Reviewed by Whitley Kaufman, Department of Philosophy, University of Massachusetts Lowell.
Email: whitley_kaufman [at] uml.edu.
Leo Katz is among the most original and interesting thinkers in legal philosophy; his wide-ranging and eclectic interests in philosophy, economics, and other disciplines have helped break out of the narrow confines of legal formalism. Katz’s new book, Why the Law is So Perverse, exemplifies his interdisciplinary approach. His challenging thesis is that many if not most of the “perverse” and odd aspects of the law can be explained through an application of social choice theory. It is always interesting to hear his new ideas, even when they misfire, as I believe the present volume does.
The book is nothing if not ambitious. It takes on a wide range of puzzles in the law: why are there some things that the law does not allow us to consent to? Why is the law so full of loopholes? Why does the law insist on either/or categories as opposed to a continuum? Why do we punish only certain kinds of bad behavior, but not others? All of them, Katz suggests, can be illuminated by reference to a famous result in Social Choice theory known as Arrow’s Impossibility Theorem (and related elaborations of Arrow’s idea). The theorem, Katz tells us, is “perhaps the most surprising finding in all of twentieth-century social science” (p.5). What it says, roughly, is that there is no way to construct a perfect voting system. There are indefinitely many different ways of aggregating individual preferences, and all of them will suffer from flaws, often leading to absurd or paradoxical results. This would seem to be a very serious problem for a democratic system.
Katz thinks that the applications of this theory go far beyond issues of voting and democracy. It has, he thinks, important lessons for statutory interpretation, as it refutes the notion that the legislature has anything like a collective will expressed in the statute (with obvious implications for judicial decisions, as it undercuts the attempt to discover the legislative intent in a given law). Katz thinks that Arrow’s Theorem applies even to individual decision making, so long as it is multicriterial (i.e. there are at least 3 distinct criteria being applied, something that no doubt applies to all significant decisions). Each individual decision is “fairly analogous to that of aggregating the preferences of voters into a master ranking” (p.7). If so, there is inherent “impossibility” in all decisionmaking, individual and collective, something that pervades and affects all human behavior. However, Katz’s main interest is the law. Since all legal decisionmaking “can be thought of as a kind of multicriterial decision making” (p.7), then the law is necessarily pervaded with the problems of social choice theory. It is this connection, Katz believes, that can help us explain so many of the puzzles and perversities of the law mentioned above, e.g. loopholes, the binary character of [*96] the law, the disjunction between law and morality, and so on.
Is Arrow’s Theorem a threat to democracy itself? The significance of the Impossibility Theorem is often overstated. It does not demonstrate that democracy is inherently “impossible” or even that it is fundamentally flawed. To the contrary, a more plausible interpretation of the theorem is that it reveals the inadequacy of a particular reductionistic conception of democracy, the idea that the collective will is found by simply aggregating all of the fixed individual preferences. Katz claims that it is “impossible to speak coherently about a group having a ‘collective will’ of some sort, which of course flies in the face of what political philosophers have believed since antiquity” (p.5). In fact it does no such thing. Advocates of “deliberative democracy” have long argued that we should reject the view that the purpose of democracy is merely to sum up a set of fixed individual desires. They argue that democracy is a process of coming together to shape a collective will, a process in which individual preferences dynamically develop as a result of the deliberation itself. This seems a far more plausible account of politics than the reductionistic and mechanistic one of Social Choice theory. Anyone who attends committee meetings knows that, at their best, this is what happens: people modify their views to create a consensus on which all agree (when there are fixed, unchanging individual preferences, that is often a sign of breakdown in the process). The point is underscored by Katz’s extension of the Theorem to individual decision-making. Indeed, it constitutes a sort of reductio ad absurdum of his entire argument. Even if there are many people convinced by the reductionist account of collective choice, it hardly seems plausible that it adequately describes the process of individual choice, as if there isolated competing preferences in the mind, each of which is given a “vote” in the process. This is to treat the mind in a grossly mechanistic way, ignoring the very idea of rationality. The reductionist approach ignores the way preferences and values interact and affect each other, and the way reason and deliberation incorporates and modifies preferences to pursue higher goals.
An equally serious issue not often enough discussed in this field is the empirical question: how often do these “impossibility” problems crop up? One of the most troubling problems in the book is an evasion of this issue. There is, if I recall correctly, only a single historical example in the whole book illustrating how Arrow’s theorem can lead to paradoxical, counterintuitive results in the real world. Katz takes a case from social choice theorist Donald Saari involving the world figure skating championship in England (apparently the event was in 1995, though Katz oddly chooses not to give the date), in which the ranking system resulted in a strange result – a strong performance by the fourth place finisher had the result of causing the second and third place finishers to switch places (for details, see pp.96-97). But of course there have been thousands of figure skating events, and many more thousands of sporting events, not to mention elections and other processes involving ranking systems. How often does this sort of breakdown occur? Though Katz tells us that “this problem keeps turning up in the most unexpected places,” the fact that he can give only a single real-world example – [*97] and a fairly trivial one at that – leaves one wondering just how important the Impossibility Theorem is to the vast majority of human events.
Even more telling is the fact that if Katz is right, then even our individual decision processes should be displaying signs of breakdown. Consider Katz’s claim:
The multicriterial version of Arrow’s theorem is even more startling than the original version. After all, it is one thing for a collectivity to behave in a way that we would judge irrational if an individual behaved that way. It is another for an individual to behave that way. And yet it turns out that we are all pretty much doomed to behave this way not because of any special cognitive failings, but because of the logic of the situation precludes anything else. (p.109)
Really? We are all doomed to be perpetually irrational? Beyond being affected by well known cognitive biases, where is the evidence for a pervasive breakdown or “impossibility” in human decision-making? Surely it would show up fairly clearly if it really existed. If not, perhaps the inference goes the other way. Since individuals do not face a serious problem in decision-making, then perhaps collectivities do not either.
Katz of course thinks otherwise, and believes that social choice theory can explain all sorts of puzzling aspects of the law. Let us consider several of these. First, Katz thinks that a “great deal” of loophole exploitation can be explained by Arrow’s results on voting paradoxes (p.107). The vague qualification “a great deal” is troubling: what percentage of loopholes are explained by Arrow: 1%, 10%, 75%? And what about the ones that are not – what are they explained by? If there is a perfectly legitimate explanation of loopholes that does not require Arrow’s Theorem, then the crucial question is one of numbers. If Arrow explains only a tiny percentage, then it does not explain very much at all of the “perversity” of the law.
This issue is all the more crucial given Katz’s troubling normative conclusion that, contrary to common belief, there is nothing wrong with openly and deliberately exploiting loopholes: “Is it all right to exploit this kind of loopholes? Sure, since we are not in any way undercutting the ‘real’ wishes of the collectivity by doing so” (p.104). But this is a gross non sequitur, and would license all kinds of selfish, manipulative behavior. For one thing, it would apply only to “this kind” of loophole, i.e. the kind caused by the Arrow Theorem. But absent a method for identifying which legal loopholes are of this kind, the advice is useless, if not pernicious, since the unscrupulous will simply assume that all loopholes are Arrow loopholes (and we haven’t even established for certain that any are: Katz gives no real legal examples). More importantly, even for Arrow loopholes the moral conclusion does not follow. First, as we noted already, the faulty reductionist assumption is that there is not a “real” wish of the collectivity. Second, it commits the false-dichotomy fallacy: even if a ranking system is imperfect, that doesn’t mean there is no meaning at all to the goals of the community; there can still be actions that are more or less obviously against what the community would like. But Katz’s claim gives free [*98] license to do whatever one can get away with so long as there is the least imperfection in the process. And third, even if there is no real identifiable collective will, it does not follow that all bets are off and it is a free-for-all. Why not say instead that each person has a moral obligation to try to get as close as possible to what is best for the community, as opposed to indulging in his most selfish desires? Recall that Katz applied this argument to statutory interpretation. So if there is no genuine intention derivable from a given statute, does that mean that a judge can interpret it any way he feels like, replacing the statute with his own political views?
Consider next the claim that Arrow’s Theorem explains the “either-or” nature of the law and its tendency to use binary categories such as guilty/not guilty and consent/no consent. Why, Katz asks, is the law so rigid? Why does it not recognize that the world is more like a continuum than either/or? Katz claims that he can prove that the law “could not possibly avoid being either/or” (p.155). He takes the example of death, which might seem like a continuous process: various organs fail at different times, and so on. Katz argues that we have no choice but to treat death as an event not as a process; there is a necessary discontinuity, else we find ourselves in logical absurdity (the detailed argument is presented at pp.158-163). Death, he concludes, “is never partial; death is quintessentially an either/or phenomenon” (p.161). Moreover, the argument (he claims) can be generalized to “many other settings,” such as the beginning of life and consent. Thus using armchair logic alone, Katz claims to have proved that biological events in nature, such as dying or coming into being, must be discontinuous.
But the argument is fallacious; one cannot deduce real facts about the world based on the limitations of one’s methodologies. It is as if, absurdly, Zeno were to use his mathematical paradoxes to prove that the hare could never beat the tortoise in a real race. Nature is continuous (though admittedly, not at the quantum level); the fact that our categories do not perfectly fit does not demonstrate anything about nature, but about the limits of our categories. Moreover, even if Katz were right, it would have limited practical import, for the law could retain either/or categories but make them so fine-grained that they would for all practical purposes function as a continuum. In fact, the law already does that to a large degree; it is far from as binary as Katz suggests. Consider homicide: we could have a strictly either/or view, in which either one is guilty of homicide or not. In fact, the law has moved in the direction of ever more gradations of seriousness, including negligent homicide, reckless, intentional, and premeditated. And that doesn’t even count the indefinitely many gradations possible in sentencing, as well as parole decisions or the severity of the treatment in prison.
Katz’ argument taken as a whole is a combination of extraordinary ambitiousness and at the same time an abundance of caution. The book is full of qualifications and tentativeness: Katz continually repeats phrases such as “seems to,” “might,” “may well,” “can be,” “seems to apply in many other settings,” and “at least in part.” Admirable as such caution is, still these qualifications barely mask Katz’s [*99] excitement that he is on to something really big, as if we are on the verge of discovering some new, unsuspected grand unifying theme that will revolutionize our understanding of the law (as well as politics, human decision making, and possibly the entirety of our culture). For Katz, “There is something going on in all of these cases that judges and lawmakers dimly perceive but cannot put their fingers on” (p.210). This enthusiasm also makes him far too quick to dismiss other explanations of the legal phenomena in question. It seems unlikely that Arrow’s Theorem will provide us with a profound new understanding of the nature of the law. More likely, it teaches us about the failure of the reductionistic approach to human behavior. Katz raises intriguing possibilities worth inquiring into and they may provide some useful insights into some aspects of the law; but in all likelihood, at least in my view, the main lesson of the inquiry will be the need to give up on the vain hope for a morally neutral “science” of law and human behavior.
Copyright 2012 by the author, Whitley Kaufman.