My aim in this Article is to convince you that, like those movie monsters, ARIL is a fantasy. We can reason in law and about law in a variety of ways, but none of those ways is ARIL. Moreover, like those movie monsters, ARIL, were it to exist, would be deformed. It would be deformed by the same process that creates human monsters: bad beginnings.
In discrediting ARIL I mean also to discredit the notion that law is an autonomous discipline with a distinctive form of reasoning. Some who argue for law's autonomy are mystics:(5) they assert that the legally trained can directly grasp relevant similarities and differences among cases without any further reasoning, that is, that legal training produces a special (and presumably superior) way of seeing things. Others are less mystical; they attempt to describe ARIL in terms of familiar forms of reasoning such as deduction, induction, and abduction.(6) Nonetheless, these rationalists, like the mystics, believe ARIL is a distinctive and legitimate form of reasoning. My critique of ARIL extends to both its rationalist and mystical proponents and to the autonomous methodology of law that they seek to advance.
Finally, I attempt to show how the most promising account of ARIL is really Ronald Dworkin's account of law.(7) Dworkin's jurisprudence is an explanation and justification of both ARIL and law's autonomy. Dworkin's account of ARIL is, however, untenable.
In case I be misunderstood, I should make clear at the outset that in attacking ARIL, I am not attacking either normatively or descriptively courts' reliance on precedent. Courts do cite precedents and claim to be bound by them, and my argument discredits neither the sincerity nor the tenability of that claim. I have previously analyzed precedential constraint and concluded that precedent cases can justifiably constrain later ones to the extent the former lay down rules to govern the latter.(8) My argument here is not that courts cannot follow precedents, but that ARIL is not a satisfactory methodology for doing so, either descriptively or normatively.
I. What Is ARIL? The Views from the Literature
We first must get a clear grasp of ARIL's nature. ARIL is frequently invoked, but infrequently described with any rigor or care. We know that ARIL has something to do with comparing possible resolutions of some present legal issue with past resolutions of "similar" legal issues, and that ARIL is operating when we resolve the present legal issue in a particular manner because of the way similar issues were resolved in the past. But how do we determine whether issues and their resolutions are "similar" or "dissimilar"? And does ARIL operate only in common law decisionmaking, or does it operate as well in decisions invoking canonical texts, such as statutes and constitutions? We must answer these and other questions before we can recognize our target.
A. Sunstein's Account of ARIL
Cass Sunstein recently described and praised ARIL in an article in the Harvard Law Review(9) and in a book.(10) According to Sunstein, when ARIL is employed in common law decisionmaking, a court looks at precedent cases, grasps the norm implicit in them, and then applies the norm to the case at hand.(11) The norm need not be the rationale(s) given in the precedent cases.(12) Rather, the court employing ARIL is free to interpret the precedent cases differently from the way in which the precedent courts themselves understood those cases. That in turn means that the new rationale given for those cases by the present court is itself only provisional for, and revisable by, a future court. Cases are precedents only for the results reached (the holdings), not for the rationales given (the dicta).(13) It is the rationale that the present court finds immanent in the past results, not the express rationales given for those results by the deciding courts, that provides the basis for finding cases to be similar or dissimilar.
Sunstein not only describes ARIL but also applauds it. He sees ARIL as a method by which members of a polity who hold quite different moral and political theories can resolve conflict. Members can come to agree that Case A is relevantly similar to Case B but not to Case C, even in the presence of high-level theoretical disagreement, because they can agree on the low-level -- i.e., less abstract and general -- norms that provide the criteria for judgments of similarity and dissimilarity.(14) These low-level rationales are what Sunstein calls "incompletely theorized agreements,"(15) and they allow us to avoid the stalemates that our conflicting views of the Good and Right would otherwise produce.(16) In other words, even if we cannot agree among ourselves about the relative merits of Nozick, Rawls, and Bentham, we can agree that finding a brooch in the window frame of a requisitioned house is more like finding banknotes on a shop floor than like finding gold rings at the bottom of a pool.(17) ARIL is the legal method of choice for the modern multicultural liberal state.
Nor does Sunstein restrict ARIL's domain to common law decisionmaking.(18) For him, ARIL can be legitimately employed in both statutory and constitutional cases. If a statute forbids dogs in restaurants, but expressly excepts guide dogs, how should the statute be applied to the airport cop who brings his drug-sniffing dog into an airport restaurant? Sunstein says this is a case for ARIL.(19) Likewise, if the Constitution forbids punishing political protest but does not forbid punishing draft card-burning, how should the Constitution be applied to laws against flag-burning? Or if the Constitution demands a trial-type hearing prior to revocation of welfare benefits, but not prior to the revocation of disability benefits, what does it demand prior to suspension from public school? Again, for Sunstein, these are cases for ARIL.(20) Indeed, even in easy cases of straight deduction from clear canonical rules, interpreting those rules almost always involves applying ARIL: we see that the green Honda before us is "like" the blue Ford the legislature was picturing when it banned cars in the park.(21) ARIL, and the low-level, incompletely theorized rationales it produces, is truly the most important legal methodology for Sunstein.
B. Scott Brewer's Account of ARIL
Whereas Sunstein's actual account of ARIL is quite spare and conclusory -- we somehow just grasp the rationales and the judgments of sameness and difference they produce -- Scott Brewer, in a recent article, gives us a fuller account of ARIL and places that account within the broader framework of reasoning by analogy generally.(22) Both aspects of Brewer's piece merit close attention.
1. Argument Types, Logical Forms, and Rational Force
Brewer takes ARIL to be a special case of a more general type of argument, which he calls "exemplary" or "analogical" argument.(23) (Brewer prefers the term exemplary to analogical, but nothing hangs on the terminology.) Argument by example -- analogical argument -- requires that we make judgments of relevant similarity and dissimilarity. How we make these judgments, however, is typically left "largely mysterious and unanalyzed."(24)
We are left, then, with explanations of analogy that tend to fall
into either of two roughly divided camps: in one camp are those
who are deeply skeptical about the argumentative force of analogical
argument; in the other camp are those who evince an almost
mystical faith that, even though analogy does not have the rational
force of either induction or deduction, it still has some ineffable
quality that merits our entrusting it with deep and difficult matters
Brewer believes that analogical argument does have rational force and that it is not mystical. In support of that belief, he taxonomizes the various argument types, explains the logical form and rational force of each type, and locates analogical reasoning within that taxonomy.
a. Deductive Argument
One standard type of argument is the deductive or syllogistic argument.(26) In terms of logical form -- the relation between the truth of the premises and the truth of the conclusion -- the truth of the premises of a deductive argument guarantees the truth of the conclusion.(27) If it is true that all men are mortal and that Socrates is a man, then it is true that Socrates is mortal. Thus, deductive arguments have the most rational force of any argument type.
b. Inductive Argument
Brewer describes inductive arguments as being of two types: inductive generalizations from particular instances and inductive analogies from particular instances to other particular instances.(28) He illustrates these argument types as follows:
Imagine a chicken that reasons inductively ... about a sequence
of events: on each of 500 successive days [days are the particular
individuals in the inductive premises], when the chicken hears a bell
(characteristic F), she comes out of her coop and gets fed
(characteristic G). Thus the reconstructed inductive inference on day 501
looks like this:
(1) [x.sub.1] is a hear-bell day and [x.sub.1] is a get-fed day.
(OR: [x.sub.1] is an F (hear-bell day) and [x.sub.1] is a G (get-fed day))
(2) [x.sub.2] is a hear-bell day and [x.sub.2] is a get-fed day.
([x.sub.2] is an F and [x.sub.2] is a G)
(3) [x.sub.3] is a hear-bell day and [x.sub.3] is a get-fed day.
([x.sub.3] is an F and [x.sub.3] is a G)
(500) [x.sub.500] is a hear-bell day and [x.sub.500] is a get-fed day.
([x.sub.500] is an F and [x.sub.500] is a G)
The two typical inductive conclusions to be drawn from these
premises are as follows:
Inductive generalization: Therefore, [probably] all hear-bell
days will be get-fed days. ([Probably] every x that is an F,
is a G.)
Inductive analogy: Therefore, [probably] day 501 [or some
other particular day], which is a hear-bell day, will
[probably] be a get-fed day. (This particular x, which is an F,
[probably] is a G.)(29)
Brewer points out that, in contrast to deductive arguments, "the truth of the premises never guarantees the truth of the conclusion"(30) of an inductive argument. Rather, "the truth of the premises [merely] makes the truth of the conclusion more probable."(31) Therefore, the rational force of inductive arguments is less than the rational force of deductive arguments.
c. Abductive Argument
An abductive argument typically has three steps.(32) At step one, the abductive reasoner notices a phenomenon, P, that calls for explanation. At step two, the abductive reasoner notes that P could be explained by factor H. That is, the reasoner formulates an explanatory hypothesis that could explain P, such as, "if H were the case, then P would be explicable." At step three, the reasoner settles on H as the tentatively correct explanation of P.
Brewer points out that the logical form of abductive reasoning is that of an invalid deductive argument: If H, then P; P, therefore H.(33) Nonetheless, abductive argument is important in the process of scientific discovery, which depends on generating explanatory hypotheses.
Brewer's account of ARIL begins with abductive reasoning. There is a "context of doubt.(34) For example, the court is unsure whether the case before it -- Case A -- is more "like" Cases B, C, and D, which would dictate one result, or is more "like" Cases X, Y, and Z, which would dictate another result. The court provisionally abduces a rule that would sort the existing cases -- an "analogy-warranting rule" or "AWR."(35) It then confirms or disconfirms the AWR by reference to explanatory or justificatory rationales -- "analogy-warranting rationales" or "AWRas."(36)
The court may adjust the AWRs to fit both the AWRas and the cases, adjust the AWRas to fit both the AWRs and the cases, or adjust both the AWRs and the AWRas until the court reaches a justificatory or explanatory equilibrium.(37) Finally, the court applies the AWRs it has thus confirmed to the case before it, Case A.(38)
Perhaps the single most important feature of argument by
analogy is this: in order for an argument by analogy to be compelling
-- to have what I have called rational force -- there must be
sufficient warrant to believe that the presence in an "analogized"
item of some particular characteristic or characteristics allows one
to infer the presence in that item of some particular other
characteristic. It is this sufficient warrant that I have labeled
`analogy-warranting rule.' An analogy-warranting rule states the logical
relation between those characteristics of compared items that are known to
be shared and those that are inferred. Another important
component in a compelling argument by analogy is what I have called the
`analogy-warranting rationale.' Without undue linguistic legislation,
one may distinguish rules from rationales in this way: rationales
stand to rules in the two closely associated relations of explanation
and justification -- that is, rationales explain and justify rules.
Accordingly, AWRs stand in these relations to AWRas. An
analogy-warranting rule states the logical relation that obtains between the
shared characteristics, on the one hand, and the inferred
characteristics, on the other. An analogy-warranting rationale explains why,
in the "eyes of the law" (when the analogical argument is legal), or
for the purposes of the argument (when the analogical argument
is nonlegal), the logical relation among the characteristics
articulated by the analogy-warranting rule either does obtain or should
It is clear that rules -- the AWRs -- play a large role in Brewer's account of ARIL. Without the AWRs, we could not determine the relevant ways in which cases are similar or dissimilar. It is the rule "Drive fifty-five miles per hour" that makes my driving my Toyota sixty-five miles per hour "like" your driving your Acura sixty-five miles per hour and "unlike" her driving her Honda fifty miles per hour. Yet Brewer distinguishes his account of ARIL from accounts like Westen's(40) or Schauer's(41) in which rules do all the work and ARIL is nothing but rule-based reasoning. As Brewer points out: "Rules play a vital role in exemplary argument, but they do not comprise the whole process. Two other critical elements in the process are the discovery (what I have been referring to as `abduction') of proposed analogy-warranting rules, and the confirmation or disconfirmation of those rules."(42)
In other words, for Brewer, ARIL is not reducible to the application of rules. If it were, then in our reasoning we could dispense with the precedent cases -- the examples -- and merely apply the rules to the case at hand. What makes ARIL different from rule application is the abductive search for an AWR in a context of doubt about the rule, and the confirmation of that rule by reference to AWRas (as well as the search for any "disanalogy-warranting" rules and rationales(43)).
3. The Applications of ARIL
Like Sunstein, Brewer believes that ARIL is appropriate not only in common law decisionmaking but also in statutory and constitutional cases.(44) He gives as two examples of ARIL's use in common law decisionmaking the case of MacPherson v. Buick,(45) a case that Edward Levi called attention to as a paradigm of ARIL,(46) and the case of Adams v. New Jersey Steamboat Co.(47) In MacPherson, the question for ARIL was whether a wooden automobile tire was more "like" a horse-drawn carriage, a circular saw, an oil lamp, a steamboiler, and a soldering lamp (no liability without privity), or was more "like" a coffee urn, a bottle of hair wash, a bottle of aerated soda, a bottle of poison labeled as medicine, and a scaffold (liability without privity).(48) In Adams, the question was whether the theft of goods from a passenger's cabin on a steamboat was more "like" the theft of goods from a guest's room at an inn (strict liability for the innkeeper), or was more "like" the theft of goods from a railroad sleeper car (no strict liability).(49)
In statutory cases, the ejusdem generis canon of construction requires resort to ARIL.(50) When general words follow specific words in a statute, the general words are to be given a "`sense analogous to that of the particular words.'"(51) Thus, in McBoyle v. United States,(52) the statute requiring interpretation, which forbade the knowing interstate transport of a stolen "motor vehicle," defined "motor vehicle" to include "an automobile, automobile truck, automobile wagon, motor cycle, or any other self-propelled vehicle not designed for running on rails."(53) The defendant had transported interstate a stolen airplane. The Supreme Court, applying ejusdem generis, and thus ARIL, held that an airplane was not within the scope of the general words "any other self-propelled vehicle not designed for running on rails."(54)
Finally, in constitutional cases, a good example of ARIL's role is determining which classifications are "like" racial classifications for purposes of applying the Equal Protection Clause.(55) Another example is found in California v. Carney,(56) where the Supreme Court had to determine whether, for purposes of the Fourth Amendment's warrant requirement, a motor home was more like a house or a car.
4. ARIL and Reflective Equilibrium
The method of ARIL that Brewer describes and endorses is, according to him, very close to the example-based form of reasoning in normative philosophy known as the method of reflective equilibrium.(57) When employing that form of reasoning, the reasoner might begin with normative judgments about particular examples, or she might begin with some general moral principle that she accepts. She may abduce a principle that would cover her examples, deduce outcomes from that principle (or from the principle with which she began) for other examples covered by the principle, and then test those outcomes against her principle-unaided judgment regarding how those examples should be resolved. If there is an inconsistency between the principle she has abduced and her judgments about some principle-dictated outcomes, she might search for a different principle. Alternatively, if the principle seems attractive for other reasons, she might reconsider her judgments about particular examples, examining them for biases, irrelevant considerations, and other sources of judgmental error.(58) Moreover, the principle abduced must not only be consistent with judgments about particular examples, but also it must be consistent with background metaphysical and empirical theories about human nature and social interaction.(59)
Now it is easy to see why Brewer believes the method of ARIL he endorses is very similar, if not identical, to the method of reflective equilibrium. As he describes the process of holistic adjustment of the abduced AWRs and the background AWRas:
The reasoning device [here] is "reflective adjustment" of the sort
familiar from the work of Goodman and Rawls. Multiple "holistic"
adjustments are possible among an abduced AWR, a proposed
application of that AWR, and AWRas that might explain and justify
it. For example, an abduced AWR might be rejected because,
although it may be an attractive solution in some ways, it does not,
as applied to some particular cases, cohere sufficiently with explanatory
and justificatory rationales that the reasoner is unwilling to
amend. Or the AWR might be so compelling that the reasoner
chooses to hold onto the AWR and effect a modification of the rationales
so that they can indeed provide an explanation of [sic]
justification of that AWR. Or it may be that an abduced AWR, although
adjudged to be adequately explained and justified by AWRas
that the exemplary reasoner takes as her guide to true (or reliable)
judgments, turns out to yield particular results that are, at least
prima facie, unacceptable to the reasoner. Here again, two kinds
of adjustment are possible -- revision of the AWR (and, if necessary,
the AWRa) to accommodate the rejection of this application of the
AWR, or holding fast to the AWR (and AWRa) while revising the
judgment that the application of the AWR in the contemplated
particular cases is, all things considered, unacceptable.
... "An analogy-warranting rule is amended if it yields a conclusion
we are unwilling to accept; a proposed conclusion is rejected
if it violates an analogy-warranting rule we are unwilling to
5. The Rational Force of ARIL
Brewer distinguishes his position on the rational force of ARIL from those of two other camps: the mystics and the skeptics. Sunstein is a mystic,(61) as are Anthony Kronman(62) and Charles Fried. (63) They are mystics because they identify ARIL as involving the intuitive grasp -- of which perhaps only the wise, prudent, and legally trained are capable -- of principles immanent in precedent cases without validating those immanent principles by recourse to plausible higher-level principles from which the former follow deductively.(64) According to Brewer, there is "no better succinct statement of the traditionalist-mystical view of analogy" than that of Fried:(65)
So what is it that lawyers and judges know that philosophers and
economists do not? The answer is simple: the law. They are the
masters of "the artificial Reason of the law." There really is a distinct
and special subject matter for our profession. And there is a
distinct method .... It is the method of analogy and precedent.
Analogy and precedent are the stuff of the law because they are the
only form of reasoning left to the law when general philosophical
structures and deductive reasoning give out, overwhelmed by the
mass of particular details. Analogy is the application of a trained,
disciplined intuition where the manifold of particulars is too
extensive to allow our minds to work on it deductively. This is not
a denial of reason; on the contrary, it is a civilized attempt to stretch
reason as far as it will go.(66)
The skeptics are represented by, among others, Richard Posner,(67) Fred Schauer,(68) and Peter Westen.(69) Posner is a skeptic because, while he agrees with Fried on what method lawyers employ, he denies its rational force.(70) Schauer and Westen are skeptics because they believe that although ARIL may have rational force, it does so only to the extent it is reducible to reasoning deductively from rules.(71)
Brewer describes himself as a "Modest-Proposal Rationalist."(72) He sides with the mystics insofar as his account of ARIL involves an "uncodifiable imaginative moment," the moment of abductive discovery of an AWR.(73) On the other hand, insofar as ARIL includes deduction from AWRs to particular cases, Brewer sides with the skeptics in recognizing the importance of rules and deductive reasoning in ARIL.(74) He concludes that ARIL has rational force that is a blend of the rational force of abduction and the rational force of deduction.(75)
II. What is ARIL? A Fresh Look at the Terrain
A. The Possibilities
I would like the reader to summon up her basic knowledge of common law cases, statutes, and constitutions, stripped to the extent possible of the encrustations of her philosophical theories of legal reasoning. My proposal is that when a case is decided, the decision, if it is to be justified, must be justified in one of the following ways.
First, the decision may be justified by showing that it is morally justified. The best method for establishing moral justifiability may indeed be the method of reflective equilibrium. Thus, a decision in a case may be justified by showing that it follows from moral principles that we (the justifiers) hold in reflective equilibrium. That the decision so follows may require a very complex account, one that not only looks at the facts of the case and the moral principles, but that also considers extant rules, entrenched errors, institutional concerns, epistemic limitations, reliance, and so forth. Nonetheless, argument to moral principles via the method of reflective equilibrium, and from those principles to results in particular cases via the facts of the case and many other facts about the world, is one possible way of justifying a result in a case. And to the extent that the method of reflective equilibrium plays a role here, then, as Brewer correctly maintains, reasoning from examples plays a role.
Second, a decision in a case may be justified by showing that it follows deductively from an authoritative rule that governs the case. The rule may be a constitutional rule, a statutory rule, or an administrative rule; or the rule may be one laid down by a court that has authority to bind others to its rules. The principal task in cases of this kind is not making the deductive inference; rather, it is establishing what the rule means in the context of the case, i.e., interpreting the rule.
I contend that these two types of justification exhaust the field for law. Of course, there are complexities. Reasoning of the first types -- traightforward moral reasoning -- may lead to the conclusion that an authoritative rule, properly interpreted, is immoral or morally suboptimal and should (or perhaps should not) be disregarded. Or, moral reasoning might dictate that a court create a rule (legislate) and apply that rule to the case before it, rather than apply moral principles unaided by more specific (though morally blunter) rules. (The rule that moral principles dictate may even produce a result in the case before the court that is different from the result those principles would dictate if unmediated by rules.)
This story about rules and moral principles is an old and familiar one,(76) but my claim is that it is the only story about legal justification that makes such justification carry any rational warrant. If we lawyers are to avoid the skeptics' charge of mysticism, then moral reasoning and the interpretation of rules must exhaust our possibilities for legal methodologies. Moreover, although both of these methodologies have room for analogical reasoning of the type Brewer endorses, the role for analogical reasoning in these methodologies cannot be what Brewer, Sunstein, and others seek for ARIL.
Most importantly, if I am correct about the limited range of justifiable methodologies in law, then analogical reasoning has absolutely no role to play in common law decisionmaking based on precedent cases, or in common law decisionmaking that treats statutes or constitutional provisions like precedent cases. Yet, such common law decisionmaking is supposed to be ARIL's natural home. If ARIL is not there, then ARIL truly is a phantasm.
Therefore, I shall tentatively add to my list of legal methodologies a third -- the discernment of principles immanent in precedent cases (or in statutes or constitutional provisions functioning like precedent cases). In the next two Sections, I shall show why the role of examples in moral reasoning and in the interpretation of canonical rules cannot satisfy those who advocate ARIL. Then in Part III, I shall turn to the third proposed methodology, the discernment of immanent principles, and show that while this is the natural home of ARIL, it fails to reckon with a problem that afflicts many monsters, including ARIL: the problem of bad beginnings.
B. ARIL and Ambiguous Beginnings
1. ARIL and Common Law Decisionmaking
I want to assume the most favorable circumstances for ARIL and that all the legal materials that the ARIL-employing lawyer or judge must work with might be morally justified. That is, I want to assume that the cases all might have reached morally correct decisions, even if the stated rationales for those decisions are not morally correct.
The reason for severing the decisions from their stated rationales is important. As Brewer recognizes, ARIL depends on the distinction between decisions and their rationales;(77) if the rationales were controlling, the reasoner could deduce the result in the present case from the rationale laid down in the precedent case and would not have to abduce AWRs and AWRas. In other words, when precedent courts' rules rather than results constrain, only deductive reasoning is required: the very point of Westen's and Schauer's skeptical critiques of ARIL.(79)
(Using as precedent statutory and constitutional rules, whose interpretations are undisputed, to decide cases outside the scope of those rules needs a slightly different account. Obviously, the statutory or constitutional rule can be viewed as nothing more than the sum of the possible cases that the rule governs. In that sense, there is no distinction between the rule and the decisions in particular cases that the rule produces. The structural analog, then, to severing court decisions from their stated rationales is to sever the statutory or constitutional rule from its background rationale(s) that otherwise could -- by deduction -- do all the work in deciding cases beyond the statutory or constitutional rule's scope.)
So, to recap briefly, we are to assume that all decisions to date might be morally justified even if their stated rationales are not, and that all statutory and constitutional rules might be morally justified even if their actual background rationales are not. Further, we are to assume that the case before us is not to be decided based on the stated rationales of the cases or based on the actual rationales behind any rules, and that the case does not come within the scope of the (by hypothesis) possibly morally justified statutory or constitutional rules. The question now is: can ARIL play a role in deciding the case at hand?
My answer is that it cannot. In the scenario I have presented, there is no reason not to use moral reasoning to discover what the morally justified result is in the case at hand. But ARIL is not moral reasoning.
Why is ARIL not just plain ordinary moral reasoning applied to deciding a legal case? If the method of reflective equilibrium is the methodology for discovering moral truths, then, as was pointed out in Part I.B.4 above, in employing it we will make use of judgments about particular cases. Is not reflective equilibrium reasoning from examples, and is it not therefore a candidate for ARIL?
Reflective equilibrium is indeed a form of reasoning from examples, but it cannot be ARIL. Consider these distinctions between reflective equilibrium and ARIL.
First, the reasoner employing reflective equilibrium ultimately tests the moral principles she is considering against her own moral judgments about particular examples. Yet ARIL gives prominence to others' judgments about particular examples, namely, the judgments of the precedent courts and the legislators. To be sure, the court deciding the instant case must itself decide what the moral principles are to which those judgments of others point. But it must accept those moral judgments of others about particular examples as fixed points in its own reasoning.
I do not mean to imply that the moral judgments of others have no role in reflective equilibrium. Quite the contrary is true. Nonetheless, those judgments of others are evidentiary for the reasoner; they bear on the confidence she has in her own particular judgments. The judgments of others are never canonical for her. Moreover, she takes good evidence where she finds it. For her, the particular moral judgments of the Harvard Philosophy Department might carry significantly more weight than the judgments of the Pennsylvania Court of Common Pleas or the Pennsylvania legislature. Yet ARIL subordinates the former to the latter. ARIL precludes her from deciding that the particular judgments of the courts or the legislature are incorrect, even if the inconsistent judgments of the Harvard Philosophy Department, as incorporated into her own moral judgments, would so declare. ARIL gives the particular judgments of others -- indeed, the particular judgments of particular others, namely, courts and legislatures -- a different status from the status they would have under reflective equilibrium.
A second difference between ARIL and reflective equilibrium lies in the rigidity or fixity of the particular judgments with which the reasoner must deal. If the reasoner were employing the method of reflective equilibrium to determine the morally best decision in the case before her, she might adjust not only her abduced moral principles to fit with her particular judgments, but also her particular judgments to fit with her moral principles. Whether she would adjust her principles or judgments would depend on the relative degree of confidence she had in each. As Brewer correctly notes, the adjustments that occur in the method of reflective equilibrium are holistic and at least potentially bidirectional.(79)
ARIL, however, does not permit adjustment of the particular judgments that serve as precedents. The particular judgments about carriages, saws, urns, lamps, and so forth that the MacPherson court had to deal with were fixed for it. Those judgments could not be revised to fit the court's moral principles. Nor could they be disregarded. The same holds true, of course, for the particular judgments of legislatures embodied in statutes that a court wishes to treat like precedent in reasoning, per ARIL, to a result in a case not covered by the statute. This fixity or rigidity of the particular judgments of others that ARIL must work with distinguishes ARIL quite clearly from reflective equilibrium.
The third difference between ARIL and reflective equilibrium lies in the rationale-filtered manner in which a court has access to precedent cases. Precedent cases do not present themselves in all their particularity to the court employing ARIL. Rather, the particularity with which a present court can perceive a precedent case is entirely dependent on the precedent court's opinion and the detail about the case contained therein.(80) The amount and the selection of that detail in the opinion will be a function of the precedent court's rationale for deciding the case. The precedent court may have favored a standard under which a large number of adjudicative facts in the case were deemed material and worth mentioning in the opinion. On the other hand, the precedent court may have favored a broad, blunt rule that rendered most adjudicative facts in the case entirely immaterial.(81)
Now as I have said, ARIL dispenses with the rationales of precedent cases. The reason again is that ARIL is supposed to be something other than deductive reasoning from rationales already established. Nonetheless, ARIL cannot capture past cases in their full particularity, but rather must confront them as stylized, rationale-encrusted judgments of other courts.
Reflective equilibrium, on the other hand, allows and encourages the moral reasoner to add facts to and to subtract facts from imaginary situations in order to discern how her moral judgments are affected and to identify the governing moral principles. The cases that the moral reasoner considers can be as detailed and as customized as she wishes because they are products of her moral imagination, not actual past events whose description has been forever fixed by others.
These three ways in which ARIL differs from the method of reflective equilibrium are, of course, related. The fact that ARIL depends on judgments of others explains why those judgments are fixed and why they are rationale-encrusted. The bottom line is that ARIL cannot be the method of reflective equilibrium, even though the latter, like the former, relies on examples.
I do not wish to be misunderstood to be saying that a court cannot take account of precedent cases and statutes while employing reflective equilibrium to reach the morally best decision in the case before it. If reflective equilibrium is the proper method for determining the morally correct course of action to take, then reflective equilibrium will have to take account of facts about the world, including past court decisions and their stated rationales and existing statutes. These facts matter morally, but they do not matter in the way ARIL describes. When we reason morally, we take account of those facts.(82) When we employ ARIL, we are supposed to be anchored to them.
2. ARIL and the Interpretation of Legal Rules
Brewer asserts that ARIL is used in statutory interpretation, as in the ejusdem generis cases like McBoyle.(83) I agree that a method similar to ARIL can be employed to interpret statutes and other canonical legal texts -- just as a method similar to ARIL (reflective equilibrium) can be employed in common law decisionmaking -- but ARIL itself cannot be so employed.
There are two basic approaches to interpreting canonical legal texts. One is the approach of moral reasoning, which seeks to make "the statute" (or "the Constitution" or "the rule in MacPherson") morally justified. If reflective equilibrium is the proper methodology for finding moral "truths," then reflective equilibrium -- taking into account the language of the rules, common understanding of that language, the effects of various institutional arrangements, and other such factors -- is the correct moral approach to "interpreting" canonical legal texts.
Although I would not call this moral approach to legal texts "interpretation," and would instead refer to it as "doing what is morally best in light of the legal texts," this is not the appropriate forum in which to enter that debate.(84) Rather than reject the moral approach to interpreting legal texts, I wish here merely to set it aside. I do so because this approach to interpretation cannot be ARIL for reasons articulated in the previous Section. I said there that if common law decisiomaking were merely discovering the morally correct course of action, then the method of reflective equilibrium -- which is not ARIL -- is what is required. This point is equally applicable if interpretation of legal texts is discovering the morally correct course of action in light of those texts.
I thus put the moral approach to interpreting canonical legal texts aside and turn to the other approach, which I shall call the empirical approach. Crudely put, the empirical approach to interpretation seeks to discover some nonnormative fact in the world that holds the key to the legal text's authoritative meaning. The most important of such empirical approaches are: 1) the approach that locates a text's authoritative meaning in the intention of its author(s), and 2) the approach that locates the meaning in the understanding of its audience.
The relationship between the second and the first is complex. Whatever their differences, however, they are similar with respect to ARIL. I shall focus exclusively on the first approach (author's intent); but my analysis of it will likewise be applicable to the second approach (audience understanding) insofar as ARIL is concerned.
If we locate the authoritative meaning of a legal text in a fact about the world, such as the intent of the text's author, then interpretation of the text is an empirical matter, and our method is, broadly speaking, scientific. For example, if we want to know what Congress might have intended with respect to airplanes in a case such as McBoyle, we look at the specific examples of motor vehicles in the statutory text, abduce a hypothesis about Congress's intent, and test that hypothesis by reference to possible rationales that Congress might plausibly have found justificatory of that intent. Note the parallels with ARIL: the particular examples, the abduction of an AWR and the testing of the AWR by reference to an AWRa. Note also that the AWRa here is explanatory, not justificatory. We seek to explain what Congress has done, not justify it. Our only interest in the justificatory power of the AWRa is as it bears on the existential plausibility of the AWR -- here, Congress's intent -- that it supports. The fact that one version of Congress's intent is supported by a stronger moral justification than another may strengthen -- though it surely does not guarantee -- the inductive inference that Congress did in fact possess the more justified intent.
Apparent similarities notwithstanding, this empirical approach to legal interpretation is not ARIL, even in an explanatory rather than a justificatory mode. ARIL gives the examples from which it operates a significance that no examples enjoy if we are merely reasoning abductively and inductively to some fact in the world, such as legislative intent. If we want to know Congress's intent in McBoyle, then we treat the statute's various examples of motor vehicles as evidence of that intent. Furthermore, we may look beyond the statute for other evidence of legislative intent, such as legislative history. Remember, we are seeking to discover some fact about the world, and anything that affects the probabilities surrounding that fact is relevant. The words of the statute, although important evidence of the intent we seek, are, for all that, merely evidence.
Under ARIL, however, the words of the statute are the evidentiary universe. They are authoritative for interpreters in a manner unparalleled by evidence in the purely empirical world, where all evidence can be considered and none is incorrigible. To put it differently and more bluntly, if interpretation of a legal text is the search for a fact like legislative intent, and if that search is ARIL, then ARIL is nothing special: it is merely the ordinary scientific method of abduction and induction applied to the search for past facts of legal significance.(85)
C. Tracking ARIL to Its Lair: Immanent Legal Principles
Thus far, I have attempted to show that ARIL cannot be ordinary moral reasoning that takes into account past legal decisions, and that ARIL cannot be ordinary empirical inquiry into legislative intent (or public understanding) that takes into account examples in legal texts. If we look for ARIL in those precincts, we shall not find it.
There is an alternative location where I believe ARIL can be found: ARIL is the method of divining legal principles immanent in the case decisions and in the decisions of legislative bodies. ARIL is not the method for understanding canonical legal texts, as I have explained. Nor is it the method for applying canonical texts, which is merely deductive. Rather, ARIL functions when a case is not controlled by a canonical text, as when no extant text covers the case, or when the text that covers it is not canonical and can be disregarded, such as some accounts of judicial texts. What ARIL directs a court to do in such situations is to find the legal principles immanent in the legal materials and apply those principles to the case at hand.
Now, what are immanent legal principles? I am going to give the best account of them I can, but because I am quite skeptical about the normative status of immanent legal principles, my account should be viewed warily.
Immanent legal principles are legal norms that are not posited by any legal decisionmaker. As I said, if they were posited norms, they would be subjects not for ARIL but for interpretation -- empirical inquiry -- and deduction. Rather than being posited norms, immanent legal principles arise from the posited legal materials, and justify them. For example, the court in MacPherson could be described as having found in the array of past cases -- though not in their stated rationales -- the legal principle that it announced regarding negligence and privity, which principle justified the decisions in both MacPherson and in the precedent cases.
Just as immanent legal principles are not posited norms, neither are they moral norms. Although they "justify" past legal decisions, they do not do so morally, at least not straightforwardly. If they were moral norms, then, as I have argued, they would be discovered through the method of reflective equilibrium, not through ARIL.
ARIL as the method for discovering immanent legal principles squares rather well with Brewer's account of ARIL. A court faced with an array of past decisions, shorn of their nonauthoritative rationales, abduces a principle (the AWR) that would fit with those past decisions. It tests that principle for plausibility against background moral considerations (the AWRas). If the principle both fits the decisions and is morally attractive, the principle is the principle immanent in those decisions and is the gauge of relevant "likeness" and "unlikeness" in analogical reasoning from the cases.
I think that this story about immanent legal principles is the best account that can be given of ARIL. Notice that on this account ARIL appears to be, as many of its advocates contend, not some mystical, intuitive grasping of "likeness" and "unlikeness," but a form of practical reasoning. Notice also that on this account ARIL explains how past cases can constrain future decisions -- as the principle of stare decisis requires -- even though the past cases are shorn of their rationales. Notice finally that on this account ARIL truly is a distinctive methodology of practical reasoning, one that can support the autonomy of law.
Even as I have presented this account of ARIL so far, it has problems. For example, because ARIL is supposed to operate on judicial decisions and not the rationales for those decisions -- else it would be merely interpretation and deduction -- we need some way to overcome the problem to which I earlier adverted: judicial decisions are accessible to later courts only through their rationales. Nevertheless, I wish to assume that ARIL can overcome this problem. A far greater problem confronts it -- past mistakes.
III. BAD BEGINNINGS
A. The Possibility of Moral Error in Law
Judges may decide cases in ways that are morally unjustifiable. Legislatures may legislate in ways that are morally unjustifiable. It is difficult to imagine that anyone would gainsay these statements. For how would one explain judges' and legislators' moral infallibility? Even when courts are unanimous, we believe they may be morally mistaken. And courts and legislatures are, quite frequently, seriously divided over the justice of their decisions, sometimes with one vote making the difference between one moral view and another.
Further, if we were to assume that courts -- even divided courts -- always succeeded in doing justice in the particular case, we would have to explain why they were not similarly infallible in stating rationales. For if a court, employing ARIL, disregards past rationales but treats past decisions as morally justified, it must assume that moral infallibility in deciding cases does not translate into the moral infallibility of the stated rationales. How could we account for this difference in courts' abilities?
Moreover, to assume courts' moral infallibility in deciding cases, we must go beyond attributing to them moral infallibility with respect to the cases before them standing in isolation. Because each court is employing ARIL and looking at past cases as well as the case before it, its decisions about the array of cases must also be morally infallible, even though its view of past cases is filtered through the prism of those cases' fallible rationales.
The moral infallibility account -- no past moral mistakes embedded in the law -- not only plunges ARIL into mysticism, but also renders ARIL unnecessary. If courts have infallible "case sense," then ARIL -- which purports to be a method of reasoning, which is always fallible -- is quite unnecessary and counterproductive. Why employ a fallible reasoning method when one can infallibly "grasp" the correct decision without it?
In light of these considerations, it seems quite clear that we must assume ARIL is to operate in a world in which some past judicial decisions, statutes, and so forth, are morally mistaken. We now face an apparent paradox: ARIL is the method for finding the legal principles that "justify" the morally unjustified.
B. Moral Error and Other Methodologies
The problem of moral error embedded in past judicial decisions, statutes, and other legal materials does not affect the other methodologies I have described: ordinary moral reasoning to determine the morally best decision in a case (the method of reflective equilibrium), and ordinary empirical methodology to discover the meaning of a legal rule qua legislative intent (or public understanding). Ordinary moral reasoning will take account of all the present effects of past decisions, including morally erroneous ones. Because we operate morally in the world as we find it -- a world in which many of its features are the residue of past moral mistakes -- we must take account of those mistakes in order to act morally. For example, we surely must take into account present reliance on past decisions, even if those decisions were morally mistaken. And we must take into account present institutional features when we reason morally about what to do, even if those institutional features are not morally optimal.(86)
Although moral reasoning must take account of past moral error in the ways just described, it need not abandon correct moral principles in doing so. In other words, it need not resort to any norms other than the correct moral norms it discovers through proper moral reasoning. Those norms dictate what to do in light of past moral mistakes. Thus, past moral mistakes do not alter the norms themselves, or the methodology for discovering them.
The same relation or lack of relation holds between past moral error and the discovery of legislative intent. To the extent we can explain past moral mistakes, we can be more confident that we have correctly ascertained what some legislative body intended by its text if our tentative conclusion is that the intent embraces the moral error. Our method is always abductive and inductive in the ordinary manner of empirical inquiry. We are still looking for the legally significant fact -- intent or understanding -- and moral mistakes are just part of the evidentiary stew. We do not seek principles to justify them, but rather facts to explain them. And, of course, once we have interpreted the rule, deduction is the only reasoning we need, for deduction operates just as well from morally mistaken rules as from morally correct ones.
To summarize, past moral mistakes can be handled by ordinary empirical and normative methods. They are, of course, problems to those adversely affected by them. They are not, however, problems for the methodologies.
C. Dworkin's Domain
1. The Problem of Bad Beginnings
We have concluded that the most plausible account of ARIL is one in which ARIL is neither ordinary moral reasoning nor ordinary empirical reasoning. Rather, it is one in which ARIL seeks legal principles immanent in past legal decisions, that is, principles that justify those decisions. Those principles are what give us the criteria of likeness and unlikeness required for analogical reasoning.
Our path has also shown us that, almost surely, many of the legal materials in which the justifying legal principles are supposedly immanent will turn out to be morally mistaken. The situation of a judge or lawyer seeking the legal principles immanent in the legal materials is therefore similar to that of a zoologist asked to continue a project of sorting animals into "fish" and "mammals." The previous zoologist who had begun the project had classified the animals she had studied in the following manner:
The new zoologist now must classify a seal, and he must accept his predecessor's classifications as authoritative. The question for him thus becomes whether -- given the predecessor's classifications -- a seal is more "like" a porpoise (fish) or more "like" a bat (mammal), or, put differently, how the principles of mammaldom and fishdom "immanent" in and "justificatory" of his predecessor's classifications work in the case of seals.
It should be obvious that if the zoologist must treat his predecessor's list as authoritative, he will find it impossible to classify seals nonarbitrarily. He will, of course, be able to point to ways in which seals are like and unlike both porpoises and bats, as he will when he moves on to sea otters, river otters, beavers, raccoons, and bears. (Note how the direction of progression through these examples will affect his answers, which should not, but likely will be, path-dependent.) All things are like and unlike all other things in myriad ways. The zoologist needs the correct principles for classifying fish and mammals, for only those principles will establish the relevant bases for judgments of likeness and unlikeness. Yet the predecessor's bad beginning has made that impossible. Does not the same hold true for a judge or lawyer employing ARIL in the real-world context of past legal error?
2. Dworkin the Magician
ARIL requires that we justify the unjustified. We obviously need powerful magic to accomplish that feat. Fortunately, the law has such a powerful magician in its service -- Ronald Dworkin. Dworkin has claimed to be able to "justify" mistakes and make the law the best it can be.(87) Surely, Dworkin's domain is ARIL's natural home. Indeed, ARIL is the methodology that has secured Dworkin his empire in the law.(88) Although Dworkin did not create ARIL, he took ARIL and harnessed it to his cause. Dworkin has, of course, put forward a jurisprudential theory in which legal principles immanent in the legal materials play the major role.(89) Those principles must "fit" the legal materials (shorn of their rationales) -- as the AWRs must fit with the past cases -- and must be the most morally acceptable of the principles that satisfy the criterion of "fit," with moral acceptability gauged by reference to correct moral principles (the AWRas).
Because of moral mistakes embedded in the legal materials, however, the legal principles immanent in the legal materials -- those that score highest on the fit and acceptability axes -- will differ from correct moral principles. (In Brewer's terms, the legal AWRs will not cohere completely with the moral AWRas.) Legal principles for Dworkin can be characterized counterfactually as those principles that would be correct moral principles in a world in which the morally incorrect past decisions were morally correct.(90) To illustrate the claim graphically, for a judge confronting, say, Plessy v. Ferguson(91) and similar decisions, the legal principles immanent in those decisions are those principles that would be morally correct in a world in which "separate but equal" were morally correct.
I have argued in several other works that Dworkin's account of legal principles, even if coherent -- compare it with "What would seals be in a world in which porpoises were fish?" -- renders legal principles normatively unattractive and ontologically queer.(92) Legal principles do not have the virtues of bright-line rules, which, even if not morally optimal, provide clear guidance. Nor do legal principles have the virtue of correct moral principles, since they are not necessarily morally correct. Moreover, the best answer to the counterfactual question that describes legal principle -- what would be morally correct in a world in which moral errors were not errors -- is "act on correct moral principles except in past cases of moral error."(93) That injunction is no different in any practical sense from an injunction to follow correct moral principles, which requires the method o reflective equilibrium, not ARIL.
Correct moral principles will never dictate their own abandonment. They might dictate that we follow bright-line rules, even at some moral cost in particular cases, in return for greater moral benefits generally (moral costs and benefits calculated by reference to the correct moral principles). Correct moral principles, however, will reject all other purported justificatory principles as counterfeit.
Finally, cannot the value of equality support ARIL? In other words, even if some past legal decisions are mistaken, does not the moral principle that requires us to treat past and present litigants "equally" thereby require us to abandon correct moral principles in favor of principles that "fit" the past cases, including the mistaken ones@ Is not Dworkin's method, which is the best account of ARIL, really what is morally required of us after all, rather than a perverse departure from morality?
The argument from equality is Dworkin's strongest argument, but it is fatally defective. To begin with, all conceptions of equality are theory-dependent.(94) That is, equality is not a free-floating moral value, but instead must be cashed out in terms of a general normative theory. There are Rawlsian, utilitarian, libertarian, and other conceptions of equality. One does not abandon correct moral principles to honor the demands of equality. Rather, one must refer to correct moral principles to know what equality demands.
Even if that is so, could not the correct conception of equality require us to acknowledge that past mistaken decisions exert gravitational force on present litigants? In other words,could not the correct conception of equality trump all the other moral principles hat would otherwise dictate renouncing past mistakes rather than perpetuating their effects?
Again, the answer is no, and for two reasons. First, if in the past there have been both morally incorrect decisions and morally correct ones, as would be expected, equality of past and present litigants points in opposite directions. One cannot treat present litigants "equally" with those in the past mistaken cases without also treating them unequally,, with those in the past correct cases, and vice versa.
Second, and more importantly, past moral mistakes just do not exert any equality claim on behalf of present similar decisions.(95) If it were otherwise, then putting one innocent person to death would create some moral reason to put a second to death. And if enough innocent persons were put to death, then the balance would be tipped in favor of "equality," that is, in favor of putting still another to death. But to argue this way would be absurd.
For these reasons, equality does not dictate that we abandon correct moral principles in favor of legal principles immanent in the cases. And without such legal principles, ARIL as a form of reasoning collapses.
When we employ ARIL and ask which past cases are like and unlike our own case, we need some norm that we can reference in order to answer that question. If we are searching for a posited rule, then we must look at court-stated rationales or legislative enactments. In such a case, our reasoning will be deductive (from the rule to the facts of our case), except when we need empirical reasoning (for example, to determine legislative intent) to interpret an ambiguous rule. If, on the other hand, we are searching for a norm that justifies the past cases, then, if correct moral principles do not do so, nothing else will. Correct moral principles may be discovered through exemplary reasoning, but not, as in ARIL, exemplary seasoning based on the cases. Past mistakes -- bad beginnings -- make ARIL impossible or perverse.
Although ARIL is perverse -- a monster -- it is also a phantasm. It does not really exist, though the belief that it does can lead to methodological error.(96)
On the subject of ARIL, we lawyers are like the man who, when asked if he believed in adult baptism, replied, "Believe in it? Hell, I've seen it done." We lawyers believe we have seen ARIL. In truth, however, while we have seen moral reasoning, empirical reasoning, deductive reasoning, and confused reasoning, we have not seen the phantasm that makes lawyers dream of an autonomous law's empire. The autonomy of law and law's methodology can only be a dream. And when ARIL is the agent of law's autonomy, the dream is a nightmare. ARIL, like Freddy Krueger,(97) does not exist, and for that we should be thankful.
(1) See Peter Westen, on "Confusing Ideas": Reply, 91 Yale L.J. 1153 (1982).
(2) See Frederick Schauer, Playing by the Rules (1991).
(3) See Cass R. Sunstein, On Analogical Reasoning, 106 Harv. L. Rev. 741 (1993).
(4) See Scott Brewer, Exemplary Reasoning: Semantics, Pragmatics, and the Rational Force of Legal Argument by Analogy, 109 Harv. L. Rev. 923 (1996).
(5) The term is Brewer's. See id. at 951 (describing mystics as those "who place a high degree of confidence in analogical argument even though they neither have nor feel the need for an explanation of its characteristic concepts of 'relevance' and `similarity'").
(6) See id. at 926 (noting the "methodological reciprocity" between analogical reasoning and the forms of argument associated with natural and demonstrative sciences).
(7) See Ronald Dworkin, Law's Empire (1986).
(8) See Larry Alexander, Constrained by Precedent, 63 S. Cal. L. Rev. 1 (1989).
(9) See Sunstein, supra note 3, at 742 ("[A]t least for law, [analogical reasoning] has distinctive advantages over forms of thought that seem to be far superior.").
(10) See Cass R. Sunstein, Legal Reasoning and Political Conflict (1996).
(11) See id. at 77-79.
(12) See id. at 32.
(13) See id. at 32, 71-72, 203.
(14) See id. at 3-9, 62, 64.
(15) Id. at 14.
(16) See id. at 3.
(17) See Hannah v. Peel, 1 K.B. 509, 520-21 (1945) (holding that a brooch found in a crevice on top of a window frame belongs to the finder, not to the never-resident owner of the house); South Staffordshire Water Co. v. Sharman, 2 Q.B. 44, 47 (1896) (holding that two gold rings found at the bottom of a pool belong to the landowner, not the finder of the rings); Bridges v. Hawkesworth, 21 L.J.Q.B. 75, 75-78 (1851) (holding that banknotes found on a shopkeeper's floor belong to the finder, not the shopkeeper).
(18) See Sunstein, supra note 10, at 79-90.
(19) See id. at 124-25; see also id. at 183.
(20) See id. at 80-81.
(21) See id. at 83-90, 203 n.26.
(22) See Brewer, supra note 4, at 925-42.
(23) Id. at 929.
(24) Id. at 933.
(26) See id. at 943.
(27) See id. at 942-43.
(28) See id. at 944.
(29) Id. at 944-45 (all but first alteration in original) (footnotes omitted).
(30) Id. at 945.
(32) See id. at 947-48 (describing the three steps of an abductive argument).
(33) See id.
(34) Id. at 962.
(37) See id. at 963.
(38) See id.
(39) Id. at 965 (footnote omitted).
(40) See Westen, supra note 1, at 1162-63 (arguing that legal rules of decision are necessarily formulated before, not after, any evaluation of a particular case's similarity to other cases).
(41) See Schauer, supra note 2, at 183-87 (explaining how one can derive a rule from either a precedent case's canonical description of the facts, its canonical rationale, or a "seemingly natural category" into which its facts fall).
(42) Brewer, supra note 4, at 975.
(43) The term is Brewer's. See id. at 1008. (44) See id. at 936 ("[T]his type of exemplary, analogical reasoning ... is obviously also thoroughly familiar ... in statutory and constitutional cases.").
(45) 111 N.E. 1050 (N.Y. 1916) (applying liability without privity to a car manufacturer and reasoning that a wooden automobile tire is similar to poisons, explosives, etc., because all are reasonably certain to place life and limb in danger if they are negligently made).
(46) See Edward H. Levi, An Introduction to Legal Reasoning 6-19 (1949) (describing MacPherson as representing the third phase of the cyclical movement in legal concepts during which a concept is built up, fixed for a period of time, and then broken down).
(47) 45 N.E. 369 (N.Y. 1896) (holding that the operator of a steamship is strictly liable to passengers for theft of their personal property in the same way that a hotel is strictly liable to its guests).
(48) See MacPherson, 111 N.E. at 1052.
(49) See Adams, 45 N.E. at 371.
(50) See Brewer, supra note 4, at 937.
(51) Id. (quoting Osborn v. Wilson & Co., 193 N.Y.S. 241, 242 N.Y. Sup. Ct. 1922)).
(52) 283 U.S. 25, 27 (1931) ("It is impossible to read words that so carefully enumerate the different forms of motor vehicles and have no reference of any kind to aircraft, as including airplanes under a term that usage more and more precisely confines to a different class.").
(53) Id. at 26.
(54) Id. at 26-27.
(55) See Brewer, supra note 4, at 936-37.
(56) 471 U.S. 386 (1984) (holding that a warrantless search of a motor home does not violate the Fourth Amendment because, although the motor home possesses some attributes of a home, it is substantially similar to a vehicle and thus falls within the vehicle exception).
(57) See Brewer, supra note 4, at 938-39.
(58) See John Rawls, A Theory of Justice 46-53 (1971).
(59) See Norman Daniels, Wide Reflective Equilibrium and Theory Acceptance in Ethics, 76 J. Phil. 256, 25&-61 (1979) (discussing the role of background theories in the method of reflective equilibrium); Howard Klepper, Justification and Methodology in Practical Ethics, 26 Metaphilosophy 201, 205-06 (1995). Klepper describes the holistic aspect of reflective equilibrium as follows: [Using the methodology of reflective equilibrium] we move back and forth through the hierarchy of theory, principles, rules, and judgments, with warrant to make changes at any level in order to produce coherence at and among all levels. Where we choose to make a change when we discover some inconsistency will be determined by pragmatic considerations such as conservatism, simplicity, fecundity, and scope, in a manner analogous to the contemporary Kuhnian view of theory acceptance in science.... A particular judgment gains credibility from its coherence with a rule, which in turn gains credibility from its coherence with both particular judgments and more general principles, and so on. .... In reflective equilibrium theory it is possible that a recalcitrant judgment could force a change in our deepest ethical theory, or our ethical theories force a change in the most strongly held particular judgment. In practice, change is likely to be absorbed at the intermediary level of rules, or, especially in the case of those judgments not so strongly held, at the level of particular judgment. This need not imply that we feel greater certitude for theory than we do for rules or particular judgments. It can be explained by the pragmatic preference for conservatism that we display throughout. Changing ethical theory or principles will likely necessitate greater change at all other levels in order to maintain coherence than will change in particular judgments, just because the theoretical claim is much more general. The most popular technique of adjustment looks to the level of rules, where we accommodate a recalcitrant judgment by creating an exception to or further specification of the rule -- one that coheres with both that judgment and our accepted principles. Id.
(60) Brewer, supra note 4, at 1023 (footnotes omitted) (quoting Nelson Goodman, Fact, Fiction, and Forecast 64 (4th ed. 1983)).
(61) See id. at 952 ("The Wittgensteinian analogists are joined by Young Turk neo-Aristotelian analogical mystics, whose leading figure among legal theorists is Cass Sunstein.").
(62) See Anthony T. Kronman, The Lost Lawyer 170-85 (1993); see also Brewer, supra note 4, at 952.
(63) See Charles Fried, The Artificial Reason of the Law or: What Lawyers Know, 60 Tex. L. Rev. 35, 57 (1981) ("Analogy and precedent are the stuff of the law.... [and] [a]nalogy is the application of a trained, disciplined intuition...."); see also Brewer, supra note 4, at 952.
(64) See Brewer, supra note 4, at 952-53.
(65) Id. at 952.
(66) Fried, supra note 63, at 57 (footnote omitted); see also Brewer, supra note 4, at 952 (quoting the same language).
(67) See Richard A. Posner, The Problems of Jurisprudence 86-87 (1990) ("[R]easoning by analogy... has an impeccable Aristotelian pedigree, but no definite content or integrity." (footnote omitted)).
(68) See Schauer, supra note 2, at 183-87 ("What distinguishes reasoning from precedent from reasoning from rule, however, is the necessity in precedential reasoning of constructing the generalization/factual predicate that already exists in the case of a rule.").
(69) See Westen, supra note 1, at 1163-64 (arguing that once one identifies the prevailing legal rule, one can analogize and distinguish among cases using logic).
(70) See Posner, supra note 67.
(71) See supra notes 1-2.
(72) Brewer, supra note 4, at 954.
(74) See Id.
(75) See Id. at 954-55.
(76) See generally Larry Alexander, The Constitution as Law, 6 Const. Commentary 103 (1989); Alexander, supra note 8; Larry Alexander, The Gap, 14 Harv. J.L. & Pub. Pol'y 695 (1991); Larry Alexander, Law and Exclusionary Reasons, 18 Phil. Topics 5 (1990); Larry Alexander, Pursuing the Good -- indirectly, 95 Ethics 315 (1985); Larry Alexander & Emily Sherwin, The Deceptive Nature of Rules, 142 U. Pa. L. Rev. 1191 (1994).
It is obvious that I am rejecting such "rational" methodologies as "decide the case in a way that maximizes the judges' (or the ruling class's) welfare." I am assuming that law is at bottom a moral enterprise, one that has a moral point. A rational justification in law is one that is rational within such a framework, not one that is rational within some competing framework, such as that of advancing the judges' self-interest. In any event, ARIL is an implausible candidate for a rationally justified decisionmaking procedure within these competing frameworks.
On the other hand, nothing in my account of the relation between moral principles and posited authoritative rules precludes the rules from being quite broad, as for example: "In deciding all cases, courts should try to maximize economic efficiency and should ignore competing moral considerations." Such a rule might even be a morally optimal rule for judges to follow, though I strongly doubt it. In any event, as long as these broad rules function as formal rules and not as "principles," see infra Part III.C.2., they are not problematic for my account of rationally justified legal methodologies.
In the end, law and its methodologies are a reflection of one central dilemma: how to determine what is morally justified in particular situations. Whether because people disagree about ultimate moral principles, or because they agree about ultimate principles but not about what those principles require in concrete cases, moral principles are not best implemented through each individual's deciding for herself what those principles require case by case. Moral principles, rather, are best implemented by rules -- authoritative determinations of what ought to be done (substantive) and who ought to decide (procedural). Law is what we call those authoritative determinations or rules. But since all such rules are both fallible and also, even if ideal as rules, over- and underinclusive relative to the moral principles they are designed to implement, the moral principles and the rules are always in actual or potential conflict. In cases of conflict, both citizens and legal officials must choose between following the moral principles and following the rules. And even if the moral principles require taking into account the morally bad consequences of rejecting rules whenever they conflict with moral principles -- that is, undermining the rules' ruleness -- the gap between what the moral principles require and what the rules require cannot be eliminated. That "gap" is the dilemma of law and the source of the intractable natural law/positivism dispute. Natural law focuses on the moral supremacy of correct moral principles; positivism is concerned with the moral advantages of implementing moral principles indirectly, through rules. And there is no third way, no via media, between the horns of the dilemma. Any "weighing" of the rules will be in terms of the correct moral principles; which will never dictate their own abandonment. That is why rules cannot be weighed against moral principles, nor procedure against substance. In the end, I view ARIL as a misguided attempt to chart a nonexistent via media between moral principles and rules.
(77) See Brewer, supra note 4, at 977-78.
(78) See supra text accompanying notes 67-69.
(79) See Brewer, supra note 4, at 939, 1023.
(80) See Alexander, supra note 8, at 42-44 ("[A] spare recitation of facts characterized at a high level of generality ... seriously impedes the constrained court's determination of whether the constrained case is an a fortiori case or is instead distinguishable on its facts from the precedent case.").
(81) An example of such a broad rule would be: "John Doe entered into an agreement with Jane Roe. There was consideration present. John Doe did not do what he agreed to do. We hold him liable to Jane Roe for damages caused by his breach." The end.
(82) See Alexander, supra note 8, at 15; Larry Alexander, Striking Back at the Empire: A Brief Survey of Problem in Dworkin's Theory of Law, 6 Law & Phil. 419, 433 (1987); Larry Alexander & Ken Kress, Against Legal Principles, in Law and Interpretation 279, 299-300 (Andrei Marmor ed., 1995).
(83) See Brewer, supra note 4, at 937-38.
(84) For my position on this debate, see Larry Alexander, All or Nothing at All? The Intentions of Authorities and the Authority of Intentions, in Law and Interpretation, supra note 82, at 357 (arguing that the texts of legal authorities are the authorities' determinations of what is morally required of citizens, and that ascertaining what those determinations are is ultimately a factual, rather than an evaluative, inquiry).
(85) I shall say nothing here about the "practical reasoning" approach to interpretation, which I find to be an incoherent "blend" of the normative (what would be good or just) and the descriptive (what the legislators intended, how the courts have interpreted, etc.). See Larry Alexander, Practical Reason and Statutory Interpretation, 12 Law & Phil. 319, 328 (1993) ("[N]orms and facts may constrain each other but cannot `combine.'"). As far as l know, however, its proponents do not claim that it exemplifies analogical reasoning. See also id. at 319 ("[I]ts proponents suggest that in interpreting a statute, we should look at the words' meanings, . . . norms regarding institutional relationships, rule of law virtues, social norms, efficiency, and justice.").
(86) Some argue that past moral mistakes give rise to equality claims in the present; I criticize that view infra Part III.C.2.
(87) See Dworkin, supra note 7, at 238-50.
(88) See Alexander, supra note 8, at 31-32, 37-42.See also Alexander, supra note 82, at 426-34.
(89) See Ronald Dworkin, Taking rights seriously 22-39 (1977) (arguing that judges use principles, not rules, to decide cases); see also Dworkin, supra note 7, at 238-50 (explaining the process that judges use to discover immanent legal principles.)
(90) See Alexander & Kress, supra note 82, at 288 n. 47 (arguing that "[s]uch a counter-factual approach is in fact the best description of the dominant methodology for dealing with precedential constraint").
(91) 163 U.S. 537 (1896)
(92) See sources cited supra note 82.
(93) See Alexander & Kress, supra note 82, at 304-06.
(94) See id. at 294-95 (arguing that "equality" is not a free-standing concept, but rather is theory-dependent); Alexander, supra note 82, at 426-31 (arguing the same point).
(95) See Alexander, supra note 8, at 9-12 (arguing that past incorrect judgments should not be repeated simply to satisfy "equality" between past and present cases); see, e.g., source cited supra note 91 and accompanying text.
(96) I may seem to be equivocating here. Is ARIL impossible? Or is ARIL possible, but perverse? Is my critique of ARIL ontological or normative? Is ARIL a phantom or a real monster?
I believe it is both. Although I would be content with my normative critique of ARIL -- after all, ARIL purports to be the method by which courts justify their decisions -- I believe my critique casts doubt on ARIL's existence as well, at least in the following sense: A court may point out the similarities and dissimilarities between the case before it and prior cases and conclude, based solely on the similarities and dissimilarities, that the case before it is more "like" one set of prior cases than other sets of prior cases. Because the relevance of those similarities and dissimilarities remains unjustified, however, this version of ARIL lacks any rational force. Alternatively, a court may establish criteria of relevance by citing either to a rule laid down in the precedent cases or to correct moral principles (which are consistent with precedent cases only if the cases were decided morally correctly), both of which can be justifiable, but neither of which is ARIL. Finally, the court may establish criteria of relevance by citing to "principles" immanent in the precedent cases. Such principles are, as I have shown, normatively unattractive. Beyond that, normatively unattractive normative principles have a queer ontological nature. See Alexander & Kress, supra note 82 at 303 n. 96. Although we can speak, for example, of the "existence" of moral principles that we reject -- the principle of utilitarianism might be a good example -- such references make sense primarily when the principles, to which they point operate in rule-like fashion, either as a master rule or as a lexically ordered set of rules. When, however, the morally incorrect principles to which these statements refer are supposed to have a dimension of weight -- or (what I believe translates as the same thing), when these principles are supposed to answer the question What would be the morally correct outcomes in a world in which some morally incorrect outcomes in our world were morally correct?" -- I believe they become incoherent and thereby lack existential plausibility.
(97) See Nightmare on Elm Street (New Line Cinema 1985).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||analogical reasoning in law|
|Publication:||University of Pennsylvania Law Review|
|Date:||Nov 1, 1996|
|Previous Article:||Rediscovering Blackstone.|
|Next Article:||From "publish or perish" to "profit or perish": revenues from university technology transfer and the s. 501(c)(3) tax exemption.|