Consequentialist justifications of the adversary system

Truth

The question whether the adversary system is, all in all, the best way of uncovering the facts of a case at bar sounds like an empirical question. I happen to think that it is - an empirical question, moreover, that has scarcely been investigated, and that is most likely impossible to answer.

The kind of empirical research that can be done, then, is laboratory simulations: social psychology experiments intended to model the adversary proceeding. Obviously, there are inherent limitations on how closely such experiments can correspond to actual trials, no matter how skillfully they are done. In fact, the only experiments of the sort I know of are those of Thibaut, Walker, and their associates, and these are far from perfect modelings of the adversary and “inquisitorial” - meaning French- and German-style - systems that they are comparing.^[49] Even so, the results are instructive: they show that in some situations the adversary system works better while in others the inquisitorial system does, and furthermore, that the participants cannot tell which situation they are in. This would hardly surprise us: it would be much more astounding to discover a greater difference in veracity between the Anglo-American and continental systems, for surely such a difference would after so many centuries have become a commonplace in our folklore.

Given all this, it is unsurprising to discover that the arguments purporting to show the advantages of the adversary system as a fact-finder have mostly been nonempirical, a mix of a priori theories of inquiry and armchair psychology.

Here is one, based on the idea, very similar to Sir Karl Popper’s theory of scientific rationality, that the way to get at the truth is a wholehearted dia­lectic of assertion and refutation.^[50] If each side attempts to prove its case, with the other trying as energetically as possible to assault the steps of the proof, it is more likely that all of the aspects of the situation will be presented to the fact-finder than if it attempts to investigate for itself with the help of the lawyers.

This theory is open to a number of objections. First of all, the analogy to Popperian scientific methodology is not a good one. Perhaps science pro­ceeds by advancing conjectures and then trying to refute them, but it does not proceed by advancing conjectures that the scientist believes to be false and then using procedural rules to exclude probative evidence.^[51]

The two adversary attorneys are each under an obligation to present the facts in the manner most consistent with their client’s position - to prevent the introduction of unfavorable evidence, to undermine the credibility of opposing witnesses, to set unfavorable facts in a context in which their importance is minimized, to attempt to provoke inferences in their client’s favor. The assumption is that two such accounts will cancel out, leaving the truth of the matter. But there is no earthly reason to think this is so; they may simply pile up the confusion.

This is particularly likely in cases turning on someone’s sanity or state of mind. Out comes the parade of psychiatrists, what Hannah Arendt once called “the comedy of the soul-experts.”^[52] Needless to say, they have been prepared by the lawyers, sometimes without knowing it. A clinical law teacher explained to a class that when you first contact a psychiatrist and sketch the facts of the case, you mention only the favorable ones. That way, he or she has an initial bias in your favor and tends to discount the unfavorable facts when you finally get around to mentioning them.

The other side, of course, can cross-examine such a witness to get the truth out. Irving Younger, in his time the most popular lecturer on trial tactics in the country, tells how. Among his famous “Ten Commandments of Cross­Examination” are these:

• Never ask anything but a leading question.

• Never ask a question to which you don’t already know the answer.

• Never permit the witness to explain his or her answers.

• Don’t bring out your conclusions in the cross-examination. Save them for closing arguments when the witness is in no position to refute them.^[53]

Of course, the opposition may be prepared for this; they may have seen Younger’s three-hour, $425 videotape on how to examine expert witnesses. They may know, therefore, that the cross-examiner is saving his or her conclusions for the closing argument. Not to worry! Younger knows how to stop an attorney from distorting the truth in closing arguments. “If the opposing lawyer is holding the jury spellbound... the spell must be broken at all cost. [Younger] suggests the attorney leap to his or her feet and make furious and spurious objections. They will be overruled, but they might at least break the opposing counsel’s concentration.”^[54]

My guess is that this is not quite what Sir Karl Popper had in mind when he wrote, “The Western rationalist tradition... is the tradition of critical discussion - of examining and testing propositions or theories by attempting to refute them.”⁴³

A skeptic, in fact, might try this scientific analogy: a beam of invisible electrically charged particles - charge and origin unknown - travels through a distorting magnetic field of unknown strength, then through an opposite field of unknown, but probably different, strength. The beam strikes a detector of undeterminable reliability, from which we are supposed to infer the nature and location of the beam’s source. That is the adversary system at its worst.

There is, however, one legal context in which the Popperian defense of the adversary system approximates reality and in which the adversary system is indeed strongly justified.

It makes sense to assign each advocate the task of arguing one side’s interpretation of the law as forcefully as possible, and doing everything possible to undermine the adversary’s arguments. With no facts to hide and everything out in the open, only the arguments and counter-arguments remain. Judges invariably attest that the better the advocates arguing before them, the better decisions they make. Adversary advocacy helps ensure that no arguments or objections get overlooked.^[55]

Now, the same thing will often be true when lawyers argue over the interpretation of evidence in a trial of facts, so it may appear that my defense of the adversary system of arguing questions of law proves too much, and provides a defense for adversary arguments about facts as well. To the extent that the lawyers are arguing the interpretation of evidence in the record, that is true. But the problems with the adversary system I have highlighted lie in the fact that trial lawyers view one of their main jobs as keeping damaging information out of the record, or - as in Younger’s recommendation that lawyers disrupt their adversaries’ closing arguments - clouding the decision­making process. Consider, for example, complaints by the president of a lawyers’ organization about a recent American innovation in which jurors are permitted to question witnesses directly. “You work very hard to keep certain information out of the trial. Then all of your finesse and art and technique are thrown out the window when a juror comes in and asks, ‘Where were you on the night in question?’ ”^[56] It is hard to defend adversary fact-finding on the ground that it is the best way of ensuring that judges and juries get the most information, when the lawyer’s “finesse and art and technique” consists of keeping awkward facts out of court.

Even worse, adversarial tactics sometimes include efforts to ensure that cases never even make it to the stage of fact-finding. Defense counsel for corporate defendants use procedural delays to exhaust their opponents’ funds. When they can, lawyers resort to intimidation tactics. A particularly egre­gious example occurred repeatedly during litigation over the Dalkon Shield, an intrauterine contraceptive device that pharmaceutical manufacturer A. H. Robins marketed during the 1970s to over three million women. Because of a design flaw, the Dalkon Shield caused an estimated 66,000 miscarriages and sterilized thousands of women by infecting them with pelvic inflammatory disease (PID). Faced with staggering liability exposure, Robins and its counsel decided on a scorched-earth defense. One tactic of Robins’s counsel soon acquired the nickname “the dirty questions list.” Defense lawyers taking depositions asked plaintiffs very specific, very graphic questions about intimate details of their personal hygiene and sexual practices - questions that one plaintiff described as “more like an obscene phone call” than a legal interrogation. Firm lawyers argued that the “dirty questions” were relevant to the lawsuits because they might reveal alternative sources of PID infection. The questions mainly served, however, to intimidate plaintiffs into dropping their lawsuits or settling them for inadequate amounts. The message was clear that they might have to answer the same questions in open court. Among other things, defense lawyers asked plaintiffs for the names of all their past and present sexual partners (“besides your husband”), with the clear impli­cation that the partners’ names might be revealed and their testimony elicited for purposes of impeaching plaintiffs’ answers to the “dirty questions” about what they like to do in bed. Potential plaintiffs filed affidavits indicating that they had dropped their own lawsuits because of the questions other plaintiffs had been asked.^[57]

A similar example is the rise of the so-called “SLAPP suit” - “Strategic Lawsuit Against Public Participation.” In a typical SLAPP suit, citizens protesting corporate policies or actions are sued for defamation or tortious interference with business.

The point of these examples is plain: you cannot defend the adversary system on the basis of its truth-finding function when it licenses (or even requires) behavior designed to ensure that the truth never comes out, because litigants are intimidated into abandoning legitimate cases.

One final difference between “pure” argument, paradigmatically appel­late argument of legal issues, and the adversary system of fact-finding appears in the ethics rules themselves. Ordinarily, lawyers are required to keep facts confidential, and in the adversary system they must never reveal damaging facts to a court unless they are compelled to do so. Matters are very different when we turn from facts to law. Here, the fundamental rule requires lawyers “to disclose to the tribunal legal authority in the controlling jur­isdiction known to the lawyer to be directly adverse to the position of the client and not disclosed by opposing counsel.”^[59] This rule, which law students and lawyers often find counter-intuitive (“why should I do my adversary’s legal research for them?”), highlights what makes argument about legal questions different. The idea is to ensure that judges reach the best resolutions they can of questions of law. Their resolutions, after all, become precedents. Getting to the best resolutions requires total transpar­ency, and if my adversary has overlooked a favorable case on point, the rule requires me to throw myself on the sword by telling the judge about the case, to ensure that the judge does not overlook it. By contrast, we have seen that adversarial advocacy on factual matters places lawyers at war with transparency.

There is a sophisticated response to these arguments, which was offered in 1958 by Lon Fuller and John Randall in a semi-official report by a conference of the ABA and the AALS (the law schools’ association). In the Joint Conference Report, Fuller and Randall argue that even though the adversary system apparently requires lawyers to obscure the truth on behalf of their clients, the system is actually more likely to arrive at the truth than an inquisitorial alternative in which the judge (or jury) rather than the lawyer conducts the inquiry. They base the argument on the psychological impos­sibility of a single mind formulating the strongest version of two contra­dictory positions:

Any arbiter who attempts to decide a dispute without the aid of partisan advocacy... must undertake not only the role of judge, but that of representative for both of the litigants. Each of these roles must be played to the full without being muted by qualifications derived from the others. When he is developing for each side the most effective statement of his case, the arbiter must put aside his neutrality and permit himself to be moved by a sympathetic identification sufficiently intense to draw from his mind all that it is capable of giving... When he resumes his neutral position, he must be able to view with distrust the fruits of this identification and be ready to reject the products of his own best mental efforts. The difficulties of this undertaking are obvious. If it is true that a man in his time must play many parts, it is scarcely given to him to play them all at once.^[60]

This argument, plausible though it appears on the surface, should set off an alarm in the minds of readers, because what Fuller and Randall claim is psychologically impossible turns out to be daily practice in civil law systems - and Fuller was a good enough comparativist to have known better.^[61] Sybille Bedford, who observed trials in several countries, wrote this about a German criminal trial:

It was a strange experience to hear this presentation of a case by both sides, as it were, in one; not a prosecution case followed by a defence case, but an attempt to build the whole case... as it went. A strange experience... to hear all questions, probing questions and soothing questions, accusatory and absolving questions, questions throwing a favourable light and questions having the opposite effect, flow from one and the same source, the bench...⁵¹

What Mrs. Bedford found “strange” she nevertheless found extremely effective as well; and there is not a shred of evidence that continental systems are worse fact-finders than their Anglo-American counterparts. Perhaps a trained judge can play all parts at once.⁵² If so, then where is the error in Fuller and Randall’s argument?

One problem is that it begs the question. When Fuller and Randall write, “Each of these [representative] roles must be played to the full without being muted by qualifications derived from the others,” they presuppose that inquiry proceeds best by unmuted adversary presentation - in which case, of course, an inquisitorial investigation becomes by definition a mere copy of the real thing. Isn’t it equally possible, however, that a decision-maker can form a more reliable picture if the opposed positions are muted by qualifi­cations derived from each other? After all, the strongest form of each side’s case may be strongest because it is exaggerated and misleading. Sometimes the opponent may be able to smoke out the distortions and half-truths, but there will inevitably be cases in which the decision-maker simply cannot sort

appeared, and there can be little doubt that Fuller, who had a strong interest in German legal thought, knew their work. Benjamin Kaplan, Arthur T. von Mehren, & Rudolf Schaefer, Phases of German Civil Procedure I, 71 Harv. L. Rev. 1193 (1958); Kaplan, von Mehren, & Schaefer, Phases of German Civil Procedure II, 71 Harv. L. Rev. 1443 (1958). There is, however, reason to believe that Fuller was deeply suspicious of continental procedure. See Fuller, The Adversary System, supra note 28, at 36.

⁵¹ Sybille Bedford, The Faces of Justice: A Traveller’s Report 117 (Simon & Schuster, 1961).

⁵² Of course one might respond that continental systems actually are adversarial. Judges don’t proceed in a vacuum; rather, they work from written, partisan submissions by the litigants’ attorneys.

This response is unconvincing. The judges still take the active role in questioning wit­nesses and eliciting further submissions and further evidence; and the lawyers assume a role considerably more passive than their American counterparts. (To offer one telling example, German ethics rules discourage lawyers from interviewing witnesses and forbid lawyers from preparing them, whereas an American lawyer who does not prepare witnesses for trial has done an inadequate job.) The German judge takes the written submissions as a beginning point, but need not confine the inquiry to those submissions. If Fuller’s argument were sound, it would apply to the half-inquisitorial, half-adversarial systems of continental procedure. through the exaggerations, strategic omissions, and false implications, and as a result decides wrongly.

In addition, Fuller and Randall’s argument proves too much: it proves the impossibility not merely of reliable inquisitorial investigation, but of partisan advocacy as well! Any skilled lawyer preparing a case tries to anticipate the strongest arguments available to the adversary, preferably in their most devastating form. When she sizes up her witnesses, she puts herself in her opponent’s shoes and probes for weaknesses in the witness’s story; she digs for damaging information the opponent might unearth about the witness. Then she tries to construct counter-arguments to the opponent’s best shot, and to anticipate counter-arguments to her counter-arguments. In short, she employs precisely the mental progression - from sympathetic identification with her own position, to detachment from it, to distrust of it, then back again - that Fuller claims is psychologically impossible for the inquisitorial judge.

The Joint Conference Report employs two additional psychological arguments against inquisitorial tribunals. The first is that the adversary system will “hold the case... in suspension between two opposing inter­pretations of it,” so the finder of fact won’t jump to hasty conclusions.^[62] The inquisitorial judge, by contrast, inevitably forms preliminary conceptions of the case, and will quite naturally become so invested in these working hypotheses that he may hang on to them even after they turn out to be false leads.

The second argument is that if the judge and not the lawyer had to “absorb” the embarrassment of her initial theory of the case being exploded in court, she would be “under a strong temptation to keep the hearing moving within the boundaries originally set for it”; that would turn a fair trial into a mere “public confirmation for what the tribunal considers it has already established in private.”⁵⁴

These arguments, unlike the previous one, have a well-confirmed psy­chological basis. The theory of cognitive dissonance holds that when we perform an action, our beliefs become more congruent with the action. An inquisitorial judge, pursuing her theory of the case, will call witnesses, request evidence, ask questions. To abandon the line of inquiry is tantamount to admitting that she has been wasting everyone’s time. Here, she eliminates cognitive dissonance by continuing to believe that her theory of the case is plausible even when it should be abandoned.

The problem with the argument, of course, is that the shortcomings of inquisitorial procedure don’t necessarily put it at a comparative disadvantage to adversarial procedure. The adversary system has its own epistemic short­comings, which derive from the fact that zealous advocates can sometimes hide facts successfully to win a weak case. In addition, many “inquisitorial” courts use multi-judge panels, which mutes the psychological distortions of cognitive dissonance: if one judge becomes overly invested in a fruitless line of inquiry, the other judges can take the reins.

Ultimately, the Joint Conference Report seems to take as a premise the idea that truth is served by self-interested rather than disinterested investi­gation. “The lawyer appearing as an advocate before a tribunal presents, as persuasively as he can, the facts and the law of the case as seen from the standpoint of his client's interest” [emphasis added].^[63] The emphasized phrase is accurate, but it gives the game away. For there is all the difference in the world between “the facts seen from X’s standpoint” and “the facts seen from the standpoint of X’s interest.” Of course it is important to hear the former - the more perspectives we have, the better informed our judgment. But to hear the latter is not helpful at all. It is in the murderer’s interest not to have been at the scene of the crime; consequently, the “facts of the case as seen from the standpoint of [the] client’s interest” are that the client was elsewhere that weekend. From the standpoint of my interest, the world is my cupcake with a cherry on top; from the standpoint of yours, its streets are paved with gold and you own the streets. Combining the two does not change folly to truth.

All this does not mean that the adversary system may not in fact get at the truth in many hard cases. I suppose that it is as good as its rivals. But, to repeat the point I began with, nobody knows how good that is.^[64]

Legal rights

It is sometimes said, however, that the point of the adversary system is not that it is the best way of getting at the truth, but rather the best way of defending individuals’ legal rights. Freedman points out that if the sole purpose of a trial were to get at the truth we would not have our Fourth, Fifth, and Sixth Amendment rights; that improperly obtained evidence cannot be used against us and that we cannot be required to testify against ourselves indicate that our society considers other values more central than truth.^[65] And, according to the theory we shall now consider, these other values have to do with legal rights.^[66]

The argument is that the best way to guarantee that an individual’s legal rights are protected is to provide him or her with a zealous adversary advocate who will further the client’s interest.

This argument, we should note, is slightly different from Freedman’s, according to which counsel by a zealous advocate is not merely the best way of defending one’s legal rights, but is itself one of those rights. That, of course, would make the adversary system necessary for the defense of legal rights, but only in the tautological sense that taking away counsel infringes a person’s right to counsel and you can’t defend a right by infringing it. Freedman suggests that adversary advocacy is a constitutional value, but this is not obvious.^[67] The Constitution makes no explicit mention of the adversary system. In fact, US courts of equity used inquisitorial procedure that only gradually metamorphosed into adversary procedure, not completing the process until the fusion of equity courts and law courts in the twentieth century.^[68] Now, it may be, as Theodore Koskoff says, a “fact, so basic that the Constitution does not even mention it, that our system of justice is an adversary system.”^[69] Certainly the Supreme Court has asserted many times that we have an adversary system.^[70] But the Court has also explicitly held that due process can be satisfied through non-adversarial procedures, and its assertion that we have an adversary system means only that under an adversary system, due process requires adversary advocacy.^[71] It is clear that the Court tolerates tinkering with the adversary format - for example, the Court approved modified discovery rules that require lawyers to turn over more information, even though three justices complained that the new rules are “contrary to the nature of our adversary system.”^[72] While the Sixth Amendment certainly gives persons accused of crimes the right to counsel, it says nothing about the rules of engagement. If we used a nonadversarial system, the Sixth Amendment right could be fulfilled by giving the accused a nonadversary advocate.

The argument we are considering is rather that, right to counsel aside, adversary advocacy is the best defense of our other legal rights. The no- holds-barred zealous advocate tries to get everything the law can give (if that is the client’s wish) and thereby does a better job of defending the client’s legal rights than a less committed lawyer would do.

Put this way, however, it is clear that the argument trades on a confusion. My legal rights are everything I am in fact legally entitled to, not everything the law can be made to give. For obviously a good lawyer may be able to get me things to which I am not entitled, but this, to call a spade a spade, is an example of infringing my opponent’s legal rights, not defending mine. The “dirty questions list” and SLAPP suits are adversarial tricks of the trade used to do opponents out of their legal deserts.

It might be replied that looking at it this way leaves the opponent’s lawyer out of the picture. Of course, the reply continues, no one is claiming that a zealous adversary advocate is attempting to defend legal rights: he or she is attempting to win. The claim is only that the clash of two such adversaries will in fact defend legal rights most effectively.

But what reason do we have to believe this, other than a question-begging analogy to eighteenth-century economic theories of the Invisible Hand, the­ories that are themselves myth rather than fact? Every skill an advocate is taught is bent to winning cases no matter where the legal right lies. If the opponent manages to counter a lawyer’s move with a better one, this has precisely nothing to do with legal rights. In the Middle Ages lawsuits were sometimes tried by combat between hired champions. Each was charged with defending the legal right of his employer, but surely the fact that one swordsman successfully filleted the other did not mean that a right was established. Of course, judicial combat did not involve argument about rights. But neither does graymailing, “dollaring to death,” driving up opponents’ costs by getting their law firms disqualified, peremptorily challenging jurors because they seem too smart, or even masking bad arguments with what Titus Castricius called “the orator’s privilege to make statements that are untrue, daring, crafty, deceptive and sophistical, provided they have some semblance of truth and can by any artifice be made to insinuate themselves into the minds of the persons who are to be influenced.”^[73]

It is obvious that litigators pride themselves on their won-lost record. The National Law Journal describes “the world’s most successful criminal lawyer - 229 murder acquittals without a loss!” and describes the Inner Circle, a lawyer’s club whose membership requirement is winning a seven­figure verdict.^[74] (That was before the day of the nine- or ten-figure verdict.) You never know, of course - maybe each of these cases really had legal right on its side. And when a coin comes up heads 229 times in a row it may be fair, but there is another explanation. Lawyers themselves do not see the point of what they do as defending their clients’ legal rights, but as using the law to get their clients what they want.

It is true, of course, that one way for society to guarantee that lawyers do their best to defend their clients’ rights is to commit them to defending every claim a client has to a right, whether valid or not. That kind of overkill is reassuring to each client, of course. But suppose we look at it from the point of view of the whole process, rather than of the individual clients. It is hard to see then why an adversary system is the best defender of legal rights. Why not, for example, a system in which both attorneys are committed to defending the legal rights of both parties, if they seem to be getting trampled? I am not recommending such a system: my point is only that we have no reason at all to believe that when two zealous advocates slug it out, the better case, rather than the better lawyer, wins.

Let me be clear about the objection. It is not that the flaw in the adversary system is overkill on the part of morally imperfect, victory-hungry lawyers. The objection is that under the adversary system an exemplary lawyer is required to indulge in overkill to obtain as legal rights benefits that in fact may not be legal rights.

At this point an objection can be raised to my argument. The argument depends on a distinction I have drawn between what a person is in fact legally entitled to and what the law can be made to give. But this is a suspect distinction because it is based on the notion that there are legal entitlements other than what the law in fact gives. American realism, the dominant jur­isprudential theory of the twentieth century, was primarily responsible for throwing cold water on the notion of entitlements-in-themselves floating around in some sort of noumenal never-never land. The law is nothing other than what the courts say it is.

The objection fails, however, for it cuts the ground out from under itself. If legal rights are strictly identical with what the courts decide they are, then it is simply false that the adversary system is the best defender of legal rights. Any system whatsoever would defend legal rights equally well, as long as on the basis of that system courts decided cases.

There is, however, a legitimate insight concealed in the realist objection. Whether or not legal rights are anything beyond what the courts say they are, it is the courts that are charged with adjudicating them. And - the point continues - if lawyers were given discretion to back off from zealous advocacy, they would have to prejudge the case themselves by deciding what the legal rights actually are in order to exercise this discretion. Lawyers would be usurping the judicial function.

Now, it must be said that this insight cannot be used to defend the innu­merable tactics lawyers use to force favorable settlements of cases outside of court (dirty questions, SLAPP suits, delaying tactics); if anything, the argu­ment should condemn such practices inasmuch as they preempt the adjudi­catory process. Nor does it militate against requiring lawyers to disclose adverse information and arguments, since doing so does not usurp the judicial function. But I do not wish to focus on these points, for I think that the insight contains an important argument for the adversary system that we have not yet considered.

Ethical division of labor

This argument is no longer that the excesses of zealous advocacy are excused by the promotion of truth or the defense of legal rights. Rather, it is that they are excused by what Thomas Nagel calls “ethical division of labor.” He says, in a discussion of the peculiarly ruthless and result-oriented role morality of public officials,

that the constraints of public morality are not imposed as a whole in the same way on all public actions or on all public offices. Because public agency is itself complex and divided, there is a corresponding ethical division of labor, or ethical specialization. Different aspects of public morality are in the hands of different officials. This can create the illusion that public morality is more consequentialist or less restrictive than it is, because the general conditions may be wrongly identified with the boundaries of a particular role. But in fact those boundaries usually presuppose a larger institutional structure without which they would be illegitimate. (The most conspicuous example is the legitimacy conferred on legislative decisions by the limitation of constitutional protections enforced by the courts.)^[75]

The idea is that behavior that looks wrong from the point of view of ordinary morality can be justified by the fact that other social roles exist whose purpose is to counteract the excesses resulting from role-behavior. Zealous adversary advocacy is justified by the fact that the other side is also furnished with a zealous advocate; the impartial arbiter provides a fur­ther check.

This is in fact one of the most commonly heard defenses for pugnacious advocacy: “He had a lawyer, too”; “I’m not supposed to do his lawyer’s job for him”; or quoting Sharswood once again, “The lawyer, who refuses his professional assistance because in his judgment the case is unjust and inde­fensible, usurps the functions of both judge and jury.”^[76]

The idea is really a checks-and-balances theory, in which social engi­neering or “wise legislation” is supposed to relieve some of the strain on individual conscience. A functionary in a well-designed checks-and-balances system can simply go ahead and perform his duties secure in the knowledge that injuries inflicted or wrongs committed in the course of those duties will be rectified by other parts of the system.

Will this do the trick? The answer, I am afraid, is no. Suppose that a lawyer is about to embark on a course of action that is unjustified from the point of view of ordinary morality, such as attempting to win an unfair, lopsided judgment for a client from a hapless and innocent party. Or think of our second graymailing example, in which lawyers for a corporation involved in a merger advise their client to fire employees a few at a time to blackmail federal authorities into permitting the merger to go forward. A zealous adversary advocate will do whatever she can to avoid the opposing counsel’s attempt to foil her designs. For that reason, she surely cannot claim that the existence of the opposing counsel morally justifies these actions. Certainly the fact that a man has a bodyguard in no way excuses you for trying to kill him, particularly if you bend all your ingenuity to avoiding the bodyguard.

The problem is this. The checks-and-balances notion is desirable because if other parts of the system exist to rectify one’s excesses, one will be able to devote undivided energy to the job at hand and do it better. It is analogous to wearing protective clothing in a sport such as fencing: knowing that your opponent is protected, you can go all out in the match. But in the adversary system the situation is different, since the attorney is actively trying to get around the checks and balances. Here the analogy is to a fencer who uses a special foil that can cut through the opponent’s protective clothing. To put the point another way, the adversary advocate attempts to evade the system of checks and balances, not to rely on it to save her opponents.

There is another problem with the notion of ethical division of labor. It attempts to justify a system of roles by the fact that the system is self­correcting, in other words that injuries perpetrated by one part of the system will be rectified by another. Rectification, however, carries with it high transaction costs in terms of money, time, worry, energy, and (generally) an arduous passage through the bureaucratic straits. These transaction costs create a general background “noise” in the system, a penalty imposed on one simply for becoming embroiled in it. This can be justified only if the system itself is justified, but then the checks-and-balances argument seems merely to gild the lily. Had we found a justification for the adversary system on other grounds, we would not have needed to turn to the ethical division-of-labor argument to begin with.

Division-of-labor arguments raise a very troubling and difficult topic. The structure of bureaucratic institutions such as the legal system lends itself to divided responsibility. Those who write the rules, those who give the orders, and those who carry them out each have some basis for claiming that they are not at fault for any wrong that results. But this is unacceptable. If moral agency divides along lines of institutional authority, it seems to me that every agent in the institution will wind up abdicating moral responsibility. It is for this reason that division-of-labor arguments must walk a thin line between the legitimate notion that different roles have different duties and the unacceptable notion that moral responsibility is itself diminished or “divided down” by institutional structure. In chapters 6 and 7 we shall examine more closely the ways in which institutional structure serves, in an illusory way, to divide down moral responsibility.^[77]

A final division-of-labor argument exists which is different from those we have just been considering. This is the general line of argument of the Joint Conference Report. It is based on a point emphasized by the realists, namely that lawyers spend very little of their time or attention on actual litigation. Mostly they are involved in other activities: document-drafting, deal-making, negotiation, giving advice, and so forth. The Joint Conference Report seizes on this fact to argue for a separation of lawyerly functions, with a corre­sponding separation of norms of professional behavior in accord with the nature of those functions. The report restricts no-holds-barred zeal to the role of advocate, a role that, to repeat, lawyers do not occupy very much of the time. The real key to the lawyer’s function in society, according to the report, lies not in litigation but in wise counsel and airtight draftsmanship, which make litigation unnecessary. As to the morally troubling cases, the lawyer is permitted or even required to advise the client against “a course of conduct technically permissible under existing law, though inconsistent with its underlying spirit and purpose.”^[78] This the lawyer does by reminding the client of the “long-run costs” of such conduct.⁷¹

I do not think we need to take this argument very seriously, for it trades on a sleight-of-hand and a key omission. The sleight-of-hand lies in the tricky phrase “long-run costs.” Costs to whom? Society at large? I suppose some clients engaged in morally shady projects may be dissuaded by being told how they are harming society, but surely these are just the people least likely to listen. Perhaps the long-run costs are to the client, costs in the form of loss of respect in the community, hard feelings, inability to do business with people in the future, etc. But why suppose that these inevitably accompany morally unworthy litigation? It is a commonplace that we live in a litigious society, and the fact that a person or corporation makes effective use of an arsenal of legal weapons is not often held against him. We have, for better or worse, learned to expect such behavior, and ruthless, hard-driving entrepre­neurship that eagerly goes to the legal mat is more likely to win respect than enmity if it is successful. You’d be surprised (or maybe not) what a lot of money will do to make people like an amoral wheeler-dealer. The Joint Conference Report's ominous rumbling about long-run costs is Panglossian piety, which harmonizes society’s loss with the client’s, when in fact society’s loss is often the client’s gain.

The argument also omits the key point that, after lawyers have offered their “quiet counsel,” they will still have to press forward with the repre­sentation if the client won’t be dissuaded. Perhaps the lawyer can say that she gave morality the old college try, and her heart is pure. Our worry, however, was not about impure hearts, but about dirty hands. And those haven’t become any cleaner.

Thus, the division of functions within a lawyer’s own professional life fares no better than the division of functions within the legal system as a whole: neither is sufficient to provide the moral timbering of adversary advocacy.