Law and Legal Reasoning

AN INTRODUCTION TO
LAW AND LEGAL

REASONING
\_, l B 1< 1
1- ',S'
""t[··~~-
BENJAMlN KEELE
1'0 ~~
'1t tH E L\ \\ "\'-
EDITORIAL ADVISORY BOARD
LITTLE, BROWN AND COMPANY
LAW BOOK DIVISION
Richard A. Epstein
James Parker Hall Distinguished Service Professor of Law
University of Chicago
E. Allan Farnsworth
Alfred McCormack Professor of Law
Columbia University
Ronald J. Gilson
Charles J. Meyers Professor of Law and Business
Stanford University
Marc and Eva Stern Professor of Law and Business
Columbia University
Geoffrey C. Hazard, Jr.

Sterling Professor of Law
Yale University
James E. Krier
Earl Warren DeLano Professor of Law
University of Michigan
Elizabeth Warren
William A. Schnader Professor of Commercial Law
University of Pennsylvania
Bernard Wolfman
Fessenden Professor of Law
Harvard University
AN INTRODUCTION TO
LAW AND LEGAL
REASONING
Second Edition
Steven J. Burton
William G. Hammond Professor of Law
University of Iowa
LITTLE, BROWN AND COMPANY

Boston New York Toronto London
Copyright© 1995 by Steven J. Burton
All rights reserved. No part of this book may be reproduced in any form
or by any electronic or mechanical means including information storage
and retrieval systems without permission in writing from the publisher
except by a reviewer who may quote brief passages in a review.
Library of Congress Catalog No. 94-74423
ISBN 0-316-11489-8
Third Printing
ICP
Published simultaneously in Canada

by Little, Brown & Company (Canada) Limited
Printed in the United States of America

For Jacob Harlan Stier
SUMMARY OF CONTENTS
Contents IX
Preface to the Second Edition XI
Introduction 1
Chapter 1. Cases and Rules 11
Chapter 2. Analogical Legal Reasoning 25
Chapter 3. Deductive Legal Reasoning 43
Chapter 4. Combining Analogies and Deductions 59
Chapter 5. Legal Reasons and Conventions 79
Chapter 6. Purposes 97
Chapter 7. The Judge's Perspective 117
Chapter 8. The Lawyer's Perspective 135
Chapter 9. Legitimacy 155
Index 175
Vil
CONTENTS
Preface to the Second Edition Xl
Introduction 1
A. The Rule of Law 2
B. A Practical Approach 4
C. Scope 8
Chapter 1. Cases and Rules 11

A. Cases 11
B. Rules 13
C. Legal Reasoning and Legal Problems 14
D. Using Rules and Cases 18
1. Rules in Problem Cases 18
2. Decided Cases in Problem Cases 20
Chapter 2. Analogical Legal Reasoning 25

A. The Analogical Form 25
B. Analogical Legal Reasoning 27
1. Precedents 28
2. Factual Similarities and Differences 29'
3. The Judgment of Importance 31
Chapter 3. Deductive Legal Reasoning 43

A. The Deductive Form 44
B. Deductive Legal Reasoning 46
I. The Rule 47
2. The Facts 50
3. The Judgment of Importance 52
IX
x CONTENTS
Chapter 4. Combining Analogies and Deductions 59

A. Common Law Rules 60
l . Easier Cases 60
2. Organizing the Precedents 62
B. Cases and Enacted Rules 65
C. Detour on the Judicial Role 74
Chapter 5. Legal Reasons and Conventions 79

A. Is Legal Reasoning Scientific? 79
B. The Normative Function of Law 82
C. Family-Style Relations Among Cases 85
D. Formalism, Skepticism, and Conventionalism 89
Chapter 6. Purposes 97
A. The Judgment of Importance 98
B. General Purposes of Law 102
C. Purposes Embedded in the Legal Experience 105
l . Purposes in the Common Law 106
2. Purposes in Statutes 111
Chapter 7. The Judge's Perspective 117

A. Easier Cases 117
B. A Harder Case 121
C. Deciding Harder Cases 124
I . Webs of Belief 124
2. Webs of Belief About Law 127
Chapter 8. The Lawyer's Perspective 135

A. How Judges Get Started 135
B. How Lawyers Get Started 140
C. Advocacy in Harder Cases 144
Chapter 9. Legitimacy 155

A. The Legal Conversation 156
B. Legitimacy and Legal Reasoning 159
C. The Legitimacy of Convention 164
l. Systemic Legitimacy 165
2. The Legitimacy of the Legal Community 169
Index 175
PREFACE TO THE SECOND
EDITION
IT IS COMMONLY SAID that a beginning law student's first goal is to learn

how to "think like a lawyer." For the most part, however, law teachers do
not deal explicitly and systematically with legal reasoning. The first-year
curriculum in U.S. law schools consists largely of courses on selected
basic areas of the law, such as contracts, torts, property, criminal law, and
civil procedure. Students read examples of legal reasoning (mostly as
expressed in judicial opinions) within the subject of study. Law teachers
challenge students to explain and criticize the examples to shed light on
the law. Students are left largely to their own devices to extract
worthwhile lessons about legal reasoning from the examples and
discussions. This book tries to help students learn legal reasoning as it is
practiced conventionally in the United States.
This second edition results from a complete rewntmg and

simplification of the original. I owe debts of gratitude to many people
who helped this book reach its present state of maturity. I reaffirm the
acknowledgements in the first edition, especially those to Owen Fiss,,
Serena Stier, Christopher D. Stone, and my father. I add a special thanks
to my colleague Ken Kress; many conversations with him over the last
decade have helped me to refine my thoughts about legal reasoning. I
thank Colleen Halliman, David Jung, Linda Meyer, Serena Stier, and Jill
Van Wormer for helpful comments on the manuscript for the second
edition. Anne Brafford' s extensive editorial comments and suggestions
were exceptionally valuable and refreshingly honest. I am grateful as
always for the continuing generous support of Dean N. William Hines
and the University of Iowa Law School Foundation. Finally, I thank the
more than 2,500 students at Iowa who have been through my introductory
course on legal reasoning over the last dozen years. Many of them
XI
x11 PREFACE TO THE SECOND EDITION
stimulated improvements from the first edition. They provide me the best
of reasons for teaching and writing in this area.
AN INTRODUCTION TO
LAW AND LEGAL
REASONING
Introduction
A LAWYER'S PRINCIPAL TASK, among several, is to help a client get
somewhere the client wants to go. To do so, lawyers make
predictions and arguments. They attempt to predict what judges
(or other adjudicators) will do so their clients can avoid running
afoul of the law. They try to convince judges to apply the law
favorably in cases involving their clients' interests. Judges, of
course, decide what the law permits or requires of people in cases
that come before courts. 1
What is the law? Philosophers have debated the question for
centuries. For present purposes, let us think of the law as the
collection of precedents, rules, principles, and policies employed by
judges when justifying their decisions. By all accounts, judges are
under a duty to uphold the law. They should apply the law to the
facts of a case to yield legal reasons, which are reasons for action by
law-abiding people. For example, a red light plus a rule requiring
motorists to stop at red lights is a legal reason for Mitchell
Motorist to stop. It is also a reason for a lawyer to predict that a
judge would fine Mitch if he did not stop, a reason for a prosecu-
tor to urge a judge or jury to convict him, and a reason for a judge
or jury to do so. Legal reasoning is the process of using legal
reasons in legal arguments.
Confusion about law and legal reasoning leads to confusion
about almost everything else in the study of law. Remarkably, few
books introduce law and legal reasoning explicitly and systemati-
cally, as does this one. More remarkably, few are directed at begin-
ning law students, who often find it frustrating to learn how to -
"think like a lawyer." The conventional wisdom is that you learn
legal reasoning by doing it. Consequently, students in the first year
of law school are left to proceed largely by trial and error.
I have found over nearly two decades of teaching that master-
ing law and legal reasoning requires patience, persistence, and
practice over an extended period of time. A short and direct
explanation, however, can help spur the process along and provide
1
On the influence of courts on lawyering activities outside the courthouse, see
Chapter 1 §C.
1
2 INTRODUCTION
a broad perspective, perhaps reducing accompanying anxiety. A

book like this one can call significant features of law and legal
reasoning to your attention, draw connections you might otherwise
miss, and place law and legal reasoning in a context where they
might make better sense. Of course, as with any worthwhile
activity, it will remain your responsibility to develop your own
approach through hard work over time.
A. The Rule of Law
A contrast between the ideal and the practice of law and legal
reasoning often bedevils beginning law students. The ideal consists
of those ways in which lawyers and judges should think and
communicate their thoughts. The practice consists of those ways in
which lawyers and judges do think and communicate their
thoughts. Few observers claim that the practice conforms to its
ideal. The contrast is troubling because there are good reasons for
upholding the ideal.
~·-
of law." Americans believe as a matter of principle _!hat t}le
----::..
In U.S. society, the ideal reflects a set af values called the "rule
~oercive powers of governl!len_!~oulcflie limifec:f~y)aw. The (LS.

Constitution, for example, provides that "[n]o person shall ... be
deprived of life, liberty, or property, without due process of law." 2
Much of what government does, of course, deprives people of life,
liberty, or property. A court's judgment may require one person to
pay money or transfer property to someone else, may incarcerate
a person or otherwise restrict her freedom, or may impose the
death penalty. A court may approve or disapprove actions by the
police or other officials with similar effects on people. By the most
widely accepted accounts, the rule of law requires coercion to be
used by_officials~nl~_!-"le; an-aE_~_uth~!J-~eq.___!2y_!he law~-~
Accordingly, people want to know by what rignr·someone
sitting behind a bench or s_t_9-ndi~g_!?e~i_n._~~ u?es ,t estate's
-power-fo-depr1vesomeone else. The law is supposed toprov1aecrn
answer to that query. _Legal reasoning is_fil!P~d~ !9_ dr~~ _a
connection between the law and a particular action, implementing
--· --·· -------------
- . ----
--------~-- -- --~--------
2
U.S. Const., amends. V, XIV.
INTRODUCTION 3
the rule of law. Faulty legal reasoning signifies a faulty connection,

raising the question whether governmental power is being used
without proper justification.
One version of the rule of law, often called "Jegal formalism," )
insists th_c!-tfegal reasoning should determine all specific actions
requir~~ by the 19w based only on objective fa~cts, unaml5iguotrs
rules~ and logic. Were the law to operate this way, legal reasonlllg
wouta1eadto the same decision no matter who is doing it. Justice
would-nofvarywiththe vagaries of human-personalitf Law and
legal reasoning-woulcfsuffice for lawyers to predicrwith confi-
dence what government officials will do. Jud es could decide ~?§es
with no need for judgmen_!. And critics could say with confidence
which judicial decisions were lawful.
It takes little exposure to law study to begin doubting whether
the practice is anything like the formalist version of the rule of
law. Close analysis of many legal decisions shows them to rest on
debatable facts, ambiguous laws, or incomplete logic. The "real"
grounds of the decision then are unclear. Comparisons of decisions
reached by different officials in similar situations may suggest that
the officials' personalities, politics, or prejudices had more to do
with the decision than did the law. Such a reaction to legal
forll'!cllism is called "legal skepticism."
Clegal skepticis19 is troubling in two ways. First, how can a
genuine skeptic function as an effective lawyer if legal reasoning
cannot be relied on to connect law with official action? The law
game may seem to go on~ithQl!!_rules or with rules that can be
.chill1ge._~ignorec:i a_t_an umpire's whim. Second, why should
such a game go on? The law makes some people rich and some,
people poor, some people free and some people fettered, some
people live and some people die-in the real world. A game that
does so for no reason or for the wrong reasons is not a game we
should want to play in a democratic society committed to freedom
and equality.
Consequently, we may confront a legal problem with conflict-
ing instincts. We may try to solve it as required by the formalistic
version of the rule of law. At the same time, we may rely on
whatever intuitions we hp.ve about the nonlegal causes of legal
decisions. The implications of each approach differ frequently. We
may be confused because it is hard to give up either instinct. We
then may not know what we should do.
4 INTRODUCTION
B. A Practical Approach
Notice, however, that formalism and skepticism both depend

on formalism's view of the law as a matter of rules and logic
dicfatiiig--resufts-111 a}l cases. ForrnalTsm holds that law and legal -
·reasoning must satisfy its standard and claims that this standard
can be satisfied in practice. Skepticism agrees that law and legal
reasoning ,nust satisfiJ formalism's standard, but it claims that,
realistically, that standard is not or cannot be satisfied in practice.
I suggest that the successful study of law requires __a third a1:d
_more p~acti~~J -'~1?££()ac.h to legal problems-one that abandons
formalism's standard completely.
To lay a basis for this alternative, consider .a crucial di~~rence
l?etween law and the empirical sdences. In the physical and social
_ s ~ \ , ~ a general description_Q[~he __~e&~Iaritie_s in the
behavior of persons, events, or things. A descriptive law in
principle is falsified by one replicable instance in contradiction to
it. The law of gravity, for example, predicts that my pencil will
drop to the ground every time I release it (under normal circum-
stances). If once I released my pencil and it floated to the ceiling,
however, this law would come into question. Formalism's standard
!!:.eats__l_e_g_a]_ laws in a_ si_!ll.ila!"__'.'Y~Y- It requires the rules, faCts~and
logic to fix lawfu.T·conduct in any case; otherwise, the skeptic says,
the law is indeterminate and has no effect at all.
. -
They say how people ought to act. When Mitch Motorist runs a red
--
Legal laws, however, dg_ nqt__.9-escril?~ _:rt;g_ulaijti~s in the
behavior oranything. Rather, these laws prescribe l}uman conduct:
-~-
light, even repeatedly; whether or not he gets caught, the relevant

traffic law does not lose its validity. Mitch is properly cited and
fined if and when he gets caught. He violated the law by failing to
do what the l ciw prescribes. The law might lose its force if
disobedience were persistent and widespread, as when the laws of
Eastern European communist regimes were widely flouted in 1989.
Occasional violations, however, are simply occasions for officials
to apply the law.
Practical choices-those concerning human action-must be
rp.a
ob~~~t
nd understood from an actor's oint of view, not an
s. When Mitch encounters a red light, an observer mig t
--------
be stationed on the curbside, recording the correlations between
red lights and stopping cars, and observing an apparently causal
INTRODUCTION 5
connection. The observer, like an empirical scientist, is interested

in describing the regularities and making predictions of future
behavior by motorists. For her, the observations yield reasons for
beliefs.
Mitch, however, is in the driver's seat. He is uninterested in
descriptions of the regularities in motorists' behavior or causal
connections between red lights and stopping cars. To Mitch, the
_red ligl)t is not a sign that he will stop. Together with the law, it
is a reason for him to sto-P-. 3 Unlike the observer, he has reasons for
~ based in the law. Beliefs stemming from empirical observa=
tions, such as the rate at which police detect violations, may
supply relevant information to him: Given an impatient character,
Mitch might regard a low rate of detections as part of a reason to
run the light. A judge and jury can fine him nonetheless for
violating the law when he fails to do what the law prescribes.
Whether acting, advising an actor, or judging an actor's
conduct, what matters are the reasons indicating what the actor
should do. Notice that we can understand the actor's point of view
significantly even when we are not in the driver's seat and do not
identify with the actor. 4 For example, a Christian may say to his
Muslim friend who is about to eat some pork fried rice unwitting-
ly, "You ought not eat that." The Christian thereby speaks from
the point of view of one who accepts the Muslim ban on eating
pork without endorsing it as right and proper. Similarly, law
teachers and practitioners can speak from the point of view of law-
abiding citizens or judges and say what the law requires without
thereby endorsing it as just and moral. From that point of view,.
the law will provide reasons for action, as in the example above.
Reasons for action are not subjective. A red light is a reason for
Mitch to stop only if it is a reason for anyone to stop under the
same circumstances. If Mitch were an ambulance driver with siren
screaming, the law would allow him to run the red light. So, too,
anyone driving an ambulance with a screaming siren could run the
red light. Moreover, reasons allowed by the law-legal rea-
sons-may not have the final word on what anyone should do.
When Mitch is rushing his seriously ill wife to the hospital at 3:00
3
H.L.A. Hart, The Concept of Law 87-88 (1961).
4
Joseph Raz, Practical Reason and Norms 171-177 (Princeton Univ. Press ed.
1990).
6 INTRODUCTION
A.M., he may be morally justified in running a red light even while

he is legally liable to pay a fine.
The law may provide competing reasons that must be weighed.
When deciding to come to law schoot for example, you probably
identified the pros and cons, weighed the pros against the cons,
and acted according to the stronger reasons in your best judgment.
Lawyers and judges often deliberate exactly the same way when
engaged in legal reasoning. While you could take into account any
reasons you wanted, however, judges are constrained when
adjudicating to consider only the reasons allowed by the law. 5
Lawy:_er~Jal<~ into account a wider rang~ 9J_reasons when predict-
- i!!g_Qf_fici_al action~ Their r~rgiim-ents_toJudg~s, however, are limited
t°-_the rea.-sons- aTlo_~~g_to the__j.udges. Their predictions of judicial
behavior are informed centrally by the same reasons.
On this basis, th~.L.@y~judging can bf explain~d
i,_I}__~ractical way _that av_oi_@_ the ___~t_ark choice between legal
formalism and legal skepticism. This approach requires that we
·~deri-aw-·anctlegaCfea's~_ning with attention to the point, of
view of legal a_<;:jor_.s, including citizens and those who advise and
judge them. It also requires that legal reasoning be considered in
_li_ght of its functions in the real-world contexts in which it is used.
In the abstract, legal reasoning does nof differ significantly from
other kinds of practical reasoning. In real-world settings, it is used
by lawyers and judges to deal with distinctively legal problems, is
expressed in a specialized language used by a professional
community, takes certain principles and goals for granted, and
employs artificial devices to circumvent practical problems.
The chief feature of legal reasoning is that it is used in the
process of anticipating or settling important disputes in advanced
societies. People who live in proximity to one another will find
themselves at times in disputes-controversies in which two or
more persons claim incompatible rights. Thus, one person may sell
another a cow that both had thought was barren. When the cow
turns out to be with calt the seller may try to get the cow back.
The buyer may refuse. Or one person may be hunting foxes and
begin to chase one. When another person intrudes, and kills and
carries off the prey, the first may demand possession of the fox or
5
Steven J. Burton, Judging in Good Faith (1992).
INTRODUCTION 7
payment of its value. The second may refuse. Or one person may
be separating fighting dogs with a stick. When that person
carelessly hits another person, the victim may request payment of
compensation for an injury. The injurer may refuse. These and
countless other situations lead two or more persons to differ over
practical matters that will not wait or go away. One person has
something the other person claims. Each may refuse to settle the
dispute.
Every society develops methods for settling persistent disputes
among its members. 6 Different methods are employed by different
societies in different times and places. In some, disputes in which
the parties persist may be settled by organized combat between the
parties (duels), their champions (jousts), or their clans (feuds). They
may be settled by ritual appeals for God's judgment (trial by
ordeal; consulting an oracle). They may be settled by a third
person's command (father knows best; the divine right of kings) or
by appeal to chance (flipping a coin). Disputes also may be settled
by appealing to a third person's reason, an intellectual search for
the fair or right in the matter (arbitration, adjudication).
Important disputes in developed societies can be settled, if
need be, by appealing to judges to apply the law. This method is
better than the alternatives when disputes should be settled with
finality in peaceful and justifiable ways. Dispute settlement by law
is more peaceful than duels, jousts, or feuds. It is more just than
trial by ordeal, consulting an oracle, flipping coins, or letting a
"wise" man decide. And adjudication is more public, accountable,
and revisable in cases of error. It seems obvious that law and legal,
reasoning enable judges to reach final, peaceful, and justifiable
dispute settlements better than would the alternatives. What is not
obvious is how law and legal reasoning do so.
h Negotiation, mediation, and conciliation are methods of dispute settlement

that may help the disputants settle their differences voluntarily. They do not
impose a settlement if the disputants persist. They are increasingly important
supplements to adjudication in the U.S. legal system.
8 INTRODUCTION
C. Scope
This book is organized into nine chapters. Chapter 1 introduces

two basic concepts-cases and rules-from which an under-
standing of law and legal reasoning should be built. Chapters 2
through 4 introduce and analyze the two main forms of legal
reasoning employed by good lawyers and judges. Chapter 2
presents analogical legal reasoning, which is most closely associat-
ed with reasoning from precedents. Chapter 3 presents deductive
legal reasoning, which is most closely associated with reasoning
from rules. Chapter 4 analyzes how analogical and deductive legal
reasoning can be combined. These chapters show that these two
forms of legal reasoning are useful in several respects and are
crucial for expressing the conclusions of legal reasoning effectively.
It also shows that, in many cases, they are not adequate to
determine legal answers, as required by legal formalism. They
leave a judgment of importance unconstrained by the rules and
precedents, making it unclear whether legal reasoning can
implement the rule of law.
Chapters 5 through 8 supplement precedents and rules with
legal principles, policies, and conventions used to make the
judgment of importance. Chapter 5 examines the nature of re-
lations among cases and introduces the conventions of the legal
community. Chapter 6 introduces the roles of principle and policy.
Chapter 7 highlights a judge's perspective on legal reasoning in
harder cases. Chapter 8 brings the lessons of preceding chapters to
bear on the practical problems of making lawyerly predictions and
legal arguments. These chapters show that, though legal reasoning
does not produce the certain answers required by legal formalism,
it nonetheless produces lawful predictions, arguments, and de-
cisions with a significant degree of regularity. This approach
therefore rejects legal skepticism.
Chapter 9, the concluding chapter, responds to a serious ob-
jection that might be raised against conventional legal reasoning as
presented in the preceding chapters. It addresses the problem of
legitimacy in a democratic government under the rule of
law-whether the law is compatible with political morality such
that it should be obeyed. This book does not conclude that law and
legal reasoning achieve legitimacy as currently practiced in the
INTRODUCTION 9
United States. It suggests, however, the conditions under which a

practice like ours could deserve the respect that obedience confers.
Some caveats: This book emphasizes legal reasoning as
practiced by lawyers and judges who are worthy of praise by the
legal profession. Proceeding this way can be misleading for three
reasons. One is that I make judgments to emphasize features I
think are important, allowing room for bias to distort the picture.
The remedy for this unavoidable problem is not to dismiss what
I say due to possible bias or for other superficial reasons. Rather,
the remedy is to read the book critically, accepting only those
assertions for which convincing reasons are given. Even then, you
should believe these lessons only tentatively, pending further
thought and experience.
A second risk of misdirection is that the legal system in
operation is a good bit messier than the picture I present. Not all
lawyers and judges are good at their jobs. (I assume you want to
be a good lawyer.) The pressures of heavy case loads lead even
good lawyers and judges to speak and write in truncated ways.
Often the version of legal reasoning presented in this book is only
implicit in the practice of law and judging. Sometimes the evidence
of practice is too scant to bear any coherent interpretation.
The third, and perhaps most important, caveat is that good
legal reasoning is not all there is to good lawyering and judging.
The successful practice of law and judging involves ethics,
imagination, common sense, and knowledge in the ways of the
world, as well as interpersonal, rhetorical, political, and other
skills, along with an understanding of the law and proficiency in_
legal reasoning. A complete theory of lawyering or judging would
explain the role of legal reasoning in relation to these other skills;
this book suggests only that legal reasoning plays a major role
therein.
Upon further study in jurisprudence and legal theory, you
would find that much that is said here is controversial among
academic theorists. I have chosen to ignore academic controversies
except when they bear on matters that should be of concern to
beginning law students. Even then, I hope only to identify issues
and suggest fruitful avenues of thought. You must do the hard
work of spinning a valuable web of beliefs about law for yourself.
CHAPTER ONE _ _ _ _ _ _ _ _ _ _ __
Cases and Rules

nk of the law as a system of rules,
: large. The study of law in U.S. law
:gins with law cases. You will study
primarily trom cases throughout law school. Parsing the cases will
be an important intellectual activity throughout your legal career.
To be sure, legal rules, principles, and policies play an essential
role. Cases, however, are the grist for the legal reasoning mill.
Before undertaking the main discussion of legal reasoning,
three orienting questions will be posed in this chapter: What is a
case? What is a rule? Why do we study primarily cases?
A. Cases
In this book, a case is a short story of an incident in which a

court acted or may act to settle a dispute. The term is used in other
ways for other purposes,1 but the purpose here is to say some-
thing useful about legal reasoning. This definition will help.
Treating a case as a short story emphasizes that it has a begin-
ning, a middle, and an end. The story begins when two or more
people get into a dispute or one person gets into a dispute with
the community as represented by law enforcement officials. In the
initial phase, the parties may interact informally or through
lawyers. A middle phase commences when one party (the plaintiff·
or prosecutor) files a complaint against another (the defendant)
with a trial court, calling on the court to settle the dispute in favor
of the complaining party. It continues in the trial court until a
judgment is entered by the presiding judge. If either party believes
the judge erred on a point of law, it may appeal to an appellate
court, where the last phase takes place. There, arguments are heard
by a group of judges who issue a final judgment settling at least
1
For a different perspective, see Owen M. Piss, The Forms of Justice, 93
Harv. L. Rev. 1, 28-44 (1979).
11
12 1. CASES AND RULES
part of the dispute. The appellate court writes and publishes an

opinion that summarizes the facts of the dispute and the proceed-
ings in the trial court, announces a decision on the issues before
the appellate court, and gives the reasons for its decision. For the
most part, a lawyer's knowledge of decided cases comes from
what the appellate courts write in their opinions, which may or
may not be accurate stories of the events in question.
Treating a case as a short story of an incident emphasizes that
every case is unique in all of its particulars. A case involves two
or more parties, both of whom are unlikely ever to repeat the very
actions that led to the dispute between them. We can describe the
events in one case in terms general enough to encompass other
disputes between other parties in other times and places. We can
also describe them in terms specific enough to encompass only the
dispute that occurred between these parties at one time and place.
However we describe cases, each occurs once.
Treating a case as a short story of an incident in which a court
acted or may act emphasizes that we will focus mainly on two sorts
of disputes: (1) disputes in the past that were settled, at least in
part, by the coercive dispute settlement machinery of the state
(decided cases, or "precedents") and (2) unresolved or foreseeable
disputes that might be settled in that way (problem cases). A court
is a central part of the state's dispute settlement machinery, but
only a part. For example, a court's judgment in favor of the
plaintiff in a civil case-generally one in which a private plaintiff
initiates a lawsuit-may be enforced against a noncompliant
defendant by the sheriff. Some of a losing defendant's property
may be taken, by force if need be, and sold by the sheriff, with the
proceeds going to the plaintiff to pay a money judgment. In
criminal cases, the police may initiate the dispute on behalf of the
community, and an official prosecutor requests court action. On
conviction, a criminal defendant may be fined or incarcerated by
the sheriff, using force if necessary. The use of coercion by state
officials usually is not necessary because most citizens are law-
abiding by disposition and, for others, the threat of force is
sufficient to induce compliance. Nonetheless, every request to a
court is a request that the state use force, if necessary, to settle the
dispute in favor of the complaining party. A court "acts" whether
such a request is granted or denied.
1. CASES AND RULES 13
Treating a case as a short story of an incident in which a court

acted or may act to settle a dispute emphasizes the social function
of the law. Formal methods of peaceful dispute settlement are
needed because disputes will arise whenever people live in close
proximity to one another. Some disputes will not be settled by
voluntary means. The availability of formal dispute settlement
procedures may lessen the level of violence and other disruptions
in society by affording the disputants an alternative to fighting.
The law courts, by their responses to concrete disputes, also
contribute to society's understanding of its values. The law both
reflects and helps mold social values, substantially through judicial
judgments in cases resolving disputes.
B. Rules
To explain why we emphasize cases, we should identify the
principal alternative-rules. The contrast between cases and rules
is not difficult to grasp.
In this book, a rule is a general statement of what the law
permits or requires of classes of people in classes of circumstances.
Again, the term is used in other ways for other purposes. This
definition is even more rough than the definition of a case because
rules are notoriously slippery characters. Some would draw
distinctions between rules and principles or rules and policies, but
such distinctions can be deferred to Chapter 6.
Treating a rule as a general statement emphasizes that all rules
are cast in language. Because all language (save proper names) is
general, our words do double- and triple-duty. Consider the word
bar. You may want to join the bar but may be barred if you fail the
bar and wind up tending bar or selling candy bars. A rule that
uses the word bar, or any other word, requires interpretation to
sort out the various possible referents. Legal rules characteristically
require interpretation, which is an important part of legal reason-
ing.
Treating a rule as a general statement of what the law permits or
requires emphasizes that a rule is normative. That is, a rule guides
conduct by saying something about what people in general should
or should not do. A rule might include a descriptive part, as when
it states the circumstances under which an obligation comes into
play. For example, a rule might say that "One who kills another
without excuse or justification shall be punished .... " The part
"One who kills another" purports to describe facts. The part
"without excuse or justification" is normative because it requires
a judgment of the rightfulness or wrongfulness of a killing. The
part "shall be punished ... " states the legal consequence attach-
ing to unexcused and unjustified killings of people by people. This
rule implies an obligation for people not to kill other people
wrongfully, though it says little about which killings are wrongful
or rightful.
Treating a rule as a general statement of what the law permits
or requires of classes of people in classes of circumstances emphasizes
that rules apply to groups of people in similar situations. Rules
operate from a position of generality; that is, they apply to more
than one case. Legal rules are supposed to affect what people do
by bringing obligation, and often force or the threat of force, to
bear on their behavior. Rules consequently should be announced
before people engage in relevant behavior and before a case
governed by the rule materializes. Because foresight is limited and
language imperfect, a rulemaker speaks in generalities. A state-
ment of what the law permits or requires of one person under one
set of circumstances-in one case-would be called an order.
In sum, a rule stands in contrast to the cases it governs. A rule
is a general statement of what the law permits or requires of
classes of people in classes of circumstances. A case is a short story
of an incident in which a court acted or may act to settle a dispute.
C. Legal Reasoning and Legal Problems
By the case method of instruction, you will be required in each

course to read many opinions in cases decided by appellate courts.
In class, you will be asked to describe what the court did in each
case, to explain why it did it, to analyze the implications of each
case for possible future disputes, and to synthesize the lessons
from groups of cases into a general understanding of each topic
under study. This process of analyzing and synthesizing cases
seeks to develop your capacity to use legal reasoning to solve legal
problems. Developing this skill implies a study of what judges do
and ought to do-a study of adjudication.
A large part of what lawyers do involves this same skill. But

a larger part involves other skills. Much lawyering is an effort to
anticipate possible disputes and then to plan a client's activities so
that disputes are not likely to arise or can be settled advantageous-
ly if they do arise. Much lawyering is an effort to settle existing
disputes by negotiation. Indeed, most of the disputes that reach a
lawyer are settled before or soon after they reach a court. Most of
those that proceed to trial and judgment are not appealed. Lawyers
in practice spend the better part of their time planning, counseling,
drafting, negotiating, or preparing for trials-anticipating and
processing disputes from the law office.
Throughout, however, an important part of lawyering is court-
oriented. This part consists of predicting what an appellate court
would do if a case were to materialize and be taken to the highest
available court, or persuading someone of what an appellate court
will or should do in that event. That much more is involved does
not mean that predicting and persuading court action are not
involved. Some examples illustrate the point while introducing
some key functions of a lawyer.
Consider a lawyer who is counsel to a business firm that
manufactures and sells a product-say, a chain saw. The sel-
ler-client wishes to market a new, improved model. The client
seeks legal advice on whether to include a certain kind of chain
guard that would reduce the risk of injuries to operators, but only
at some increase in the price of the saw (and consequently some
reduction in the client's volume of sales and the amount of its
profits). The lawyer will want to determine whether omitting the .
chain guard makes it significantly more likely that the seller will
be held liable (that is, will be forced to pay compensation to users
of the chain saw for injuries that could have been prevented by a
chain guard). The lawyer's prediction of what the law will do in
the end-what an appellate court would decide-is a significant
element in that determination.
If the lawyer concludes that omitting the chain guard is likely
to make no difference before an appellate court, she may advise
the client that a decision whether to include a chain guard need
not be influenced by legal: factors. Further questions will arise if
she concludes that omitting the chain guard makes liability more
likely: What are the courts likely to do should the seller omit the
chain guard but include in the instruction booklet a warning of the
danger? Will the courts enforce a contract clause by which the

buyers agree to assume the risk of such injuries (a disclaimer of
liability)? Will it matter how the language of such a warning or
disclaimer is drafted? The lawyer's predictions of judicial action in
each variation on the anticipated dispute will influence her advice
on whether to include the chain guard. The predictions will also
influence her drafting suggestions on language for inclusion in the
instruction booklet or contracts of sale.
Assume now that the chain saw is marketed without the chain
guard, warnings, or disclaimer. A buyer is injured in a way that
would not have occurred with a chain guard. The buyer seeks
compensation for his injuries from the seller; the seller does not
want to pay. At this point, the seller's lawyer may try to negotiate
a favorable settlement while preparing for trial, should a lawsuit
be initiated.
The lawyer again will want to predict what a court will do
based on the facts then available. The lawyer's advice on whether
to settle out of court will be influenced by the prediction in the
following way, to give a simplified example. 2 If the buyer claims
compensation in the amount of $10,000 and the seller's lawyer
predicts that the seller has a 70 percent chance of losing for that
amount in court, the lawyer should advise the seller to settle for
any amount less than $7,000. The client would be better off settling
for less than $7,000 now than taking a 70 percent chance of being
forced to pay $10,000 later. On the other hand, if the buyer's
lawyer predicts that the buyer has a 40 percent chance of winning
in court, the lawyer should advise the buyer to settle for any
amount more than $4,000. The client is better off with more than
$4,000 for certain than with a 40 percent chance of getting $10,000
later. Given these stakes and these predictions by the two lawyers,
the conditions are ripe for settlement somewhere between $4,000
and $7,000.
The lawyers' skills at negotiation will determine the amount of
the settlement. Though many factors will influence the negotiation,
one will be the lawyers' skills at persuasion. If the seller's lawyer
can convince the buyer's lawyer that the probability of a buyer's
2
This example assumes that the people involved are rational and wish neither
to gamble nor to avoid risks unreasonably. For purposes of discussion, it also
ignores lawyer's fees and other litigation expenses.
victory in court is less than the buyer's lawyer had thought, the
buyer's lawyer may advise his client to settle for a lower amount.
The converse is also true. The lawyers' persuasive skills may make
a settlement possible even if their initial predictions did not set the
right conditions. As the lawyering process unfolds, new informa-
tion is acquired and new arguments are developed. The lawyers
should revise their predictions and negotiate further.
Assume now that no settlement was reached and the dispute
goes before a trial court. The trial judge will be required to make
a large number of decisions, including rulings on pretrial motions,
admissibility of evidence, and in some cases instructions to the
jury. Each decision affects each party's prospects of winning the
lawsuit. These decisions should be made according to the law. If
the trial judge errs either by applying the wrong law or, as
sometimes happens, by applying the correct law wrongly in the
case, the final judgment may be reversed on appeal to a higher
court. The lawyers representing the parties will be allowed to
make arguments to the trial judge on what the law permits or
requires in each decision. These arguments by counsel in large part
will seek to persuade the trial judge of what the appellate court
would decide on the point if an appeal were taken.
Assume now that the case reaches the highest appellate court
in the relevant jurisdiction,3 the end of the dispute anticipating
and processing sequence. The lawyers and the lower court judges
wanted to predict what judges further along the sequence would
do and persuade others of what those judges would do, to avoid
moving to the next stage of the sequence unnecessarily. The
highest appellate court does not care to predict what it will itself
do. The highest appellate court cares about what it should do.
The highest appellate court's concern may be said in general to
have two components. First, this court will care that the law be
reasonably stable and predictable so that lawyers and lower court
judges can do their jobs and people can plan their activities to
remain within the law. This will lead the court to give weight to
established understandings of what the law is because lawyers and
lower court judges rely on these understandings. Second, the
3
In some jurisdictions, there are two levels of appellate courts, in which argu-
ments before the intermediate court will focus on persuading the judges of what
the highest court would decide.
highest appellate court will care that the law be justified in light of
contemporary social, historical, and cultural circumstances and
evolving notions of justice. Therefore, the court will be open to
arguments that the law should be changed to improve its justifica-
tion; it will care what the law ought to be.
The "is" and "ought" components, however, are not so easily
separated in practice. Because the lawyers and lower court judges
know that the appellate court may interpret or change the law,
their reliance on established law will be incomplete. Their predic-
tions of and arguments about what the highest appellate court
would do in the case, like the court's final judgment itself, should
be based on the established law and an evaluation of it under
contemporary circumstances and in light of evolving notions of
justice. Therefore, the lawyers' predictions and legal arguments
depend in part on what lawyers think the highest appellate judges
will think the law ought to be. The appellate decision depends in
part on what the highest appellate judges think lawyers (common-
ly) think is the law as it stands.
Consequently, decisions of the appellate courts are important
throughout the dispute anticipating and processing sequence. From
Chapter 2 to Chapter 7, we will emphasize legal reasoning in
adjudication, focusing on the law that judges have a duty to
uphold. Legal reasoning in counselling and advocacy settings will
be reexamined in Chapter 8.
D. Using Rules and Cases
Statements of the law take the form of both rules and cases. To
develop skill in legal reasoning, a study of cases is crucial. Rules
are deceptively simple in appearance while cases are complex and
rich in variety. Sooner or later, you will find that reasoning from
rules alone is often unreliable in a world that is more complex and
varied than even the cases.
1. Rules in Problem Cases
Using a rule often requires intensive analysis and considerable

interpretation. Consider, for example, a simple rule providing that
1. CASES ANO RULES 19
"No person shall sleep in a city park." 4 Imagine two problem

cases. 5 In the first, a gentleman was found sitting upright on a
park bench at noon-his chin was resting on his chest, his eyes
were closed. The gentleman was snoring audibly. In the second, a
disheveled tramp was found lying on the same bench at mid-
night-a pillow was beneath his head, and a newspaper was
spread over his body as a blanket. The tramp, however, had
insomnia. Both were arrested under the rule and brought before a
court for trial. Would you predict that the gentleman will be
convicted and the tramp acquitted? That the gentleman will be
acquitted and the tramp convicted? That both will be convicted?
That both will be acquitted? Does your answer follow from the
language of the rule itself?
Consider, for a second example, a rule providing that "No
person shall bring a vehicle into a city park." 6 Imagine some
possible problem cases. An arnbulance was driven into the park to
reach a stricken jogger. The local Jaycees put a World War II tank
in the park as a monument to the town's war dead. Some teenag-
ers held a car race, a go-cart race, a bicycle race, or a roller-skating
race in the park. A tree surgeon drove his truck into the park to
load and remove the branches of a dead tree under contract with
the city. Surely some of these incidents would not be violations of
the rule. Yet all might be said to involve a person bringing a
vehicle into the park. Consequently, merely saying in any of these
cases that a person brought a vehicle into a park would not be
sufficient to persuade someone else to believe that the rule had been
transgressed.
In almost any case, knowing the rules leaves more intellectual
work to be done because rules are expressed imperfectly and
projected into an uncertain future. A rule alone does not determine
whether many cases come within the class it designates. Whoever
states a rule to govern future cases rarely, if ever, will anticipate
all of the future situations that might plausibly be described in the
4
This example is adapted from Lon L. Fuller, Positivism and Fidelity to Law:
A Reply to Professor Hart, 71 Harv. L. Rev. 630, 662-664 (1958).
5
The shorthand descriptions of cases may be thought of as "short short
stories," which rely on common images to fill in the details.
6
This example is adapted from H.L.A. Hart, Positivism and the Separation of
Law and Morals, 71 Harv. L. Rev. 593, 607 (1958).
language of the rule but that should not be within the class
designated by the rule. 7 Consequently, you should analyze and
interpret the rules in light of possible cases, using legal reasoning
to reach a sound conclusion. Karl Llewelyn overstated the point
colorfully in his lectures of 1929-1930 introducing Columbia
students to the study of law:
We have discovered in our teaching of the law that general

propositions are empty. We have discovered that students who
come eager to learn the rules and who do learn them, and who
learn nothing more, will take away the shell and not the substance.
We have discovered that rules alone, mere forms of words, are
worthless. We have learned that the concrete instance, the
heaping up of concrete instances, the present, vital memory of a
multitude of concrete instances, is necessary in order to make any
general proposition, be it rule of law or any other, mean anything
at all. Without the concrete instances the general proposition is
baggage, impedimenta, stuff about the feet. It not only does not
help. It hinders. 8
2. Decided Cases in Problem Cases
Decided cases also can be used in legal reasoning to solve

problem cases, with some advantages. Lawyers in practice
generally will care less about the law in the abstract than in its
practical implications for particular existing or possible future
disputes involving a particular client. Judges do not generally
enact rules; instead, they decide cases.
Effective legal planning requires a keen sense of the variety of
disputes that can arise in the future. A lawyer planning a client's
activities is engaged partly in imagining the disputes that might
arise from the client's activities and in taking precautions to
minimize losses in such disputes. Elegantly drafted contract
language may be worthy of praise for its style; it will earn none if
a foreseeable dispute arises that was not anticipated in the
7
Consider, also, cases that could not plausibly be described in the language
of the rule but that should be within the class.
8
Karl N. Llewelyn, The Bramble Bush 12 (1951).
drafting. The world has a habit of confounding even those with

perspicacious insights and vivid imaginations. A study of the cases
stimulates and supplements the imagination so that better
precautions can be taken for a greater variety of disputes that
might arise from a client's activities.
Once a dispute has arisen, the lawyer will care what a court
will or should do in that case. Pyrrhic victories, as when a court
accepts a lawyer's preferred general rule but concludes that his
client loses under that rule, usually are of little interest to the
lawyer and even less to the client. To predict what the court will do
in one case, you can look to what courts have done in other,
similar cases. To persuade a court of what it should do in one case,
you can point out what courts have done in other, similar cases. A
practice of comparing and contrasting cases may supply particular-
ities that general rules leave untreated.
Consider again the rule prohibiting any person from bringing
a vehicle into a city park. Assume that you are a judge who must
decide whether a tree surgeon violated that rule by bringing his
truck into the park to load and carry away dead tree branches
under contract with the city. You could say that the tree surgeon
violated the rule because he is a person, and he brought a vehicle
into the park. You could also say that the city, not a "person"
within the meaning of the rule, brought the vehicle into the park
through its contract with the tree surgeon. The language of the rule
allows both interpretations. How should you think your way to a
decision?
Assume further, then, that from a previously decided case you _
know, or from common sense all would agree, that an ambulance
can legally enter the park to administer aid to a stricken jogger.
Assume also that you know or we can agree that teenagers violate
the rule by racing cars in the park. Now you can ask: Is the tree
surgeon's truck case more like the ambulance case or more like the
racing case?
It would seem that you can reach a reasoned answer to the
question, though the reasons do not line up to yield that answer
as numbers line up to yield a sum. The ambulance case tells you
that the statute should not be interpreted literally, as prohibiting
all persons from bringing any vehicle into any city park under any
circumstances. There is room for interpretation. The racing case
tells you, by inference in light of common sense, that the rule seeks
to protect the park and those who seek rest and recreation in the
park from noisy and dangerous activities involving vehicles. You
might think that, like the rule, the city and the tree surgeon were
maintaining the park for the benefit of people who use it. The
short intrusion on the tranquillity of the park is small in relation
to the benefits. It would be absurd in this day to require the tree
surgeon to carry the tree branches out of the park by manual labor
or horse-drawn cart (if that is not a "vehicle"!). Therefore, you may
conclude, the tree surgeon should not be punished under the rule
as properly interpreted.
From a different perspective, a further reason can be given for
emphasizing the study of cases: The action of a court to settle a
dispute-the decision in a law case-is coercive action by the state
involving at least the threatened use of physical force. The law that
determines when that power may be used-when the sheriff may
deprive an individual of liberty or property pursuant to judicial
decree-also determines the limits of our freedom from a major
form of state compulsion. The law at the same time partly defines
permissible uses of force by the state and the scope of individual
liberty. It seeks to rule out arbitrary and oppressive uses of power
by the state while allowing justifiable uses of that power. Lawyers
are society's experts on when (in what cases) the state may use its
coercive powers within the rule of law.
Cases, much more than rules, press us to think hard about
justice, the limits of proper governmental power, and the scope of
individual freedom. It is easy, for example, to agree on a general
rule that promises should be kept. But should a court enforce your
promise to pay me $500 for the Eiffel Tower, should you decide
not to honor it? Would it make a difference if the Eiffel Tower in
mind were a stage prop I had delivered to you for use in a play?
Again, should a court enforce your promise to join me for dinner
at my club, should you decide not to keep it? Would it make a
difference if I had paid you to give an after-dinner speech or that
my club discriminates against women? What, exactly, should it
mean to "enforce" that promise when the dinner is history?
Cases display the complexities with which the law should deal.
Comparing and contrasting cases supplies the particularities
needed for lawyers to predict intelligently what a court will do or
persuade a court of what it should do in a case, and for judges
to make reasoned decisions in problem cases. Comparing and
contrasting cases requires hard, rigorous thinking about justice and

the proper role of government in a democratic society. Let us turn,
then, to the principal form of legal reasoning from cases-the
analogy.
CHAPTER TWO _ _ _ _ _ _ _ _ _ _ __
Analogical Legal
Reasoning
LEGAL REASONING TAKES two principal forms: One is analogical;
the other is deductive. These forms perform important practical
functions; some means of organizing the mass of legal materials is
essential. They can help you to identify the proper starting points
for reasoning, locate the relevant materials, and formulate issues
to focus deliberation. You will find that poorly formed legal
arguments are easily dismissed because they are hard to under-
stand. Chapters 2 and 3 will introduce these forms, beginning with
analogies.
The forms of legal reasoning, however, cannot guarantee the
soundness of a legal argument. Many well-formed arguments are
wrong, while others are less persuasive than competing well-
formed arguments. A form is like an empty vessel: Its usefulness
resides in the space where there is nothing. As a vessel can carry
wine or water, a form can carry sense or nonsense. The soundness
of a legal argument depends on how the forms of legal reasoning
are filled in-on the content of the statements in an argument.
A. The Analogical Form
Most legal problems in the United States are governed by _

either the common law or the law enacted by a legislature or an
administrative agency. The common law is the law made by judges
through their decisions in cases within their authority. Common
law making is on a case-by-case basis. The law of contracts, torts,
and property is largely common law.
The central tenet of the common law is the principle of stare
decisis: Points of law once decided in an appropriate case should
not be reopened in other cases involving the same point in the
same jurisdiction (unless something has changed that justifies
changing the law). Accordingly, decided points of law are
normally binding or authoritative and are referred to as legal
25
26 2. ANALOGICAL LEGAL REASONING
authority. The principle of stare decisis supports the common law

doctrine of precedent, which treats previously decided cases as
authorities for the decision of later cases. Reasoning under the
doctrine of precedent is mainly by analogy, which requires
basically that like cases be decided alike. 1 This is a requirement of
formal justice.
Analogical reasoning is familiar in everyday nonlegal situa-
tions. For example, Mother may allow Older Brother to stay up
until 9:00 P.M., and Younger Brother may demand the same
treatment. Younger Brother may make an argument for his view
by claiming he is like Older Brother because both are children.
Therefore, he thinks, they should be treated alike. When Mother
rejects his claim, explaining that older children need less sleep than
younger children, she is arguing that there is an important
difference between her two children. Therefore, she thinks, they
should not be treated alike. On reflection, you can imagine any
number of familiar situations in which people argue by analogy.
To make any analogical argument, you should take three steps.
First, you identify a base point situation from which to reason (e.g.,
Older Brother's bedtime). The base point consists of the relevant
facts together with a decision about what someone should do.
Second, you describe those factual respects in which the base point
situation and the problem situation (Younger Brother's bedtime)
are similar or analogous (childhood status) and different or dis-
analogous (age). Third, you judge whether the factual similarities
or the differences are more important under the circumstances.
Thus, if childhood status is more important than age in the
example, the analogy suggests that Younger Brother's bedtime
should also be 9:00 P.M. If age is more important, however, the
disanalogy suggests that an earlier time would be justified.
The second and third steps are made necessary by the simple
logic of analogies. No two people, acts, or things are alike in all
respects. The claim that two people, acts, or things are alike is not
a claim that they are identical. If identical, they would not be two
and could not be compared and contrasted at all. Nor will any two
people, acts, or things ever be different in all factual respects. If
different in all respects, they could not both be people, acts, or
1
On common law rules, see Chapter 4 §A.
2. ANALOGICAL LEGAL REASONING 27
things; comparing or contrasting them would be pointless.

Therefore, analogical reasoning requires careful consideration of
both the similarities and the differences between two situations.
The crucial point is judging whether the similarities or differences
are more important.
Analogical reasoning is highly dependent on context. It is silly
to ask whether Older Brother and Younger Brother are alike or
unalike in the abstract: They are both. Significant likeness or
unlikeness makes sense only in a concrete setting, as when bedtime
is disputed. Even then, likeness or unlikeness may change with the
circumstances. Given the fact of a three-year difference in age, for
example, the two brothers may be unalike for purposes of bedtime
at ages 3 and 6 but alike for the same purposes some years later.
And they may be different for purposes of bedtime while similar
for purposes of distributing Christmas presents fairly. (This
dependence on context is one reason why it is hard to formulate
rules that can be applied formalistically without producing some
absurd results.)
In most familiar nonlegal settings, analogical reasoning is
highly informal. The individuals involved decide what will count
as a base point, or an important similarity or difference, for
whatever reasons they find appealing. Rarely are they quite aware
that they are reasoning "by analogy" or are analyzing (1) the
propriety of a base point, (2) the factual similarities and differences
between two situations, and (3) the relative importance of the
similarities and differences under the circumstances.
B. Analogical Legal Reasoning
Analogical legal reasoning is not fundamentally different from

analogical reasoning in familiar situations. It is, however, more
formal, rigorous, and uniform in expression. What can count as a
base point or an important factual similarity or difference is
constrained by the law, in principle if not always in practice.
Underlying good legal reasoning of this kind are well-accepted
rules identifying the authoritative base points, a vocabulary and
method encouraging rigorous consideration of both similarities and
differences, and a form of expression for framing the issue to be
decided. Strict analogies, however, leave the crucial judgment of
importance-determining whether the factual similarities or

differences should control the outcome-unconstrained by the law
and open to abuse.
1. Precedents
The first step in analogical reasoning is identifying a proper

base point. In law, the doctrine of precedent gives a special status as
base points to law cases decided in the past by the highest court
in the relevant jurisdiction. 2 The U.S. legal system includes a
federal jurisdiction in which federal courts are primarily responsi-
ble for matters of national interest. The U.S. Supreme Court is the
highest court for these matters. It also includes fifty state jurisdic-
tions in which state courts carry primary responsibility for most
other matters. Each state has a highest court, usually called the
supreme court of the state. 3 The cases decided in the past by these
courts are the most authoritative precedents for deciding future
cases within the respective jurisdictions.
Other cases, though of less significance, can serve as persuasive
base points. You can use the precedents of any court (including
foreign courts) as base points for analogical argument before any
other court. If the case is not a binding precedent, it nonetheless
may be a well-reasoned decision that a court will find convincing.
You also can use hypothetical cases found in the scholarly
literature or the American Law Institute's Restatements of the Law,
and hypotheticals of your own construction when the result is
obvious to reasonable people.
This explanation of the base points in common law adjudica-
tion simplifies the rules of the game. For example, it is open to the
highest court in a problem case to overrule its own precedents. The
case that overrules then supplants the case that was overruled as
the authoritative base point for future cases of that kind, making
a change in the law. Overruling is not a common occurrence,
2
Jurisdiction refers to the scope of a court's authority to decide cases lawfully.
It may be described in territorial, citizenship, functional, or other terms.
Jurisdiction is normally established by the statute creating the relevant court.
3
In New York, the highest court is called the Court of Appeals, while the trial
and intermediate appellate courts are called the supreme courts.
though the possibility is ever-present. Moreover, the complex

relationships among courts are treated in specialized law school
courses in far greater detail than would be appropriate in this
introduction. Here, you only need understand that legal reasoning
proceeds on the basis of formal rules that identify authoritative
base points for analogical legal reasoning.
2. Factual Similarities and Differences
The second step in analogical reasoning is identifying factual

similarities and differences between the base point situation and a
problem situation. Legal reasoning by analogy uses a vocabulary
and rhetoric that emphasize the need for rigorous attention to both
relationships. In controversial cases, which are the problem cases
that require lawyerly skill, there will be many precedents that are
somewhat similar to the problem case but seem to cut both ways.
You should, by rigorous analysis of the facts of the cases, proceed
to identify the many plausible points of factual similarity and
difference. You can make a good judgment whether the similarities
or differences are more important only after you have identified all
plausible points of comparison and contrast.
We say that a judge or decision follows precedent when the facts
of a precedent are so similar to those of a problem case that the
same outcome is required (unless the earlier case is overruled). A
judge or case distinguishes precedent when the facts of a precedent
are so different that a different outcome is required. There is no
reason to presume in advance of rigorous analysis that a superfi-
cially similar precedent should be followed or distinguished. The
idea that like cases should be decided alike implies that unalike
cases should be decided unalike if the differences are more
important under the circumstances. Stare decisis requires judges to
distinguish dissimilar precedents as much as it requires them to
follow similar precedents.
In principle, then, the doctrine of precedent requires a judge to
treat each relevant authoritative precedent in one of three ways: A
judge may follow a prece.dent, distinguish it, or (if on a high
enough court) overrule it. A judge may not in good conscience
ignore a relevant authoritative precedent, though this sometimes
happens and becomes a reason for criticizing the judge. In their
arguments, lawyers similarly should advocate that each relevant

authoritative precedent be followed, distinguished, or overruled.
Legal arguments are seriously vulnerable if they ignore a relevant
and unwelcome authoritative precedent. A lawyer who knowingly
fails to disclose such a precedent can be subject to discipline by the
bar. 4
Whether a precedent should be followed or distinguished
depends in part on a careful analysis of the facts of the precedent
in relation to the facts of a problem case. The facts of a case consist
of a description of the events in the world that set the stage for the
dispute, how the parties came to find themselves in dispute, and
sometimes what the parties did to try to resolve the dispute on
their own. These are all events that normally occur before a court
is asked to settle the dispute and mostly can be described in
ordinary, nontechnical language. The facts of a case also include a
description of the legal proceedings in the lower courts, if any, as
necessary to identify the legal point that was or may be appealed
to the higher court. Most important, the legal issue on appeal
always involves the question whether the trial judge erred in
making some particular decision under the factual circumstances
in the case. The facts and legal issue establish the context in which
we make the judgment of importance at the third step.
When working by analogy, you should compile the facts of the
problem case as they have been or may be proved in court.
Analyze them until you understand them in great detail. Mastery
of the legally provable facts is crucial. The presence or absence of
a particular fact may become the point on which a precedent is
determined to be alike or unalike in an important respect. At the
early stages of legal analysis, you never know which facts will
matter. Good lawyers err on the side of compiling more facts in
greater detail than they are likely to need in the course of legal
proceedings.
Then you should locate (by legal research) the facially similar
precedents and analyze their facts. Here again, a masterful
understanding of the facts is needed, though the facts as summa-
rized in judicial opinions normally are sufficient for this purpose.
4
Model Rules of Professional Conduct Rule 3.3(a)(3) (1992).
You do not yet know which facts will or should matter when a
court decides to follow or distinguish each precedent. A good
judge will want to hear about all plausible similarities and
differences before deciding which are more important under the
circumstances.
Only then can you list the factual similarities and differences
between each precedent and your problem case. Doing so, of
course, is not a mechanical matter of finding identical statements
in the descriptions of the cases. You may summarize the facts in
your own language. This allows room for insight into relationships
among the facts that overly technical or thoughtless descriptions
sometimes mask. It also allows room for distorting the facts,
though opposing counsel and the judges can be expected to expose
such distortions. 5
3. The Judgment of Importance
The third step for analogical reasoning is determining whether

the factual similarities or the differences between the two situa-
tions are more important under the circumstances. In law, the
analogical form requires a judgment whether precedents should be
followed or distinguished (assuming away any question of over-
ruling). Unlike a layperson, however, a judge in a law case is not
free to assign importance for any reason whatever. A judge's duty
is to decide that question according to the law. But it is most difficult
to give a satisfactory account of what this might mean in common
law adjudication.
You might think that judging according to the law means
following the common law rules at this third step and, therefore,
departing from the analogical form of legal reasoning. Common
law rules are rules announced by judges in their opinions in cases
governed by the common law. These rules, you might think,
should perform two functions: They should identify the legally
important facts in advance of a case and the required legal
consequences when the important facts are presented. For example,
a common law rule might provide, "Whenever a man dies without
5
An advocate has an ethical duty of candor when before a court. Id. Rule 3.3.
a will, his property shall become the property of his eldest son."
A man's death without a will would be a possible fact. The legal
consequence of its occurrence would be the transfer of his property
to the eldest son. The law would make the judgment of importance
in advance of a case materializing; the only important fact would
be that of a man's death without a will. The judge in such a case
would not be free to decide that any other fact is important
enough to justify a different legal consequence.
To elaborate, such a common law rule would take the form (or
be translatable into the form) of a "when ... , then ... " state-
ment. For example:
When facts a, b, and care present but fact dis not, then the
defendant shall compensate the plaintiff for harms caused.
Such a legal rule would tell us that the presence of generic facts a,
b, and c together with the absence of d shall result in the defen-
dant's liability. The "when ... " clause would state the factual
conditions that require a court to order the legal consequence
stated in the "then ... " clause. Common law rules often are
stated in a form approximating this simple scheme.
This rigid deductive form of expression can be illustrated, and
its serious deficiencies in common law adjudication exposed, by an
extended example. 6 Five hypothetical cases will be described as
they might come up in one common law jurisdiction. The state-
ments of fact and reasoning are simplified representations of what
a court might state in a published opinion. Each story starts with
Costello, the original owner of five horses and a somewhat naive
and too-trusting friend of Abbott. Each horse came by theft or
fraud into Abbott's hands. Costello sought a court's aid to recover
possession of each horse from Abbott or someone who came into
possession of the horse after Abbott sold it in turn. To help the
reader keep the facts straight, a diagram of the factual relation-
ships in each case appears as Figure 2-1.
6
The illustrations are adapted from Lon L. Fuller, The Forms and Limits of
Adjudication, 92 Harv. L. Rev. 353, 375-376 (1978). See also Chapter 3.
FIGURE 2-1
Case 1: Costello ➔ Abbott by theft
Holliday by innocent purchase
Case 2: Costello ➔ Abbott by fraud
Holliday by purchase;
Holliday a party to the fraud
Ball by purchase;
Ball had heard of the fraud
CASE 1
Abbott stole Costello's horse and sold it to Holliday, who

did not know and had no reason to know it had been
stolen from Costello. Costello sued Holliday to recover the
horse. Costello won.
The court in Case 1 might state a general rule in its opinion: A

person who purchases property from a seller who did not own the
property does not acquire ownership and must return the property
to the rightful owner. The court then might apply the rule to the
facts of the case: Holliday purchased the horse from Abbott, a thief
who did not own it. Logically, this yields as a conclusion: Holliday
did not acquire ownership of the horse from Abbott and must
return it to Costello.
CASE 2
Abbott bought Costello's horse, giving as payment a forged

check on another person's account. Abbott knew the check
was forged. After delivering the horse to Abbott, Costello
discovered the fraud and sued Abbott to recover the horse.
Costello won.
The court in Case 2 might state another general rule: A person

who acquires possession of property by fraudulent purchase does
not acquire ownership and must return the property to the rightful
owner. Application of the rule: Abbott acquired possession of the
horse by purchase with a fraudulent check. Conclusion: Abbott did
not acquire ownership of the horse and must return it to Costello.
So far, so good.
CASE 3
The facts are similar to Case 2, except that Abbott sold the
horse to Holliday. Holliday knew that Abbott had bought
the horse from Costello, but she did not know or have any
reason to know that Abbott paid with a forged check.
Costello sued Holliday to recover the horse. Holliday won.
Now, if the rules stated in the precedents were "the law" that
determines the results in future cases, Case 3 would have to go the
other way. The combination of the rules from Cases 1 and 2 would
seem to require that Costello win. Case 2 states that a person who
fraudulently acquires possession of property does not acquire
ownership; Case 1 states that a person without ownership of
property cannot transfer ownership to another. In Case 3, Abbott
did not acquire ownership of the horse, under the rule in Case 2.
By finding for Holliday, the court seems to ignore the rule stated
in Case 1. It finds that Holliday owns the horse when Abbott did
not. (There are many cases on the books that decide Case 3 for
Holliday in jurisdictions that decide Cases 1 and 2 for Costello.)
The court in Case 3 could create and state a new rule that
requires the results in all three cases. It could announce, for
example, that an owner of property is entitled to possession when
he loses possession by another's wrongful act and seeks to recover
the property from the perpetrator or from a third party purchaser
when the wrongful act was theft. But surely this is quite a change
from the rule given in Case 1. Moreover, this new rule would not
justify the result in Case 3 based on the law stated in the prece-
dents. Clearly, the judgment of importance-that Holliday's
innocence in Case 3 is an important fact requiring a different result
from the first two cases-would be made by the court in Case 3,
not by the rule stated in previous cases.
How reliable, then, would be the rule stated in Case 3 for
Cases 4 and 5?
CASE4
The facts are similar to Case 3, except that Holliday had

helped Abbott perpetrate the fraud on Costello. Costello
sued Holliday to recover the horse.
CASE 5
The facts are similar to Case 3, except that, after buying the
horse from Abbott, Holliday sold and delivered it to Ball.
Ball had heard rumors of the fraud worked on Costello by
Abbott. Costello sued Ball to recover the horse.
The court in Case 3 modified the rule of Case 1 to decide Case 3.

The court in Case 4 similarly can modify the rule of Case 3 to
decide Case 4. And the same is true in Cases 5 and so on.
Consequently, the common law process is not one of simply
tracing the logical consequences of preexisting rules stated in the
precedents. As Edward H. Levi put it, "(t]he rules change as the
rules are applied. More important, the rules arise from a process
which, while comparing fact situations, creates the rules and then
applies them." 7 Judges may write as if the common law has
always been what it has come to be. Some legal thinkers, such as
Blackstone, thought of the common law like this. Others, such as
the great Oliver Wendell Holmes, Jr., belittled views of law as a
"brooding omnipresence in the sky." 8 Most of us today would
reject Blackstone's view. 9
It is misleading in another way to regard the judgment of
importance as a logical function of common law rules. Reasoning
by analogy would be analytically superfluous if there were
previously stated rules that dictated the judgment of importance.
Such a rule could simply be applied to the facts of a new case.
There would be no need to search out the precedents or analyze
7
Edward H. Levi, An Introduction to Legal Reasoning 3-4 (1948).
8
Southern Pac. Co. v. Jensen, 244 U.S. 205,218 (1917) (Holmes, J., dissenting).
9
See, e.g., Melvin A. Eisenberg, The Nature of the Common Law (1988).
the facts of the precedents to identify similarities and differences

between the precedents and the problem case. Studying cases
would not be so important. Instead, we vvould study rules and
collect them in textbooks that contained little other than the rules.
A basic principle of common law adjudication is that a judge
is empowered to decide the case before the court and only the case
before the court. A judge has no authority at common law to enact
an authoritative general rule to govern parties and situations that
were not before the court. Thus, the judge in Case 1 could not
decide the outcome in Case 3, however broadly she may craft a
rule to explain the decision in Case 1. In all likelihood, the judge
in Case 1 did not consider the facts of Cases 3 et seq. when
crafting the general rule. Its mechanical application in later cases
may yield a thoughtless and arbitrary result. More important, the
parties in the later cases are entitled to their days in court. They
should not have their rights adjudicated on the basis of the facts
and arguments put before the court by others in Case 1, due to the
happenstance that the language employed by the court in Case 1
was sufficiently general to be so used.
These characteristics of common law adjudication are rein-
forced by the practice of distinguishing between the holding of a
precedent and its dicta. "Holding" is comrnonly used in either of
two senses. A broad holding is much like a common law rule,
which states its factual part in general terms. 10 To minimize
confusion, let us call this a "ruling." A narrow holding is more
case-specific and difficult to get a handle on. For our purposes, a
holding is a statement that captures in a sentence or two the
probable significance of a single precedent as a base point for
reasoning by analogy in future cases.
A holding summarizes the important case-specific facts in the
precedent case and states the legal consequences then attached to
those facts. The important facts are those that are likely to become
a point of important similarity or difference between the precedent
and a problem case. For example, the holding of Case 1 above is
that an owner of property who is the victirn of a theft can recover
possession of the property from a third party who bought it from
the thief, even if the third party did not know or have any reason
10
See Chapter 4 §A.
to know that it was buying stolen property. Consider variations on

the facts in Case 1. Should the result be the same if Costello is the
victim of a fraud? Case 3 suggests not, so the law in Case 1 should
be confined to cases involving a theft. That is all the judge had
authority to decide.
Accordingly, without casting doubt on the holding in Case 1,
the court in Case 3 can properly hold that an owner of property
who is the victim of a fraud cannot recover possession of the
property from a third party who, not kno-wing or having reason to
know of the fraud, bought it from the perpetrator of the fraud. It
thus decides that the difference between fraud and theft is an
important difference that requires a different outcome. Similarly,
the court in Case 4 can hold that a purchaser who helped perpe-
trate a fraud on the original owner is not like the buyer in Case 3,
which decided only that the owner cannot recover possession of
the property from a third party who did not know and had no
reason to know of the fraud. It thus decides that Case 4 is more
like Case 2, in which the owner sued the perpetrator of the fraud
himself, than it is like Case 3, in which the owner sued a subse-
quent purchaser who was not complicit in the fraud.
A holding will be used as a base point in reasoning by analogtj- It
summarizes what was decided in a single case in which the judge
did not and could not decide a class of cases. A holding does not
determine the result in other cases involving different parties
under different circumstances. Those cases will be decided when
they come up. They will be treated according to the holding of a
precedent if they then are determined to be like the precedent in
the more important respects. A holding may have implications for
a class of future cases. But whether a problem case is within that
class will be decided when the case is brought before a court.
Common law rules stated in precedents are dicta to the extent
they are broader than a case-specific statement of the important
facts and the legal outcome. (Statements on points of law that need
not be decided in the case are called "obiter dicta.") Dicta lack the
status of legal authority because they purport to exceed the powers
of a common law judge. Dicta can be useful when lawyers predict
what the court will do in future cases because they often express
the court's inclinations on matters expected to come before it. But
the holding of a case has the privileged status of "the law." Dicta
forecast, in a vague and less reliable way, w·hat the law is likely to
become.
In each case as the law unfolds, then., the general rule an-
nounced in a prior case need be given effect only to the extent that
the (narrow) holding of the precedent requires. The general rule
announced in Case 1 (" A person who purchases property from a
seller who did not own the property does not acquire ownership
and must return the property to the rightful owner") can be
disregarded properly in Case 3. It is dictum to the extent it is more
general than the holding(" An owner of property who is the victim
of a theft can recover possession of the property from a third party
who bought it from the thief, even if the third party did not know
or have reason to know that it was buyilng stolen property").
Therefore, the best statement of a holding will change as subse-
quent cases are decided.
The analogical form captures significant aspects of legal
reasoning at common law (and, we will see, in other settings). The
Abbott and Costello sequence illustrates how the analogical form
in law provides a vocabulary and frames an issue for decision,
contributing to the rationality of legal thought. In Case 5, for
example, a judge should decide whether one who, having heard
rumors of the fraud, buys a horse from an innocent purchaser who
bought it from one who took possession by fraud, is more like one
who bought a horse fraudulently (Case 2) or from a thief (Case 1),
or more like one who, not knowing or having reason to know of the
fraud, bought a horse from one who took possession by fraud
(Case 3). We would expect legal arguments in Case 5 to address
that legal issue, posed in the analogical form, by parsing the
analogies and disanalogies among the cases.
But what leads a court to decide, for example, that the
difference between fraud and theft in Cases 1 and 3 is an impor-
tant difference requiring different results? Fraud and theft are alike
in some respects; both are wrongful in the eyes of the law. They
are unalike in other respects: Fraud at common law is a civil
wrong, not punishable by imprisonment, but theft is a criminal
wrong, punishable by imprisonment. The courts often consider the
difference here to be more. important than the similarity, at least
when the subsequent purchaser is an innocent one. Similarly,
Costello is the victim of wrongful behavior in all of these cases.
The precedents are unlike Case 5 in other respects, however,

because none involved a purchaser who had heard of the wrongful
behavior and bought the horse from a purchaser who was wholly
innocent and herself would win in a suit by the original owner.
The resolution of the key issues remains a mystery, so far as the
form of analogical legal reasoning suggests. 11
Analytically, analogies leave the crucial third step-the
judgment of importance-wholly unconstrained. As Professor
H.L.A. Hart put it:
[T]hough "Treat like cases alike and different cases differently"

is a central element in the idea of justice, it is by itself incomplete
and, until supplemented, cannot afford any determinate guide to
conduct. ... [U]ntil it is established what resemblances and
differences are relevant, "Treat like cases alike" must remain an
empty form. To fill it we must know when, for the purposes in
hand, cases are to be regarded as alike and what differences are
relevant.12
To be sure, in practice, some cases are so obvious that an analogy

seems to suffice without supplementation. But the interesting cases
will not be those.
To summarize, analogical legal reasoning-reasoning from

(narrow) case holdings-is but a formal version of the analogical
reasoning used in everyday life. It is governed by the principle that
like cases should be decided alike and unalike cases decided
unalike if the differences are important. This form of reasoning
requires three steps: (1) identifying an authoritative base point,
or precedent; (2) identifying factual similarities and differences
between the precedent and a problem case; and (3) judging
whether the factual similarities or the differences are more
important and, therefore, whether to follow or distinguish the
precedent. The analogical form contributes significantly to the
rationality of legal thought by providing a framework for analysis,
identifying starting points for reasoning, and framing a legal issue.
11
For further discussion of these examples, see Chapters 4 §A and 6 §C.
12
H.L.A. Hart, The Concept of Law 155 (1961).
However, judging which facts are more important remains a

mysterious activity.
I will take up this key problem-the judgment of impor-
tance-in later chapters. Let us first consider deductive legal
reasoning in an enacted law context. Deductive legal reasoning
also requires a judgment of importance. We can better confront
that difficult problem after you consider whether it is avoidable.
CHAPTER THREE _ _ _ _ _ _ _ _ _ __
Deductive Legal E~easoning

LEGAL REASONING IN THE deductive form is rnost closely associated
with reasoning from enacted law, which usually consists of general
rules. Such rules are found in a variety of official legal documents,
such as constitutions, statutes, codes, regulations, and executive
orders. 1 These laws are enacted and published by groups of
people who are authorized to make law, such as Congress, a state
legislature, or executive and administrative agencies. For conve-
nience, I will primarily discuss statutes enacted by Congress or a
state legislature in this chapter. The lessons, however, largely
apply to other enacted rules (and, as you will see in the next
chapter, common law rules).
Deductive legal reasoning differs from analogical legal
reasoning in a number of key respects. First,. reasoning starts from
a rule, not a case. Rules are enacted before cases governed by the
rule have materialized. Second, the principle of legislative suprem-
acy generally requires judges to play a subordinate role to the
more democratic branches of government. 2 Consequently, the
authoritative statement of an enacted rule remains static, as it was
enacted, until the nonjudicial lawmaker amends or repeals it.
Courts have no authority to reformulate enacted rules as the case
law interpreting the rules unfolds. Nor rnay courts ignore an
applicable enacted rule, which normally has superior authority to
any inconsistent common law. Third, the static statement of
enacted rules leads legal reasoning from such rules to focus
heavily on problems of rule interpretation. The judicial task is to
classify problem cases in classes of cases designated by the rules.
1
This form of reasoning also is employed with rules "enacted" by private
persons in the form of contracts, wills, corporate charters, and the like. What is
said in this chapter applies to private law of this kind with some modifications.
2
The principle of constitutional supremacy and the doctrine of judicial review
make out an important exception to this statement. The special interpretive
problems of constitutional law are generally set aside from the discussion
throughout this book.
43
44 3. DEDUCTIVE LEGAL REASONING
Casting legal arguments in the deductive form serves a number

of valuable functions. Legal rules establish the framework of law
within a legal system, setting up an elaborate classification scheme
that specifies the conclusions that judges might reach. Reasoning
from settled rules makes the law manageable, transforming vague
questions of what is "just" into more concrete questions that are
often uncontroversial. Even in contested cases, the rules at least
establish the starting and ending points for argument, pose the
legal issues, and establish a common language for legal discourse.
Judges are duty-bound to order remedies only when and as
authorized by the rules. Accordingly, plaintiffs and prosecutors
bear the burden of invoking a legal rule that, if applied, will vin-
dicate their legal claims. However, deductive legal reasoning can
be highly misleading: It may seem that the rules dictate the result
in a case when this is not so. Like analogical legal reasoning,
deductive reasoning does not avoid the need to judge importance.
A. The Deductive Form
Like analogical reasoning, deductive :reasoning3 is familiar in

a variety of everyday nonlegal contexts. A common point of
comparison is the rules of a game. In the board game "Monopoly,"
for example, a rule provides that any player who passes "Go"
receives $200 unless that player is on his way to "Jail." It is easy
to determine when a player passes "Go" and is not on his way to
"Jail." That player's entitlement to $200 follows automatically.
Deductive reasoning is also used extensively in more serious
settings. A teacher may announce that (only and all) students who
answer 90 percent or more of the questions correctly on a test will
receive an "A." If Martha then answers 92 percent correctly, it
should follow that she is entitled to an "A." It also follows that she
is not so entitled if she answers fewer than 90 percent correctly.
Logicians call this form of reasoning a syllogism. Its three steps
distinguish deductive from analogical reasoning. First, having
taken a first look at the factual situation, you should identify a
3
The term deductive reasoning is used here to refer to deduction in the form
of a syllogism. Other kinds of deduction are far less important in legal practice.
3. DEDUCTIVE LEGAL REASONING 45
major premise (e.g., students who score 90 percent or better on a test

shall receive an "A"). Second, you should formulate a minor
premise in the language of the major premise (Martha scored better
than 90 percent). Third, you should infer the conclusion (Martha
shall receive an "A"). The major premise is a general rule describ-
ing a class with many members, allowing a conclusion that places
an individual in the class designated by the rule. By contrast,
wholly analogical reasoning starts from an instance and dispenses
with such general categories. A conclusion in analogical reasoning
is a matter of judgment, while a deductive conclusion follows
logically from the premises.
So, to use a familiar example,
MAJOR PREMISE: All men are mortal;

MINOR PREMISE: Socrates is a man;
CONCLUSION: Socrates is mortal.
Crucially, a syllogism is valid when the conclusion must be true if

the premises are true. Only someone who is crazy or foolish would
deny that Socrates is mortal while accepting both that all men are
mortal and that Socrates is a man. A syllogism is sound, moreover,
when it is valid and the premises in fact are true. A valid syllogism
transfers the truth of the premises to the conclusion. When a prem-
ise is false, the conclusion will be true only by coincidence.
When used properly, deductive reasoning is a powerful form
of reasoning-much more powerful than the analogy. The
deductive form, however, can be easily misused. It may, by its
form, leave the impression that a conclusion is sound when this is
far from so. Look at the following silly syllogism:
MAJOR PREMISE: A foot has 39 inches.

MINOR PREMISE: Susan has a foot.
CONCLUSION: Susan has 39 inches.
The conclusion is clearly·. nonsense. But the language of the

premises yields the language of the conclusion, mimicking a
proper argument. The major premise is false. The minor premise
is uncertain i( for example, Susan in fact has two feet and that
matters in the context. The inference is not valid, in any event,

because the meaning of foot is different in the major and minor
premises (as, less obviously, is the meaning of has). Any one such
error robs the conclusion of its soundness.
Legal reasoning commonly is expressed in deductive form. But
a syllogism, however logical in appearance, is only as good as its
premises and the logical relationship between them. Validity is
necessary in legal reasoning but in itself is of trivial importance.
The key problems are: (1) identifying an authoritative major
premise; (2) formulating a truthful minor premise; and (3) drawing
a sound conclusion.
B. Deductive Legal Reasoning
Deductive legal reasoning, like its analogical cousin, is more

formal and rigorous than similar reasoning in most everyday
nonlegal contexts. For example, law practice relies on well-accepted
rules that identify the authoritative major premises. Federal law is
superior to state law; among federal laws, the Constitution is
superior to statutes and statutes are superior to administrative
rules. Among state laws, the state constitution is superior to
statutes and statutes are superior to admLinistrative rules. Due to
the principle of legislative supremacy, enacted laws in principle are
superior to a state's common law, which is made by judges. In
general, later enactments prevail over earlier enactments when
they are inconsistent. 4
Nonetheless, serious problems can be :involved in adopting the
premises and establishing an appropriate relationship between
them. At any point, judgments of importance akin to those
required in analogical reasoning can arise. Consequently, you
should learn the deductive form most carefully; if you do not,
others may use it to deceive you with its pretense of iron logic.
4
This neat picture of a hierarchy of laws can be misleading. There may be
several plausibly applicable laws within the same plane of the hierarchy and even
within the same enactment.
1. The Rule
The first step in ordinary deductive reasoning is identifying a

relevant major premise. Similarly, the first step in deductive legal
reasoning is identifying the legal rules that govern the case at
hand. Conflicts between rules aside, enacted rules themselves
describe the classes of cases in which they apply. On first encoun-
ter, you may think that a rule is easy to apply. Upon careful
reading and analysis, however, you will often find that this
impression is misleading.
To illustrate, assume a problem case as it might appear to a
lawyer from an initial interview with a client in the law office. We
will work through part of the relevant statute as the lawyer might.
Note that the discussion begins with the facts before searching for
the applicable rule. These facts are stated wholly in ordinary,
nontechnical language and give only a superficial account of what
happened. After locating a plausibly applicable rule, a lawyer
might have to search out additional facts in greater detail to
formulate minor premises that are suitable for reasoning deduc-
tively to a conclusion. The first impression that a particular rule
applies could easily be wrong.
Assume that Franny Farmer grows peaches at her small,
family-run orchard in rural Georgia. She contacted Morris Auster,
a buyer of produce for distribution in the Atlanta area. Farmer and
Auster met on May 28 at the orchard. After inspecting the young
fruits, they reached an agreement for Farrrler to sell Auster 200
boxes of peaches per week for three weeks in July, from Farmer's
orchard, at a price of $40.25 per box. They shook hands and
departed, not having put their agreement into writing. On June 3,
rural Georgia was struck by severe rain, winds, and flooding,
damaging much of the state's peach crop for the season. Thinking
the shortage of peaches would cause the price to rise, Auster sent
a signed letter to Farmer on June 10, stating that it was "to remind
you of our agreement for you to sell me 200 boxes of peaches per
week for three weeks in July. $40.25 per box." Farmer did not
reply. In July she sold all of her peaches to Bert Berkowitz, at
$50.75 per box.
Auster sees his lawyer. He wants to sue Farmer to enforce the
contract. To determine whether it would be ,worthwhile to litigate,
Auster's lawyer needs to predict what a court is likely to do in this
case. This is a commercial transaction governed by a statute known

as the Uniform Commercial Code (UCC). 5 On consulting the UCC,
the lawyer will find dozens of statutory sections grouped in ten
general categories, called articles. Which provisions might govern
Auster's case? Auster's lawyer will see that Article 2 is titled
"Sales." His attention will focus here.
The lawyer will read the first provisions of Article 2 to
determine whether it governs Auster' s case. He will find that an
early section provides that "this Article applies to transactions in
goods" (§2-102). In the deductive form, he will ask if peaches are
"goods'' within the meaning of this provision. This is not a
difficult legal issue. A nearby section of Article 2 provides that
goods means "all things which are movable" and "includes growing
crops" (§2-105(1)). The peaches in Auster's case were "growing
crops" and therefore "goods" within the meaning of Article 2. It
applies. Auster's lawyer has narrowed the search to the class of
commercial transactions involving sales of goods and therefore to
the provisions of Article 2 of the UCC. He has excluded the many
rules that govern sales of land, sales of corporate stock, sales of
services, other kinds of sales, and other commercial transactions.
Auster's lawyer will then read through Article 2 to further
narrow the scope of his problem. On reading §2-201(1), he might
be disappointed to find that a class of contracts for the sale of
goods must be in writing to be enforceable:
Except as otherwise provided in this section a contract for the

sale of goods for the price of $500 or more is not enforce-
able ... unless there is some writing sufficient to indicate that a
contract for sale has been made between the parties and signed
by the party against whom enforcement is sought.
It might appear from this rule that the Auster /Farmer contract will
fall in the class of unenforceable contracts, much to Auster's
chagrin. The contract is for the sale of goods at a price of more
5
The Uniform Commercial Code has been enacted as statutory law in forty-
nine states and the District of Columbia. It was drafted and promoted by two
nongovernmental organizations of the nation's leading lawyers, judges, and legal
scholars-the National Conference of Commissioners on Uniform State Laws and
the American Law Institute.
than $500. Farmer is the "party against ,Nhom enforcement is

sought" in the action by Auster to enforce the contract. She did not
sign a writing of any kind. However, this would be a hasty
conclusion: The rule applies "except as otherwise provided" in
§2-201. Auster's lawyer should read on.
The second subsection of §2-201 identifies a class of contracts
that are enforceable even if they are for sales of goods for $500 or
more and there is no writing signed by the party against whom
enforcement is sought. The rule is a fairly complicated one. It
amounts to saying that some people wlho receive a writing
confirming a deal must object to the confirnnation promptly or be
precluded from using §2-201(1) to defeat enforcement of the deal.
Section 2-201(2) provides:
Between merchants if within a reasonable time a writing in

confirmation of the contract and sufficient against the sender is
received and the party receiving it has reason to know its
contents, it satisfies the requirements of subsection (1) against
such party unless written notice of objection to its contents is
given within ten days after it is received.
This rule might apply in Auster's case. Auster sent Farmer a letter
to remind her of the deal. She did not reply. Auster's lawyer might
use the rule in §2-201(2) as a major premise and formulate a minor
premise that is faithful to the facts while satisfying each element
of the major premise. If so, he could argue that Auster's case falls
within the class of enforceable contracts designated by §2-201(2),
not the class of unenforceable contracts designated by §2-201(1).
Auster's lawyer knows from these rules that to get a court to
enforce the contract (so far as the requirement of a writing is
concerned) he must show that the Auster /Farmer contract was
"between merchants," that Auster's letter was a "writing in confir-
mation," that it was "received" by Farmer "within a reasonable
time," and so on. Farmer's lawyer knows as well that she can
block enforcement by showing that the Auster /Farmer contract
was not "between merchants," or that Auster's letter was not a
"writing in confirmation," or that Farmer did not "receive" the
letter "within a reasonable time," and so on. A judge similarly
knows that she must make a finding on each of the several issues
to decide whether the requirements of §2-201(2) were met in
Auster's case.
The rules as stated in an enactment, however, only identify the

legal issues. Framing the issues is no small contribution. Until
these issues are resolved, however, the lawyers do not know
whether the Auster /Farmer contract is unenforceable by Auster
because it is not in writing. The rules do not determine the scope
of their own applications.
2. The Facts
The second step in deductive legal reasoning is stating the facts

in a way that allows an inference leading to a valid conclusion.
Here you may confront a problem of considerable complexity: The
facts in any case can be described in a variety of terms. Some
plausible descriptions will fit the terms of an authoritative rule
with little or no discomfort, as Farmer's peaches seem clearly to be
"growing crops" and therefore "goods" vvithin the meaning of that
term in Article 2. Others, however, will fit the rule only awkward-
ly or with incompatible implications.
I will use Auster's case to illustrate this step as well. To focus
your attention on characterizing the facts, let us simplify the issues.
From the previous discussion, we can derive a somewhat unwieldy
rule that can be stated as a major premise:
MAJOR PREMISE: Whenever a contract for the sale of goods

is between merchants, and within a reasonable time a
writing in confirmation of the contract and sufficient
against the sender was received, and the party receiving it
had reason to know its contents, and notice of objection to
its contents was not given within ten days after it was
received, then the requirement of UCC §2-201(1) is satisfied
against the recipient.
The statement of facts stipulates that Farmer did not reply to

Auster's letter of June 10. Assume further that Auster's letter was
a "writing in confirmation of the contract,." was "sufficient against
the sender," was "received" "within a reasonable time" by Farmer,
and that Farmer had "reason to know· its contents." A more
manageable, if artificial, major premise now can be formulated for
illustrative purposes:
MAJOR PREMISE: When and only when a contract is between

merchants and ... , then the requirements of §2-201(1) are
satisfied against the recipient.
To reason from this premise deductively,. Auster's lawyer must

formulate one of two possible minor premises: Either the
Auster /Farmer contract was or was not bernreen merchants. Given
the major premise, no other minor premise allows an inference to
a valid conclusion. Because the two possible minor premises
contradict each other, one of only two conclusions is possible:
Either the requirements of §2-201(1) were or were not satisfied
against Farmer. Auster will win the point only if the first minor
premise is found to be true as a matter of fact.
MAJOR PREMISE: When and only when a contract is

between merchants and ... , then the requirements
of §2-201(1) are satisfied against the recipient.
The
MINOR PREMISE 1: MINOR PREMISE2 :The
Auster /Farmer contract Auster /Farmer contract
was between merchants. was not between mer-
chants.
CONCLUSION 1: The require- CONCLUSION 2: The re-

ments of §2-201(1) were quirernents of §2-201(1)
satisfied against Farmer. were not satisfied
against Farmer.
Which minor premise is true? Auster's lawyer can say that

either is true; he can pick the one that favors his client's interest.
Insofar as anyone else is concerned, however, this only identifies
a logically possible minor premise. The opposite minor premise
and conclusion is also logically possible. Therefore, merely saying
that one of them is true will not convince anyone else, even if his
logic is valid. In particular, Auster's lawyer will not convince a
judge or Farmer's lawyer unless he establishes a sound argument
based on facts that can be evidenced at a trial.
The problem, then, is to determine whether the Auster /Farmer

contract in fact was "between merchants." But which of the two
characterizations of the facts is proper? What facts would be
relevant to a decision? The facts of a case do not come prepack-
aged in the language of the rule. The rule may not even contain
criteria for its own application. Interpretation is surely required in
any case.
3. The Judgment of Importance
Interpretation, in the narrow sense, is the intellectual process of

giving meaning to linguistic symbols, such as the words or
combinations of words that are found in legal rules. Certain
common problems of language make interpretation necessary. Few,
if any, words have one and only one meaning so that they refer to
some object in the world with certainty. Proper names come closest
to that ideal. Most words, however, suffer from a lack of clarity in
one or more of several ways. Words rnay have two or more
identifiable meanings-as previously mentioned, the word bar is
an example. Such words are called ambiguous. Words may have
meanings that shade continuously from one to another with no
line of demarcation-orange and yellow are examples. Such words
are called vague. And groups of words in a sentence may cause
similar clarity problems. In the sentence The house had a gazebo in
the yard which was yellow, was it the house or the gazebo or the
yard that was yellow? Such a lack of clarity may be called sentence
ambiguity. All three kinds of unclear meanings can require
interpretation of a legal rule.
In everyday life, we normally think of the meaning of a word
as its definition. Legal definitions of legal terms can be useful in
easy cases. However, interpretation through definitions often
transforms the problem of interpreting a rule into a problem of
interpreting the definitions of its words, transforms a problem of
interpreting the definitions into one of interpreting the definitions
of a definition, and so on. When the last of the available definitions
is found, there may still be vagueness or ambiguity.
In Auster's case, the term between nzerchants is defined in
§2-104(3) of Article 2 of the UCC:
"Between merchants" means in any transaction with respect to

which both parties are chargeable with the knowledge or skill of
merchants.
The case is a bit unusual because we have a statutory definition of

a part of a definition. Merchant is defined in §2-104(1) in relevant
part as follows:
"Merchant" means a person who deals in goods of the kind or

otherwise holds himself out as having knowledge or skill
peculiar to the practices or goods involved in the transaction ....
These definitions are helpful in further narrowing the scope of

the problem. But they do not solve it. Auster is probably charge-
able as a "merchant" because he "deals in goods of the kind." As
a middleman between the growers and the Atlanta market, he
probably makes several hundred produce purchases and sales and
several dozen peach purchases and sales in any year. He is
reasonably expected to know his trade well. Consequently, we can
safely conclude, Auster should be chargeable as a "merchant."
For the contract to be enforceable against Farmer, however,
"both parties [must] be chargeable with the knowledge and skill of
merchants." It is far from clear whether Far:mer is chargeable as a
"merchant." She may make as few as one peach sale for a few
deliveries in a year. Her orchard was described, you will recall, as
a "small, family-run" business. She may not be a very sophisticat-
ed businessperson at all. Perhaps no one vvould expect that she
should be. Farmer might "deal in goods of the kind" or might
"otherwise hold [her ]self out as having kno wledge or skill" of the
1
kind described. Or she might not.

The definitions are like a rule in a significant respect: The
definitions themselves may require interpretation because they
suffer from the problems of language. In the first part of the
definition of merchant, the phrase goods of the kind may refer only
to peaches or also to fruits, produce, or foodstuffs in Auster's case.
This phrase is ambiguous. Moreover, Farmer might or might not
deal in peaches. The word deals is vague because it is not known
how much or what kind of business activity would make Farmer
a dealer; her level of business activity is on a continuum of sorts
from a consumer (who is not a merchant) to a big-time middleman
like Auster (who surely is). And the phrase between merchants
means "in any transaction with respect t:o which both parties are
chargeable with the knowledge or skill of merchants," not more
simply when both parties "are merchants." This suffers from a sort
of sentence ambiguity: One who surely "deals in goods of the
kind" might or might not be "chargeable with the knowledge or
skill of merchants" if she does not also "hold [her]self out as
having knowledge or skill peculiar to the practices or goods
involved in the transaction." The definition of "between mer-
chants" might refer to both parts of the definition of merchant, or
it might refer only to the second part.
A judgment of importance is required: What facts would justify
charging Farmer as a merchant because she "deals in goods of the
kind"? Is Farmer more like Auster the merchant, or more like
Bessie, a consumer who may make more peach purchases in a year
than Farmer makes sales? Is the number of sales or purchases the
important fact? The quantities involved? Is the difference between
sales and purchases important? Is the fact that Auster both
purchases and sells important? What about the fact that Auster
and Farmer act for a profit, while Bessie does not? Is the fact that
Auster deals in produce while Farmer deals-if at all-only in
peaches important?
Similarly, what are the important facts that would establish
whether Farmer is chargeable as a merchant because she "holds
[her ]self out as having knowledge or skill peculiar to the practices
or goods involved in the transaction"? Must she say to Auster that
she knows the business practice of sending and responding to
confirming letters-the "practices ... involved in this transac-
tion"? Must she say to Auster that she knows produce well-the
"goods involved in the transaction"? Or would it be enough that
she introduced herself to him as a businesswoman who grows and
sells peaches in small quantities, or that she has a sign at the
entrance to her orchard announcing to the world "FRANNY'S OR-
CHARDS-EXPERTS IN PEACHES"?
The rules and their definitions do not point directly at the
important facts in the case-those that establish whether Farmer
is a merchant such that the Auster /Farmer contract was "between
merchants." The rules and their definitions do not provide the
particularities you need to resolve the controversy by deduction
from the rule stated in §2-201(2). Moving from the rule to the
definition of terms in the rule, to definitions of the definitions,

keeps you in the realm of generalities. Eventually, you exhaust the
rules and definitions that purport to determine which argument in
a case is the correct one by deduction. 6 Then you must make a
judgment of importance to classify a problem case.
To help a statutory rule perform its functions, deductive legal
reasoning employs a number of source materials other than
definitions. These other materials also do not supply the particular-
ities we need. For example, some say that an unclear statute
should be interpreted to give effect to the intention of the legisla-
ture. This "intention" is well known to be a fictional thing. A
legislature is a group of people; groups themselves do not have
intentions. The remarks in debate by a bill's floor leader or the
report of a single legislative committee or a statement by a
legislator at a hearing each do not necessarily represent the
intention of the deliberative body as such. Lacking such necessity,
deductive reasoning from such statements will not yield a
necessary conclusion. Only the language of the rules is approved
by the group, and that returns us to the starting point of our
analysis, caught in a circle.
Moreover, if a reliable picture of the legislature's intention
could be pieced together, what form would it take? It could be
another general statement of what the law permits or requires in
a class of cases-a rule-which circles back into the problems of
language. It could be an example of the intended effect of the rule
in a hypothetical or historical situation, leaving one to reason
analogically and with the need to judge importance. Or it could be
a statement of the rule's objective or purpose, though these, too,
will be like rules or cases and circle back into the problem of
importance. In any event, the "intention of the legislature" will not
6
Lawyers sometimes consult a dictionary of English language which supplies,
at best, another rule, which leads to the same problem one syllogism down the
road. It more often supplies a number of definitions among which one must
choose. The drafters of statutes often do not intend for the ordinary meaning of
a word to be employed, especially when a term has been given a statutory
definition that turns it into a term of art within the statute. As in the example,
lawyers sometimes are less concerned with a single word than with a phrase or
sentence or, as will be seen in Chapter 7, much more.
supply the particularities we need to reason deductively to a

necessary conclusion in a controversial case.
A number of other kinds of guidance may be available for
interpreting a statutory text. Perhaps statutes should be read as a
whole to give them coherent meanings, avoiding contradictions
between provisions or leaving any provision without a point.
Perhaps the legislative history, including the floor debates,
committee reports, and hearings before its enactment, should be
taken into account. Perhaps statutes should be interpreted in light
of legal history-how the common lavv and other legislatures
treated the problem in other times and places. And perhaps
statutes should be interpreted in light of the historical, economic,
and social circumstances at the time of its enactment.
All of these sources of information are often helpful and in fact
are used in good legal reasoning, as we will see in Chapter 4.
However, the commonly expressed comjmandment to "take into
account" such a wide variety of things generates considerable
frustration. We feel that we have been told to think about every-
thing before doing anything. That is a formula to ensure that we
do nothing. We need to know what we are looking for in the enacted
text as a whole, the complex legislative history, the centuries of
legal history, or the multitude of historical, economic, and social
circumstances at the time of enactment.
Surely none of these sources of information will supply the
particularities we need to formulate minor premises allowing
deductive reasoning from an unclear legal rule to a sound
conclusion. Some may provide examples of situations (historical or
hypothetical) and the "intended" effect of the general rule in those
situations. Again, however, we can use that information analogical-
ly. Some may provide statements that are abstract and refer to
classes of cases. That, again, will stop short of logically determin-
ing a case's membership in the abstract class. In either event, we
are left with the need to judge importance.
The judgment of importance requires us to go beyond the
problems of language that inhere in legal rules, definitions, and
similar expressions. This should not be surprising. As indicated at
the outset, a rule is an abstract or general statement of what the
law permits or requires of classes of people in classes of circum-
stances-in classes of cases. A rule stands in contrast to a case-a
short story of an incident in which the state acted or may act to
settle a particular dispute. Accordingly, the language of an enacted

rule, announced before any case governed by the rule has material-
ized, describes an abstract class. The staternent of conditions (the
"when ... " clause) points at the class of cases, not at the particu-
lar facts of any problem case.
When a word like merchant is merely copied from the condi-
tioning part of a major premise and placed in the position of a
minor premise, it does not suddenly point at the particular facts in
a case. Instead, it continues to point at a class of cases. Such a use
of words leaves the minor premise, so to speak, dangling in the
air. The connection between the abstract class and the case remains
to be drawn; in the example, it remains to determine whether
Farmer, in particular, is a member of the class of people "charge-
able as merchants." And it should be dravvn by some process of
reasoning that is not dependent on other rules or definitions that
themselves will similarly dangle. As Justice Holmes said long ago
from the Supreme Court, "[g]eneral propositions do not decide
concrete cases. " 7
Drawing this all-important connection--placing a case in a
legal class-requires a judgment of importance to mark the
particular facts that justify the classification. Deductive legal
reasoning does not indicate how to make this judgment in the
real-world situations that interest lawyers and judges. The form of
deductive legal reasoning promises a conclusion backed by logical
necessity. At least in controversial cases, the problem of character-
izing the facts of a case in the language of the rule robs the
syllogism of its necessity.
Like analogical legal reasoning from cases, deductive legal
reasoning from rules has its uses and abuses. Good lawyers and
judges neither accept these forms nor rejec1t them as forms. They
use them to benefit from their strengths and supplement them to
avoid their weaknesses. You should understand their plusses and
minuses if you are to make good arguments and see through bad
ones. Mastering them requires you to understand, crucially, the
points in reasoning where a judgment of importance remains
unconstrained by officially stated legal rules or definitions.
We still must confront the judgment of importance. First,
7
Lochner v. New York, 198 U.S. 45, 76 (1905) (Holmes, J., dissenting).
however, let us consider the relationships between the deductive

and analogical forms. The combination of the two principal forms
of legal reasoning contributes further to the rationality of legal
thought. But it does not constrain the judgment of importance.
CHAPTER FOUR _ _ _ _ _ _ _ _ _ __
Combining Analo~~ies
and Deductions
To SIMPLIFY THIS introduction to law and legal reasoning, I have
separated analogical and deductive reasoning too completely. True,
analogical legal reasoning is characteristic of the common law, and
deductive legal reasoning is characteristic of enacted law. But good
lawyers also use common law rules deductively and apply enacted
rules through analogical reasoning from cases and other base
points. This chapter will show you how to combine the deductive
and analogical forms.
Combining the two forms of legal reasoning serves some
highly useful functions. When you are working through a common
law problem, the analogies become increasingly unwieldy as the
number of precedents increases. Common law rules allow you to
express yourself economically and to organize the precedents.
When you are working through an enacted law problem, you will
find, the rules often lack particularities needed to determine the
class to which a problem case belongs. The context, however, may
include examples of the rule's intended or accepted effect. You can
use these situations as base points for reasoning analogically to
interpret and apply the rule.
But combining the two forms does not allow you to dodge the
need to judge importance. If your resources were limited to rules,
precedents, and logic, legal reasoning would send you back and
forth between analogies and deductions in a regress, as when you
stand between two tilted mirrors. You could start with the
analogical form, reach the judgment of importance, and shift to the
deductive form using a common law rule. The problem of
importance, however, would reappear in the deductive form. If
you start with the deductive form, reach the judgment of impor-
tance, and shift to the analogical form, the judgment of importance
remains as well. This conundrum reflects how the two forms
amount to two ways of expressing the same thought: Treating like
cases alike, as in analogies, is much like applying rules consistent-
ly, as in deductions. Fortunately, you can escape the conundrum
59
60 4. COMBINING ANALOGIES AND DEDUCTIONS
by using additional resources in legal reasoning, as we shall see in

Chapters 5 and 6.
A. Common Law Rules
In Chapter 2, we considered analogical legal reasoning in

common law cases. You learned that, in this kind of reasoning, we
(1) identify the relevant precedents, (2) compare and contrast the
facts of a problem case with the facts of the precedent cases, and
(3) judge whether the factual similarities or differences are more
important in the circumstances. The third step is the troubling one:
Judges may claim to be following a previously stated common law
rule when deciding a case. But the rule is dicta, and thus not law,
to the extent it goes beyond the previous judges' authority to
decide the case before them. Consequently, we saw, rules stated in
common law precedents do not determine the outcome in subse-
quent cases.
1. Easier Cases
It would be a mistake, however, to think that common law

rules serve no useful purpose. Unlike enacted rules, common law
rules have roots in the case law: They stand for the holdings of a
group of precedents. You can refer to the group by using the
words and phrases of the common law rule. With this in mind,
you can use the rules in deductively formed arguments for
convenience in easy cases. There is no point to articulating many
obvious analogies when a simple deductive argument will satisfy
those who are concerned.
To illustrate, consider again the Abbott and Costello sequence. 1
To refresh your memory, all of the cases concerned the rights of
the original owner of a horse (Costello) 1to recover possession of
the horse when possession was taken by a person (Abbott) using
1
See Chapter 2 §B.
4. COMBINING ANALOGIES AND DEDUCTIONS 61
wrongful means (theft or fraud). The first three cases held that (1)
an owner of property who is the victim of a theft can recover
the thief, even if the third party did not know or have any reason
to know that the property had been stolen; (2) an owner of
property who is the victim of a fraud can recover possession of the
property from the person who perpetrated the fraud; and (3) an
owner of property who is the victim of a fraud cannot recover
the perpetrator when the third party did not know or have any
reason to know of the fraud.
The court in Case 3 might state a general rule that summarizes
these three holdings: An owner of property is entitled to posses-
sion when he loses possession by another's V11rongful act and seeks
to recover the property from the perpetrator or from a third party
purchaser when the wrongful act was theft. We know from
Chapter 2 that this rule would not be reliable in some possible
future cases. There are cases, however, where the rule can yield an
uncontroversial conclusion-one that passes without objection
among those concerned. The deductive form may be the most
simple way to express the law in such cases.
For example, assume Abbott steals a horse from Costello.
Costello sues Abbott to recover possession of the horse. Of course
Costello should succeed, though no precedent is directly on point.
Costello's lawyer may present his argument to Abbott's lawyer in
the deductive form.
MAJOR PREMISE: Case 3 ruled that an owner of property

is entitled to possession when he loses possession by
another's wrongful act and seeks to recover possession
from the perpetrator of the wrongful act or from a third
party purchaser when the wrongful act ·was theft.
MINOR PREMISE: Costello lost possession of his horse to

Abbott by theft, a wrongful act, and seeks recovery of
possession from Abbotti the perpetrator.
CONCLUSION: Costello is entitled to recover possession of

the horse from Abbott.
62 4. COM131NING ANALOGIES AND DEDUCTIONS
Abbott's lawyer is likely to advise Abbott to return the horse

even though he knows that this conclusion is not required by the
rule stated in Case 3. That rule is dicta insofar as it is broader than
its roots in the precedents. Case 1 held only that the original owner
can recover possession from a third party who bought it from a
thief. Case 2 held only that the original owner can recover
possession from the perpetrator of a fraud. Neither precedent held
that the original owner can recover possession from a thief, which
is the issue now presented. Abbott's la"'ryer, however, could not
argue convincingly that this case is different from the precedents
in any important respect. Rather, this case is an even clearer case
requiring recovery by the original owner. Abbott's lawyer should
concede the point because there is no plausible argument with
which he can contest it.
The deductive reasoning in such a case is short and sweet and
to the point. Full expression of the analogical reasoning yielding
the same conclusion could show the obvious in a more complex
and pedantic way. To do so would be unnecessary, confusing, and
time consuming for all concerned.
What makes this case easy, however, is not the deductive form
of Costello's argument. No court can stop the law's evolution by
announcing a rule in a single case. Rather, a common law rule
summarizes a number of precedents. The precedents have the
superior authority: They retain their significance as base points for
analogies. If a stated common law rule leads logically to a different
result from the one supported analogically by its underlying
precedents, then so much the worse for the common law rule.
2. Organizing the Precedents
Common law rules have other useful functions, too. Not the
least of these is providing a framework for organizing what may
be dozens or hundreds of precedents. Reasoning by analogy from
each and every precedent would be over"'rhelming. Composing an
intelligible argument would be humanly impossible. To make the
thought process manageable, we simplify by organizing the large
number of relevant precedents into subclasses of precedents, each
with fewer members. We then name each subclass with a phrase
in a common law rule, which should capture the significance of the
precedents it covers. You can then locate (by legal research) or

refer (in argument) to that subclass of precedents quickly and
easily by using the right phrase.
Consider, for example, the general common law rule governing
cases in which a plaintiff seeks compensation from a defendant
whose careless behavior harmed the plaintiff or the plaintiff's
property. The traditional lawsuit arising at common law from an
automobile accident is a familiar example. The rule, known as the
common law negligence rule, may be stated as follows:
If (1) the defendant was under a duty to use reasonable

care to avoid a harm, and (2) the defendant breached that
duty by a negligent act or omission, and (3) the breach of
duty was a cause-in-fact and (4) a legal cause of (5) damage
to the plaintiff's person or property, then the defendant
shall compensate the plaintiff for such damages (unless
certain exceptions not relevant here apply). 2
In a single jurisdiction, there could be a few hundred prece-

dents supporting this negligence rule. These precedents are
organized into five subclasses, each of which is named by an
element of the rule. Thus, in shorthand form,. there are the (1) duty
cases, (2) breach of duty cases, (3) cause-in-fact cases, (4) legal
cause cases, and (5) damage cases. These subclasses are distin-
guished from one another by the kind of issue that was decided by
the court. The precedents in each subclass should be more like
each other (in respect to the issue that was decided) than they are
like the precedents in the other subclasses (in the same respect).
The negligence rule indicates that, to recover compensation, the
plaintiff in a negligence case must show that her situation is like
that of the plaintiffs who won in all five subclasses. Assume that
Georgio's car collides with Susan's when Georgio is admiring the
scenery instead of watching the road. Susan's lawyer knows that,
strictly speaking, five analogical arguments are necessary. She
2
The distinction between ca11sc-injacl and legal cause (also called proximate
cause) is not important for present purposes, as long as the reader accepts that
there is a difference.
must show that (1) Susan's situation is like that of the plaintiffs
who succeeded in showing that the defendant had a duty in cases
where that question was decided. She must show in addition that
(2) Susan's situation is like that of the plaintiffs who succeeded in
showing that the defendant breached its duty in the cases where
that question was decided. Susan's lawyer also must show that
Susan's situation is like that of the plaintiffs who succeeded in
showing that the defendant's breach of duty was (3) the cause-in-
fact and (4) the legal cause of (5) the damage suffered by the
plaintiff. Georgia's lawyer can defeat Susan's claim by showing
that Susan's situation is unlike those of the victorious plaintiffs, or
that Georgia's situation is like those of the victorious defendants,
in any one of those five subclasses.
Most problem cases do not require the court to analyze the
analogies in all subclasses named by a rule. The only contested
issue in Susan's case might be whether the defendant's negligence
was a legal cause of Susan's injury. That is, the parties might
accept that Susan's situation is like that of the plaintiffs who won
in the precedents encompassed by the other four elements of the
rule. The court now must decide if Georgia's breach of duty was
the legal cause of Susan's injury; that is, ,Nhether Susan's situation
is like that of the plaintiffs who won on the legal cause issue in the
legal cause cases.
Consequently, an argument at comn1on law can be stated in
the form of a legal syllogism in any case:
MAJOR PREMISE: If (1) the defendant was under a duty to

use reasonable care to avoid a harm, and (2) the defendant
breached that duty by a negligent act or omission, and (3)
the breach of duty was a cause-in-fact and (4) a legal cause
of (5) damage to the plaintiff's person or property, then the
defendant shall compensate the plaintiff for such damages.
MINOR PREMISE: Georgia (1) had a duty to operate his car

with reasonable care, (2) he breached this duty by failing to
maintain a proper lookout, (3) which in fact caused a
collision with Susan's car, (4) his breach was the legal cause
of injury to Susan, and (5) it damaged Susan's foot.
CONCLUSION: Georgia shall compensate Susan for the

damage to her foot.
This represents a deductive argument from the common law rule

only in a limited way. The logic of the rule as stated establishes
that Susan's lawyer must make five successful arguments, setting
aside any question of overruling. Assuming that Susan's lawyer
makes all five arguments, however, the soundness of the conclu-
sion cannot be established by the rule and the logic alone. Because
soundness depends on the underlying analogies, you must look
and see whether the analogies support the minor premise.
Therefore, the legal syllogism does not represent the full reasoning
required to decide Susan's case. Rather, the syllogism represents
abstractly the results of a highly complex web of analogical argu-
ments from a large number of precedents.
In sum, the common law precedents are organized in a
classification scheme described by the common law rules. Like
enacted rules, the common law rules refer to classes of cases.
Unlike enacted rules, however, common law· rules refer to classes
of precedents already decided by courts. Common law rules are
used deductively to express conclusions economically and to
identify precedents to consider analogically. Analytically, however,
the action is with the underlying analogies, even when they are left
unstated. Consequently, combining common law precedents and
rules does not constrain the judgment of importance.
B. Cases and Enacted Rules
You learned in Chapter 3 that legal reasoning in the deductive

form requires lawyers and judges to (1) identify an authoritative
and applicable rule that serves as a major premise, (2) describe the
facts of a problem case in a minor premise, and (3) interpret the
rule in relation to the facts to reach a sound conclusion. In many
situations, however, a conclusion deduced from an enacted law is
not sound. This is because, when enacted, the rules refer to classes
of cases that have not yet materialized. They consequently may not
supply the particularities you need to place a problem case in these
classes. Classification depends on a judgment of importance that
is not determined by the general language of the rule.
You can employ analogical legal reasoning to help interpret an

enacted rule in relation to the facts of a problem case. Enacted
rules, however, differ from common law rules because they usually
lack roots in the case law. Consequently, judicial precedents using
the rule are not available as base points until some time after
enactment. Therefore, reasoning analogically requires other kinds
of base points with which to compare and contrast a problem case.
Relevant base points include both real and hypothetical situations.
But not just any situation will shed light on the meaning of a rule.
There should be a good reason for thinking that the situations are
examples of what the rule permits or requires. Several kinds of
base points can be employed meaningfully, though such base
points do not have quite the authority of a precedent at common
law.
The text of an enacted rule is the starting point. The text has a
context that includes examples of its intended or accepted effect
in concrete situations. Most obviously, the context includes the
ordinary meanings of the terms used in the text. Examples that
normally come to mind can be used as base points, though it is
often not necessary to spell this out. Thus, if a rule prohibits
anyone from bringing a "vehicle" into the city park, a normal Ford
Taurus can be used as a base point for analogies to identify less
obviously prohibited vehicles.
The ordinary meanings, however, tend to be helpful only in
easy cases. The interesting problems are those in which the
ordinary meaning of the text is vague, ambiguous, or leads to an
absurd conclusion. To illustrate, consider the Speech Clause of the
First Amendment to the U.S. Constitution: "Congress shall make
no law ... abridging the freedom of speech .... " Any lawyer
reasoning from this general rule deductively can be seriously
misled. For example, the language on its face expresses a prohibi-
tion only on certain acts by Congress . But, together with the
Fourteenth Amendment, it also prohibits such acts by the states.3
There is no need to digress to explain how that came about, since
it is settled law for now. A more straightforward problem will be
posed.
3
The Speech Clause is "incorporated" into the Due Process clause of the
Fourteenth Amendment, which applies by its terms to the states. See Gitlow v.
New York, 268 U.S. 652 (1925); Whitney v. California, 274 U.S. 357 (1927).
4. COMBINING ANALOGIIES AND DEDUCTIONS 67
Assume two problem cases.
PROTEST'S CASE:
A state enacted a statute making it a crime for anyone to

hang the governor in effigy. Peter Protest was prosecuted
under the penal statute for hanging the governor in effigy,
which he admitted doing. Protest was convicted and
sentenced to a jail term. He argued unsuccessfully in state
court that his conviction was unconstitutional and should
be set aside because the statute violated the Speech Clause.
Protest appealed to the U.S. Supreme Court on that ground.
NIHIL'S CASE:
A state enacted a statute making it a crime with special

penalties for anyone to hang the governor. Neil Nihil was
prosecuted under the statute for hanging the governor,
which he admitted doing. Nihil was convicted and sen-
tenced to a jail term. He argued unsuccessfully in state
court that his conviction was unconstitutional and should
be set aside because the statute violated the Speech Clause.
Nihil, too, appealed to the U.S. Suprerne Court on that
ground.
By coincidence, the two cases reach the Supreme Court at the same
time. Certainly, the Supreme Court would and should set aside
Protest's conviction but not Nihil' s. A s1tatute that penalizes
hanging the governor in effigy abridges the freedom of speech, but
a statute that penalizes hanging the governor does not. The
problem is to explain the reqsoning that justifies these conclusions.
This may not be easy. You cannot justify setting aside Protest's
conviction by an argument that also requires setting aside Nihil's.
Nor can you justify convicting Nihil with an argument that also
requires convicting Protest. The rule of law disallows this kind of

serious inconsistency. Therefore, you nn.ust draw a distinction
between the two cases.
You cannot use the text of the Speech Clause deductively to
draw the needed distinction. The text leads deductively to the
conclusion that both convictions should be upheld. Neither
hanging the governor in effigy nor hanging the governor is speech
(if "speech" is interpreted in its ordinary sense to mean talking or
using words). By this approach, neither act would be protected by
the First Amendment.
Fortunately, the law allows you a richer array of resources. In
addition to the ordinary meanings of a text, its context includes
precedent cases interpreting and applying the text, uncontroversial
hypothetical cases, cases or situations governed by other rules
within the same enactment, historical events or situations linked
to enactment of the rule, contemporary historical, economic, and
social practices at the time of enactment,. and examples given in
the legislative history. As we will no~, see, these features of
context can shed light directly on the legal meaning of an enacted
text.
The Supre1ne Court, for example, has given "freedom of
speech" a broader interpretation than the ordinary meaning of
"speech"-talking or using words. In Tinker v. Des Moines School
District,4 the Court held that a high school student who wore a
black armband to a public school in 1965 to protest the Vietnam
War could not be suspended from school for that reason. This
protest was "symbolic speech" within the protection of the Speech
Clause. In other cases, the Court has said that the First Amend-
ment generally protects freedom of expression. Paintings, photo-
graphs, music, and other forms of expression thus may be
constitutionally protected, along with lectures criticizing official
policies, campaign speeches, and other mLore obviously protected
kinds of expression.
But you cannot justify the Protest and 1\Jihil decisions by giving
"speech" the meaning expression (or political expression) and then
reasoning deductively from that interpretation. You would have to
conclude that both convictions should be set aside:
4
393 U.S. 503 (1969).
MAJOR PREMISE: When a state imposes a penalty for

engaging in expression, then the penalty shall be set
aside as a violation of the First Amendment.
MINOR PREMISE]: Protest's MINOR lPREMISE2 : Nihil' s

conviction was a penalty conviction was a
for engaging in expression. penalty for engaging in
express10n.
CONCLUSION 1: Protest's CONCLUSION 2: Nihil' s

conviction shall be set conviction shall be set
aside. aside.
Both hanging the governor in effigy and hanging the governor are
ways of expressing political views. The latter is but a way to
express them forcefully. The mere fact that expression, even
political expression, is a part of Protest's purpose does not justify
setting aside Protest's conviction. The same argument requires
setting aside Nihil's conviction, which surely would be absurd.
You might better distinguish the two convictions by introduc-
ing some analogies. For example, you could interpret the Speech
Clause in light of an analogy between the hvo problem cases and
a precedent case that construed the same enacted rule, such as
Tinker. Thus, Protest's conviction might be set aside because
hanging the governor in effigy is like wearing a black armband to
protest the Vietnam War. Tinker and Protest's case both involve
expressions of dissenting political views by symbolic means that
should be allowed in a democracy. The differences between
wearing a black armband and hanging a governor in effigy do not
seem important. Nihil's act of hanging the governor, by contrast,
was violent, physically harmful, and antidemLocratic. It is different
from Tinker and Protest's case in a way that justifies a difference
in outcome.
Additionally, you might reason by analogy from uncontro-
versial hypothetical cases suggested by your imagination. Hanging
the governor in effigy is like giving a public lecture criticizing the
governor's policies. Such a lecture is surely protected by the First
Amendment. By contrast, hanging the governor is more like

hanging one's spouse, usually a clear case of murder. Hanging
one's spouse, too, is expression in some respects: Ask a psychoana-
lyst. Yet the First Amendment does not protect murder.
A third kind of base point for analogies-cases or situations
governed by other rules within the same enactment-might also
support the needed distinction. The Speech Clause is joined in the
First Amendment with the Press Clause: "Congress shall make no
law ... abridging the freedom of speech, or the press. . . . "
Neither Protest nor Nihil was exercising freedom of the press, so
that clause does not apply. It probably vvould apply, however, to
protect a published political cartoon depicting the governor
hanging from a tree, in effigy or in the flesh. Hanging the governor
in effigy is in important respects like such a published political
cartoon. Hanging the governor is not. You could thus look to cases
or situations governed by other rules vvithin the context of the
Speech Clause. Perhaps the entire Constitution should be interpret-
ed to give it a coherent meaning, either because the drafters should
be presumed to have intended a coherent document or because a
coherent constitution is a better constitution.
You might find other useful base points in historical events or
situations linked to the enactment of a rule like the Speech Clause.
The history of events that inspired the First Amendment is part of
its context. A colonial government's suppression of Peter Zenger's
expressions of political criticism, for exaimple, is the sort of thing
the First Amendment was designed to prevent. The suppression of
Zenger's press was recalled during the debates preceding ratifica-
tion and often compared to unwanted English practices in the
colonies. 5 You might justify the two decisions by reasoning that
hanging the governor in effigy is like what Zenger did but hanging
the governor is not. Some such reasoning might underlie Tinker.
Wearing a black armband to protest the Vietnam War is much like
what Zenger did.
Analogies to situations in the context can support decisions at
surprisingly sharp odds with the ordinary meaning of a legal text.
Perhaps no case illustrates this possibility better than the one that
s Zaccariah Chaffee, Free Speech in the United States 21 (1969).

follows, which also illustrates additional base points from the

context. In 1885, Congress enacted a statute making it
unlawful for any person ... or corporation ... in any manner

whatsoever, to prepay the transportation, or in any way assist or
encourage the importation or migration of any alien or aliens,
any foreigner or foreigners, into the United States ... under con-
tract ... to perform labor or service of any kind in the United
States .... 6
The Church of the Holy Trinity, a corporation in New York, made

a contract in 1887 with E. Walpole Warren, then an alien living in
England. The contract required Warren to move to New York and
enter into service as the rector and pastor of the church. The
chnrch was prosecuted for violating the statute, which prescribed
a fine as the penalty for its violation.
The lower court reasoned that the church had prepaid the
transportation for Warren to come to the United States to perform
the services of rector and pastor. Therefore, it concluded, the
statute had been violated.7 The deductive validity of the lower
court's reasoning was impeccable.
Nonetheless, the U.S. Supreme Court reversed. 8 Writing for the
Court, Justice Brewer conceded that "the act of the corporation
[was] within the letter of this section, for the relation of a rector to
his church is one of service. . . . " 9 Hov\rever, he made two
analogical arguments to justify the Court's conclusion that
performing pastoral services in New York was not performing
"labor or service of any kind in the United States," within the
meaning of the statute.
First, Justice Brewer considered "contemporaneous events, the
situation as it existed, and as it was pressed upon the attention of
the legislative body," 10 to identify the perceived problem that the
statute was intended to rectify. He found that
6
23 Stat. 332 ch. 164 (1885). '
7
United States v. Church of the Holy Trinity, 36 F. 303, 303-304 (1888).
8
Church of the Holy Trinity v. United States, 143 U.S. 457 (1892).
9
Id. at 458.
10
Id. at 463.
[t]he motives and history of the act are matters of common

knowledge. It had become the practice for large capitalists in this
country to contract with their agents abroad for the shipment of
great numbers of an ignorant and servile class of foreign laborers,
under contracts, by which the employer agreed, on the one hand,
to prepay their passage, while, upon the other hand, the laborers
agreed to work after their arrival for a certain time at a low rate
of wages. 11
Second, Justice Brewer considered the statute's legislative history.

He found that the responsible Senate committee recommended
enactment to the full Senate, but in its report expressed a prefer-
ence for amending the language to substitute the words "manual
labor" or "manual services" for the expressions "labor" or
"service" wherever they appeared. The committee did not so
recommend, the report said, because the committee expected the
courts to interpret the statute to include only manual laborers, and
the committee wanted a quick adoption of the statute in the final
days before congressional adjournment. So it would appear that
the Senate sought to prohibit importation of manual laborers, but
not others. The responsible House com.mittee had made similar
remarks.
The Court concluded that the church's contract with Warren
was not legally like the contracts to import manual labor. Justice
Brewer quoted from the House committee's report in describing
the important features of the base point situation:
[T]hey come here under contract to labor for a certain number of

years; they are ignorant of our social condition, and that they
may remain so they are isolated and prevented from corning in
contact with Americans. They are generally from the lowest
social stratum, and live upon the coarsest food and in hovels of
a character before unknown to American workmen. They, as a
rule, do not become citizens, and are certainly not a desirable
acquisition to the body politic. The inevitable tendency of their
presence among us is to degrade American labor, and to reduce
it to the level of the imported pauper labor. 12
11
Id. at 463 (quoting United States v. Craig, 28 F. 795, 798 (1886)).
12 Id.
He pointed out a distinction for ministers: "It was never suggested

that we had in this country a surplus of brain toilers, and least of
all, that the market for the services of Christian ministers was
depressed by foreign competition." 13 Consequently, his opinion
concluded: "So far, then, as the evil which was sought to be
remedied interprets the statute, it also guides to an exclusion of
this contract from the penalties of the act." 14
In other words, Justice Brewer identified from contemporary
historical, economic, and social practices and from the legislative
history a relevant base point for reasoning by analogy. "Cheap
immigrant labor" is the kind of situation prohibited by the statute.
He identified the similarity between that situation and the problem
case: Both involved contracts to bring aliens to the United States
to perform labor or services here. He also identified the differences
between the two situations: Warren's arrival would not "degrade
American labor." The Court concluded that the differences were
more important than the similarity. Accordingly, the lower court's
decision was reversed, despite its airtight logic.
In sum, analysis of enacted law begins in any case with an
authoritative rule as stated-the text. The rule, however, does not
necessarily govern deductively according to its ordinary meaning.
Cases may be decided by reasoning analogically to interpret and
apply the enacted rule. Because enacted rules lack roots in the case
law, however, such analogical reasoning requires base point
situations found in the context of the enactn1ent. There are at least
seven features of the context that may contain useful base points:
(1) the ordinary meanings of the words of an enactment, (2)
judicial precedents applying the same enacted rule, 15 (3) noncon-
troversial hypothetical cases, (4) cases or situations governed by
other rules in the same enactment, (5) historical events or situa-
tions linked to the enactment, (6) contemporary economic and
social practices at the time of enactment, and (7) the legislative
history.
13
Id. at 464.
14
Id. If the language of the statute and parts of the Supreme Court opinion
strike you as poor English, be assured that I agree.
15
Note, however, that the force of precedent need not be the same in common
law, statutory, and constitutional cases. See Edward H. Levi, The Sovereignty of
the Courts, 50 U. Chi. L. Rev. 679 (1983).
Of course, these resources may yield conflicting analogies, and

they may not be helpful in some cases. Even when they help,
however, you do not avoid the judgm.ent of importance. Each
analogy presents or re-presents that crucial judgment. Combining
the forms of legal reasoning simply moves the judgment to another
place in the reasoning. Even in combination, rules and precedents
do not determine legal results based on facts and logic alone.
C. Detour on the Judicial Role
From the foregoing discussion, you rnay wonder whether the

courts have too much power to expand or restrict the apparent
meaning of a duly enacted law. In the examples, the Court
expanded the Speech Clause to protect expression or political
expression when the First Amendment said "speech." The Court
restricted the immigration statute to prohibit importing only
manual labor and services when it said "'labor or services of any
kind." But the principle of legislative supremacy generally requires
judges to play a subordinate role to the rnore democratic branches
of government. Similarly, principles of constitutional democracy
require the judges to interpret and apply the Constitution, not to
make new constitutional law. How can these cases be right?
The need to interpret general enactments can allow judges to
exceed their lawful authority even when starting with authoritative
rules. Judges are reasoning much like a conscientious legislator
when they consider all possible analogies and decide on the basis
of how all of them should be treated. Overuse of analogical
reasoning, therefore, can elevate the judiciary to a more powerful
position than most of us consider acceptable in our system of
government. This brings the legitimacy of the judicial practice into
question.
Legitimacy will be considered in Chapter 9. Here, we observe
that the constitutional doctrine of a separation of powers often
leaves it unclear whether a question should be resolved by the
judicial or legislative branch. In practice, it seems, the courts
decline to act on all possible analogies ·when doing so tends to
establish what the law should be, not what it is. 16 Courts may
decline for this reason to apply an enacted law as called for by an
intellectually appealing analogical argument,, applying instead the
enacted law using the ordinary meaning of the text.
In McBoyle v. United States, 17 for example, the defendant was
convicted of violating a 1919 federal statute prohibiting any person
from knowingly transporting a stolen motor vehicle across state
lines. 18 McBoyle transported from Illinois to Oklahoma an air-
plane he knew to have been stolen. The statute defined the term
motor vehicle as follows:
The term "motor vehicle" shall include an automobile, automo-

bile truck, automobile wagon, motorcycle, or any other self-pro-
pelled vehicle not designed for running on rails. 19
An airplane is not commonly called a "motor vehicle." It probably

would not fall within the above definition, according to the
ordinary meaning of the words in 1919.
You could argue that an airplane should be treated as a "motor
vehicle" for purposes of this statute. An airplane can be analogized
to an automobile or truck and is a "self-propelled vehicle not
designed for running on rails." Here is the analogy: Automobiles
and trucks were included in this federal enactment because they
can so easily be taken outside the state in which they were stolen.
When this happens, state law enforcement officials are hampered
because their jurisdiction ends at the border. This is not a problem
for the feds, whose jurisdiction covers the nation. Like automobiles
and trucks, a stolen airplane can easily be taken outside the state
in which it was stolen. Consequently, it may be argued, airplanes
should be classified as "motor vehicles" because they are legally
like automobiles and trucks for purposes of this statute.
Justice Holmes, writing for the U.S. Supreme Court, acknowl-
edged the force of this analogy. He refused, however, to uphold
Mc Boyle's conviction:
16
This Introduction is not the place to consider the highly complex practices
of judicial restraint in constitutional cases.
17
283 U.S. 25 (1931).
18
41 Stat. 324 (1919).
1
" Id. at §2(a).
[T]he statute should not be extended to aircraft, simply because

it may seem to us that a similar policy applies, or upon the
speculation that, if the legislature had thought of it, very likely
broader words would have been used. 20
Holmes reasoned that "vehicle," in everyday speech, calls up the

picture of a thing moving on land. He insisted that, when criminal
penalties are involved, a fair warning should be given to the
world, in language common people understand, of what the law
intends to do. For this reason, Holmes refused to follow the
analogy and expand the statute's coverage beyond its apparent
meaning.
Holmes's approach to the statute in lVfcBoyle stands in contrast
to Justice Brewer's in Church of the Holy Trinity. You may be
tempted to think that either Holmes or Brewer has the better
position in general. This would be a mistake. The restrained
judicial role exemplified by McBoyle may be appropriate to several
kinds of statutory cases. Our tradition suggests that criminal
statutes like the one in McBoyle should not be expanded by the use
of analogies. 21 As Holmes pointed out, the rules of the criminal
law should give reasonable notice to the public of what will
constitute an infraction. Keeping to the ordinary meaning of the
words furthers that goal. Some spending statutes are another
example. The power of Congress to control the federal budget is
basic to the U.S. system of government. The judiciary rightly
hesitates to intrude on Congress's prerogative to authorize
expenditures of taxpayers' dollars. ivforeover, many judges
occasionally follow a rule of interpretation requiring a court to
follow the plain meaning of a statute that is clear and unambigu-
ous on its face. 22
The more expansive judicial role exemplified by Church of the
20
283 U.S. 25, 27 (1931).
21
Note that Justice Brewer did not expand the criminal statute's scope in
Church of the Holy Trinity.
22
See, generally, Karl N. Llewelyn, Remarks on the Theory of Appellate
Decision and the Rules or Canons About How Statutes are to be Construed, 3
Vand. L. Rev. 495 (1950); Arthur M. Murphy, Old Maxims Never Die: The "Plain
Meaning Rule" and Statutory Interpretation in the "Modern" Federal Courts, 75
Colum. L. Rev. 1299 (1975).
Holy Trinity also is appropriate to several kinds of statutory cases.

For example, commercial questions, such as those raised by
Auster's case in Chapter 3, do not enjoy much political priority on
the legislative agenda. For this reason, legislatures often want the
courts to continue developing areas like co1mmercial law, as they
have traditionally. Cases involving statutes protecting minority
rights, such as the civil rights laws, may require a more active
judicial role. It is one thing to say to the U.S. Department of
Justice, as Justice Holmes effectively did in lvfcBoyle, that it should
make its argument to Congress. It is quite another to say the same
thing to a lone victim of racial or other similar discrimination.
My purpose is not to delineate a theory about when the courts
should apply statutory rules according to the ordinary meanings
of the words, leaving a wider range of arguments to the more
political branches of government. Enough has been said to suggest
that both approaches play-and should play-significant roles in
statutory cases. A judge's approach is often informed, to a large
extent, by her conception of the proper judicial role in relation to
the legislature's. That role is not the same in all statutory cases. In
fact, judges often subordinate their role to that of the legislature,
as Holmes did in McBoyle.
To summarize the main points of this chapter, the deductive

and analogical forms of legal reasoning may be combined in cases
governed by both common law and enacted law. Common law
rules may be used in the deductive form to express the conclusion
in easy cases economically. They also are used in the deductive
form to identify the legal issues by organizing the relevant
precedents. However, legal reasoning in the analogical form
remains the underlying mode of thought. Similarly, analogical
reasoning may be used to help interpret and apply an enacted rule.
The analysis begins with the enacted text. It may help to find base
points in the context that can be used to reason analogically in a
problem case.
Even together, however, the two forms do not solve the key
problem that neither solves alone. Switching at every difficult point
to the other form of reasoning does not solve the problem of
judging importance. Rules and cases, used deductively or by
analogy, do not seem able to constrain that judgment. Holmes was
eloquent in describing the need for additional legal resources:
[T]he logical method and form flatter that longing for certainty
and for repose which is in every human mind. But certainty
generally is illusion, and repose is not the destiny of man. Behind
the logical form lies a judgment as to the relative worth and
importance of competing legislative grounds, often an inarticulate
and unconscious judgment, it is true, and yet the very root and
nerve of the whole proceeding. You can give any conclusion a
logical form .... But why do you [do] it? 23
He was less helpful in explaining how the judgment of importance

can be constrained by the law so that judicial decisions are made
within the rule of law. Let us now confront that crucial problem.
23
Oliver Wendell Holmes, Jr., The Path of the Law, 10 Harv. L. Rev. 457, 466
(1897).
CHAPTER FIVE _ _ _ _ _ _ _ _ _ _ __
Legal Reasons an,t

Conventions
THE UNA VOIDABLE judgment of importance requires a judge to
determine which of the many facts in a case should count in
judicial deliberations as reasons for placing a case in the classes
designated by legal rules. A legal reason consists of a fact together
with a legal standard that makes that fact irnportant as a guide to
action for law-abiding people. Thus a red light, together with the
legal rule requiring motorists to stop at red lights, is a legal reason
for a motorist to stop. It is also a reason for a judge to classify a
motorist who ran a red light as a violator of the traffic code, and
a reason to fine the motorist as provided therein. Judges should
consider the legal reasons, pro and con, bearing on each decision
they make. They should decide based on the stronger reasons in
the circumstances.
The rules and precedents considered in preceding chapters are
among the legal standards that can generate legal reasons. We
have seen that these resources for reasoning may not suf-
fice-separately or in combination-to constrain a judgment of
importance within the law. In the next three chapters, you will find
that additional resources are available for legal reasoning. In this
chapter, I will introduce the professional conventions that support
legal principles and policies, which will then be presented in the
next two chapters. Together, these chapters show you how judicial
decisions may draw on all of these legal resources to decide cases
coherently according to the law. Your legal arguments should
draw on these additional legal resources as well as on rules and
precedents.
A. Is Legal Reasoning Scientific?
We begin this phase by dipping briefly into the philosophy of

law and legal reasoning. In, my experience, many beginning law
students (and others) have difficulty grasping law and legal
79
80 5. LEGAL REASONS AND CONVENTIONS
reasoning because they expect something that law and legal

reasoning cannot and should not deliver. Specifically, they expect
the law and legal reasoning to function like a modern scientific
enterprise. The remarkable successes of the physical and mathe-
matical sciences understandably inspire people in other fields to
follow or adapt the methods of reasoning they think are employed
in the most successful fields. Often, however, it is a romanticized
version of scientific reasoning that dominates thinking about what
is rational. 1 Without digressing to discuss more accurate views of
science, let me suggest that the laws of a society should not be
used as a scientist might use the lav\rs of nature. When we
distinguish scientific and legal reasoning carefully, we make
additional resources available to help you solve legal problems.
A scientific kind of reasoning may seem implicit in the
deductive and analogical forms of legal reasoning. The popular
conception of scientific reasoning focuses on a potential relation-
ship of correspondence between a word or symbol in a scientific
proposition and an objectively ascertainable (observable) fact in the
empirical world. The proposition is supposed to state the condi-
tions under which natural events occur. For example, one may
hypothesize that water changes from liquid to gas when heated to
212° F, using a thermometer to gauge the temperature. Following
a number of such observations, one might formulate the proposi-
tion that water always boils when heated to 212° F. If on some
occasion water does not boil at 212° F, one might conclude that the
thermometer is broken or that another relevant condition, such as
atmospheric pressure, has changed. In the latter case, the proposi-
tion might be reformulated to say that water always boils at 212°
F when the atmospheric pressure is the same as at sea level.
Predictions then could be made with greater confidence by looking
for two observable facts corresponding to the two conditions in the
proposition. In a sound scientific proposition, on this conception,
the two conditions should be necessary and (jointly) sufficient
conditions for water to boil. All cases of boiling water would have
1
"It is a commonplace of the philosophy of science that evidence is incom-
plete, that alternative hypotheses and possibilities can be imagined, that theories
are held tentatively until a better one is produced, and so on." Robert Nozick,
Philosophical Explanations 23 (1981 ). See also Thomas Kuhn, The Structure of
Scientific Revolutions (2d ed. 1970).
5. LEGAL REASONS AND CONVENTIONS 81
two empirical facts-temperature and atmospheric pressure-in

common.
The two forms of legal reasoning might seem to involve a
similar kind of reasoning. Deductive legal reasoning uses a rule as
a major premise that appears to state the necessary and sufficient
factual conditions for any case to fall within the class designated
by the rule. The common law negligence rule, for example,
requires a defendant to compensate a plaintiff if the defendant (1)
was under a duty to use reasonable care, (2) breached that duty,
which breach was the (3) cause-in-fact and (4) legal cause of (5)
damage to the plaintiff. This seems to suggest by its form that
compensation is required in any case in which five empirical facts
are present and not when one or more are absent-that all cases
of negligence have these facts in common. (Note, however, that the
five conditions state legal conclusions, not facts in any empirical
sense!)
The analogical form also seems to presuppose a kind of
reasoning like the popular conception of scientific reasoning. You
might think that two cases are alike and should be treated alike
under stare decisis if they have material facts in common. 2
Assume that a precedent case includes facts a, b, and c but not-d,
and the court holds that the plaintiff must compensate the de-
fendant. A problem case also includes facts a, b, and c but not-d.
You might say that the plaintiff should compensate the defendant
because the two cases have facts a, b, and c but not-d in common;
therefore, they are like cases that should be treated alike. However,
you could as easily say as a rule that the plaintiff must compensate
the defendant whenever facts a, b, and c but not-dare present. The
rule is implicit in the analogy.
Despite these formal similarities, you should resist any
temptation to believe that legal reasoning should mimic this
conception of scientific reasoning. Below, I ·will offer two argu-
ments for this. First, scientific and legal reasoning have such
different functions that the formal similarities are overwhelmed by
2
See, e.g., Harry W. Jones, John M. Kernochan, & Arthur W. Murphy, Legal
Method 6 (1980); E. Levi, An Introduction to Legal Reasoning 3 (1948); Jack L.
Landau, Logic for Lawyers, 13 Pacific L.J. 59, 77-78 (1981); Herman Oliphant, A
Return to Stare Decisis, 14 A.B.A.J. 71, 72 (1928).
the differences. Second, law cases that are members of the same
legal class need not have all-or even any-facts in common.
B. The Normative Function of Law
Respected legal scholars have conceived of legal reasoning

along the lines of the popular conception of scientific reasoning
described above. Most prominently, Christopher Columbus
Langdell, originator of the socratic method of law instruction in the
late nineteenth century, sought to place the law on a scientific and
conceptual footing. The lore is that, for Langdell, the law consisted
of objective, neutral, and determinate rules that dictate single
correct results in all possible cases. 3 Holmes rebelled against
Langdell's conceptualism, proclaiming that "the life of the law has
not been logic, it has been experience.'" 4 He treated the law as
predictions of judicial behavior and aspired to place it on an
empirical scientific basis. In the early twentieth century, "legal
realists" reconceived law as a social science that should capture the
observable regularities in official behavior. 5 More recently,
members of the critical legal studies movement have argued,
among other things, that because legall reasoning fails to obey
causal laws, the law is pervasively indeterminate. 6
This intellectual history is recounted because of its lega-
cy-skepticism about law and legal reasoning among many legal
academics. For different reasons, beginning law students tend to
find themselves similarly skeptical early in the first year of law
school: Good students learn quickly that little law involves the
straightforward application of legal rules. Though that lesson is
valuable, you might jump from there to skepticism because you
think that law must either operate scientifically or else count for
3
See Thomas C. Grey, Langdell's Orthodoxy, 45 U. Pitt. L. Rev. 1 (1983).
4
Oliver Wendell Holmes, Jr., The Common Law 5 (Mark DeWolfe Howe ed.
1963).
5
E.g., Karl N. Llewellyn, Some Realism About Realism, 44 Harv. L. Rev. 1222
(1931); Walter W. Cook, Scientific Method and the Law, 13 A.B.A.J. 303, 309
(1927). This approach to law, a part of "legal realism," sterns from Oliver Wendell
Holmes, Jr., The Path of the Law, 10 Harv. L. Rev. 457 (1897).
h E.g., Robert W. Gordon, Critical Legal Histories, 36 Stan. L. Rev. 57 (1984).
nothing. In my view, however, that jump would be unwarranted.

My first argument for rejecting the popular scientific model is
that scientific and legal reasoning serve different functions
requiring differences in how they are used. Propositions of the
empirical sciences function to describe, explain, and predict events
and relationships in the empirical world. Hypotheses are offered
and tested by observation to confirm or refute their accuracy qua
descriptions, explanations, and predictions,. as suggested by the
boiling water example in the preceding section. Accordingly,
empirical sciences use a language of data, measurement, probabili-
ty, verification, statistical significance, and the like, reflecting their
interest in empirical events and causal relationships.
The empirical sciences stand in contrast to the mathematical
sciences, which do not function to describe, explain, and predict
events and relationships in the empirical world. Instead, ordinary
mathematics works out the analytical relationships among num-
bers. It proceeds in terms of units, equality, proof, transitivity,
infinity, recursion, and the like, reflecting the entities and relation-
ships of interest to mathematicians. Mathernatics in principle re-
quires warrants for all inferences in a rigorous proof.
It would, of course, be silly to criticize a n1athematical claim for
lacking empirical support. Empirical investigations can neither
confirm nor refute mathematical truths, which make no empirical
claims. Such a criticism would be beside the point due to the
difference in function. Similarly, it may be that legal (and ethical)
reasoning is not vulnerable to the criticism that legal rules fail to
dictate right results or describe causal laws. As math is not the
kind of thing to be criticized for lacking e1npirical support, law
may not be the kind of thing to be criticized for lacking compre-
hensive determinacy of results. If this were so, skepticism about
law and legal reasoning would not flow so easily from the logical
and causal indeterminacies in the law.
Propositions of law differ fundamentally from propositions of
the empirical sciences: They do not describe, explain, or predict
anything. Instead, laws guide conduct by prescribing lawful
behavior; that is, they are normative. 7 For example, the law says
7
Normativity here has nothing to do with the social scientist's norm, which
refers to the average or typical behavior in a group. In some communities, the
use of cocaine may be common enough to be "the norm" in a social scientific
that motorists should stop on encountering a red traffic light and

go on a green one. Imagine yourself in the driver's seat when you
encounter a traffic signal-not on the curbside observing the
correlation between changing traffic lights and traffic in the
aggregate. Now, you have a choice. The colored light, together
with the relevant legal rule, is a reason for you to stop or go as
indicated. It is not a description, explanation, or prediction of your
behavior. It is normative because it guides your conduct.
To be sure, lawyers sometimes need to predict how judges will
decide cases. Their predictions, however, should take into account
that, from the judge's point of view, the law prescribes judicial
conduct. Judges are not involved in predicting what they them-
selves shall do. They have the capacity to deliberate on what they
ought to do and to act rationally and intentionally as a practical
matter. The outcomes of lawsuits are supposed to be reached by
legal reasoning-identifying and considering the legal reasons for
and against each proposed action. A judge's judgments and orders
are not determined by scientific laws, such as those that determine
the paths of comets.
Accordingly, to perform their function, law and legal reasoning
use a language of rights, duties, principles, responsibilities,
excuses, and the like. Reference is made basically to reasons for
action, not reasons for belief, as in the sciences. Legal reasoning
mainly involves argument about what someone should do, not
proof or experimentation, as in the sciences. It focuses on the best
of the available alternative actions in concrete circumstances, not
verifiable generalities, as in the sciences. A scientist can conclude
that the answer to a question remains unknown for decades or
centuries; however, a judge should decide a law case according to
the law and without undue delay. Because legal reasoning is used
to solve problems when decisions must be made, scientific criteria
for its success are as inapposite as empirical findings are for a
mathematician.
It need not bother you that the law· does not identify legal
results as laws of nature identify outcornes of experiments. The
difference leaves us free to formulate laws to help make our world
sense. Surely that does not make it lawful. Norrnativity also is distinct from
feelings or beliefs about appropriate conduct. See Chapter 9, note 8.
a better place in which to live. Distinguishing legal and scientific

laws, however, only opens the door for a practical approach to law
and legal reasoning. It does not show how the judgment of
importance can be constrained by the law. First, however, let us
consider a second argument for abandoning the popular scientific
model of legal reasoning.
C. Family-Style Relations Among Cases
My second and more important argumlent for rejecting the

popular scientific model is that law cases can be members of the
same legal class even when they do not have all material facts in
common. It would make sense to adapt that scientific model to the
legal function only if two criteria are satisfied: (1) all law cases that
should be decided alike must have empirical facts in common; and
(2) the common facts must correspond to the words of legal rules
so that, by their presence, they signal how cases should be
classified. Alas, the facts of law cases do not organize themselves
so neatly.
To illustrate, consider two easy cases. Each case is governed by
a common law rule requiring all contracts to be performed in good
faith. The court in each was asked to decide whether a contract
party breached by failing to perform in good faith. Though the
cases were decided by different courts at different times, the pair
of cases should be alike in the important respects to justify
decisions that both parties breached by performing in bad faith.
In the first case, 8 the Vanadium Corporation contracted to
purchase from Horace Reddington certain rnining rights to land
belonging to the Navaho Tribe on its reservation in Arizona.
Because federal law required the Secretary of the Interior's
approval before such rights could be transferred, the contract
parties agreed that Vanadium could call off the deal if the
Secretary did not approve it within six months. Vanadium
promptly entered negotiations with two other people to purchase
8
Vanadium Corp. of America v. Fidelity & Deposit Corp., 159 F.2d 105 (2d
Cir. 1947).
other rmn1ng rights to the same lands, without which use of

Reddington's mining rights would be hampered. These negotia-
tions fell through, and Vanadium's interest in the Reddington
mining rights cooled. Vanadium refused to give requested
assurances to the Secretary that it would cooperate with the
owners of the other mining rights, ·withdrew its request for
approval of the contract, and asked the Secretary to disapprove its
application formally. The Secretary did so. Vanadium claimed that
it therefore had the right to call off the deal.
In the second case,9 a Mr. Fry entered a contract to purchase
a residential home in California. As is usual in such contracts, the
deal was concluded before Fry had obtained a mortgage loan. He
was given a right to call off the deal if he could not obtain a loan
on terms described in the contract. Fry was advised by the real
estate agent that he could not get a loan on the required terms
from a bank, but probably could get one from a mortgage
company. Fry delayed for some time before he applied for a loan,
and he then applied only to two banks, both of which denied his
applications. He told the real estate agent that he had lost all
interest in the home because he had decided to move to Hawaii.
Fry claimed he had the right to call off the deal because he could
not obtain a loan as required by the contract.
The courts found that neither Vanadium nor Fry had a right to
call off the deal. Each court said that the buyer was required by
the law to perform its contract in good faith-Vanadium to
attempt in good faith to obtain the Secretary of the Interior's
approval, Fry to put forward a good faith effort to obtain a
mortgage loan on the terms specified. Each court found that the
buyer had failed to perform its contract in good faith and therefore
was in breach.
Lawyers and judges would generally agree that Vanadium and
Fry were both in breach for failing to perform in good faith. The
problem is to explain why this is so. Significantly, there is no
single nontrivial empirical fact that the Vanadium and Fry cases
have in common and that signals their proper classification as
contract breaches. If it were regarded as a fact that each buyer had
a change of mind after concluding the deal, and for that reason
9
Fry v. George Elkins Realty Co., 162 Cal. App. 2d 256 (1958).
tried to call the deal off, the search for an observable fact in
common would not end. Descriptions of mental states depend on
inferences drawn from observable facts, such as what the subject
says or does; the statement that each buyer had a change of mind
cannot be justified by pointing to observable facts that the two
cases have in common. Hence, the judicial decisions cannot be
justified by pointing to a common mental state (if that were
conceivable) as the common fact that signals a breach of contract.
Good faith, like most legal concepts, is not an observable fact, nor
can it be reduced (without remainder) to necessary and sufficient
conditions that are observable facts. 10
As indicated in Section A above, the popular scientific model
requires cases to be classified together vvhen they have the
important facts in common. It breaks down when Vanadium and
Fry are juxtaposed. But what does make sense of their common
classification?
I suggest that cases falling within a legal class are alike the way
members of a family are alike, insofar as we may know.11 No two
members of most families will be alike in all respects, nor must any
two members of most families be alike in any distinctive respect.
In the nuclear family, the parents usually do not share any
observable physical features in common such that the presence of
those features signals their membership in the same family. Two
siblings from those parents, however, will probably share some
features with each parent and some features 1with each other. They
are recognizable as members of one family, though all four do not
have any distinguishing characteristic in cornmon. A family-style
relation can be modelled as follows: A shares characteristics with
B; B shares characteristics with C; A does not share any nontrivial
characteristic with C. A and C may belong in the same classifica-
tion.
Now reconsider the Vanadium and Fry cases in light of a
hypothetical third case. Ms. Vanafry concluded a contract to
purchase a retail store from which she planned to sell groceries. As
is common in such contracts, Vanafry was given a right to call off
10
See generally Steven J. Burton & Eric G. Andersen, Contractual Good Faith
(1995).
11
See Ludwig Wittgenstein, Philosophical Investigations §§65-76 (G.
Anscombe trans. 1958).
the deal if she could not obtain a mortgage loan on terms specified
in the contract. Vanafry duly applied for a loan. She then entered
negotiations with the owner of the neighboring property in order
to purchase it. This was necessary to her plan because she needed
a parking lot. Vanafry called off negotiations when the owners of
the neighboring property refused to sell but offered a long-term
lease. Vanafry had won the state lottery in the meantime. She then
refused to give the lender requested assurances that she would
acquire adequate parking space, withdrew her application for a
loan, and requested the lender to make a formal determination that
the loan request was denied. The lender did so. Vanafry claims
that she has a right to call off the deal because she could not
obtain a loan as required by the contract:.
You will quickly see that the hypothetical case has some facts
in common with the Vanadium case, others in common with the Fry
case, and yet others that are not shared ·with either. It seems clear
that Vanafry, too, performed her contract in bad faith. The Vanafry
case can link the Vanadium and Fry cases, strengthening our
intuition that Vanadium and Fry are akin. It is like a child in a
family that links the parents as members of the family. Thus, the
three cases may be members of the same legal class, even if there
are no common facts that signal their proper legal classification.
The example, of course, is simplified for purposes of illustra-
tion. For each legal class of cases, all possible members may be
related as are the members of a large extended family viewed over
many generations past, present, and future. There are an unknown
number of possible members, and the mernbers that are known are
not necessarily typical of the entire membership. Additionally, each
case can be linked to several legal classes, as each person is a
member of several extended families. Some cases are even like
stepchildren, adopted children, and children born out of wedlock.
Moreover, though legal classes are like families in the foregoing
respects, they are also unlike many families in important respects.
There is nothing that links cases as genes link most members of
most families: Membership is assigned by human beings (judges).
Nonetheless, you may find it helpful to 1think of the me1nbers of
legal classes as the members of a family. It is more realistic, and it
steers you away from the popular model of scientific reasoning.
D. Formalism, Skepticism, and Conventionalism
The idea of family-style relations among cases captures what

you need to grasp if you are to understand legal reasoning well.
The family metaphor should help you see why legal rules and
precedents often do not dictate the correct result in a case: The
cases are too richly different for legal rules to dictate results
without leading to absurdities. At the same time, the members of
a class are too related for their grouping to be arbitrary. Therefore,
you need to understand how a grouping can be nonarbitrary when
it is not dictated by rules and precedents. Put another way, family-
style relations are composed of webs of analogies linking the
members of a legal class. The problem of importance, now seen in
a more complex and elaborate way, arises pervasively within the
web.
This way of framing the problem is not universally shared by
law teachers. In general, there are two alteirnatives, which I will
pose starkly. First, you could insist dogmatically that legal rules
must dictate the results in all cases decided ,-vithin the rule of law.
This would endorse classical legal formalism. It would require
sharply restricting the domain of law-governed activity from what
it has been throughout the last several centuries. In particular,
taken seriously, it would require judicial abstinence in almost all
of the great cases generally thought to have contributed centrally
to greater social justice in the United States.12 It would also
restrict much of the statutory and common law. For this reason,
even those who uphold legal formalism in the abstract seldom
follow through on its full implications in all cases. 13
Second, you could take family-style relations to indicate that
legal rules and precedents count for little or nothing. Rather, as
skeptics have recently urged most significantly, law should be
considered a form of politics. Judges, then, would decide cases
according to their own views of social justice, manipulating the
rules and other legal authorities as necessary to create the illusion
12
E.g., Brown v. Board of Education, 347 U.S. 483 (1954).
13
See, e.g., Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. Chi. L.
Rev. 1175 (1989).
of a rule of law. 14 This skeptical approach expands the judicial

power far beyond its historic scope, setting up judges as philoso-
pher kings to rule without effective legal constraints. It gives up on
the rule of law in the hope that judges will see social justice
acceptably on a regular basis. Probably n1.ost of us doubt that this
is a safe bet.
A third approach, elaborated in subsequent chapters and
elsewhere, 15 does not suppose that the lffw must dictate results or
count for nothing, as do legal formalism and legal skepticism.
Instead, I will build on the idea that the law constrains the
judgment of importance by allowing and disallowing reasons to
count in judicial and other legal deliberations. Reasons allowed by
the law may compete. Judges must ·weigh them under the
circumstances in each case to reach a lawful conclusion in that
case. But other kinds of reasons-such as those stemming from a
judge's personal interests, prejudices, and religious or moral
views-would be ruled out.
Once we refocus attention from results to reasons, a richer and
more practical understanding of law and legal reasoning comes
into view. The law may be seen to constrain official decisions by
restricting the allowable reasons for official action, even when it
does not dictate outcomes. Legal reasoning may be seen as the
process of using legal reasons in legal arguments, even when the
arguments leave a judge to make a judgm.ent. Legal standards can
latch onto facts to yield legal reasons without necessarily requiring
results. The law thus may include precedents and legal rules of the
kind introduced earlier, though with tempered force. As suggested
in the next chapter, the law may also include principles and
policies embedded in the law, which do not pretend to dictate
outcomes. And the legal standards-precedents, rules, principles,
and policies-may come together in a decision constrained as to
its lawfulness by the background conventions of the practice. This
third approach may be called "conventionalism" because it shifts
from the abstract form of law and legal reasoning to the functions
they perform as they are used by members of the legal community.
14
See, e.g., Duncan Kennedy, Freedom and Constraint in Adjudication: A
Critical Phenomenology, 36 J. Leg. Educ., 518-562 (1987).
15
E.g., Steven J. Burton, Judging in Good Faith (1992).
Despite important differences, lawyers and judges share much

in common. As Roscoe Pound observed,
It is an everyday experience of those who study judicial decisions

that the results are usually sound, whether the reasoning from
which the results purport to flow is sound or not. The trained
intuition of the judge continually leads him to right results for
which he is puzzled to give unimpeachable legal reasons. 16
Similar observations are commonplace. While a judge on the New

York Court of Appeals, the highest court in New York, Benjamin
Cardozo reported:
Of the cases that come before the court in which I sit, a majority,
I think, could not, with semblance of reason, be decided in any
way but one. The law and its application alike are plain. Such
cases are predestined, so to speak, to affirmance without opin-
ion.17
More recently, Judge Jon 0. Newman of the U.S. Court of Appeals

for the Second Circuit reported that dissenting votes were cast in
only 3.7 percent of all cases decided by federal courts of appeal in
the year ending June 30, 1983. 18 Even a highly critical observer
like Professor Duncan Kennedy reports that "It's endlessly the case
that the judge's approach to the problem strikes me as intelligent
and that ... the judge comes up with a rule which, when I think
about it, strikes me as a good idea for the very reasons that he
gave for it." 19
This suggests that judges and those who study judicial
decisions share understandings that often lead them to the same
result, even when that result is not a necessary product of the rules
and precedents. This is most plausible when we consider all cases
16
Roscoe Pound, The Theory of Judicial Decision, 36 Harv. L. Rev. 940, 951
(1923).
17 Benjamin Cardozo, The Nature of the Judicial Process 164 (1921).
18
Jon 0. Newman, Between -.Legal Realism and Neutral Principles: The
Legitimacy of Institutional Values, 72 Calif. L. Rev. 200, 204 (1984). Of course, a
few judges may not dissent every time they disagree.
19
Duncan Kennedy, Address at the AALS Annual Meeting, Section on
Jurisprudence Gan. 7, 1986).
that arise or could arise, instead of focusing only on the few cases
decided by appellate courts, much less the fewer in law school
casebooks. The controversial cases discussed in the popular press
and law school classrooms do not fairly represent the extent of
agreement on settled points of law. In fact, the law is applied
every time a lawyer advises a client and, for that matter, every
time a motorist stops at a red light. Few people face difficult
questions when deciding, for example, whether they are legally
required to file an income tax return. In Vanadium and Fry, lawyers
and judges generally would agree that the buyers breached their
contracts by performing in bad faith. They might give differing
explanations or no articulate explanation at all. But the results are
conventionally accepted within the professional community.
You could understand this extensive agreement in practice by
focusing on the relative homogeneity of the legal profession in
contrast with the general populace. Perhaps the profession shares
a self-interested ideology that accounts for widespread agreement
within it. 20 Surely the profession remains less than fully reflective
of the general populace in terms of race and gender. However, it
is too diverse in terms of politics, religion, economic views, and
moral philosophies for homogeneity to be a good explanation.
Lawyers and judges commonly distinguish their personal views
of what the law should be from their professional views of what
the law now in force permits or requires. There may be no
controversy in our time so intractable as the one concerning
abortion rights. Nonetheless, friends and foes of abortion rights
agree that the prevailing law guarantees ·women a right to choose
abortion in many circumstances. One side finds in this a strong
motivation to seek legal change. The other resists. Both agree that
the issue is one of legal change because they agree on what the
law now is, despite their moral conflict. Similarly, within the legal
community, you will find far more agreement on what the
prevailing law permits or requires than on its justice. Accordingly,
the law can be interpreted and applied as a professional matter
even while its wisdom is in dispute.
20
See, e.g., Joseph W. Singer, The Player and the Cards: Nihilism and Legal
Theory, 94 Yale L.J. 1 (1984).
Let us call the points of general agreement among members of

the legal community legal conventions. The parties to these conven-
tions are judges and lawyers most centrally and, in appropriate
cases, legislators, executive or administrative officials, and
sometimes others. The conventions consist: of the practices and
dispositions of the parties. The practices are the collection of
solutions that have been reached for legal problems in the past.
The dispositions are the members' propensities to agree on the
legal outcomes in possible cases in the future. 21 Dispositions are
revealed only after lawyers have immersed themselves in the facts
and law of cases and thought their way through to deliberate
conclusions. For this reason, decided cases,. published scholarly
articles, treatises, and similar works better indicate professional
dispositions than opinion polls of lawyers. Because they include
dispositions, conventions can guide decisions even when the cases
were not the subject of any explicit agreement before they arose for
decision.
Conventions support interpretations of the legal materials,
allowing lawyers and judges to select conventionally accepted
meanings and implications from among all logically possible
meanings and implications. The existence of legal conventions is
evidenced by the many clear and uncontroversial cases, especially
those in which results are not dictated by rules and precedents.
Legal conventions are necessary for a legal system to exist under
the rule of law. By most accounts, the rule of law requires more
regularity in official conduct than rules and precedents can
establish alone.
Conventional practices and dispositions characterize the legal
community adequately to treat it as a distinctive interpretive
community whose job is to interpret the law (among its functions).
As you will see, professional conventions support the use of legal
precedents, rules, principles, and policies to decide problem cases,
thereby coordinating decisions in the aggregate to implement the
rule of law. The law in a case is developed from the legal materials
21
For a different conventionalist view, see Owen M. Fiss, Objectivity and
Interpretation, 34 Stan. L. Rev. 739 (1982); Owen M. Fiss, Conventionalism, 58 S.
Cal. L. Rev. 177 (1985) (conventions as disciplining rules). See also Jules L.
Coleman & Brian Leiter, Determinacy, Objectivity and Authority, 142 U. Pa. L.
Rev. 549, 622-632 (1993).
consulted in the course of legal research-case reports, constitu-

tions, statutes, administrative regulations, treatises, law review
articles, and the like. These materials represent the practices and
evidence the dispositions of the legal co1mmunity.
An important caveat to avoid misunderstanding: On this view,
a legal decision may be right in relation to the conventions of the
legal community. But it does not follow that the decision is right
in any stronger sense. The decision also can be evaluated in
relation to various moral, religious, economic, and political conven-
tions or abstract theories of justice and political morality. An act
may be lawful whether or not the law is just. 22
Legal reasoning aims to identify lawful actions according to the
law in force in a society, leaving it to critical reasoning whether
that law is deserving of respect by obedience. Separate but equal
educational facilities were constitutional in the United States until
1954. 23 Legal reasoning could then be used to determine what the
law permitted and required, as it can be used today for the same
purpose with a different result. Legal reasoning has practical value
to those who want to know when they or another run afoul of the
law in force.
From a critical standpoint, moreover, legal reasoning can be
used to identify important objects for criticism. The law in force,
as it is applied, is surely a more important target than any
professor's pet theory. It may be highly deserving of debunking.
In my view, for example, racial segregation was all the more
odious when it was required by the lavv and enforced by legal
means. That experience is a clear object lesson for us: The law is
not always just, and it can be used as an instrument of oppression.
Legal reasoning is not adequate to either defend or debunk
prevailing legal practices.
The interpretive conventions support a practice that calls
attention to the facts in a problem case that are likely to be
regarded as important facts by other members of the legal
community. The conventions can help you make reasonably
reliable predictions and effective legal arguments, as they help
judges make responsible decisions that may earn respect as law, at
22
For elaboration, see Chapter 9 §B.
23
Brown v. Board of Education, 347 U.S. 483 (1954); Plessy v. Ferguson, 163
U.S. 537 (1896).
least within the profession. We can allo\v that the currently

prevailing legal conventions may support practices that are unjust
and deserving of change-even radical change. Nonetheless, the
prevailing conventions of the legal community may be useful
because they can help you get somewhere you want to go.
CHAPTER SIX _ _ _ _ _ _ _ _ _ _ __
Purposes
LET ME NOW ILLUSTRATE how judgments of innportance depend on
what the relevant law is trying to do. Rules and precedents are not
only efforts to settle disputes. We could do that by flipping coins,
at considerably less expense. Rules and precedents set standards
of lawful conduct that are supposed to hel]P make our world a
better place in which to live. To do this, rules and precedents
implement the law's vision of a better society. Put differently, laws
have purposes: They implement, and should be justified by,
desirable principles and policies. 1 The distinctions among rules,
principles, and policies are controversial among legal philosophers.
I will not get into the technical debates here. However, let me
suggest the following as a working hypothesis.2
In a legal system, the framework of law is described by legal
rules which, as you now know, designate classes of cases and affix
concrete legal consequences to membership in the classes. Like
legal rules, principles and policies say soimething about how
people should act, though in a more general way. Principles and
policies, however, do not set up legal classes and fix legal conse-
quences. Rather, they provide justifications for the rules and also
legal reasons for placing cases in legal classes set up by the rules.
A judgment of importance may be constrained by the law when it
furthers the purposes that a rule is supposed to serve (that is,
implements the principles and policies that justify the rule).
Consequently, you can use principles and policies from within the
law to interpret and apply legal rules, treating problem cases as
occasions for implementing the law's vision of a better society.
1
For convenience, I will often refer to "purposes" instead of "principles and
policies."
2
For contrasting views of rules, principles, and policies, see Ronald Dworkin,
Taking Rights Seriously (1978); Ronald Dworkin, Law's Empire (1986); Frederick
Schauer, Playing by the Rules (1992).
97
98 6. PURI'OSES
A. The Judgment of Importance
Without attending to a law's purposes, judicial decisions might

be based on any facts that happen to loom largest to a judge at the
time. Often, importance would flow from the judge's political
preferences and interests, identification with a party, or from the
happenstance of the language from a legal rule that was an-
nounced in some other context. Decisions on these bases are
arbitrary in relation to any conception of justice. The law's
purposes, by contrast, can be used to coordinate the activities of
judges so their decisions together might have the regularity
required by the rule of law. Principles and policies can call some
facts to the foreground of the totality of the circumstances, leaving
others in the background as context.
The judgment of importance is the decision on which of the
many facts in a case justify placing that case in a legal class.
Absent a theory about what the rules and precedents are trying to
do, the congeries of facts in a case are puzzling. There are so
many! Each fact by itself is just that-a fact. Any fact might be im-
portant or unimportant. It merits special attention only if it counts
more than others, and it counts more than others only if it is
valued for more than its truth. A fact can be so valued if it plausi-
bly fits with a normative theory about what the law is trying to do.
To start with a nonlegal example, inrragine that you are asked
to tell a true story about the Mississippi River. 3 If you are told
nothing more, you might be puzzled. You can tell stories about the
Mississippi's length, the volume of water it carries, its contribu-
tions to irrigation, the drainage basin it serves, the facilities and
sites along its banks, its propensity to flood, its role in political
boundaries, its effects on human settlen1ent patterns, the traffic it
carries, the pollution it carries, its geollogical history, its role in
Huckleberry Finn, the hazards of navigating it, and so on. An
almost infinite number of facts about the Mississippi could be
gathered. You may not know where to begin, where to end, or
how to organize your story intelligently.
Now imagine that you are asked to tell a story about the
3
This example is suggested by James B. White, The Legal Imagination: Studies
in the Nature of Legal Thought and Expression 3-34 (1973).
6. PURPOSES 99
Mississippi River because you want to transport grain from St.

Louis to New Orleans by barge. Given this purpose, you will at
once turn attention to the river's length, the navigational hazards
one might encounter, and the facilities along its banks for servicing
barge transports. These and related facts will merit special notice
in light of your purpose. Other facts fall away from attention. They
are not important for this purpose.
By contrast, imagine that you are asked to tell a story about the
Mississippi because you want to locate oil deposits. You will turn
attention now to its geological history. Given this purpose, you
should focus on the transport of sediments into the Gulf of Mexico
eons ago. Geological theories indicate that there consequently
might be oil there. Now these and related facts about the Mississip-
pi will come to the foreground of the totality of the circumstances.
These facts are important for this purpose.
A lawyer is similarly puzzled when he first approaches a law
case. So many facts can be gathered and described in so many
ways! Without a theory about the relevant law's purposes, it is
hard to know where to begin, where to stop, or how to organize
a statement of the facts that will communicate all and only what
is needed to do the job.
Suppose that you are a lawyer who is visited by a prospective
client, Marilyn Mother. Mother tells you that she is divorced,
works in the home, and had one child until last week. Tommy
Trucker then was driving a truck down Main Street. As he
approached Chestnut Avenue near her home, her son pursued a
ball into the street. A collision occurred and the boy was killed.
Mother wants to sue Trucker and wants you to represent her.
Before deciding to do so, you will want to get some idea of what
is involved by gathering the relevant facts. You should begin by
questioning Mother.
What questions will you ask? At first you might be puzzled.
How old was the boy? What was his name? Why was he given
that name? Why did he enter the street? Was it nighttime or
daytime? Did he look both ways? What was he wearing? Was he
a good student? Did Mother see the collision? Is she capable of
bearing more children? What kind of truck ,Nas Trucker driving?
What color was it? How fast was he going? Did he wear glasses?
What is his religion? Was he watching where he was going? Did
he brake? Were his brakes in good working order? Did he carry
l 00 6. PURPOSES
insurance? Was an ambulance promptly called to the scene? Did

the ambulance reach the hospital quickly? Was medical care
properly administered? Absent a purpose, any of the facts found
in answer to these questions, and many others, might be impor-
tant. The fact that the child was a boy, not a girl, looms as large or
as small as any other fact.
Consider the probable effect of two of the several possible
purposes that may justify the rules in the relevant area of the
common law. One theory (a deterrence theory) supposes that the
law in cases like this should discourage people from behaving
unreasonably when they might cause harm to others. To imple-
ment this purpose, the law forces people to compensate those who
are avoidably injured by unreasonably dangerous behavior. This
purpose calls your attention to Trucker's behavior. In its light, it is
probably important that Trucker was not wearing his glasses, was
not watching where he was going, did not keep his brakes in good
repair, failed to slow down when he sa-w the boy's ball roll into
the street, and the like. If the law's purpose is to discourage
unreasonably dangerous behavior, these behaviors should be
discouraged by forcing Trucker and others to pay compensation to
their victims on such a basis. The fact that Trucker was driving a
truck, however, is probably not important. The law does not try to
discourage people from driving trucks.
By contrast, a second possible theory (a loss spreading theory)
supposes that the law should spread the losses resulting from
accidents among large numbers of people instead of leaving the
victim to cover the losses alone. This theory might suggest that
people who are more likely to have insurance should compensate
victims who are less likely to have insurance. In its light, it is
probably important that Trucker was driving a truck and the
victim was a child. Truck drivers generally carry insurance for
accidents. Children do not, and many parents do not carry
insurance to cover their children when they are pedestrians.
Trucker's insurance company can pay conapensation to Mother and
spread the cost among a large group of subscribers. For this
purpose, it is not important whether Trucker was wearing glasses,
was watching where he was going, kept his brakes in good repair,
slowed down when a child's ball rolled into the street, and the
like.
Assume it were to turn out that Trucker was driving a truck
6. PURPOSES 101
and the victim was a child, and also that Trucker was wearing his
glasses, was watching where he was going, had just inspected his
brakes, braked as soon as any good driver would, and in all other
respects was proceeding safely down the street. If the courts
employ the deterrence theory, you should conclude that Trucker
probably will not be held liable. The facts indicating that he was
behaving safely are important because he did nothing that was
unreasonably dangerous and should be discouraged. If the courts
employ the loss spreading theory, however, you should conclude
that Trucker probably will be held liable. The fact that he was
driving a truck that hit a child is important because Trucker would
seem to be in the best position to spread the loss among a large
group of people through insurance.
Thus, the principles and policies that justify the relevant rules
establish a perspective that calls the important facts into promi-
nence. The law's purposes therefore should play a crucial role
when you decide what questions to ask M[arilyn Mother. They
should play a similar role when you try to predict what a court
would do or to persuade others of what a court should do in the
case.
In addition to purposes, the legal rules themselves are sup-
posed to call the important facts into prominence. Some elements
of some rules do a reasonable job of capturing in language the
significance of the class of cases for which they stand. These rules
perform this function well in easier cases. Assume that Trucker
had behaved in a less than perfectly safe manner in that he had
failed to have his brakes inspected when they were due for an
inspection a week before the accident. Assume further that the
common law negligence rule governs this case: Trucker is liable if
he was under a duty to use reasonable care and breached that
duty, which breach was a cause-in-fact and legal cause of damage
to Mother.
The language of this rule makes it fairly clear that you should
focus attention on the potential causal relationship between the
accident and Trucker's failure to have his brakes inspected.
Because Trucker behaved safely in all respects except in failing to
have his brakes inspected, that is the only possible breach of duty.
The rule tells the lawyer that the breach of duty must be the
"cause-in-fact" of the accident if Trucker is to be held liable under
this rule. In other words, he will not be held liable if the accident
102 6. PURPOSES
would have occurred even if the brakes had been inspected.

Accordingly, important facts include whether a defect in the brakes
would have been discovered on inspection and whether, if
discovered and repaired, the truck would have come to a stop
sooner, before hitting the child. In an easy case, there may be no
need to resort to the law's purposes to apply at least some
elements of the rule.
Often, however, the elements of legal rules do not do a good
job of capturing the important aspects of the cases they govern. As
stated, for example, the language of the negligence rule does not
call attention to the facts that matter when you decide whether
Trucker's failure to have his brakes inspected was a "breach of
duty to use reasonable care." The language of this element of the
rule suggests that less than perfectly safe behavior is not a breach
of duty if it nonetheless was reasonable behavior. But it does not
indicate how much less than perfect is unreasonable.
If you interpret the language in light of the deterrence theory,
you might focus on the fact that people commonly treat the due
date for a brake inspection somewhat casually. A week's delay is
not generally regarded as unreasonably unsafe behavior that the
law should discourage; what people generally do in such circum-
stances is probably important for this purpose. However, if you
interpret the language in light of the loss spreading theory, as
courts sometimes do, you would reach a different result. The
theory may lead a judge to seize on the irnperfect behavior at least
to allow a jury to find Trucker liable. What people generally do in
such circumstances is not so important. That the loss probably falls
on Mother alone, but could be spread by Trucker, is important.
Again, the law's purpose should be a cirucial ingredient in your
legal reasoning: The principles and policies that justify a law
provide a normative perspective that calls the important facts into
pronunence.
B. General Purposes of Law
To avoid misunderstanding, I should comment further on the

law's purposes. In the aggregate, legal ru1es, principles and policies
should describe a possible world to be brought into existence by
the coordinated actions of citizens and the state. The traffic laws,
6. PURPOSES 103
for example, represent a complex set of coordinated actions that

should be taken by motorists in response to various colored lights
and signs with various shapes. It is not hard to imagine that
world. To a remarkably large extent, motorists are guided by the
law to take those actions and thus to realize the traffic laws' vision
in the real world. Of course, other laws, like those prohibiting the
sale or use of illicit drugs, are notoriously ineffective. The social
world is brought into conformity with such laws, if at all, when
legal sanctions (enforced losses of liberty or property imposed for
disobedience) are imposed on violators.
We assume that society would be different if the law did not
exist. People generally abide by the law due to a habit of compli-
ance and a sense of obligation, or to the practical threat of its
sanctions, or from both kinds of motive. People in many situations
would not behave in the same ways if there were no formal ex-
pressions of what the society regards as 1minimally honorable
behavior and no organized criticism, such as sanctions, for mis-
behavior. We also assume that society generally is better off
because the law exists. The laws should encourage valued be-
havior, discourage disvalued behavior, and guide conduct accord-
ingly. If the laws have an effect on society, they should help make
it a better society in which to live.
The abstractness of this broad perspective can be reduced
somewhat by identifying two frequently competing general
purposes underlying the largest part of the law. 4 First, the law
should contribute to a more orderly society. People at times will
find themselves in conflict in any society with any form of social
organization. Serious conflicts sometimes would result in fights,
feuds, and the useless destruction of life and property; if left
unsettled, even nonviolent disputes can be highly disruptive to the
disputants and others. Society would be better off if serious
conflicts more often were resolved peacefully. An offer of peaceful
dispute settlement procedures, such as adjudication, may induce
many people to seek peaceful resolutions of serious conflicts that
are not settled by other means.
4
These, in turn, will help generate more concrete purposes tied to specific
laws, which purposes are of far greater practical value iin particular cases. See §C,
below.
104 6. PURPOSES
A society could settle persistent disputes in a number of

peaceful and less costly ways not involving cases, rules, lawyers,
and courts. Many societies in many times and places have used
such alternatives, and some are being used increasingly in the
United States for some kinds of disputes. For example, we could
flip coins, appoint a "wise" man to rule by fiat, leave disputants to
work out their differences by negotiation, or leave them to agree
to employ third parties to mediate or arbitrate privately, insofar as
they can without a compulsory backup. The law, however, is
trying to do more than settle disputes peacefully. There is a reason
why, when the disputants persist, much more costly public and
compulsory dispute settlement procedures are available.
The reason is that the law should also contribute to a more just
society. Absent the law, the distribution of valued liberties,
properties, and dignities presumably would be a function of the
relative power of each person. Society is better off if the strong at
least sometimes are restrained from dominating the weak. We
hope that some of our laws are effective in helping to give all
people a fair opportunity to pursue their chosen ends, consistent
with the liberties, properties, and dignities of others.
One or the other of these values may be emphasized in
different legal systems or in a single legal system at different times.
A legal system that pursued order without justice would tend
toward a police state-Mussolini's Italy is an example. A legal
system that pursued justice without order would tend toward
chaos, as did China's Cultural Revolution. At present, the U.S.
legal system is committed to both values in some sort of balance.
Though these general purposes of the law are obvious and
abstract, they have practical implications for law and legal
reasoning. Most members of the legal con1munity hold as an ideal
that each legal rule and each law case should further the purposes
of the law as a social institution. Rules and cases thus should
contribute to the maintenance and growth of a more orderly and
just society. We know that we often fall short of these ideals. We
nonetheless strive for them, one case at a time.
Judges are striving to promote these broad ideals of the law as
well. When trying to predict judicial decisions, you should gauge
how your client's activities will appear to the judge in light of
these ideals, as they are implemented by the relevant rules and
precedents. When persuading judges and others, you can show
6. PURPOSES 105
how a judgment for your client helps to promote these ideals while
a judgment for the opponent tends to hinder them. A judge can
safely assume that members of the legal community, and the
broader society, value a legal system that contributes to a more
orderly and just society. Judicial decisions that so contribute are
more likely to earn respect as law.
You may wonder whether the ideals of the law are too broad
and too vague to help in a particular case. How do we know which
decision will enhance social order or contribute to a more just
society? Must we not first believe that an ideal Orderly and Just
Society somehow "exists"? Must we first ans\lver the questions that
philosophers have failed to answer in the 1nillennia since Plato
asked them? Do we not, in fact, disagree too widely on the most
basic terms of a good social life?
Let us distinguish two sorts of questions. The question "What
is the ideal Orderly and Just Society?" involves many difficult and
controversial philosophical questions that are of little immediate
value to practicing lawyers and judges. As I see it, a major point
of law is to relieve us from the daunting task of resolving our
disputes by solving those problems. Contrast the practical ques-
tion, "What will members of the legal comn1unity generally take
to be required by an orderly and just society in this case and like
cases?" This is a question about the considered judgments of
people in a legal culture concerning a practical problem. The
practical question may be of great value to practicing lawyers and
judges: It directs us toward tentative answers in the conventions
of a highly practical profession, not in philosophy.
C. Purposes Embedded in the Legal Experience
Given a case, you can develop useful theories about the law's
purposes in a number of ways. The most useful purposes are
embedded in the legal experience-the existing common law
precedents and rules and the enacted texts with their contexts. You
will find other useful purposes in the concurring and dissenting
opinions of judges or the secondary literature, including treatises,
restatements of the law, and law reviews. These sources, too, for
the most part draw their inferences from the legal experience,
though some advocate legal change and others are academic.
106 6. PURPOSES
Embedded purposes are reliable indications of the legal conven-

tions. Like judges, you need not judge irnportance on the basis of
your own value preferences or philosophical views. Instead, you
can judge importance in a way that furthers the law's purposes,
implementing its principles and policies, for better or worse.
1. Purposes in the Common Law
To illustrate the play of purpose in common law cases, recall

that the principle of stare decisis requires judges to follow prece-
dent analogically. Having identified the relevant precedents and
the factual similarities and differences between the precedents and
a problem case, the analogical form leaves judges to decide
whether the similarities or differences are more important under
the circumstances. You now have seen that principles and policies
help make the judgment of importance by calling attention to the
important factual relationships between cases.
Recall also the line of horse trading cases in Chapter 2. I will
use two of them as if they were precedents for the purpose of
discussing a third as if it were a proble1m case. For your conve-
nience, the facts are repeated. 5
CASE 2
(Precedent)
Abbott bought Costello's horse, giving; as payment a forged

check on another person's account. Abbott knew the check
was forged. After delivering the horse to Abbott, Costello
discovered the fraud and sued Abbott to recover the horse.
Costello won.
5
You may find it helpful to refer back to Figure 2-1, page 33.
6. PURPOSES 107
CASE 3
(Precedent)
The facts are similar to Case 2, except that Abbott sold the
horse to Holliday. Holliday knew that Abbott had bought
the horse from Costello, but she did not know or have any
reason to know that Abbott paid with a forged check.
Costello sued Holliday to recover the horse. Holliday won.
CASE 5
(Problem)
The facts are similar to Case 3, except that, after buying the
horse from Abbott, Holliday sold and delivered it to Ball.
Ball had heard rumors of the fraud worked on Costello by
Abbott. Costello sued Ball to recover the horse.
Analogically, the legal issue in Case 5 is \t\rhether it is more like

Case 2, where Costello recovered possession of the horse from
Abbott, or more like Case 3, where Costello did not recover
possession from Holliday. Case 5 may be more like Case 3 if the
fact that Ball bought from an innocent purchaser is more important
than the fact that Ball had heard rumors of the fraud. It may be
more like Case 2, however, if the fact that Ball had heard rumors
of the fraud is more important than the fact that Ball bought from
an innocent purchaser. The decision in Case 5 turns on whether it
is more important that Ball had heard rumors of the fraud or that
Ball had bought from an innocent purchaser.
It is not obvious whether Costello should be more careful about
being defrauded or Ball should look into the rumors of fraud and
refrain from buying tainted property. The judge in Case 5,
however, is unlikely to judge importance in Case 5 apart from the
legal experience. Judicial decisions would not be coordinated if the
judges were to make their decisions without legal basis. The law
would not implement any social vision at alll, and people would
find it difficult to plan their transactions to avoid running afoul of
108 6. PURPOSES
the law. In case of disputes, fewer settle1ments would occur while

the court docket would grow, delaying the delivery of justice to all.
Each case should not be decided ad hoc in isolation.
1•
Moreover, the law would not be just if judges were to decide

each case ad hoc. Basic to any concept of justice is formal justice,
the principle of equal treatment under the law, which requires that
like cases be treated alike and that rules be applied consistently.
Dispensing with this principle leaves judges free to indulge their
personal values, prejudices, ideologies, and whims without pause.
Consequently, some judges might use the power of the state to
oppress the odd, the unpopular or the weak; others might use
official power to rob from the rich and give to the poor. One might
question whether such a society settled its disputes according to
law at all.
The judge in Case 5 is more likely to ask, if only implicitly,
what members of the legal community generally would take to be
required by an orderly and just society in Case 5 and like cases.
Cases 2 and 3, of course, are like Case 5 in some respects. What
the judges decided in those cases (together with the rest of the
legal experience) is a fairly reliable indication of the legal com-
munity's considered judgments. The judges in those cases held that
the defrauded owner can recover possession of the horse from the
perpetrator of the fraud (Abbott), but not from an innocent
purchaser from the perpetrator (Holliday). The judge's objective is
to decide Case 5 in a way that permits Cases 2, 3, and 5 to be
reconciled so that the law does not require Ball to act differently
from how anyone should act, given the same circumstances.
Accordingly, all three cases would be decided consistently under
principles and policies that together may contribute to a more
orderly and just society. Because Cases 2 and 3 were decided by
earlier courts, the judge in Case 5 should vvork presumptively with
purposes furthered by those decisions, if any. The judge in Case 5
should make the judgment of importance in a way that makes
sense for the entire family of cases.
The principles and policies implemented by Cases 2 and 3 may
not be hard to discern. Preliminarily, it should be noted that the
holding in Case 2 "rights the wrong" perpetrated by Abbott on
Costello by requiring Abbott to return the horse to Costello. But
this will not do as a principle for Case 5. A principle that the law
should "right all wrongs" hardly survives Case 3. Costello was as
6. PURPOSES 109
wronged in Case 3 as in Case 2, but Costello did not recover the

horse.
The better theory in Case 2-one that is also compatible with
Case 3-is that the law here tries to enhance the security of an
owner of property as against others with no claim of right. One
function of a system of property rights is to encourage owners to
invest in maintaining and improving their property, increasing the
aggregate wealth of society. Some people would be less likely to
spend time, money, and other resources on maintaining and
improving property if another person could take the property
without the voluntary consent of the investor. In Case 2, forcing
Abbott to return the horse allows Costello to reap the benefits of
ownership as expected. Abbott has no claim of right because there
is no apparent reason why the perpetrator of a fraud should retain
possession of the horse.
This understanding of Case 2 is consistent: with Case 3 because
there are countervailing reasons why Holliday should retain
possession of the horse. The law also seeks to encourage people to
buy and sell property, thereby increasing the aggregate wealth of
society through trade. Purchasers like Holliday would be less likely
to buy property if they could lose possession 1to a prior owner who
had been defrauded by another, perhaps long ago. In Case 3
Holliday paid a price for the horse and relied reasonably on
receiving ownership and possession. She did not know and had no
reason to know that Abbott obtained possession of the horse by
fraud. Allowing Holliday to keep possession of the horse furthers
the goal of encouraging trade; allowing Costello to recover
possession might discourage some people like Holliday from
trading. Consequently, Holliday should receive the benefit of her
bargain.
A judge in Case 3 need not deny that holding for Holliday
might tend to discourage owners like Costello from maintaining
and improving their property. In this case, however, there is no
way to avoid hurting one of them unfairly. (Abbott cannot be sued
because, typically, he has flown the coop.) As between Holliday
and Costello, Costello might have protected himself while Holliday
could not. People like Costello should be encouraged to be more
careful about detecting fraud in the transactions to which they are
parties. Therefore, Holliday stands in greater need of the law's
protection.
110 6. PURPOSES
To complete the story, in light of this understanding of the

law's purposes, Costello has a plausible argument for recovering
possession in Case 5. Ball had heard rmnors of the fraud. Unlike
Holliday in Case 3, it was possible for her to protect herself.
Consequently, Costello's appeal to the security of ownership is
stronger against Ball than it was against Holliday in Case 3.
Ball, however, has a stronger argument than Costello in Case
5. The law in these circumstances seeks to increase the wealth of
society by enhancing the security of owners and innocent purchas-
ers of property. When these two goals conflict, as in Case 3, the
law gives preference to the security of innocent purchasers when
the original owner is defrauded by another. The preference for
protecting the security of innocent purchasers is likely to carry
over to Case 5, to Ball's advantage. Consequently, the fact that Ball
had bought from an innocent purchaser is more important than the
fact that Ball had heard rumors of the fraud. Costello should not
be able to impair Holliday' s ability to sell the horse by giving
general publicity to the fraud or spreading rumors. Moreover, it
would hamper trade if a potential buyer bore the burden of
checking out every rumor of a defect in the title of personal
property, running back perhaps many years. Again, Costello is in
a better position than Ball or Holliday to protect himself against
fraud; on balance, Ball and Holliday stand in greater need of the
law's protection.
A judge writing on a clean slate might reach the opposite result
in Case 5. Ball's argument depends in substantial part on the pre-
sumption that Case 3 was decided correctly for Holliday. The legal
community's practice of stare decisis, however, supports the pre-
sumption that Ball should be treated like Holliday if her case is
more like Holliday's than it is like Abbott's in Case 2. The law
could easily have evolved differently. But, having evolved as it
did, a change is unlikely unless underlying social circumstances
change or notions of justice mutate.
To review, the decision in each common law case elaborates on
the legal experience (absent overruling). A judgment of importance
should not be made ad hoc: It should flovv from the principles and
policies embedded in the precedents, according to the conventions
of the profession. Each decision is creative to the extent that each
problem case presents a new question. So the law's purposes
should be elaborated as necessary to justify the decision and its
6. PURPOSES 111
forebears as members of a family-to show how the decisions are

consistent with each other and may contribute to a more orderly
and just society.
2. Purposes in Statutes
To illustrate the play of purpose in statutory cases, recall that

the principle of legislative supremacy generaHy requires judges to
subordinate their personal views to the enacted views of the
people's representatives in the legislature. The rules enacted by
legislatures, however, do not dictate the conclusion in a case
entirely by deduction. A judgment of importance is required to
place a problem case in a class of cases designated by an enacted
rule. The statute's purposes help make the judgment of importance
here, similarly to the common law.
Recall the case of Franny Farmer and Morris Auster in
Chapter 3. Farmer reneged on a deal to sell peaches to Auster.
Auster wanted to sue Farmer to enforce the contract. Section
2-201(1) of the Uniform Commercial Code stood as a possible
barrier to Auster's action. It provided that, to be enforceable, a
class of contracts for the sale of goods must be in writing and
signed by the party to be charged (Farmer). The Auster /Farmer
contract was oral. Auster's prospects in court consequently
depended on an exception to §2-201(1), set forth in §2-201(2). This
exception might allow a court to enforce the Auster /Farmer
contract because Auster sent Farmer a confirmatory letter to
remind her of their deal, and she did not object. This would be so,
however, only if the contract was "between 1nerchants"-if both
parties were "chargeable with the knowledge or skill of mer-
chants" (§2-104(3)). Though it is clear that Auster is chargeable as
a "merchant," it is not at all clear that Farmer is chargeable as one
for the purposes of this case. A judgment of importance is
necessary to determine which of the facts in the case will or should
lead a judge to decide whether the Auster /Farmer contract was
"between merchants."
A judge in Auster's case should decide whether the law's
purposes are better implemented by requiring Auster to put such
deals in writing or by requiring Farmer to object to confirmatory
letters. The judge, however, is unlikely to judge importance in
112 6. PURPOSES
Auster's case apart from the legal experience. Principles of

constitutional democracy and legislative supremacy are deeply
ingrained in our legal culture. What matters is what the legislature
was trying to do in Article 2 of the Uniform Commercial Code
and, specifically, in §2-201 and the relevant definitional provisions.
Given a general background in the law of sales contracts, the
legislature's purposes can be discerned from the statutory text
together with its context. A judge in Auster's case should show
how a decision can be reconciled with the statutory materials so
that Auster is not required to act differently from how anyone
should act in the same circumstances. The judgment of importance
in Auster's case will flow from the principles and policies embed-
ded in the legal experience, elaborated as necessary to justify the
decision and in a way that implements the statutory scheme.
A central policy of the law of sales contracts encourages people
to promise to buy and sell goods, such as the projected exchange
of peaches for money between Auster and Farmer. Exchange
benefits society because trade contributes to economic prosperity.
The law assumes that people promise fe,wer such exchanges when
they hesitate to rely on others to keep their promises. Enforcing
contracts may encourage exchange by encouraging people to keep
their promises. Accordingly, the law generally enforces promises
to buy and sell goods.
As the requirement of a writing for enforcing a class of
contracts suggests, however, the law does not enforce all promises
to exchange goods. Voluntary trade contributes to economic
prosperity. Consequently, the law should enforce promises that
really were made and hold people to the terms of their promises
as made. The writing requirement may prevent unscrupulous
people from alleging promises or terms of promises falsely. The
tradition, at least, holds that written contracts are less likely to be
fraudulent or make a fraud easier to discover and prove. Accord-
ingly, some contracts are enforceable only if in writing and signed
by the party to be charged.
The policy of promoting exchange suggests that people should
be protected by the law when they reasonably expect and probably
rely on others to keep promises made as part of an exchange. The
principle supporting the requirement of a writing suggests that
sales contract law also protects people iNho allegedly made such
promises from being forced to keep promises they did not make.
6. PURPOSES 113
Note that the second standard allows some oral contracts that
really were made to go unenforced. It is usually easy for the
parties to put it in writing, and perhaps it is generally understood
in the trades that important contracts should be in writing. On the
other hand, perhaps the writing requirement should be abolished
because it invalidates too many contracts in a time of telephone
shopping and electronic mail; until it is abolished, however, a
judge is duty-bound to work with the law in force, attributing to
it the best justification that the legal community can muster.
These two purposes help to determine whether Farmer is
"chargeable as a merchant" such that the Auster /Farmer contract
was "between merchants" and that §2-201(2) will remove the
requirement of a writing signed by Farmer so that Auster can
enforce the deal. When Auster sent the letter to Farmer, reminding
her of their deal, he probably expected her to reply and object if
there in fact was no deal or if he got the terms wrong. Not
receiving a reply, he probably relied on a belief that the deal was
on and conducted himself accordingly. The purposes of this law
suggest that his expectation should be protected if it was reason-
able-if encouraging people like Auster to so rely would encour-
age exchange and without allowing the unscrupulous to stick
people like Farmer with phony deals too easily.
With this background, the more specific purpose embedded in
§2-201(2) can be brought to light. The statutory text says that the
law would consider Auster's reliance to be reasonable only if
Farmer was chargeable as a merchant, and a merchant is defined
as one who "deals in goods of the kind or otherwise holds [her]self
out as having knowledge or skill peculiar to the prac-
tices ... involved in the transaction." The W'ord otherwise in the
definition suggests that all people who deal in goods of the kind
are deemed to hold themselves out in the manner described. The
more pertinent textual definition of "between merchants," requir-
II
ing both parties to be chargeable with the knowledge and skill of
II
merchants," confirms that deals in goods of the kind" should be
interpreted this way. People who so hold themselves out-by
dealing or by representations-lead others to expect and probably
to rely on them to use their knowledge and skill, much as people
who make promises lead others to expect and rely on them to keep
their promises. In §2-201(2), then, a contract is "between mer-
chants" if the parties, by the nature of their businesses or by
114 6. PURPOSES
representations, lead others to expect an objection if a confirmatory

letter is received and is not an accurate confirmation of a real deal.
The precise question in Auster's case, then, is whether, by the
nature of her business or by representations, Farmer led Auster to
reasonably expect her to object to the confirmatory letter if it was
inaccurate. From the facts as stated in Chapter 3, there is no reason
to believe that Farmer made such representations. There are facts
that characterize the nature of her business: Her orchard was
described as a "small, family-run" business; she makes as few as
one peach sale for a few hundred boxes in a year. These facts are
important because they suggest Farmer 1nay not be a sophisticated
businessperson. On the other hand, as the UCC' s official commen-
tary suggests, almost everyone in business would be expected to
answer the mail. 6 On balance, therefore, Auster's expectation
seems reasonable. Farmer should have objected to protect herself
from fraud. The Auster/Farmer contract should be held to be
"between merchants." Section 2-201(2) applies to this transaction.
And the lack of a writing signed by Farmer would not bar
enforcement of the contract by Auster.
The statutory context confirms this interpretation of the text.
Article 2 as a whole contains many provisions looking to ordinary
business practices as a source of reasonable expectations. For
example, §1-102 provides that the statute should be liberally
construed and applied to promote its underlying purposes, among
which are the promotion of commercial practices. Section 1-205
provides that a contract should be interpreted in light of any usage
of the trade, which is defined as "any practice or method of
dealing having such regularity of observance in a place, vocation
or trade as to justify an expectation that it will be observed with
respect to the transaction in question." Numerous provisions
require the parties to sales contracts to conform to commercially
reasonable practices unless otherwise agreed. Moreover, the
drafting history of Article 2 and the secondary literature confirm
that this statute sought generally to bring the law of sales contracts
into conformity with reasonable commercial practices and expecta-
f, U.C.C. §2-104, Comment 2.

6. PURPOSES 115
tions. 7 Therefore, interpreting the definition of "between mer-

chants" to protect reliance on reasonable commercial practices
concerning confirmatory letters furthers the purposes of the law as
a whole.
The statutory definitions themselves are puzzling and inade-
quate if applied to the facts of a case deductively, based only on
the ordinary meanings of the words. As indicated in Chapter 3,
contradictory conclusions are possible and of equal logical validity.
Having generated a conventional justification from the text, its
context, and the secondary literature, however, the problem is far
less perplexing. The principle of legislative supremacy and the
policies of sales contract law involving the protection of reasonable
expectations and the promotion of voluntary trade reflect the legal
conventions as they bear on cases like Auster's. The conventions
of the legal community select from all logically possible conclu-
sions those that implement the law's purposes. A judge may make
a judgment of importance by synthesizing the legal experience in
light of the legal conventions.
In sum, members of the legal community generally consider

the proper judicial role to be limited by principles of stare decisis
and legislative supremacy. Judges acting in good faith do not make
the judgment of importance in a case in isollation. They take the
common law precedents or statutory materialls to be authoritative.
They work, if sometimes only implicitly, with the principles and
policies embedded in the legal materials, which are reliable
indications of the pertinent legal conventions. The law's purposes
are elaborated as necessary to decide new questions, reconciling
each decision with the relevant precedents or statutory matedals
and, together with the legal conventions, contributing to the law's
vision of a more orderly and just society.
7
E.g., U.C.C. §2-104, Comments 1 & 2; Richard Danzig, A Comment on the
Jurisprudence of the Uniform Commercial Code, 27 Stan. L. Rev. 621 (1975);
Zipporah Batshaw Wiseman, The Limits of Vision: Karl Llewellyn and the
Merchant Rules, 100 Harv. L. Rev. 465 (1987).
CHAPTER SEVEN _ _ _ _ _ _ _ _ _ __
The Judge's Perspective

THE PRECEDING CHAPTER highlighted the law's purposes-the
principles and policies that should justify legal rules and yield
reasons for classifying cases within the framework of law described
by the legal rules. To bring these roles out, however, I simplified
matters again, though not unrealistically. In many cases, only one
purpose lies behind a law or several relevant purposes have a
single implication for the decision of a case. Given agreement on
the facts, lawyers and judges should corne to the same legal
conclusion in such cases, even if they might give different explana-
tions for so doing. In many other cases, however, competing
purposes have conflicting implications. These are harder cases, in
which the law is unsettled and the outcome vvill be uncertain, even
when the facts are not in dispute.
The harder cases are more interesting, whether you seek profit,
public service, or fun from lawyering. They are more likely to
reach appellate judges and shape the law. The difference between
harder and easier cases is one of degree. Both are capable of
reasonable treatment in light of the legal conventions-what
members of the legal community generally vvill take to be required
by an orderly and just society in a case, as indicated by the legal
experience and the law's purposes.
A. Easier Cases
As a practical matter, it is not necessary to appeal explicitly to

the law's purposes in easier cases. The legal rules, by their
ordinary meanings, may do an adequate job of calling the impor-
tant facts into prominence. Or, the purposes may converge to
suggest a single result, making it superfluous to give the full
argument explicitly. The role of purpose is hidden either way.
However, understanding the hidden role of purpose in easier cases
is helpful to understanding its more complex and explicit role in
harder cases.
117
118 7. THE JUDGE'S PERSPECTNE
Imagine a simple case: Old Man Featherstone was a wealthy

landowner. He had no surviving immediate family of his own,
only nieces, nephews, and cousins. He also had a grandson born
out of wedlock, whom he barely knew. Featherstone was thought
to be a bit of a crank, and he deeply resented the frequent
self-ingratiating kindnesses of his relatives. He knew they were
motivated solely by greed. Following his death, the hopeful
relatives gathered to hear the will read. They were shocked to find
that the Old Man left his entire estate 1to his grandson. The will
was typed in proper form, duly signed by Featherstone, and
witnessed by two people whose signatures appeared.
Some of the relatives were upset enough to challenge the will.
Let us assume that the estate would go to two cousins if there
were no will or if the grandson had died before the Old Man.
Assume also that the relevant statute (the Statute of Wills) provides
that a person's property shall go to wh01never is named in a valid
will and that, to be valid, a will must be signed by the testator
(Featherstone), while sane and not under fraud or duress, and
in the presence of two witnesses whose signatures appear. To
apply this statute, a judge would reason deductively. Thus,
Featherstone's will named his grandson. It was signed by him and
witnessed by two people whose signatures appeared. From the
facts as stated, there is no reason to believe he was insane or a
victim of fraud or duress. Therefore, the will is valid. The grand-
son will inherit the estate. The case is easy.
You might think that purpose has nothing to do with this case
because deduction from the statute is wholly adequate. 1 So it
appears. But two variations on this case will show that the law's
purposes, too, justify this decision. Moreover, many easier cases
can be justified adequately only with recourse to purpose. Conse-
1
See Benjamin Cardozo, The Nature of the Judicial Process 112-114 (1921);
Ronald Dworkin, Taking Rights Seriously 81,337 (1978); H.L.A. Hart, Positivism
and the Separation of the Law and Morals, 71 Harv. L. Rev. 593, 607-608 (1958);
Neil MacCormick, Legal Reasoning and Legal Theory 100 (1978). Both Dworkin
and Hart seem to have had subsequent doubts about the logic of easy cases. See
Ronald Dworkin, Law as Interpretation, 60 Texas L. Rev. 527, 545 (1982); H.L.A.
Hart, Problems of Philosophy in Law, in 6 Ency. Phil. 264, 271 (R. Edwards ed.
1967). See also Frederick Schauer, Easy Cases, 58 S. Calif. L. Rev. 399 (1985).
7. THE JUDGE'S PERSPECTIVE 119
quently, deduction does not adequately justify easier cases 1n

general.
For the first variation, imagine Featherstone left two wills. One
gave the estate to two cousins; the other gave the estate to the
grandson. The will favoring the grandson bore a date subsequent
to the will favoring the cousins. Both \vills were otherwise
identical, save that the first will was witnessed by two people and
the second by three. 2
Assume the relevant statutory rules are the same as before with
the addition of a rule providing that a subsequent valid will
revokes a prior will. The grandson will inherit the estate only if the
second will is valid; that is, if it was signed by Featherstone, while
sane and not under fraud or duress, and in the presence of two
witnesses whose signatures appear. Does the estate go to the
cousins or the grandson? The fact of a third witness on the
subsequent will is the only possibly significant problem, though it
seems as clear as before that a judge should award the estate to the
grandson.
You cannot justify this result by deduction from the statutory
rule, which requires the will to be witnessed "by two witnesses."
The statute does not say "at least two witnesses" or "at most two
witnesses" or "two and only two witnesses." "Three witnesses"
surely is not the same as "two witnesses." You cannot possibly
deduce either that the will is valid or not valid. It is not known
whether the fact of a third witness is important. A judge, too,
would be puzzled if only rules, facts, and logic were considered.
The law's purposes lurk behind the easiness of this case: Why
should two witnesses be required for a vvill to be valid? The
conventions of the profession assign several purposes to such a
requirement. 3 In theory, the witnessing ceremony serves to bring
home to the testator the legal significance and possible finality of
what he is doing; the requirement tries to assure that he acts with
due deliberation. The witnesses also serve to reduce the likelihood
of some person fabricating a will for another, on the assumption
that it is harder to get two people to witness a fraudulent docu-
2
This illustration is adapted from William R. Bishiin & Christopher D. Stone,
Law, Language and Ethics 724-726 (1972).
3
See generally Ashbel G. Gulliver & Catherine J. Tilson, Classification of
Gratuitous Transfers, 51 Yale L.J. 1, 5-15 (1941).
120 7. THE JUDGE'S PERSPECTIVE
ment and escape detection. (Perhaps one witness will not do

because that person could be the perpetrator of the fraud.) And the
witnesses might later provide evidence of the testator's mental
health when he signed his will and of other circumstances at the
signing. (Perhaps one witness will not do because of the greater
risk that that person will be unavailable should a dispute arise
years later.) Because three witnesses serve all of these purposes
better than two witnesses, there is no reason for the third signature
to invalidate the subsequent will.
This conventional justification of the second case underlies the
first case along with the previous justification. Two witnesses serve
adequately to assure due deliberation by the testator, to reduce the
likelihood of fraud, and to provide evidence of the circumstances
at the signing, as determined authoritatively by the legislature.
Whatever purposes the statute might serve, they have a single
implication in the first case. A judge should decide that that will
is valid (in this respect) because of both deduction and, implicitly,
the statute's purposes.
The first case can be justified by both deduction and purpose,
but the second can be justified only by looking to underlying
purposes. Now consider another variation. The facts are similar to
the first case except that the will leaving the estate to the grandson
was not witnessed but was in Featherstone's handwriting and was
signed by him. (Such a will is called a holographic will.) Assume
that, the preceding year, the highest appellate court in the relevant
jurisdiction held that a holographic will executed by some other
testator was valid even though not witnessed. A judge should
decide the third case of Old Man Featherstone's will the same way.
This, too, is an easier case.
Like the second case, you cannot justify this one by deduction
from the statute of wills. Nor can you justify it sufficiently by
analogy from the precedent ,.vithout calling on the law's purposes
to make the judgment of importance. 4 The third case differs from
the preceding cases because the plausible purposes generate
reasonable arguments for the cousins and the grandson.
The absence of a witnessing ceremony might weaken the law's
effort to assure due deliberation by the testator and to provide
4
The hypothetical assumes the precedent is based on the common law.
evidence of the circumstances at the signing. But the fact that the
will was in Featherstone's handwriting tends to indicate some
deliberation (as contrasted with an oral will) and to reduce the
possibility of frauds that could not be detected by handwriting
experts. Whether the fact of no witnesses or the fact of a handwrit-
ten will is more important under the circumstances depends on
how the reasons stemming from these novv-competing purposes
are weighed.
This could be a harder case. The preceding year, however, the
court in the precedent was presented with this very question under
similar circumstances. By holding that the holographic will in that
case was valid, the court resolved the conflict in favor of the
grandson's position. Consequently, a judge should decide that the
third will is valid in this jurisdiction because the recent decision is
stare decisis-a thing decided. The case is easier because the
purposes, as weighed to decide the recent precedent, point to one
outcome.
We have examined three easier cases. Deduction justifies only
one, but the conventionally understood purposes of the law justify
all three. Each represents a kind of case that will be found in the
legal world. We can conclude, then, that easier cases in general are
better justified by the law's purposes, whether or not they are
made explicit.
The singular implication of the relevant purposes makes easier
cases easier. This implication may result in two ways: In the first
two cases, all relevant purposes pointed in the same direction; in
the third, the purposes pointed in different directions, but a recent
precedent resolved the competition. As a practical matter, given
the precedent and stare decisis, the relevant purposes had one
implication in that case as well.
B. A Harder Case
Of course, the relevant purposes in a case can have multiple

and competing implications for judicial decision. Purposes may
conflict in ways not resolved by the precedents and enactments, as
they had been in the third case of Old Man Featherstone's will. Or
a previously unaccepted but plausible purpose may compete for
acceptance as a new convention against a traditionally accepted
purpose. In either situation, both sides in such cases will have

reasonable arguments. The judgment of importance in such cases
is harder.
Consider a fourth variation. 5 Old Man Featherstone informed
his cousins and grandson that his estate ,Nould go to the grandson
under a duly executed will. Some w·eeks later, Featherstone
discovered that his grandson had concealed his heavy debts due
to a gambling obsession. Featherstone confronted his grandson,
who told the Old Man to mind his own business. Featherstone
threatened to disinherit his grandson \t\rho, in response, drew a
knife and stabbed Featherstone to death. The grandson was
convicted of murder in a criminal court and sentenced to fifteen
years in prison. The cousins challenge the will in court because
they do not think a man's murderer should inherit his property.
Assume that the relevant statutory lavv provides that a person's
estate shall pass upon death to the person or persons named in a
valid written will. Assume also that Old Man Featherstone's will
was in writing and was valid, that no statutory or other enacted
rule denies the right of a murderer to inherit from his victim, and
that the cousins clearly would inherit the estate if there were no
will or if the grandson had died before the Old Man. It might
appear that the law gives Old Man Featherstone's estate to his
grandson. But this deduction is troubling; most of us probably
think it should matter that the grandson n1urdered Featherstone to
prevent him from changing the will. A judge's problem is to show
that the law precludes the grandson from taking the estate because
that fact is legally important.
The statute's purposes in this case indicate that both sides have
plausible legal arguments. The grandson's lawyer can argue that
the principle of legislative supremacy requires the court to follow
the law exactly as laid down by the legislature. It would promote
disorder for the court to guess what the legislature might think
about a case it had not thought about. The purpose of the statute
is revealed best by its terms, taken literally. The law thus gives the
estate to the person named in a valid will-in this case the grand-
son. This result gives effect to the revealed intention of the testator
5
This illustration is adapted from the famous case of Riggs v. Palmer, 106
N.Y. 506 (1889).
and avoids any need to guess the intentions of a dead man.

Sticking to the written document promotes the orderly passing of
property from one generation to another. Therefore, this argument
concludes, the legislature should change the law if it needs to be
changed.
On the other hand, the lawyer for the cousins can argue that
the underlying purpose of the law is to provide for the orderly and
just passing of property from one generation to the next according
to the testator's desire. In normal circumstances, giving the estate
to those named in the will furthers this purpose well. This,
however, is an unusual case in which following the literal meaning
of the will does not further the law's purposes. No reasonable
testator could want his estate to go to his murderer, and no
reasonable legislature could have intended such a result. Moreover,
justice would not be promoted by allowing one to inherit after
murdering the testator to prevent a change in a will. A basic
principle of the law is that no person shall profit by his own
wrong. Therefore, it is important that the grandson murdered the
Old Man. The statute, this argument concludes, does not require
the grandson to inherit the estate.
This is a harder case due to multiple purposes with competing
implications. The grandson's argument places greater stress on the
orderly passing of property from one generation to the next. It
emphasizes the literal interpretation of the statutory text, the
clarity of the testator's written will, and the uncertainty of guessing
a dead man's intention. That the grandson rnurdered the testator
is not important for this purpose. By contrast, the cousins'
argument stresses the just passing of property from one generation
to the next. It emphasizes the unusual circurnstances, the probable
intention of the testator and the legislature, and the injustice of
rewarding wrongful behavior. The fact that the grandson mur-
dered the testator to prevent a change in the will is important for
this purpose.
The competing purposes in this case all are embedded in the
legal experience, generating legal reasons ·with weight. A judge
should decide which are the stronger under the circumstances.
Parenthetically, note that harder cases also occur when the
implications of embedded purposes differ from the implications of
a plausible principle or policy that perhaps should back the law
but has yet to enjoy conventional acceptance within the legal
community. The latter kind of purpose may have some support

among members of the legal community and be articulated in
appellate briefs, concurring and dissenting judicial opinions, or the
secondary literature. A case that accepts such a purpose and is
decided according to its implications is likely a "landmark''
decision, which will overrule established precedents, thereby
changing the law. Judges in such cases exercise a special lawmak-
ing power that is not only a matter of legal reasoning. Our
attention here focuses on cases in v\rhich legal reasoning is
dominant.
C. Deciding Harder Cases
We have seen that easier cases in general are easier due to the
single implication of the relevant purposes in the case; harder cases
are hard due to the conflicting implications of competing purposes
in the case. The next question concerns how the judgment of
importance is made in harder cases.
1. Webs of Belief
Judicial reasoning is but a stylized and rigorous version of the

reasoning used in everyday life. Beneath the dignified bearing, the
black robes, and the air of mystical capacity, judges are human
beings who think basically as you and I do. Much depends on
judgment, though professional legal judgments are influenced by
legal education, developed by experience, and refined in each case
through legal research, reflection, and argument within the legal
conventions. Something useful can be said about legal reasoning
and judgment in harder cases if we first consider human reasoning
and judgment in simple everyday problem situations.
Imagine you and a fellow birdwatcher are walking one
afternoon along a path in a park when your friend calls attention
to a bird in a tree. She asks what kind of bird it is, and you say, "I
believe it is a crow." But your friend says she thinks it is a
blackbird. You argue for a bit, during 'Which you say that you
believe yourself to be an expert birdwatcher and that you care and
understand about these things. The disagreement is left unre-
solved. That evening, you pick up a copy of the Audubon Society

magazine, which you believe to be careful and expert about birds.
You find a picture of a bird you think is the same kind as the bird
in the park. The caption, to your relief, says "Crow." You feel
confirmed in the opinion ventured earlier and believe you were
correct.
By contrast, imagine the caption belo'v\r the picture in the
magazine had said "Blackbird." Depending on how serious a
birdwatcher you are, you might feel disappointed. In the first
situation, all of your beliefs together had a single implication
under the circumstances. You could believe that it was a crow in
the park, and that you are an expert bird watcher, that the
Audubon Society magazine is careful and expert about birds, that
the bird in the picture is of the same kind as the bird in the park,
and also that you care and understand about these things. In the
second situation, however, your beliefs have conflicting implica-
tions under the circumstances. Something must give.
You have several alternatives. You can rrtaintain that the bird
in the park was a crow if you drop the belief either that the
Audubon Society magazine is careful and expert about birds or
that the bird in the picture is the same kind as the bird in the park.
Or you ca11 decide that the bird in the park was really a blackbird,
as long as you drop the belief that you are an expert birdwatcher.
Or you can decide that it all makes very little sense, if you drop
the belief that you care and understand about these things.
However, you cannot maintain that the bird in the park was a
crow and, at the same time, hold unmodified all of the beliefs you
held before seeing the bird in the picture with its caption.
The beliefs bearing on this matter are like a web in which each
relevant belief is connected with several others that, in turn, are
connected with many of the beliefs that you hold. The web is a
"seamless web" because no one belief is the starting point from
which all others necessarily follow. Each belief is supported by a
complex network in which all, ultimately, are mutually reinforcing
in light of experience. A change in any one belief has a ripple
effect causing changes in a number of other beliefs. Any belief can
be changed without producing a tattered ,.veb if consequential
changes are made.
Accordingly, which alternative you select in the second
situation will be a function of many, if not all, of your beliefs about
yourself and the way the world works. It would be easier for me,
for example, to give up the belief I foolishly held that I am an
expert birdwatcher. I can quickly concede that the Audubon
Society is more likely to get it right than I am. Also, I would not
have noticed enough about the bird in the park to be sure it was
not the same kind as the bird in the picture. I want things to make
sense, so I would conclude that the bird in the park probably was
a blackbird.
My friend Judy, however, happens to be a professor of
ornithology. Under the same circumstances, she would find it
harder to give up her belief that she is an expert birdwatcher,
easier to challenge the accuracy of the Audubon Society magazine,
and harder to conclude that her initial belief that the two birds
were of the same kind was wrong. She, too, wants these things to
make sense. So she might conclude that the magazine had made
an error. My other friend, Clarence, thinks of himself as an expert
in everything and finds it hard ever to admit an error, most of all
to himself. He also is proud to serve on the Audubon Society's
board of directors. Perhaps he is more likely to conclude that the
bird in the park was not the same kind as the bird in the picture.
And my friend Albert, who is not very simart, is quite accustomed
to confusion. Perhaps he is more likely to hold contradictory
beliefs in this situation and, with a shrug, drop his belief that he
cares and understands about these things.
Logically speaking, anyone could select any alternative in these
circumstances if he were willing to adjust his web of beliefs as
necessary to accommodate the alternative selected. 6 Psychological-
ly speaking, however, each of us is likely to select the alternative
requiring the fewest adjustments to maintain the coherence of our
webs of beliefs.7 Moreover, we try to conserve those beliefs that
are most likely to work well given what we are trying to
do-those beliefs that, if acted on, have not and likely will not
produce disappointing experiences given our purposes. (Of course,
there is no guarantee that operating this way will provide us with
all and only true beliefs.)
See generally Willard Van Orman Quine & J. G. Ullian, The Web of Beliefs
6
(1970); Willard Van Orman Quine, Two Dogmas of Empiricism, in From a Logical
Point of View 20 (W. Quine ed. 1961).
7
Leon Festinger, A Theory of Cognitive Dissonance (1957).
Accordingly, I am likely to conclude tha1t the bird in the park

was probably a blackbird, though maintaining that it is a crow is
as possible logically. I hold my beliefs about my birdwatching
skills lightly and can give them up easily; I am too inexperienced
to be attached to them. I hold my beliefs about the reliability of
reputations of groups like the Audubon Society more deeply. I
would much prefer to alter my belief about my birdwatching skills,
which requires me to alter very few other beliefs that I depend on,
than to alter my beliefs about the reputation of the Audubon
Society. In the way I manage my general web of beliefs, I think I
am like other people. Judges, too, are people; they just have
distinctive experiences and goals.
2. Webs of Belief About Law
The birdwatcher situations are comparable to easier and harder

cases. The first birdwatcher situation is like easier cases because all
relevant beliefs, like the purposes in the first cases of Old Man
Featherstone's will, converge to suggest a single result. Harder
cases are like the second birdwatcher situation because a number
of relevant beliefs, like the purposes in the fourth case of Old Man
Featherstone's will, have conflicting implications. However, the
judicial response to a harder case should be mtore constrained than
a layperson's response in a situation of conflict. Among a judge's
core beliefs should be those concerning judicial responsibility: The
judge is under a duty to uphold the law, not: to decide according
to personal views which are not coincident ·with the law. There-
fore, a judge's decision should be accommodated coherently with
a deep commitment to the rule of law and all that this ideal
entails.
Before being confronted with the facts of a harder case, the
judge holds a web of beliefs about law. For example, the judge in
the fourth case of Old Man Featherstone's will (in which the
grandson murdered the Old Man) probably \Vould include among
the law's purposes that the law in such circumstances should
provide for the orderly and just passing of property from one
generation to the next; that the law should give effect to the
revealed intention of the testator and avoid guessing a dead man's
intentions; that the courts should follow the law as laid down by
the legislature; that following the literal meaning of the will does
not always serve the law's purposes; that no reasonable testator
could want his estate to go to his murderer; that no reasonable
legislature could have intended such a result; and that a basic
principle requires that no person should profit by his own wrong.
The previously decided cases might be reconciled coherently under
all of these principles and policies. On being confronted with the
problem case, however, the judge bec01nes aware of competition
among purposes. The law's coherence seems to fall apart.
Such an experience may lead the judge to review her web of
beliefs about the law to determine which parts should be adjusted
to accommodate a decision in the case as coherently as possible.
Logically speaking, the judge can conclude either that the grand-
son should inherit or not inherit if she is willing to make the
necessary adjustments to her beliefs about law. Like a layperson,
the judge will select from among the logical possibilities those that
can be reconciled with past experience in light of relevant pur-
poses. But the layperson considers a belief to work well if it
comports with his past experiences and probably will not lead to
future disappointing experiences to him in light of his personal
goals. By contrast, a judge is a professional with serious obligations
to serve the community. She should consider a belief to work well
if it comports with the legal community's past experiences and will
not lead to disappointing experiences for the legal community. She
will try to make the fewest adjustments necessary to accommodate
a decision coherently with the precedents and legislative materials
in light of the legal community's conventional justifications for
what has gone before.
To elaborate, on taking the oath of office in good faith, a judge
assumes a duty to support the rule of la'w, which requires regular
and coordinated action among legal officials. The judge's web of
beliefs about law consequently should include the law's conven-
tional justifications, which are crucial for regularity and coordina-
tion. Order and justice should be at the center, where they can
support the remainder of the web, which should work out the
implications of those values for legal precedents, rules, principles,
and policies. To have value in pursuing order and justice, a legal
web of beliefs should synthesize the legal experience and the law's
purposes as a coherent and valuable vvhole. The principles and
policies that give coherence to the rules and precedents should be
those accepted by the legal community as conventional justifica-

tions for the laws. The conscientious judge faced with a harder
case should reach the decision that requires the fewest adjustments
to maintain or enhance the coherence of the law, given the
centrality of order and justice. Accordingly, a judge should prefer
to abandon or modify those beliefs about law that deviate more
from the conventions of the legal community.
The judicial opinion in the case of Riggs v . Palmer,8 from which
the harder case of Old Man Featherstone's will is adapted,
illustrates well the operation of this conventional perspective on
legal reasoning. Not surprisingly, the New York Court of Appeals
held that a person who murders a testator to prevent the testator
from changing his will may not inherit property under the will,
notwithstanding a statute of the kind described above. The
interesting part is the court's reasoning in support of this conclu-
sion.
After stating the facts in that case, Judge Robert Earl, writing
for the court, noted the grandson's argument that the will was
made in proper form and must have effect according to the letter
of the law. He conceded that the relevant statutes, if literally
construed, gave the property to the murderer. He then embarked
on an extensive discussion of the principle of legislative suprema-
cy.
Judge Earl abandoned the view that the principle of legislative
supremacy requires statutes to be interpreted literally in all cases,
if he ever had held it among his beliefs about law. He retained the
view that statutes should be interpreted to give effect to the
legislature's intention. He thus maintained the principle of
legislative supremacy, which is basic in a democratic political
system, by recognizing one version of it and rejecting another.
Further, Judge Earl explained in effect, the rejected version of
legislative supremacy is not generally supported by the conven-
tions of the legal community. He quoted extensively from eminent
legal writers on statutory law:
It is a familiar canon of construction that a thing which is within

the intention of the makers of a statute is as much within the
8
115 N.Y. 506 (1889).
statute as if it were within the letter; and a thing which is within

the letter of the statute is not within the statute, unless it be
within the intention of the makers. The writers of laws do not
always express their intention perfectly, but either exceed it or
fall short of it, so that judges are to collect it from probable or
rational conjectures only, and this is called rational interpreta-
tion .... 9
Judge Earl cited or quoted from no fewer than five legal treatises
to this effect, evidencing the legal corrumunity's attitude. He also
reported two proverbial cases that exerr1plify the operation of this
view of statutory interpretation:
There was a statute in Bologna that whoever drew blood [i.e.,

engaged in sword-play] in the streets should be severely pun-
ished, and yet it was held not to apply to the case of a barber
[i.e., a surgeon] who opened a vein in the street. It is commanded
in the Decalogue that no work shall be done upon the sabbath,
and yet, giving the command a rational interpretation founded
upon its design, the Infallible Judge held that it did not prohibit
works of necessity, charity or benevolence on that day. 10
In light of such authority, Judge Earl w·as arguing, it is hard to

believe that many lawyers and judges reasonably expect judges to
follow statutes literally in all cases. Abandoning that version of the
principle of legislative supremacy would not be greatly incompati-
ble with the practices and dispositions of the legal community.
This left it necessary for him to determine what the legislature
intended in the circumstances of this case and like cases. The
thought that the legislature might have intended the murderer to
inherit under his victim's will did not fit well within Judge Earl's
web of beliefs about law:
What could be more unreasonable than to suppose that it was the

legislative intention in the general laws passed for the orderly,
peaceable and just devolution of proper1ty, that they should have
operation in favor of one who murdered his ancestor that he
9
Id. at 509.
10
Id. at 511.
might speedily come into the possession of his estate? Such an

intention is inconceivable. 11
Moreover, Judge Earl argued, application of a certain maxim of

the common law (a kind of general principle) was relevant to the
"orderly, peaceable and just devolution of property":
No one shall be permitted to profit by his own fraud, or to take

advantage of his own wrong, or to found any claim upon his
own inequity, or to acquire property by his own crime. These
maxims are dictated by public policy, have their foundation in
universal law administered in all civilized countries, and have
nowhere been superseded by statutes. 12
To illustrate, he gave a number of analogous cases exemplifying

application of the maxim. In one, it was held that a person could
not collect under an insurance policy according to its literal terms
when he murdered the insured to make the benefits payable. 13
Justice Stephen J. Field in that case had written for the Supreme
Court:
It would be a reproach to the jurisprudence of the country, if one

could recover insurance money payable on the death of a party
whose life he had feloniously taken. As well might he recover
insurance money upon a building that he had willfully fired. 14
In other cases, Judge Earl reported, the courts held without

statutory basis that a will would be set aside if procured by the
fraud or undue influence of the person in vv-hose favor it would
operate. 15 Here, the effect of Judge Earl's argument is to suggest
by analogies that members of the legal community generally are
disposed to apply the principle about not allowing a person to
11
Riggs v. Palmer, 115 N.Y. 506, 511 (1889) (emphasis added).
12
Id. at 511-512. Judge Earl cited four treatises on foreign law to support his
assertion that the maxim has its foundation in "universal law administered in all
civilized nations."
13
N.Y. Mut. Life Ins. Co. v. Armstrong, 117 U.S. 591 (1886).
14
Id. at 600.
15
Riggs v. Palmer, 115 N.Y. 506, 512 (1889) (citing two cases).
profit from his own wrong even in the face of the literal terms of
an applicable statute and will.
Judge Earl's difficulty in accommodating a decision for the
grandson with the legal experience, including the law's commit-
ment to the above-quoted maxim and the presumed wisdom of the
cited precedents, is further revealed by a number of hypothetical
cases he posed in the analogical form:
Here there was no certainty that this murderer would survive the
testator, or that the testator would not change his will, and there
was no certainty that he would get this property if nature was
allowed to take its course.... If he had met the testator, and
taken his property by force, he would have had no title to it.
Shall he acquire title by murdering him.? If he had gone to the
testator's house, and by force compelled him, or by fraud or
undue influence had induced him to will him his property, the
law would not allow him to hold it. But can he give effect and
operation to a will by murder, and yet take the property? To
answer these questions in the affirmative, it seems to me would
be a reproach to the jurisprudence of our state, and an offense
against public policy. 16
Thus, the grandson's argument depended on the view that the

principle of legislative supremacy requires a literal interpretation
of a statute in all cases. That view, however, conflicted with one
focusing on the probable intention of the legislature and the
principle reflected in the maxim. More important, it cannot be
reconciled with the many real and hypothetical cases Judge Earl
discussed. In his opinion, it seems, the policy of literal statutory
interpretation does not work well; if recognized, it would cast
doubt on much of the legal experience, displace a larger number
of accepted legal conventions, and lead to disappointing experi-
ences for the legal community. His citations to legal writers and
cases support his belief that this purpose is not thought to con-
tribute much to a more orderly and just society in this case and
like cases.
Judge Earl had less difficulty accorrunodating a decision for the
other relatives. In his opinion, it seems, the policy of "rational
16
Id. at 512-513.
interpretation" and the maxim are more strongly supported by the

legal community in similar circumstances, as evidenced by the
many cases he discussed. Retaining these principles and policies,
and therefore deciding for the other relatives, conserves the
conventions that work well. So deciding maintains greater
coherence in the law in its entirety than the available alternative,
which injects an absurdity. Accordingly, a decision for the relatives
was thought better to contribute to the orderly and just passing of
property from one generation to the next, where the implications
of order and justice were a function of the conventions of the legal
community. 17
To be clear, there was nothing inevitable about the result in
Riggs v. Palmer. It was as possible logically for Judge Earl to retain
the policy requiring a literal interpretation of statutes and to hold
for the grandson. If it were not, the case probably would not have
been litigated to the highest court in New York. Moreover, my
point here is not that literal interpretation has no place in law and
legal reasoning. Rather, literal interpretation its not required by the
nature of law or judging. Literal interpretation, and other kinds of
interpretation, should be employed when they better further the
law's purposes in the relevant context. This is not always the case.
In sum, legal reasoning in law cases is complex. We have
17
This approach to easier and harder cases is comparable to Professor
Dworkin's early approach to hard cases. See Ronald Dworkin, Taking Rights
Seriously 81-131 (1978). Dworkin would have judges in hard cases follow
principles that figure in the "best" justification of the legal experience, where the·
"best" justification is a grand theory that (1) fits a substantial part of the legal
experience and (2) is most justifiable in political morality even if less consistent
with the legal experience than an alternative theory. Ronald Dworkin, Law's
Empire (1986). The approach in this book emphasizes the judge's responsibility
to the legal community's conventions of argument and judgment. This approach,
unlike Dworkin's, does not claim that there is a "single right answer" in hard
cases. It requires judges to coordinate their decisions by conforming to the
conventions of the legal profession (including conventional principles and
policies), rather than abstract political morality. When the conventions break
down due to controversy, judges must make a little :law to settle the dispute.
Judges should do so, however, in an effort to win the support of the legal
community by building on convention, not by acting alone and without
coordination to reach a right answer as a matter of abstract morality or
idiosyncracy.
travelled far from the simple versions of deductive and analogical

reasoning that give legal reasoning its form of expression. A judge
need not decide a case by confronting the facts with a single rule
or precedent in mind or with wholly personal value preferences,
either way producing arbitrary decisions too often. A judge should
judge importance according to which decision requires the least
adjustment to accommodate the decision coherently with the facts
of the case and the legal experience, interpreted in accordance with
legal principles and policies, and supported by conventions of the
legal community.
CHAPTER EIGHT
The Lawyer's Perspective

ONCE YOU VIEW THE judicial decision in harder cases as suggested
in the preceding chapters, we can pose a number of entirely
practical questions: How does a judge cope vvith the huge mass of
information within the legal materials? Hovv do lawyers develop
a case and influence which decision a judge will settle on as the
best? How do the analogical and deductive forms of legal reason-
ing work in legal advocacy? In this chapter, I will use these
questions to probe the practical implications of conventional legal
reasoning. I hope you will see that the lawyers play a key role in
presenting a case to a court so the judges are faced with a
manageable problem. Consequently, capable lawyers can have a
substantial influence over the way in which a judge will approach
a case, including whether it will be perceived as easier or harder.
I also hope you will see that the forms of legal reasoning, together
with the law's purposes and the legal conventions, can help you
make reasonably reliable predictions and effective legal arguments,
even in many harder problem cases.
A. How Judges Get Started
In principle, the entire law is involved in each judicial decision.

The analogical and deductive forms themselves imply a need to
consider the entire law. Sound analogies not only require judges
to decide each problem case as like cases have been decided; in
principle, they also should distinguish each case from all other
cases that might require a different result. Sound syllogisms not
only require judges to decide each case by deduction from a rule
of law; in principle, they should be sure that all rules requiring a
different result are not applicable. This does not change when we
take a broader view of the resources available for legal reasoning.
We then try to accommodate a decision coherently with the facts
of the case and the legal experience, interpreted in accordance with
legal principles and policies, supported by conventions of the legal
community. This, too, involves the entire lavv.
135
136 8. THE LAWYER'S PERSPECTIVE
In practice, however, no one can negate the possible implica-

tions of all other cases and rules exhaustively. Nor can anyone
knowingly accommodate each decision vvith the whole of the law.
Consequently, suggesting that judges decide harder cases in
accordance with the law in its entirety m.ay seem a truism with no
practical implications. It may seem to enjoin you to think about
everything before doing anything-effectively ensuring that you
will do nothing. The extent of the law is surely vast. You would be
disabled from acting if you had to view and digest the entire law
before giving advice in any case. To make law and legal reasoning
manageable, you must select from the vvhole law those parts that
are most germane to a particular case.
Law and legal reasoning should be understood as they are
used in their real-world contexts, not only in the abstract as a
philosophical matter. In context, they are not so overwhelming.
Lawyers in fact make their predictions and arguments in light of
what a judge will or should do in a case. Judges in fact make their
decisions in cases that have materialized. The courts' procedural
rules limit how a case can be brought to a court and how lawyers
must present the case. The context set by the case refines the focus
of legal reasoning for each judicial decision.
For example, a civil lawsuit is initiated when the plaintiff files
a complaint with a court. In the federal courts and many state
courts, a complaint must include "a short and plain statement of
the claim showing that the [plaintiff] is entitled to relief," together
with a request for relief, such as the payment of money by the
defendant. 1 The defendant then is required to respond in one of
several ways. Among them is a defendant's right to ask the court
to decide that the plaintiff's complaint "fail[s] to state a claim upon
which relief may be granted." 2 If the court so decides, the plain-
tiff's complaint is dismissed and the la'wsuit ends immediately;
otherwise, the parties proceed to the next step in litigation. A court
will dismiss a complaint on this ground i:f, for example, it does not
1
Fed. R. Civ. P. S(a).
2
Fed. R. Civ. P. 12(b)(6).
8. THE LAWYER'S PERSPECTIVE 137
provide sufficient information to give the defendant's lawyer a

reasonable idea of what facts and law might sustain the plaintiff's
lawsuit. Otherwise, the defendant would not have a fair opportuni-
ty to prepare a defense.
Accordingly, bringing a case to court requires the plaintiff's
lawyer to announce at the outset, albeit in a truncated way, the
legal basis of the plaintiff's claim. In effect, the plaintiff's lawyer
must point to a part of the legal experience and assert that that
part entitles the plaintiff to the requested relief. At this early stage
of litigation, the complaint need not set forth a detailed legal
argument or even cite precedents and legal rules. The complaint
needs to inform the court and the defendant of the facts pointing
to a general area of law that might allow the plaintiff to prevail as
claimed. A complaint that gestures at the law as an entirety
without alleging facts constituting a cognizable legal claim would
be dismissed for failing to state a claim upon which relief may be
granted.
To illustrate, consider a case involving a property owner in the
role of defendant and a trespasser in the role of plaintiff. 3 The
property involved was an unoccupied farmhouse containing
personal property. Over a ten-year period, unknown persons had
broken into the farmhouse and damaged it. They took some of the
defendant's possessions. The defendant then set up a booby trap,
involving a shotgun, in the bedroom. The plaintiff was shot and
seriously wounded in the lower leg after entering the house and
opening the door to the bedroom. The plaintiff had pleaded guilty
to a minor criminal trespass charge, and he then brought a civil
action against the defendant for damages involving personal
injuries, medical expenses, and loss of earnings occasioned by the
shotgun wound. Other facts will appear below.
A lawyer for the plaintiff might file a comLplaint in proper form
as follows:
3
The illustration is based on Katko v. Briney, 183 N.W.2d 657 (Iowa 1971).
IN THE DISTRICT COURT

OF THE STATE OF WILLARD
IN AND FOR LUDWIG COUNTY
Terry Tress,
Plaintiff COMPLAINT
v. Civil No. 95-1949
Oscar Olner, Trial by Jury Demanded
Defendant
The plaintiff, Terry Tress, states for his complaint against defen-
dant, Oscar Olner, as follows:
1. On or about July 16, 1995, plaintiff, Terry Tress, was shot by

a spring gun trap set by defendant, Oscar Olne·r, in an unoccupied house
owned by the defendant.
2. Defendant, Oscar Olner, erected the spring gun trap with an

intention to cause death or serious bodily injury to people entering the
premises.
3. Defendant was not privileged to use deadly force or to erect

the spring gun trap.
4. As a direct and proximate cause of the defendant's wrongful

acts, plaintiff Terry Tress's leg was seriously injured, and plaintiff was
prevented from working for some time, suffen~d great pain of body and
mind, and incurred expenses for medical attention and hospitalization
in the sum of forty-two thousand, seven hundred, and twenty dollars
($42,720).
WHEREFORE plaintiff, Terry Tress, demands judgment against

defendant, Oscar Olner, in the sum of forty-tivVo thousand, seven hun-
dred, and twenty dollars ($42,720) plus interest from the date the cause
of action arose together with his costs and expenses of prosecuting this
action.
In most jurisdictions, this cornplaint states a claim upon which

relief may be granted. By its allegations, it effectively invokes
certain well-known rules and precedents in the common law of
torts. All common law jurisdictions recognize a claim for battery,
which occurs when one person intentionally causes harmful or
offensive physical contact to the person of another without excuse
or justification. If the defendant's lawyer were to ask the court to
dismiss the complaint for failing to state a claim upon which relief
may be granted, the plaintiff's lawyer could cite the common law
rules and precedents in a brief in opposition to the defendanf s
request.
Only then would the judge be in a position to decide the
question. She would not, however, have to think about everything
before doing anything. The complaint and the rules and precedents
cited in the parties' briefs would be before the court. These
documents would call the judge's attention to parts of the legal
experience that, in the lawyers' opinions, are relevant to the case.
The judge's reasoning would focus on the cited rules and prece-
dents, though she could draw as well on her own research and
general knowledge concerning the law of torts and related criminal
or property law, theories of compensation and responsibility, rules
and theories of civil procedure, and other aspects of the law. Her
decision would be manageable because the lawyers will have
presented the case in a manner that poses a discrete question and
proposes a reasoned decision.
In this respect, the judicial decision on a defendant's request to
dismiss a complaint for failure to state a claim upon which relief
may be granted is typical of almost all judicial decisions. Such
decisions are made in a context with structure. Courts generally act
in concrete cases at the specific request of a party. The request
must pose a specific legal issue. The possible requests that a party
can make are fixed in advance by the procedural forms. Each party
is afforded an opportunity to present arguments that will call
attention to relevant aspects of the legal experience. The lawyers
thus take the initiative to bring into prominence the most relevant
aspects of the law in its entirety, leaving the rest in the background
as the context for thought.
B. How Lawyers Get Started
The lawyers start the ball rolling for the judges, but no one
similarly starts the ball rolling for the la\ivyers. Clients come to the
law office with an often vague (and sornetimes misplaced) sense
that they have been wronged. They rarely provide the lawyers
with much more than a bare, often biased or misleading summary
of the facts of a dispute. They may be unclear about their objec-
tives and, in any event, rarely know whether the law permits or
requires what they want to do, much less what law is relevant.
The lawyer should start by developing the facts, based on legal
knowledge and a capacity for bringing that knowledge to bear in
a concrete dispute through legal reasoning.
Imagine, for example, that the plaintiff in the spring gun case
summarized above had started an initial interview with you by
asking, as a client might, whether it was "legal" for someone to
booby trap an abandoned farmhouse with a gun. As a competent
lawyer, you would know immediately that answering the question
simply yes or no would be unwise. Your general knowledge of the
law would be sufficient for you to have at least a vague idea of the
kinds of precedents you would find from research. You would
know that there are probably some precedents in which defendants
who set traps that injured other people 'Were convicted of crimes
and others in which such defendants were acquitted; some
precedents in which plaintiffs injured by traps recovered damages
for battery and others where injured plaintiffs did not recover. You
also would have some notion that these precedents often involved
the common law rules concerning battery and a property owner's
privilege to use force in self-defense or the defense of property.
The client's question should call your attention to these parts of the
legal experience. With a moment's reflection, you would know that
you need more information before you can give reliable legal
advice.
The needed additional information is of three kinds. You first
need to know why the client wants to know whether it is lawful
to set a spring gun. Based on the simple first question asked by the
client, you do not know whether the client wants to set a trap or
sue someone who set one. Second, you need to know a great deal
about the facts that have led or might lead to a dispute that could
come before a court. Third, you need to know much about the
precedents, rules, principles, and policies that will determine what

a court should do in a case in the jurisdiction where you would
litigate.
The second and third kinds of information are interrelated
because the law determines what facts are relevant while, at the
same time, the facts determine what law is relevant. Thus, you
should begin a process of (1) gathering facts to narrow the focus
of legal research, (2) conducting preliminary legal research, (3)
gathering more facts in light of that research, (4) conducting
further legal research in light of those facts, and so on.
You could begin by questioning the client to develop a
preliminary picture of the relevant facts. The hard part is knowing
what questions to ask. A general knowledge of the law gained
from prior study, together with skill at using the analogical and
deductive forms of legal reasoning, is indispensable here. Though
many lawyers ask good questions instinctively, a good question
can be the product of a hypothesis concerning; a possible analogical
or deductive relationship between the problem case and the
precedents or rules that might be found within the legal experi-
ence. That the question arises from a hypothesis bears emphasizing
because, at the early stages in developing a case, any one hypothe-
sis could easily turn out to be wrong. The initial questioning of a
client is like a fishing expedition-you may have some reason to
believe that you have located a good fishing spot, but you never
know whether the fish will be biting.
To illustrate, consider the following imaginary transcript of part
of an initial interview in the spring gun case. Notice how your
impression of the case changes as the lawyer elicits more facts:
Client (C): ls it legal for someone to set a trap with a gun in

a farmhouse?
Lawyer (L): Why do you want to know?
C: I was shot by one.
L: Please continue.
C: I want to sue the guy who set it.
L: Tell me how it happened.
C: I just opened the door to the bedroonn and the thing went
off. It nearly blew my foot off.
L: I'm sorry. Are you OK now?
C: I can walk all right. It still hurts. I was in the hospital for
fifteen days. In surgery twice. They almost had to take it
off.
L: You were lucky. What were you doing in the farmhouse?
C: Well, I was picking up some old jars and bottles. I'd been
there before.
L: Whose farmhouse was it?
C: Turns out it belongs to a guy narned Olner.
L: Did you know that before you went in?
C: No. I thought the place was abandoned.
L: Why?
C: I don't live far from there, and I've passed it lots of times.
It's all boarded up, overgrown-has been for years. I
didn't think I was doing anything wrong by going in and
taking some things no one cared about anymore.
L: Do the police know you went in?'
C: Sure. They busted me for burglary. I pleaded to a lesser
charge, and they fined me fifty bucks.
L: Do you know if they charged Mr. Olner?
C: Don't know.
L: Did you see any "No Trespassing" signs before you went
in?
C: No.
L: Did you see any signs warning of the trap?
C: No.
L: Did you look through the window·s before going in?
C: Couldn't. They were all boarded up.
L: How did you get in?
C: I-and my friend-we tore a board off. There was no glass
in the window, so it was easy.
L: Then what happened?
C: Joe, my friend, went into the kitchen to look around. I saw
this chair propped under a door handle, so I got curious.
I moved the chair and turned the door handle. That's when
it happened. Boom.
L: That thing could have shot some kid on a lark!
C: Sure.
L: Was anyone else in the farmhouse?
C: Just Joe. He took me to the hospital.
L: Have you spoken with Mr. Olner since this happened?
C: Yeah. He came to the hospital once. Said he wanted to see

the bastard who was driving him and his wife nuts. He's
a nasty customer.
L: Do you have a family?
C: Wife and one kid.
L: Do you work?
C: I pumped gas at Shelly's station for four years before this
happened. Couldn't work for two mLonths. I'll start again
next week.
Some of this lawyer's questions are similar to those a layperson

might ask. Significantly, however, others are not. The dissimilar
questions are a product of the lawyer's professional training and
experience-his general knowledge of the law and skill at legal
reasoning. Thus, a layperson might pass by the client's seemingly
innocent comment that he was in the house to pick up some old
jars and bottles. But the lawyer asks a series of questions revealing
that the client was a trespasser and thief. As you will see, the
precedents concerning a property owner's right to use force in
defense of his property sometimes allow an owner to use greater
force against a thief than a mere trespasser, though when asking
the question the lawyer may hypothesize only that the law might
so distinguish such cases. Similarly, a layperson might not ask
whether there were any "No Trespassing" signs, signs warning of
the trap or other ways for an intruder to learn of the trap before
being shot. But the lawyer asks such questions because the prece-
dents might permit Olner to use a trap to scare off intruders but
not to cause them death or great bodily harnr1. Again, a layperson
might not think to ask whether there were any people in the house
other than the intruders. But the lawyer might think that the cases
and rules could then allow Olner to use force lawfully in defense
of people when he could not use it lawfully in defense of property
only. Thus, a good lawyer will be using his skill at legal reasoning
even as he conducts an initial interview with a client.
After developing a preliminary account of the facts by ques-
tioning the client, you cannot yet give reliable legal advice. Except
for very accomplished specialists working within an area of
expertise, lawyers do not know the relevant rules and precedents
in sufficient detail to give legal advice with only momentary
reflection. You will have to read or reread the cases, statutes, and
legal commentaries with the problem case in mind. You may be
guided through this legal research by the forms of legal reasoning:
The legal materials are organized in a deductive framework
described by the legal rules, and your hypotheses concerning what
might be suggested by the rules and precedents can lead you
through the library. Then you can develop more facts (including
the other party's point of view) in light of your legal research and
think hard about the case.
The general knowledge possessed by every competent lawyer,
together with skill in analogical and deductive reasoning, suffices
for you to get started and to develop the case to a large extent.
However, such knowledge and skill are not sufficient for you to
make reasonably reliable predictions or effective legal arguments.
Good legal advice and good advocacy require a judgment of
importance. As you will see, your skills at reasoning analogically
and deductively can be supplemented by skills at using the
purposes of the law.
C. Advocacy in Harder Cases
Having developed the facts and conducted legal research, you

will be in a position to decide whether to advise your client to
litigate, settle, or drop the matter. Your advice should depend
partly on the result you predict if you litigate, as indicated in
Chapter 1.4 Your prediction, in turn, is largely a function of how
you think a judge will respond to argu1nents to be pressed on the
court by the lawyers on each side. To be sure, the predictive goal
may be aided by assessments of the psychologies and politics of
the players. Legal officials are by no means immune to the
influences on behavior that affect humans generally. These
influences, however, are modulated to a significant extent by
professional training, experience, and socialization, and by a
judge's duty to uphold the law.
Hence, the legal arguments are significant. Before giving sound
legal advice to your client, you should compose your best argu-
4
See Chapter 1 §C.
ment and anticipate your opponent's arguments. You should

gauge how a judge (and, in appropriate cases, a jury) will respond
to each argument. The power of the initiative gives you a precious
opportunity to wield substantial influence over a judge's first
impression of the case and, consequently, the decisions she will
make.
The components of the judicial decision are: (1) the facts of the
problem case; (2) the legal experience, including the common law
precedents and rules, and enacted texts with their contexts; and (3)
the conventional understandings of the relevant law's purposes.
Each of these components consists of a potentially vast array of
bits of information that, in interesting cases, can be combined in a
number of plausible ways. Your task as an advocate is to use these
bits to compose a picture showing how a judge can accommodate
a decision for your client coherently within her web of beliefs
about law. You can do this by (1) calling attention to particular
facts, parts of the legal experience, and principles or policies; and
(2) integrating these bits of information into a coherent legal
argument that supports a decision for your client.
When you mention a fact, cite a precedent or rule, or explain
how a principle or policy bears on the case, you call information
to the judge's attention. A conscientious judge then has an
"experience" like that of the birdwatchers in the previous chapter. 5
Such a judge will feel obligated to respond to this experience in
one way or another. She may decide that the information does not
merit attention, as when a fact is inadequately supported by the
evidence, a precedent is not authoritative, a irule is not applicable
to the case, or a principle or policy is not ernbedded in the legal
experience. Or she may accept that the bit of information matters,
in which case the fact, case, rule, or purpose is there to be dealt
with and cannot be ignored. Each bit of information that gains
attention may support or oppose the decision you prefer. How the
judge will treat the information depends on how all of the relevant
bits of information are combined.
A skilled advocate can influence the integrative process, in
addition to calling attention to the bits that go into the process.
You can use intellectual and rhetorical skills to narrate a story that
5
See Chapter 7 §C.
shows how the pieces fit together to support a decision for your
client. You can often describe the inforrnation in terms that evoke
various connotations or associations. You can treat a point in detail
to draw out the attention paid to that point and to intensify the
emotional reaction. You can place information in a broader or
narrower context to affect its perceived rr1eaning or significance. By
juxtaposing one bit of information to another, you can highlight
compatible or anomalous relationships. You can also criticize an
adversary's argument by pointing out facts or laws that challenge
the coherence of his presentation.
To illustrate, assume that the jury at a trial in the spring gun
case found the defendant liable and awarded damages to the
plaintiff. The defendant appealed, claiming the trial judge erred in
her instruction to the jury. This instruction had said that an owner
of premises is prohibited from willfully or intentionally injuring a
trespasser by using force that either takes life or does great bodily
injury, and that use of a spring gun or similar booby trap for the
purpose of thus harming trespassers is unlawful, even if the
trespasser is violating the law. It added that such a use of force
would be privileged only when the trespasser was committing a
felony of violence or one punishable by death, or when the
trespasser was endangering human life by his act.
The arguments of the two lawyers on appeal should present
the judges with a summary of facts supported by evidence in the
trial record, citations to and brief sumrnaries of the common law
rules and precedents relevant to the court's decision, and an
argument showing how a decision for their respective clients
furthers the relevant purposes of the law. In this case, like any
harder case, plural but plausible versions of the facts, the legal
experience, and the law's purposes are possible. Consider, first,
sketches of the statements of facts that each of the lawyers could
compose, with support in the trial record 6 and without contradict-
ing the facts stated by opposing counsel.
6
All of the facts that appear below were stated as fact in the majority or
dissenting opinion in Katko v. Briney, 183 N.W.2d 657 (Iowa 1971). Note that a
lawyer is subject to discipline for knowingly "'mak[ing] a false statement of
material fact or law to a [court]." Model Rules of Professional Conduct Rule
3.3(1)(1) (1992).
The defendant's lawyer might present the facts as follows:
Oscar Olner, the defendant, testified that the farmhouse

in this case belonged to his wife and had been in her
family for several generations. It contained furniture and
other valuable possessions, including a collection of antique
jars and bottles. She frequently visited the farmhouse to
enjoy the collection. The house had been broken into and
vandalized repeatedly in recent years. To stop the intru-
sions, Mr. Olner nailed the doors and sonrre windows shut,
boarded up others, locked the doors, posted seven "No
Trespassing" signs, and complained on numerous occasions
to the sheriff. When all of these efforts proved futile, he
placed a gun in a bedroom and wired it: so that it would
shoot downward and toward the door if anyone opened it.
He said he first aimed it straight at the door but later, at
his wife's suggestion, pointed the gun down in a way he
thought would only scare an intruder. He testified that he
"didn't want to injure anyone."
The plaintiff was injured while making a second visit to
the defendant's farmhouse to enter unlawfully and steal
Mrs. Olner' s antiques. He entered by tearing a plank from
a porch window and was injured when he came to a closed
bedroom door, removed a chair braced under the door
knob, and opened the door. This action triggered the gun.
The blast went through the door and struck the plaintiff
just above the right ankle.
The plaintiff's lawyer might present the facts differently:
Terry Tress, the plaintiff, was shot by the defendant

because, over the preceding ten years, a number of tres-
passing incidents occurred at the defendant's unoccupied
farmhouse. The defendants over the years boarded up the
windows and doors and posted a few "'No Trespassing"
signs on the land. However, the nearest remaining sign was
thirty-five feet from the house at the tim.e of the incident.
On June 11, 1995, following one such trespassing incident,

the defendant set a shotgun trap in the north bedroom.
After cleaning and oiling his 20-gauge shotgun, the power
of which he was well aware, the defendant took it to the
old house where he secured it to an iron bed with the
barrel pointed at the bedroom door. It was rigged with
wire from the doorknob to the gun's trigger so that it
would fire when the door was opened, no matter who
opened it. The defendant at first pointed the gun so an
intruder would be hit in the stomach but, at his wife's
insistence, lowered it to hit the legs. He admitted doing so
because "I was mad and tired of being tormented," though
he "didn't want to injure anyone." He gave no explanation
of why he used a loaded shell, set it to hit a person already
in the house, and posted no effective warning. The bed-
room window was covered so that no one could see the
trap.
Terry Tress lived with his wife and child and worked
regularly as a gasoline station attendant nearby. He had
observed the old house for several years, knew it was
unoccupied, and thought it was abandoned. In 1995, the
area around the house was covered ·with high weeds. The
plaintiff had been there with a friend and found several old
bottles and fruit jars, which he took home. On a second
trip, he and his friend entered the house through a window
that was without glass. As he started to enter a bedroom,
the shotgun went off, striking him in the right leg below
the knee. Much of the leg, including part of the tibia, was
blown away. Only with his friend's help was he taken to a
hospital for emergency surgery, and he remained in the
hospital for fifteen days.
These two versions of the facts leave very different impres-

sions. The defendant's account portrays the plaintiff as a heartless
burglar depriving the defendant's poor wife of her family heir-
looms. The defendant appears as a vigilant, if frustrated, husband
defending his wife's property from persistent and anonymous van-
dals. By contrast, the plaintiff's account portrays the plaintiff as a
hardworking family man with an interest in collecting old jars and
bottles. The defendant appears as a cold and vengeful man seeking

to violently punish a stranger, whether or not that person was re-
sponsible for past intrusions. You might be tempted to characterize
the difference as one involving the goodness or badness of the
principal players, and this is a part of what is involved in good
advocacy. But the two versions at the same time bring into
prominence facts that are relevant to the legal issues involved.
A sketch of how the defendant's lawyer nlight present the legal
experience follows:
The jury should have been instructed that the plaintiff

1n this case could not recover if the defendant set the
spring gun in his dwelling house with the intent to re-
pel, but not seriously injure, a felonious intruder. In
Scheuerman v. Sharfenberg, 163 Ala. 337 (1909), the plain-
tiff had been badly hurt by a spring gun ·while burglarizing
the defendant's storehouse. The court assimilated the store-
house to a dwelling and held that the plaintiff could not
recover because he was injured while cornmitting a felony.
In Hooker v. Miller, 37 Iowa 613 (1873), the court held that
a plaintiff who was injured by a spring gun while tres-
passing in the defendant's garden could recover, but placed
emphasis on the fact that the plaintiff was not committing
a felony and did not trespass in a dwelling house. Many
cases have held that a property owner is privileged to use
reasonable force, including devices, to repel an invader
even if there is no threat to human life or safety. E.g.,
Allison v. Fiscus, 156 Ohio 120 (1951).
The plaintiff's lawyer might present the legal experience different-

ly:
The trial court's instruction was correct to limit the

defendant's privilege to use force to situations in which
there is a threat to human life, whether or not the intruder
is a wrongdoer. A property owner is privileged to use
mechanical devices that inflict great bodilly harm only if he
would be entitled to use such force in person. State v.

Childers, 133 Ohio St. 508, 515 (1938). One person is not
privileged to use deadly force or force likely to do great
bodily harm to protect property, even from burglary,
absent an accompanying threat to the life or safety of a
person. People v. Ceballos, 12 Cal. 3d 470, 479 (1974). In
Hooker v. Miller, 37 Iowa 613 (1873),, the court held that a
vineyard owner was liable for damages resulting from a
spring shot gun although the plaintiff was a criminal
trespasser and there to steal grapes. Accord, Bird v.
Holbrook, 130 Eng. Rep. 911 (1825).
These two sketches also leave very different impressions. 7 The

defendant's account emphasizes the precedents allowing a proper-
ty owner to use force to protect his dwelling and highlights the
criminal conduct of the intruder. Accordingly, the facts as stated
above by the defendant's lawyer call attention to the defendant's
effort to protect his property, using force only after taking a
number of less aggressive measures unsuccessfully, and to the
plaintiff's intention to steal valuable antiques from the farmhouse.
By contrast, the plaintiff's account emphasizes the precedents
limiting a property owner's privilege to use force to protect
property when there is no threat to human life or safety, and
giving little or no weight to the criminal conduct of the intruder.
Accordingly, the facts stated above by the plaintiff's lawyer call
attention to the plaintiff's reasons for believing that the farmhouse
was abandoned and, without denying that the plaintiff trespassed,
treat his adventure almost as a misguided lark. The defendant is
portrayed as acting recklessly, without concern for the circumstanc-
es under which a person may be shot, if not with a malicious
intention to seek revenge for the past intrusions by others.
7
A lawyer is subject to discipline for knowingly "fail[ing] to disclose to [a
court] any legal authority in the controlling jurisdiction known to the lawyer to
be directly adverse to the position of the client and not disclosed by opposing
counsel." Model Rules of Professional Conduct Rule 3.3(a)(3) (1992). If the spring
gun case arose in Iowa, for example, both lawyers should bring the case of
Hooker v. Miller to the court's attention. Note how each legal argument does so,
but spins it in a way that supports the client's position.
A court consequently would be faced with two versions of the

facts and the law, both of which are plausible. The defendant's
version fits the facts and the legal experience together compatibly,
as does the plaintiff's. The court's decision depends on which
version fits better with the remainder of the legal experience and
conventional understandings of the law's purposes. Surely the
parties' advocates are well advised to continue their arguments to
show how their versions can be coherently accommodated with the
principles and policies that justify this part of the law.
The defendant's lawyer might continue along the following
lines:
There is no doubt in this case that the plaintiff was a

wrongdoer-a thief who entered the defendant's dwelling
house without permission. The defendant acted in defense
of his valuable property after taking a large number of
nonviolent measures unsuccessfully and with no wrongful
intention to harm intruders unnecessarily. The common law
tradition upholds the right of a homeovvner to protect his
dwelling by the use of force, including deadly force. W.
Prosser, Handbook of the Law of Torts 116 (4th ed. 1971).
"The principle that controls is that the right need never
yield to wrong, where the justification of self-defense has
shown the defendant to be in the right and the plaintiff in
the wrong. To deny the use of deadly force necessarily
allows the plaintiff to profit from his own wrong even
when the defendant is able to stop him." Epstein, Intention-
al Harms, 4 J. Legal Stud. 391,419 (1975) (citation omitted).
The plaintiff's lawyer of course would continue differently:
The law has always placed a higher value on human safety

than on mere rights in property. W. Prosser, Handbook of
the Law of Torts 115 (4th ed. 1971). Consequently, "spring
guns and other man-killing devices are not justifiable
against a mere trespasser, or even a petty thief." Id. at 116.
See also 2 F. Harper and F. James, The Law of Torts
1440-1441 (1956). It makes no difference that the plaintiff

entered a farmhouse because the privilege to use greater
force to defend a dwelling exists only when there is a
threat to the safety of people residing therein; in this case,
the farmhouse was unoccupied. See People v. Ceballos, 12
Cal. 3d 470 (1974). It makes no difference that the plaintiff
was a trespasser because the foundation of the law is "not
the criminality of the act or the turpih1de of the trespasser."
Hooker v. Miller, 37 Iowa 613, 615 (1873). The privilege of
a property owner to use force against an intruder is limited
to force that is proportional to the nature of the intrusion;
"it would seem clear that no interest which is merely one
of property can be equal or superior to the interest which
both the individual and society have in life and limb."
Bohlen and Burns, The Privilege to Protect Property by
Dangerous Barriers and Mechanical Devices, 35 Yale L.J.
525, 528 (1926).
Each of the lawyers thus would appeal to a theory about the

law's normative purposes and argue that the preferred principles
and policies are better embedded in the legal experience. More-
over, by their citations, they would effectively argue that the fa-
vored purpose is better supported by the legal conventions. The
defendant's theory calls attention to the \Vrongfulness of the plain-
tiff's conduct and requires no balancing of the interests at stake; by
contrast, the plaintiff's theory calls attention to the value of human
life and requires a balancing of interests. In retrospect, it should be
apparent that the facts and legal experience presented above by the
defendant fit well with the defendant's theory, as the facts and
legal experience presented above by the plaintiff fit well with the
plaintiff's theory. Each argument has a coherence of its own and
integrates the facts in the trial record, the precedents on the books,
and the principles and policies supported by the legal conventions.
From these sketches of arguments, it is not obvious how a
court would or should decide this case: It is a harder case. 8 It is
8
The decision of the Iowa Supreme Court on these facts is in Katko v. Briney,
183 N.W.2d 657 (Iowa 1971). The result is far less important for this discussion
than the plausible legal reasons supporting each party.
possible, however, that full argumentation ,would show that one

argument or the other can be accommodated more coherently with
the facts, the legal experience, and the law's purposes. That is what
the lawyers would seek to show in their full arguments. When
judges decide harder cases, as suggested in Chapter 7, the lawyer
who succeeds in presenting the more coherent argument-taking
into account more of the relevant facts and the legal material-will
make the more persuasive legal argument.
To review the preceding chapters in brief, the analogical and

deductive forms of legal reasoning seem to promise clear and
knowable answers to legal questions. This promise is seriously
misleading. The unavoidable judgment of importance, resulting
from the normative function of law and the family-style relations
among cases, often precludes effective legal reasoning when it is
confined to precedents and rules. Legal precedents do not them-
selves dictate the results in problem cases because each problem
case differs from its precedents in some respects; to make a sound
analogy, you must decide whether the differences between the
cases are important enough to require different results. Legal rules
do not dictate results because the language of a rule need only
designate classes of cases; to reach a sound conclusion, you must
decide which facts are important and justify placing a case in a
legal class thus designated.
To make a judgment of importance, you may supplement the
analogical and deductive forms with legal principles, policies, and
professional conventions. Principles and policies provide accepted
justifications for legal rules; you can use thern to urge a judgment
of importance that furthers the law's purposes. Legal conven-
tions-the practices and dispositions of the members of the legal
community regarding the lawful result in a case-support the
principles and policies that are implemented in this way.
CHAPTER NINE
Legitimacy
LET us CONCLUDE BY reflecting from a different perspective on
some implications of what we have studied. We have considered
law and legal reasoning from the perspectives of law students,
lawyers, and judges. We should not forget that the law affects
other people, sometimes profoundly changing their lives in ways
they do not welcome. This question may have occurred to you
already: Why should the members of the legal community have so
much to say about the legal reasoning in cases? Lawyers and
judges are not elected and accountable dem.ocratically, with few
exceptions. We surely have no credible clain1 to an inside line on
truth and justice. Yet our interpretive role in law and legal
reasoning gives us great power. Why should laypeople listen to an
elite professional community and obey our prescriptions? 1
The perspective of ordinary people calls into question the
political legitimacy of the law and legal reasoning. In this chapter,
I will cautiously suggest that the interpretive role of a properly
constituted legal community is compatible V1rith the legitimacy of
the legal system in a democratic society like that in the United
States. That is, ordinary people may have obligations to obey the
law of a legal system in which interpretation depends on the
conventions of a legal community. My purpose is not to pronounce
whether the current practice is legitimate. It is to suggest a
practical standard for gauging the legitimacy of the legal com-
munity's role and for reforming our practices to the extent they fall
short. I hope you will consider my suggestions critically and
continue to think about the problem of legitimacy as you learn
more about the law and legal reasoning.
1
See Paul Brest, Interpretation and Interest, 34 Stan. L. Rev. 765, 770-772
(1982); Sanford Levinson, Law as Literature, 60 Texas L. Rev. 373, 386 (1982).
155
156 9. LEGITIMACY
A. The Legal Conversation
Law and legal reasoning, as we have constructed it, is a matter

of practice by a distinctive professional community. A practice
exists when the members of a group generally conform their
behavior to standards of conduct and think of conforming behavior
as in some sense obligatory. Consequently, deviant behavior is
criticized, and those who make such criticisms are not themselves
criticized in turn. The standards then claim to prescribe what
participants ought to do. 2 Generally, however, the standards do
not claim to bind outsiders. Even among participants, dissenters
may claim that they are not bound by \!\That others think and do,
however much the others may agree among themselves.
We can think of a practice as a (metaphorical) conversation.3
Consider, preliminarily, a nonlegal example. A construction worker
may have no difficulty whatsoever knowing what is meant by
straight line when asked by a foreman to mark off the plan for a
wall. A mathematician working with non-Euclidian geometries
may find the concept straight line among the most troublesome.
Only a pedant would challenge the construction worker's concept
of straight line with the arguments of the mathematician. The lay
notion is wholly adequate in the construction setting to accomplish
the relevant purpose to the satisfaction of those concerned. It is as
if the concept were used in a conversation, and the participants
allow it to pass without objection. The ·worker, the foreman, the
building contractor, the architect, the owner, and the city building
inspector are the principal participants in the conversation. The
mathematician is not a participant in that conversation. His
objection is unwelcome and beside the point at hand.
Legal conversations, though more elaborate, have similarly
distinctive participants and goals. Conversations about legal
problems take place in courts, administrative agencies, the halls of
Congress and the state legislatures, law offices, legal publications,
the news media, and at the family dinner table. They usually
concern concrete problems that require practical solutions. The
2
This is adapted from H.L.A. Hart, The Concept of Law 55-61 (1961).
3
The metaphor is adapted from Richard Rorty, Philosophy and the Mirror of
Nature (1980) (without also accepting that there is nothing more to it than a
conversation). See also Hans-George Gadamer, Truth and Method (1975).
9. LEGITIMACY 157
participants vary with the problem. Notably, the conversation is

influenced by people who are not present at the actual discussion.
Substantially the same legal question moves from one setting to
another before it is settled in one case. It arises repeatedly in a
series of cases. What was said in a prior conversation, or is
expected to be said in a future conversation, influences each actual
discussion.
For example, a husband and wife, Richard and Lisa, may
discuss a divorce. To trace out the implications for their family,
they may mention what the family lawyer told one of them, or
what they expect their lawyer to tell them, about the division of
their property or child custody. They may go to their respective
lawyers' offices to pursue the matter. What the two lawyers say in
their offices will reflect what they have read in legal publications,
including case reports expressing the views of the judges, statutes
expressing the views of legislators, and scholarly works expressing
the views of legal observers. The lawyers' contributions to the
conversation will also reflect what they expect the local civil trial
court to say if the parties seek a divorce. What the trial court will
later say reflects what it expects the appellate court to say if an
appeal were taken. What the appellate court 'Will then say reflects
what the judges have read in the relevant legal publications and
the legal briefs of counsel, what the trial court said in its judgment
in the case, and what the parties said to the trial court through
their lawyers.
Metaphorically, a question concerning the implications of a
divorce is the subject of a single, extended conversation that moves
from setting to setting and, in sum, involves everyone who
participates in the actual discussions. The dynamics of a partici-
pant's voice vary from one actual discussion to another: The
appellate court's voice may be nearly inaudible at the dinner table,
faint in the law office, stronger in the trial court, and dominant in
the appellate court; Richard's and Lisa's voices may be dominant
at the dinner table, strong at the law office, faint in the trial court,
and nearly inaudible at the appellate court. But everyone who
influences what course of action will be taken as a consequence of
1,
the entire conversation, is a participant to smne degree.

Individual cases are also part of a broader conversation about
order and justice. The broader conversation extends through
history with each case providing an occasion for it to continue.
158 9. LEGITIMACY
Writings preserve the voices of the past; some judicial opinions

from fifteenth-century England continue to affect legal problems in
the United States, as do statutes from the middle ages and the
writings of Roman jurists. Though a judge today would rarely read
such materials in the original, the ideas they carry influence legal
thinking today. Each judicial opinion concludes a phase of the
legal conversation and makes a contribution that will be available
in the next round.
The legal conversation shapes the participants' statements and
actions as well as their very thoughts. Should Richard and Lisa
seek a divorce, the legal conversation ,will determine how their
property is divided and what kind of a relationship each may have
with their children. Once Richard and Lisa enter the legal system,
however, their voices grow faint as the lawyers take over. Clients
commonly dread the loss of control they associate with seeing
lawyers. Public anger at the legal profession partly reflects feelings
of disempowerment commonly experienced by clients.
Assume that, in this story, Richard and Lisa get a divorce and
a judge awards custody of the children 1to Richard alone. Assume
further that Lisa is understandably outraged at the court's harsh
judgment of her as a mother and also dreads the prospect of a
distant relationship vvith her children. lier appeal fails, and she
faces the question whether to run away with her children in
defiance of the law. Morally speaking, it hardly suffices to tell Lisa
that she should obey the law. 4 As the foregoing indicates, the law
is, in important part, a product of an extended legal conversation
to which she is an outsider. Moreover, the law that awards
custody to Richard could be unjust, as Lisa no doubt believes. Or
the law may have been just but was interpreted and applied
unjustly, though in accordance with our legal conventions. What
can we, participants in the legal conversation, say to outsiders who
resist the changes we impose on their lives?
4
Of course, it may be imprudent for Lisa to disobey because she will be
caught and suffer. Let us assume further, however, that she just might get away
with disobedience. The question for her then is a moral one: Whether she should.
9. LEGITIMACY 159
B. Legitimacy and Legal Reasoning
In general, legitimacy requires (1) that government action be

lawful and also (2) that the law be justified as a matter of political
morality. When these conditions are met, lay people are under
moral obligations to obey the law. We then could say to Lisa that
she should obey the law because it is right for her to do so. That
may or may not matter to her; it should, however, matter to us.
We should not want the law enforced unless enforcement would
be legitimate.
The demand for legitimacy is central to the American political
tradition, reflecting an historic commitment to individual liberty,
human rights, and a democratic government of limited power. The
concern is so central that it has a silent bearing on almost every-
thing else in the law, shaping the legal conventions pervasively.
Consequently, it is hard to master the basics of legal reasoning
without appreciating the problem of legitimacy in the law.
Legitimacy requires the people who hold the reins of official
coercive power to use it to implement the law-not arbitrarily, for
their own benefit, or to oppress others. For example, the police are
justified in using force, if necessary, to limit a person's freedom by
lawful arrest if they have probable cause to believe that that
person violated the law. The police are not authorized to arrest
someone just because they do not like his attitude, disapprove of
his life-style, believe him to be an immoral person, or oppose his
group memberships or political views. As a matter of principle, all
citizens are entitled to their liberty and digni~y unless they violate
the law.
Similarly, a judge issues or approves of orders backed by a
threat of official coercion. When the judge awarded custody to
Richard, he would have ordered Lisa to turn over the children. If
Lisa defies the order, Richard can get a further order instructing
the sheriff to take the children from Lisa and bring them to
Richard. If Lisa persists in defiance, the sheriff then may seize the
children forcibly. If she flees to another state, the sheriff there will
do the same. The threat of enforcing judicial orders induces most
losing parties to comply with court orders peaceably. Those who
resist, however, will find themselves on the wrong end of a
sheriff's revolver.
160 9. LEGITIMACY
Anyone who stands before a judge, too, is entitled to liberty

and dignity. We should want to know by what right a judge coerces
other individuals (or approves of police conduct). Our tradition re-
quires the judge to act for good and appropriate reasons-reasons
that justify the implicit use of coercion. At the least, judges are
under a duty to uphold the law and should act on legal reasons.
Reasons of advantage to self, friends, or groups with which a
judge identifies should be ruled out. Moral, religious, and political
reasons should also be excluded unless they happen to be
warranted by the law as grounds for judicial decisions. Judges
decide within the law when they decide on legal reasons. 5
Deciding within the law, however, does not suffice to achieve
legitimacy. A precedent, rule, principle, or policy can be legal
without being either just or required by political morality.
Consider rules: Not all rules are legal in nature. There are rules of
social etiquette, games, morals, scientific method, and other kinds.
Most people probably think of rules as legal rules when they are
in force in a society. Rules are in force when they are regularly
applied by legal institutions (mainly courts) within a legal system
that effectively guides the conduct of people to a large extent. In
contrast, social rules, such as the rule requiring a man to remove
his hat upon entering a church, some moral rules, like the golden
rule, and other kinds of rules are not applied by legal institutions
at all (or at least not regularly). These rules are not legal in
character.
Rules need not be just in order to be legal. Until recently, for
example, apartheid was the law in force in South Africa. You could
determine what the laws of apartheid permitted and required by
using legal reasoning. Assume, for exarnple, that the manager of
a municipal golf course denied entry to a black man solely because
of his race. Under the laws of apartheid, the manager would be
acting unjustly but lawfully. Under the Equal Protection Clause to
the U.S. Constitution, by contrast, the 1nanager would be acting
unjustly and unlawfully. Legal reasoning does not operate differ-
ently in the two settings. It is used to identify and apply the law
in force, for better or worse.
5
Steven J. Burton, Judging in Good Faith (1992).
9. LEGITIMACY 161
When so used, legal reasoning has practical, though limited,

value. As Holmes put it:
The reason why ... people will pay lawyers to argue for them
or to advise them, is that in societies like ours the command of
the public force is intrusted to the judges in certain cases, and the
whole power of the state will be put forth, if necessary, to carry
out their judgments and decrees. People want to know under
what circumstances and how far they will run the risk of coming
against what is so much stronger than themselves, and hence it
becomes a business to find out when this danger is to be feared. 6
From a critical standpoint, moreover, legal reasoning can be used

to identify objects for moral criticism. We can object to apartheid,
rally others to protest it, and celebrate its demise, when we dis-
tinguish law from justice and morals. Legal reasoning, however,
does not establish the justice or legitimacy of the law.
A law is just when it treats all people as they should be treated
ideally-as God, who is infinitely just, would treat them.7 The
laws in force in any society fall short of perfect justice. There are
three main reasons for this. The first is that we humans are fallible
in its pursuit. Whether acting individually or in groups, governing
by fiat or democratically, we make mistakes (and not infrequently).
The second reason is that a system of perfect justice would not be
administrable. Consider the age of maturity for purposes of voting.
No matter what age is set, there will be individuals who, in justice,
should be allowed to vote before that age and others who should
not vote even later. An age is set, despite the over- and under-
inclusiveness of the rule, for reasons of administrability, not justice.
The third reason is that we need some way to impose a final
settlement of disputes to short-circuit uncontrolled violence and to
stabilize social intercourse. The law functions to settle disputes
between people when they cannot do so on their own. For these
reasons, the law in force in every society makes practical compro-
mises with perfect justice.
6
Oliver Wendell Holmes, Jr., The Path of the Law, 10 Harv. L. Rev. 457, 457
(1897).
7
For a different view, see Ronald Dworkin, Law's Empire 165 (1986).
162 9. LEGITIMACY
Political morality is not the same as law or justice. Political

morality tells government to take its best shot at justice within
practical constraints. It guides governmental action through
principles of democracy, freedom of speech, equality, contract,
property, due process, and other principles. It also tells govern-
ment to modify the strict implications of these principles by
considering human fallibility, administrability, and the need for
finality. In addition, political morality indicates the political
obligations of government's subjects, such as a citizen's obligations,
if any, to obey the law. It tells citizens like Lisa, too, to take into
account both justice and limiting practicalities. Few indeed would
claim that government should not act, and that citizens are under
no obligation to obey the law, unless and until the law achieves
perfect justice. 8
The problem of legitimacy arises within political morality so
demarcated. Legitimacy does not signify that a precedent, rule,
principle, policy or judicial decision is lawful. That depends on
whether it is accepted as such by legal institutions within a legal
8
To forestall misunderstanding, I consider political morality to be a part of
critical morality, not conventional morality. Conventional morality contains
principles that people in a community generally accept as guides to proper
conduct. That one person or group believes a person's act to be wrong, however,
is not a reason for that person to behave differently. For example, that Joan is
outraged at Bill for working as an analyst for the CIA and believes that he should
resign his job is not a reason for Bill to resign. Bill should do so if Joan is correct
that the job is immoral or illegal. But Joan's feeling or belief then is superfluous
because Bill should resign whether or not Joan is angry or believes that he should.
We raise questions of legitimacy when we want to urge legislators and judges to
behave differently, whether or not their behavior is popular. We can do that
soundly with critical morality, which contains principles of proper conduct
independent of anyone's beliefs.
Some people, of course, are skeptical of the reality of critical morality.
Consider, however, whether you believe that slavery was wrong when and as
practiced in the South in colonial times. Conventional morality cannot reach back
and condemn it then, before much of anyone believed it wrong. The first
abolitionist can argue that slavery is wrong only on the basis of critical morality.
This is not to suggest, however, that we can know critical morality in any
verifiable way. We can exercise our best judgment on the best available
arguments, holding open at all times the possibility of error. See generally
Michael S. Moore, Moral Reality Revisited, 90 Mich. L. Rev. 2424 (1992).
9. LEGITIMACY 163
system that guides conduct effectively. Nor does legitimacy signify

that a precedent, rule, principle, policy or judicial decision is just.
That depends on whether it treats all people as they should be
treated ideally, irrespective of human limitations. Rather, legitima-
cy signifies that government is acting as it should act, all things
considered, including fallibility, administrability, and the need for
finality.
When a law binds in political morality, people have (prima
facie) moral obligations to obey. A law can bind in political
morality in two main ways. First, it can be rnorally permitted and
democratically enacted. Consider the laws requiring motorists to
drive on the right side of the road. There is nothing morally
objectionable about left- or right-sided driving; for the sake of
safety and efficiency, it is morally desirable for everyone to drive
on the same side. Morality permits the choice to be made by a
majority. For these reasons, statutes requiring everyone to drive on
the right side are justified in political morality.
Second, a law can be justified in political 1norality if, apart from
democratic preferences, it would be morally right to enforce its
obligations. Consider a law prohibiting everyone from torturing
anyone. Such a law would implement a moral principle prohibiting
anyone from intentionally inflicting unwarranted pain on another.
It may be morally allowable for the state to enforce this moral
obligation as a matter of law. The law would be justified by the
same moral principle that binds the people to whom it applies.
Consequently, a law prohibiting torture would be legitimate, even
when a majority wants to torture a minority.
Our main focus, however, is on legitimacy and legal reasoning.
In legal reasoning, the problem of legitin1acy arises after we
assume or conclude that the relevant laws are justified in political
morality. The base points for reasoning by analogy and the major
premises in deductive legal reasoning are then assumed to be
morally acceptable (as well as legally valid). The problem of
legitimacy arises if the results in cases do not flow from those
starting points. This problem of legitimacy can be understood to
arise in two ways.
Historically, the problem was conceived as one of constraining
judicial decisions, one by one, in a formalist box. This could be
done if legal reasoning allowed judges to transfer the legitimacy of
a law to each judicial decision by applying the law to facts through
164 9. LEGITIMACY
a logically airtight chain of reasoning. 9 The law, together with the

facts, then would fully determine the result in the case. There
would be no room for a judge to decide on idiosyncratic reasons
of advantage to self, friends, or groups with which the judge
identifies, or on moral, religious, and political reasons when those
reasons are outside the law. The judicial result would enjoy
whatever legitimacy was enjoyed by the law.
We have seen, however, that such a formalist scheme is not
ours in practice. Moreover, it probably could not be made to work
in any society as complex as ours without abdicating important
legal goals. 10 The judgment of importance is too often unavoid-
able; it breaks the chain of logic and blocks the transfer of le-
gitimacy. We have not, however, concluded that the judgment of
importance induces illegitimacy. To do so would be to accept the
formalist criterion for legitimacy. Instead, we have suggested a
kind of conventionalism. The problem of legitimacy arises here,
too. But it takes on a markedly different cast.
On the conventional view, the logical gaps in a chain of
reasoning are closed in accordance with the conventions of the
legal community. We have already seen that, descriptively, these
conventions guide and constrain judicial decisions. 11 Here, we
consider normatively whether the conventions themselves, as the
charge of a professional community with limited membership,
contaminate the legitimacy of legal resul1ts in adjudication.
C. The Legitimacy of Convention
By what right does the legal community play an interpretive

role in legal reasoning and, therefore, in determining how the
coercive powers of the state may be used? You might conclude
that conventions of interpretation and judgment, shared among
members of the legal community, are no better than a judge's
idiosyncratic values. This conclusion, hmNever, has profound im-
plications for the current practice. It suggests, in particular, that we
9
E.g., Duncan Kennedy, Legal Formality, 2 J. Leg. Stud. 351 (1973).
10
See Chapters 2-4.
11
See Chapters 5-7.
9. LEGITIMACY 165
should do one of three things. First, we could maintain the

demand for legitimacy and change our practices by radically
restricting legal reasoning to stay within the formalist's box. This
would leave many basic legal functions unperformed in our
society. The laws of contracts and torts-and much else-depend
extensively on standards of "reasonableness" that do not satisfy
the formalist criterion for legitimacy. Second, we can give up the
demand for legitimacy and continue current practices. This would
be lamentable. The demand for legitimacy is central to the values
at the founding of our democracy. Third, we can modify the
demand for legitimacy so that our practices, or a moderate variant
of them, can achieve legitimacy. We should not, of course, endorse
a view of legitimacy just because it blesses the status quo. The
whole point is to hold current practices up to a critical standard.
We might, however, find that the traditional formalist view of
legitimacy in legal reasoning was a bad idea all along.
In the remainder of this chapter, I will propose a response of
the third kind. I will argue that a properly constituted legal
community can contribute to the legitimacy of the legal system as
a whole. When it does so, such a community enjoys legitimacy in
its interpretive role. I will take a systemic approach.
1. Systemic Legitimacy
A systemic approach focuses less on the logic of legal reason-

ing in each case than on the benefits and burdens of a legal system
and adjudicatory practices within that system. Think of the legal
system as a coordinated set of legal precedents, rules, principles,
and policies that guide the conduct of law-applying institutions,
such as courts, and may be changed by la,Nmaking institutions,
such as legislatures.
A legal system may offer considerable benefits to a society,
including an end to self-help settlements of disputes with atten-
dant risks of arbitrary violence, consistent implementation of
principles and policies, and final settlements cutting off feuds and
the like. A legal system can coordinate people's behavior in the
interests of all, as by providing motor vehicle laws, commercial
laws, property laws, and sundry other laws, often when it is more
important that we all do the same thing than it is what we do. By
166 9. LEGITIMACY
prohibiting invidious discriminations, a legal system might foster

a mutually respectful and cooperative corrununity in which people
might better flourish. Obviously, the benefits of a legal system can
be substantial even when some of its la,,vs fall short of justice or
lack adequate justification in political morality.
Under a systemic approach, legitimacy is a property of the
legal and political system as a whole, not an individual judicial
result or a specific input into a decision. The judicial decision in
a case, or the legal community's conventions, cannot be fairly
appreciated in isolation from the institution of adjudication within
a larger legal system. The legal system, in turn, is part of a political
system that operates in a social, historical, and cultural context.
Evaluating the legitimacy of legal reasoning in cases in isolation
from the system, rather than the legitimacy of judicial practices in
their full systemic contexts, is an acadenuc exercise of little real
concern to the people with interests at stake.
In a democratic society like ours, systemic legitimacy has two
aspects. First, in its theoretical aspect, it requires that the people
generally recognize an obligation to abide by the law, because it is
the law. Whether the people generally recognize an obligation to
obey the law is a factual question, requiring observation of
attitudes and behavior in compliance with the laws of the system.
Second, in its practical aspect, legitimacy requires that the legal
and political system deserves allegiance from the people it governs.
This is a question of political morality requiring normative evalua-
tion of the system as a whole. (The old formalist view, by contrast,
loses sight of legitimacy as the sign of an obligation to obey the
law. It obsesses with the logical foundations of a judicial decision
in place of the practical implications of that decision for people's
actions.)
The theoretical question seems important for us because all
theories of democratic government require the people's voluntary
acceptance of the laws. Systemic legitimacy requires their accep-
tance of an obligation arising from the system, not their separate
assent to each law or judicial decision. There is no law and no
legal system, but instead anarchy, when each person picks and
chooses the laws he or she will obey. We live in a diverse society
in which we frequently disagree and in which few laws enjoy
universal support. No individual can claim a right to disobey
whenever he or she is convinced that a law is unjust, and grant
9. LEGITIMACY 167
others the same right, without in principle licensing disintegration

of the entire system. So legal reasoning faces the tribunal of
democracy, not on a case-by-case basis, but within a corpus of law
and politics.
In theory, legal and political systems lack legitimacy when
large numbers of people in fact cease to recognize an obligation to
abide by laws or decisions with which they disagree, as in times
of revolution, civil war, or civil disobedience. It is not easy to
determine when the factual condition for legitimacy is met by an
existing legal and political system, however. Whether people
accept the legitimacy of the prevailing system depends on
observations of social facts that would evidence recognition or
rejection of an obligation to abide by the la'w.
The fact that most of the people comply with the law most of
the time, for example, evidences their acceptance of such an
obligation. It would be hard to imagine a people accepting such an
obligation and, at the same time, generally disobeying the law. But
the fact of widespread obedience is not dispositive of the theoreti-
cal question. Some people obey the law frmn a sense of obligation,
others from a fear of sanctions for disobedience, and many from
a combination of both motives. Obedience should not count as
voluntary acceptance of the system when it is coerced by the
state's threat of force. Indeed, it may only evidence the effective-
ness of a police state. But how can one tell the extent to which
obedience is not coerced? Many of us would be hard put to
distinguish with confidence the roles of obligation and prudence
or fear in our own law-abiding behavior.
There is also a good deal of disobedience to the law in the
United States, as indicated by the crime rate and the huge number
of people who are incarcerated. It seems far-fetched to interpret the
disobedience of the typical mugger or tax cheater as a rejection of
the legitimacy of the legal and political system as a whole. But
identifying the disobedience that counts involves the observer in
subjective judgments that may defeat the purpose of looking at the
evidence. The observer's predispositions would seem inevitably to
affect his or her interpretations.
In part because of these difficulties, the practical and normative
question-whether the legal and political system deserves the
allegiance of the people-is more important. The practical question
for each individual guides his or her conduct. It precedes the
168 9. LEGITIMACY
behavior that makes up general obedience or disobedience to the

law. Normatively, if you conclude that the system deserves respect,
you probably should obey the law whether or not others generally
do so. If you conclude that it does not deserve respect, you are not
under an obligation to obey even if others generally do so.
Therefore, people within the system who are concerned about
legitimacy should conduct themselves so as to ensure that the
system deserves respect.
Asserting that the American people should reject the legitimacy
of their legal and political system is neither an idle act, a strong
form of rhetoric to be used in political posturing, or an excuse for
disobedience. Real claims of illegitimacy are fighting words. The
American revolutionaries stood on a claim of the illegitimacy of
continued rule by the British crown. Martin Luther King, Jr.'s
defiant sit-ins were civil disobedience to illegitimate Jim Crow
laws. Claims of illegitimacy may, of course, also extend to
situations that are not so dramatically evidenced, such as the
divorce of Lisa and Richard. Lisa's anger and disappointment,
however, do not in themselves justify disobedience.
In my view, claims of illegitimacy are justifiably made when
the legal and political system denies the most basic rights and
interests of significant groups in society to a large extent and roads
to changing the law within the system are substantially closed to
those groups. The system must be so bad that compliance with
wrong decisions is not justified by the benefits of the system as a
whole. Recall the American revolutionaries' view of continued
British rule and the civil rights movement's view of Jim Crow laws
in the South. The claim, in each instance, was one of persistent
oppression without hope of change through argument and lawful
action within the system. The only alternatives were to submit to
grave injustice or to resist. The oppressors, in each instance, were
hard-pressed to say credibly that the oppressed group had an
obligation to obey the law. To the oppressed group, the "law" was
indistinguishable from domination by the powerful under a
transparent cloak of legitimacy.
A crisis of legitimacy occurs when the legal and political
system breaks down and tests of physical power replace legal and
political argument. Crises of legitimacy can be avoided when the
system respects the most basic rights and interests of all significant
groups in the society and contains open avenues of legal change.
9. LEGITIMACY 169
This permits each group to believe reasonably that it can change

laws lacking a justification in political morality. A legal and po-
litical system that satisfies these conditions is engaged meaning-
fully in pursuit of a more orderly and just society with respect for
all members of the body politic. Such a legal and political system
has humility about its own fallibility. It vvelcomes criticism and
proposals for legal reform from all members of the body politic.
Even if we can imagine a better system, the venture may then
deserve our respect as law.
Especially in a pluralistic society like that of the United States,
all groups know that they will lose a fev'r battles for laws they
favor. Each will and should continue its participation in the sys-
tem-turning away from disobedience-as long as it also wins a
few and believes that it can win a few more. The U.S. system of
government, if it stands for anything of constitutional dimension,
should, in respect of all segments of society, be distinguishable
from the British colonial rule against which the American revolu-
tionaries acted.
The difference between systemic legitimtacy and illegitimacy is
one of degree in the extent to which a system treats all people
fairly. The difference is not in the success or failure of a system to
satisfy all demands-even all majoritarian demands-for reform.
By contrast with the formalist view, syste:mic legitimacy is con-
cerned with the conditions under which people have an obligation
to obey the law. We should care whether the system, by compari-
son with alternative possibilities, treats all people fairly with an
acceptable level of correctable error.
2. The Legitimacy of the Legal Community
The preceding section suggests that a functioning legal system

has value in political morality because it produces the benefits
identified therein. This section suggests that a legal system cannot
function unless supported by a legal or equivalent community that
plays the interpretive role. If this is right, it would seem, a legal
comrnunity has value in political morality when it supports a legal
system that produces benefits for the population it governs.
Our historical, social, and cultural context includes three stable
features that serve as appropriate starting points for this part of
170 9. LEGITIMACY
my argument: U.S. society is a large and complex society, a diverse

society, and a society in which the people want the legal and
political system to contribute to order and justice. These three facts
suggest that a properly constituted legal community would play
a legitimate interpretive role in that system.12 Thus, I will argue:
(1) the legal conversation focuses heavily on cases in a way that is
distinctive and crucial in a large and complex society; (2) the legal
conversation proceeds independently of majoritarian political
processes so that minority rights might be protected by law; and
(3) the legal conversation distinctively works out the implications
of order and justice for the principles, poHcies, rules, and holdings
in particular cases. Far from being a source of contamination, a
legal community that serves these functions has value as a matter
of political morality.
First, in a large and complex society like ours, it is not feasible
for the broader society to make or pass judgment on each legal
decision that must be taken. No one could read, for example, all of
the appellate court decisions that are made every day, and those
are only the uppermost tip of a gigantic iceberg. The society must
delegate responsibility for operating and rnonitoring the legitimacy
of the system in its details to a smaller group of people.
The people's elected representatives in Congress or the state
legislatures are smaller groups. As members of the legal communi-
ty, these representatives participate in the monitoring function.
But they are too few to make and monitor the huge number of
decisions taken each day. Legislators normally respond to deci-
sions brought to their attention by constituents and lobbyists, and
they monitor the judicial system from a political perspective that
enhances its legitimacy. But, as an expression of presumably
majoritarian or special interest preferences, they cannot establish
12
The present membership of our legal community hardly reflects the
diversity of the general population. Reasons of legitimacy support efforts to
achieve a more diverse membership so that points of view otherwise likely to be
neglected by privileged members of society can be better heard and taken into
account by the law. This is needed, not to make the legal community democrati-
cally representative or to distribute law jobs fairly across groups, but to enhance
the legitimacy of the legal conversation by hearing a wider range of arguments
from previously inaudible voices. See David Millon, Objectivity and Democracy,
67 N.Y.U. L. Rev. 1 (1992).
9. LEGITIMACY 171
legitimacy for the full range of cases. The legitimacy of judicial

judgment is not a question of what a court can get away with in
relation to majoritarian politics.
Lawyers and judges are society's experts on when-in what
cases-the state may use its coercive powers legitimately. The
legal community is, of course, a small part of the general popula-
tion. It involves a larger number of people than could be elected
feasibly. Most legal decisions are scrutinized by one or more
lawyers for conformity with the law that provides for legal rights.
Though lawyers are largely motivated to scrutinize decisions from
the standpoint of their clients' interests, their goal is to protect
their clients' rights. The case-orientation of the legal profession
seems well-designed to monitor the legitimacy of a legal system in
its details. Indeed, in any large and complex society concerned
about law, some such professional community would be necessary
to run and watch the legal system on a case-by-case basis.
Second, in a diverse society like ours, strictly majoritarian
politics could hardly produce a legal and political system that
satisfies the demand for legitimacy with respect to all significant
subgroups. U.S. society is composed of many groups with differing
perceptions, values, and interests. During long periods in Ameri-
can history, majorities have been able to enslave or oppress
substantial racial, religious, ethnic, political, and other minorities.
Systemic legitimacy requires that minorities, too, should accept the
legitimacy of the legal and political system on the whole.
A professional community that operates at some distance from
majoritarian politics is well positioned to maintain and enhance
unpopular rights so that all people and groups in society are
treated fairly. Adjudication provides an avenue of legal protection
and legal change to everyone at individual :initiative, regardless of
political clout (though not, as presently constituted, regardless of
financial ability). The legitimacy of the system would be enhanced
if some such avenue were available to people who seek justice but
may be unable to protect their most basic rights and interests
through majoritarian politics. Adjudication, as part of a legal
conversation about order and justice in cases, may offer a meaning-
ful alternative to disobedience.
Third, in a society that is committed to order and justice, a
legal conversation is a good idea. We should resist the temptation
to equate majority preferences, or the preferences of any particular
172 9. LEGITIMACY
person or group, with the political morality that confers legitimacy.

We know that majoritarian politics often are driven by passing
cross-currents of public opinion or by coalitions of special interest
groups; people's preferences often reflect personal experiences and
values or narrow self-interest. We need not know what Order and
Justice require, in a philosophically satisfactory way, to know that
an orderly and just society would often be different from the
preferences of people or groups. We may rightly believe that a
legal conversation is more likely than other available alternatives
to make a meaningful contribution toward that highly valued end.
An orderly and just society may encompass ample room for
majoritarian or other political preferences to be heard and felt in
the legal and political system. Yet it also n1ay encompass room for
a legal conversation using legal reasoning in a continuing effort to
work out the implications of order and justice in cases. A profes-
sional community whose job is to worry about order and justice,
as part of a larger system of law and politics, augments majori-
tarian politics so that the system as a whole includes the pursuit
of a better society through law .13
In sum, a legal conversation can contribute to the legitimacy
of the legal and political system because it serves these three
functions within the system. The legal conversation focuses heavily
on cases as needed in any large and complex society. It operates
at a modest distance from majoritarian political processes so that
minority rights might be protected. The legal conversation works
out the implications of order and justice in cases. Consequently,
legal reasoning is less influenced by political power or passing fads
than by alternative ways of implementing the law's social vision.
At the same time, legal reasoning takes place in legal conversa-
tions that are distinctively professional. These conversations focus
on what a court will or should do in a case. Judges are appointed
by elected officials or elected themselves, and they generally must
sign and publish their opinions to take public responsibility for
those decisions. Judges do not decide what cases they shall decide,
but must react to the initiatives of people who find themselves in
13
To be sure, this role could be played by philosophers, economists, clergy,
or laypeople. However, to the extent nonlawyers succeed in playing this role well,
they will be transformed into the functional equivalent of a legal community,
facing the same issue of legitimacy.
9. LEGITIMACY 173
disputes. The judges must hear and respond to the arguments of

the parties, who introduce the constraining conventions of the legal
community through citations to the legal experience and the law's
purposes. No one judge can make effective law without substantial
support among many judges; any judicial decision is subject to
overruling or subsequent neglect; and almost any legal question
surfaces in a series of cases before different panels of judges. No
group of judges can escape from the highly democratic checks of
subsequent legislation or constitutional amendment and, ultimate-
ly, removal from the bench for misbehavior.
When thus placed in context, the legitimacy of a legal com-
munity's interpretive role seems plausible. The lack of an absolute
assurance of legitimacy is troubling only when we accept the
formalist idea of legitimacy in legal reasoning. Within the formalist
view, judges and others will use any logicall room in legal reason-
ing to decide on presumably improper grounds. If judicial de-
cisions were a matter only of the judges' or the legal community's
self-interest or political views, there would be every reason to
confine adjudication to a far smaller scope than it has enjoyed.
The demand for legitimacy, however, is not a demand for
consistent and complete logic in legal reasoning. Rather, it is a
demand fur limiting the use of official coercion to instances in
which the state is enforcing its subjects' political obligations.
Political obligations do not flow only by logic from rules imple-
menting majoritarian preferences. They also flow from the practical
advantages of having a legal system-even an imperfect one. To
be sure, the benefits of a system are not always so great as to
guarantee the system's legitimacy just because it is a system. In
addition, it should have sufficient moral content to deserve respect
commensurate with the demands it makes on its subjects. Howev-
er, the systemic approach suggests that the interpretive role of a
legal community can be legitimate in a context like that now
prevailing in the United States.
I leave it to you to judge how well we are doing as you learn
more about what we do. To· the extent current practice falls short
of full legitimacy, whether due to the legal community's composi-
tion, the law's current content, or other matters, we, the members
of the legal community, should change it. The law schools are
engaged in helping you spin a valuable web of beliefs about law.
I hope that, if anything, this book has convinced you that a
174 9. LEGITIMACY
valuable web includes principles and policies, in addition to rules

and precedents. I also hope that attention to principles and poli-
cies will open a door for you to criticize the prevailing law and
proposals for legal reform on an ongoing basis. The legal commu-
nity welcomes responsible criticism.
INDEX
Advocacy. See Lawyers, functions Coercion lby law, 2, 12, 22-23, 159-
of 160., 171, 173
Analogy. See also Cases; Deduc- Common law
tion; Precedent defined,, 25
common law and, 25-41, 60-65, enacted law, contrasted with, 25,
I 06-111 60
defined, 25-27 process illustrated, 31-39, 60-62
enacted law and, 65-68, 127-133 rules, 31-39, 43, 59-65
functions of, 39-40, 59 Conventions, legal. See also Legal
generally, 25-41, 59-78, 106-111, community
135, 153-154 defined, 93
importance, judgment of, 26-27. functions of law, and, 90-95, I 05,
31-41, 59, 7 4, 106-111, 154 155-174
limitations of, 36-37, 39-40, 74- 78 judicial decision and, 106-115,
purposes and, 106-111, 127-133 117-133, 135-139
rules and, 15-23, 31-39, 43 legitimacy and, 155-174
scientific reasoning and, 79-82 limitations of, 123-124, 133n.,
Authority. See also Legitimacy 155-158
of judges, 37-39, 74-78 purposes and, 105-115, 117-133,
of precedents, 25-29 151-153
of rules, 43, 46, 65, 73, 74 Counseling. See Lawyers, functions
of
Courts, structure of, 11-12, 17-18
Blackstone, Sir William, 36 Critical legal studies movement,
82. See also Legal skepticism
Cardozo, Benjamin, 91
Cases. See also Analogy; Synthesis Deduction. See also Interpretation;
defined, 11 Rules
family-style relations among, 85- analogy and, 43-46, 59-78
89 defined, 44-46, 65
rules and, 14, 18-23, 57, 59-78 functions of, 44, 57-58, 59
175
176 INDEX
Deduction ( continued) Holding and dicta, 37-39, 60

generally, 43-58, 111, 135, 141, Holmes, Oliver Wendell, Jr., 36, 75-
153-154 78, 82
importance, judgment of, 44, 46-
58, 59, 65, 74
limitations of, 32-39, 50, 52-58, Importance, judgment of. See also
65, 68 Analogy; Conventions, legal;
scientific reasoning and, 79-82 Deduction; Rules
Democracy. See Legitimacy conventions and, 90-95, 106-111
Dispute resolution, 6-7, 11-13, 97, defiined, 27-28, 77-78, 79, 98
103-106, 164-174 purposes and, 98-113, 115-135,
Dworkin, Ronald, I 18n., 133n. I 51- I 54
Indete1minacy of law, 3, 8, 27-28,
40, 44, 46, 54-58, 59, 74, 77-
Earl, Robert, 129-132 78, 82-85, 133n., 153-154,
Easier cases, 40, 117-124 163- I 64, 173
Enacted law. See also Analogy; De- Interpretation. See also Conven-
duction; Rules tions, legal
cases and, 65-7 4 Constitutional, 43n., 65-70
generally, 43-58, 65-78, 111-115 defined, 52
interpretation of, 52-57, 65-78 resources for, 8, 59-60, 79, 93-95,
purposes and, 111-115, 115-134 127-133
text and context, 55-56, 65-72, rules and, 43, 47-57, 60-78, 79
115-134 Issue, legal, 30, 50, 145-146
Ethics. See Legal ethics
Judges. See also Judicial decision

Facts. See also Importance, duties of, 31, 37-38, 43, 74-78,
judgment of 90, 128-129, l 33n., 160
advocacy and, 145-149 functions of, I, 37-38
analysis of, 29-31, 4 7, 50-52 lawmaking power of, 124
defined, 30 Judicial decision. See also Judges;
families of cases and, 85-89 Rule of law
in legal reasons, I, 79, 98-102 coercion and, 159, 171
Field, Stephen, 131 constraints on, 6, 74-78, 90-91,
Formalism. See Legal Formalism 127-134
Forms of legal reasoning. See procedure, role of, 135-139
Analogy; Deduction reasons for action in, 6, 79, 90-91
J urisprudencc, 9
Justice, 26, 44, 98, l 03-105, 111,
Harder cases, 124-133, 135, I 53 123, 128-129, 158, 160-163,
Hart, H.L.A., 40, 156 169, 171- 172
INDEX 177
Kennedy, Duncan, 91 Legal skepticism, 3-8, 82-83, 89-90

Legal system, 93, 104, 155, 165-174
Legislative supremacy, 43, 74-78,
Langdell, Christopher Columbus, 82 115, 127-133
Law. See also Legal system; Legitimacy. See also Authority;
Legitimacy; Rule of law Rule of law
defined, I defined!, 155, 159
framework of, 44, 97, 117 democracy and, 155-174
functions of, 6-7, 13, 90-95 importance of, 2-8, 159-165
justice and, 92-93, 94, 158, 160- legal community and, 8-9, 155-
164 174
normativity of, 4-6, 17-18, 82-85, obedience to law and, 155, 158,
153 159-164, 166-169
purposes and, 97, 103-115, 127- Levi, Edward H., 36
133, 151-154 Llewelyn, Karl N., 20
Lawyers, functions of. See also Logic. See Analogy; Deduction
Conventions, legal
advocacy, 16-18, 140-154
counseling, 15-18, 47-57, 62, 135- Morality. See Political morality
144
generally, I, 9, 15-23, 117, 135-
139, 170-174 Negotiation. See Lawyers, functions
interviewing, 99-100 of
negotiating, 16-17 Newman, Jon 0., 9
Legal community. See also
Conventions, legal
defined, 93 Obedience to law. See Legitimacy
functions of, 90-95, 169-174 Order, 103-105, 111, 123, 128-129
legitimacy of, 92-93, 155-174
Legal ethics, 30, I 46n., 150n.
Legal formalism, 2-8, 89, 163-164, Policies. See Purposes
173 Political morality, 159-161, 174
Legal reasoning. See also Analogy; Politics and law, 89-90, 167. See
Deduction; Importance, Legitimacy
judgment of Pound, Roscoe, 91
defined, I, 90 Precedent
functions of, 6-7, 171-174 analogy and, 25-26, 29-30, 36-38
resources for, 8, 59-60, 79, I 27- common law rules and, 60-65
133, 135 defined, 12
Legal reasons, I, 4-6, 79, 90. See legal reasons and, 79
also Analogy; Deduction; purposes and, 97, 106-111
lmportance, judgment of Principles. See Purposes
178 INDEX
Procedure, 11-12, 30, 135-139 functions of, 44, 59-65, 77, 97

Purposes. See also Analogy; Com- holdings and, 37-39, 60
mon law; Conventions, legal; importance, judgment of, 46-58,
Enacted law; Importance, 59, 60-74, 111-115
judgment of; Law; Precedent; legal reasons and, 79
Rules limitations of, 18-23, 48, 52-58,
defined, 97 59-78, 85-89, 102
generally, 97-115, 117-134, 145, nom11ativity of, 82-85, 103-115
151-154, 174 princi pies and policies distin-
policies as, 13, 97, 117 guished from, 97
principles as, 13, 97, 117 purposes and, 97
Rule of law, 2-6, 18-23, 37, 93
Rules. See also Analogy; De-
Statutes. See Enacted law
duction; Interpretation
Syllogism. See Deduction
common law, 31-39, 59-65, 77-78
Synthesis, 105-115, 135-136, 144-
defined, 13-14
153
enacted law, 43-58, 65-78
framework of law and, 44, 62-65,
97, 144 Webs of beliefs, 124-133, 173-174
Little, Brown and Company, Publishers ...
LEADING. THE WAY

-lll
LEGAL EDUCATION
In legal education we stand out! Little, Brown is dedicated to
providing law students with expertly crafted books that are
unparalleled in their authorship, content, and design.
The Casebook Series

Offering the best teaching materials in our ,,-cll -kno\\·11 red c;1sc -
books trimmed in black ;nd gold.
The Electronic Casebook
Presenting the traditional casebook in an electronic form \\'ith
hypertext linking and search/retrinal, highlighting, outlining, and
notetaking capabilities.
Legal Research and Writing
Expertly authored paperback texts that prm·ide the hcsr guidance
;n·ailablc for .learning the intricacies of legal resc.1rch and ,,-riting.
Examples & Explanations
Study guides for the most difficult la\\' school courses. Authored
by America's foremost la\\' teachers. Designed in a st;mdard format
,,:ith clearly written text and rest-yourself questions and ans\\'crs.
Mate Cards
A quick re,·ie\\' system of key statutes, rules, and concepts for
each field of la\\' 011 notebook -size laminated ends.
We at Little, Brown are proud of our long history in serving the

needs of law students and teachers. Look for our coursebooks and
study materials at your bookstore today.
Little, Brown
and Company
34 Beacon Street
Boston, MA 02108 See us on the Web!
For information, phone:
1(800)950-5259
ISBN 0-316-11489-8
I 90000)
9 780316 114899 11

Law and Legal Reasoning

Uploaded by

Copyright:

Available Formats

Law and Legal Reasoning

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Law and Legal Reasoning

Uploaded by

Copyright:

Available Formats

AN INTRODUCTION TO

LAW AND LEGAL

Geoffrey C. Hazard, Jr.

LITTLE, BROWN AND COMPANY

Library of Congress Catalog No. 94-74423

Published simultaneously in Canada

Printed in the United States of America

Preface to the Second Edition XI

Chapter 1. Cases and Rules 11

Chapter 2. Analogical Legal Reasoning 25

Chapter 3. Deductive Legal Reasoning 43

Chapter 4. Combining Analogies and Deductions 59

Chapter 5. Legal Reasons and Conventions 79

Chapter 7. The Judge's Perspective 117

Chapter 8. The Lawyer's Perspective 135

Chapter 9. Legitimacy 155

Preface to the Second Edition Xl

Chapter 1. Cases and Rules 11

Chapter 2. Analogical Legal Reasoning 25

Chapter 3. Deductive Legal Reasoning 43

Chapter 4. Combining Analogies and Deductions 59

Chapter 5. Legal Reasons and Conventions 79

Chapter 7. The Judge's Perspective 117

Chapter 8. The Lawyer's Perspective 135

Chapter 9. Legitimacy 155

IT IS COMMONLY SAID that a beginning law student's first goal is to learn

This second edition results from a complete rewntmg and

a broad perspective, perhaps reducing accompanying anxiety. A

A. The Rule of Law

~oercive powers of governl!len_!~oulcflie limifec:f~y)aw. The (LS.

the rule of law. Faulty legal reasoning signifies a faulty connection,

Notice, however, that formalism and skepticism both depend

light, even repeatedly; whether or not he gets caught, the relevant

connection. The observer, like an empirical scientist, is interested

A.M., he may be morally justified in running a red light even while

h Negotiation, mediation, and conciliation are methods of dispute settlement

This book is organized into nine chapters. Chapter 1 introduces

United States. It suggests, however, the conditions under which a

Cases and Rules

In this book, a case is a short story of an incident in which a

part of the dispute. The appellate court writes and publishes an

Treating a case as a short story of an incident in which a court

C. Legal Reasoning and Legal Problems

By the case method of instruction, you will be required in each

A large part of what lawyers do involves this same skill. But

danger? Will the courts enforce a contract clause by which the

D. Using Rules and Cases

1. Rules in Problem Cases

Using a rule often requires intensive analysis and considerable

"No person shall sleep in a city park." 4 Imagine two problem

We have discovered in our teaching of the law that general

2. Decided Cases in Problem Cases

Decided cases also can be used in legal reasoning to solve

drafting. The world has a habit of confounding even those with

contrasting cases requires hard, rigorous thinking about justice and

A. The Analogical Form

Most legal problems in the United States are governed by _

authority. The principle of stare decisis supports the common law

things; comparing or contrasting them would be pointless.

B. Analogical Legal Reasoning