). For a succinct if now somewhat dated review of the literature for and against, see Robert E. Scott & William J. Stuntz, Plea Bargaining as Contract, 101 YALE L.J. 1909, 1909 n.4 (1992).

84. See SWEDISH PRESIDENCY OF THE EUROPEAN UNION, EUROPEAN UNION POLICY ON

DEATH

THE http://www.eurunion.org/legislat/DeathPenalty/Demarche10May.htm (“The European Union is deeply concerned about the high number of executions in the United States.”) (on file with author).

available

PENALTY

(2001),

at

85. DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS, BULLETIN NCJ-195189, PRISONERS IN 2001 11-12 (2002). More precisely, in 2001, 46.3% of the total state and federal inmate population was Black, compared to 36.1% white and 15.6% Hispanic. Id. at 11. Even more stark are comparative incarceration rates: Fully 10% of Black males between twenty-five and twenty-nine are in prison; the rate for whites is just over 1%. Id. at 12. “[36%] of all inmates were not employed during the month before they were arrested for their current offense.” DEP’T OF JUSTICE, BUREAU OF JUSTICE STATISTICS, CRIMINAL OFFENDERS STATISTICS (2002), at http://www.ojp.usdoj.gov/bjs/crimoff.htm. Curiously, the Bureau of Justice Statistics does not seem to maintain statistics on inmate income levels or other more direct indices of economic or social class.

86. “Poor people accused of capital crimes are often defended by lawyers who lack the skills, resources, and commitment to handle such serious matters.” Stephen B. Bright,

Counsel for the Poor: The Death Sentence Not for the Worst Crime but for the Worst Lawyer, 103 Yale L.J. 1835, 1836 (1994).

87. William Simon has gone a long way in that direction, suggesting that what he calls “[a]ggressive defense should be limited only to those cases that present a threat of excessive or arbitrary punishment and should be employed only to the extent it is likely to counter the threat.” SIMON, supra note 10, at 190. This approach, he plausibly argues, “would enable defense lawyers to connect their most plausible commitments more directly to their everyday practices.” Id. at 194.

88. See supra note 3.