Journal of Philosophy, Inc.

Review
Reviewed Work(s): The Definition of Law by Herman Kantorowicz, A. H. Campbell and A.
L. Goodhart
Review by: M. P. Golding
Source: The Journal of Philosophy, Vol. 56, No. 17 (Aug. 13, 1959), pp. 708-714
Published by: Journal of Philosophy, Inc.
Stable URL: https://www.jstor.org/stable/2022051
Accessed: 22-10-2023 16:41 +00:00

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Journal of Philosophy, Inc. is collaborating with JSTOR to digitize, preserve and extend
access to The Journal of Philosophy

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
708 THE JOURNAL OF PHILOSOPHY

caninot be taught, and hence not forgotten. The liking or taste for
virtue can be lost, can deteriorate, but such moral deterioration
is not a process of forgetting or losing expertness of judgment.
The general implication of Ryle's analysis is that moral experi-
ence is closer to aesthetic experience than to scientific knowledge.
Marcus G. Singer of Wisconsin, in a very able and informative
essay, argues that the distinction between principles and rules,
which is clearly recognized in legal theory, is equally important,
though often obscured, in ethics. He discusses the various kinds of
rules and their relations to principles. Genuine principles are
general and cannot conflict with each other, whereas this is not
true of rules. Local or otherwise variable rules are justified by
reference to invariant principles.
J. 0. Urmson of St. A-ndrews calls attention to the fact that
the existence of saints and heroes proves that the realm of morals
is much wider than the conventional categories of deontology (the
obligatory, the permissible, the forbidden), illustrated by Prior's
essay, would lead us to believe. He criticizes especially the failure
of Moore and the utilitarians to take these saintly and heroic di-
mensions of morality into consideration.
In reading these essays, I am impressed by their importance
for general theory as samples of various kinds of analysis. The
kinds of subject-matter do not determine the kinds of analysis
used, for in several cases essentially the same problems are ana-
lyzed in very different ways with very different results. I would
therefore recommend the study of this volume to anyone who wishes
to make a critical comparison of the various analytic methods.
These samples are excellent not only as good pieces of workmanship
but as representative experiments in method. They suggest that
the genuine problem of evaluation of methods is not the attempt to
arrange them in a general order of merit, but rather to determine
which kinds of analysis are good for which kinds of moral interest.

COLUMBIA UNIVERSITY HERBERT W. SCHNEIDER

The Definition of Law. HERMAN KANTOROWICZ. Edited by A. H.

Campbell, with an Introduction by A. L. Goodhart. Cam-
bridge: At the University Press; [New York: Cambridge Uni-
versity Press] 1958. xxiv, 113 p. $3.00.

This little book was written in 1939 by the late eminent legal
historian as an introduction to a projected Oxford History of
Legal Science. The war compelled its abandonment, and Kan-
torowicz's essay is virtually all that remains of the enterprise. Its

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
 BOOK REVIEWS 709

immediate purpose is to provide a definition of "law" which will

be broad enough to encompass all the materials which ought to be
included in a history of legal science and yet exclude extraneous
matters. Keeping this practical aim in mind, namely the formula-
tion of a technical definition which will delimit an area of research,
Kantorowicz is led nevertheless into discussions of important issues
in legal philosophy.
In undertaking a task such as Kantorowicz's, one should begin
with a fairly clear idea of what is covered by the term "legal sci-
ence." Kantorowicz has a broad conception of this "science":
studies of the relationship between law and ethics as pursued in
ancient times, the history of law, criticisms of law with a view to
legal reform, studies of the social and economic implications of
law, etc. But the "core of legal science "is dogmatics, that is, the
orderly exposition and arrangement of a legal system, the harmoni-
zation of its rules by casuistry, and the setting up of rules for
hitherto unprovided cases. The science of law, therefore, includes
not only treatments of law, but also law, or parts of legal systems.
A history of legal science, then, would contain not merely history
of the history of law, but also history of law in some of its aspects.
Perhaps this would occasion no dismay on Kantorowicz';, j.art, but
in places the reader will be somewhat puzzled as to tne exact dis-
tinction, if any, between law and its "science." This p-i-lzement
will also occur in connection with Goodhart's interesting exposi-
tion of the author's earlier pieces on legal science.
The method employed in arriving at a definition of "law'"-
the "right method"-is called "conceptual pragmatism." It es-
chews verbal realism, the looking for the essence of a thing. The
difficulty with the latter approach is that no one has been able to
state clearly what "essence" means, nor has a method for intuiting
essences been given. The conceptual pragmatist asks, rather, what
ought to be understood by a term for the purposes of a particular
science. Thus there is no correct technical definition of a term
and more than one definition is possible. The acceptability of one
proposed definition over another is determined by the test of com-
parative usefulness. There are important respects in which such
definitions are not arbitrary. Firstly, while they need not com-
pletely conform to the varied ordinary usages, it is still advisable
that they capture as many such (mutually consistent) usages as
possible. Secondly, different definitions provide different schemas
for classifying the materials under investigation, and one schema
may be better than another in that it connects what ought to be
connected and separates what ought to be separated. (Kantorowicz
admits that more than one definition might satisfy these condi-

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
710 THE JOURNAL OF PHILOSOPHY

tions.) One wonders whether Kantorowiez has not himself fallen

into the trap of verbal realism. Jurists who refuse to allow that
public international law is more than a kind of morality, and hence
is law improperly so called, are criticized for inviting appalling
dangers and making more precarious the precarious validity of the
law of nations "on which the very existence of nations and civili-
zation itself depends." Hence, it is important not to substitute
another name for "international law." But perhaps these jurists
are pointing out that there is an important distinction between the
positive law of states and public international law; they are sepa-
rating what, for the sake of clarity, ought to be separated. To as-
sociate these distinct spheres, even in the name of important prac-
tical considerations, is to rely on the verbal magic which Kan-
torowicz so roundly condemns, they would say.
For the purposes of a history of legal science, once this area is
more or less clearly staked out, the usefulness of a broad defini-
tion will surely be admitted. The usual definitions of "law" in
terms of positiveness, enforceability, state origin, or binding char-
acter, prove too narrow, Kantorowicz argues. To take one ex-
ample, pre-Justinian Roman law was not binding on any Italian or
French court when it was rediscovered by Alciatus and Cujas,
who surely have a place in a history of legal science. More-
over, Kantorowicz maintains, fixing on these characteristics pre-
vents us from understanding the judicial process, the nature of
juristic methods, and the social functions of law. It is truly a pity
that Kantorowicz was not able to develop this point in more de-
tail, although now and again he has some interesting remarks bear-
ing on these issues.
Kantorowicz proposes the following definition of "law": a body
of rules prescribing external conduct and considered justiciable.
Thus, it is not the terms "a law" or " the law" which he defines.
One suspects that many of the definitions which Kantorowicz re-
jects are designed for these terms, particularly the latter. Holmes'
Bad Man, for instance, wants to know what the law is so that he
can know what risks he may take. Certainly, for the purposes of a
history of legal science Kantorowicz has taken a wise course in
focusing upon bodies of rules, i.e., rules which have the same con1-
tent (private law, penal law), or belong to the same code, or coin-
cide in time or space (ancient law, European law), etc. Further-
more, he claims it as a virtue of his definition that it enables us
both to detect the "legal character" of any given rule and to distin-
guish law from other social systems such as collective habits and
morals.
The concept of "rule" (of conduct) must be broadly viewed,

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
 BOOK REVIEWS 711

for it is a general notion which is also applicable outside the sphere

of law. Kantorowicz's characterization of rules as "expresions
of relations that ought to exist between some human conduct, A, and
some of the possible properties of this conduct, B," is highly ob-
scure. At any rate, a rule of conduct prescribes a conduct which
ought to be actualized, though it may never be. Now the author
admits that the introduction of this ought-element involves philo-
sophical matters of great complexity, particularly the concept of
duty. The duty contained in the category of the " ought, " he
maintains, is dependent upon other duties, higher duties which
impart obligatory power to the lower duties. Ultimately presup-
posed is a "basic and absolute rule" on which the validity of all
the other rules depends, and which is accepted dogmatically by an
act of faith. As Goodhart indicates, this resembles Kelsen's notion
of the basic norm, although Kelsen explained it as the will of the
original usurper of power. However, Kantorowicz cannot accept
this view of the Ursprungsnorm, for it presupposes the still more
basic rule that we ought to obey this individual; but such an "ultra-
ultimate rule has not even been stated, still less justified." Un-
fortunately, the author does not explicitly formulate his conception
of a (or the) basic and absolute rule, so that his criticism of Kelsen
seems self-applicable. One may also wonder exactly what rules
the basic rule is said to justify. Consider all the legal rules which
comprise a "body" in that they coincide in time. The elements of
this body will originate from different nations and places. Is
there a basic rule which at once gives binding force to all these
disparate items? Is it the same as the basic rule which gives
validity to those rules having the same spatial origin and which
constitute another body? One can at least see how far we now are
from merely seeking a useful definition of "law" for a projected
history of legal science; we have embarked on the hoary quest for
the nature of law.
Kantorowicz distinguishes rules according to the ground of
their validity, although the classification is neither exclusive nor
exhaustive. Rules which are recognized as binding because they
originate from the will of a recognized authority he calls com-
mands. The basic rule for a command is: obey the supreme au-
thority. But how helpful is Kantorowiez's notion? A supreme
authority is one whom we have a duty to obey. But there is no
trick in claiming that we ought to obey those whom we have a
duty to obey. The important question concerns how we can
identify such individuals, and Kantorowicz says nothing about
this. Precepts are rules whose contents are such that conscience
recognizes our duty to conform to them. This leads to the ideal

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
712 THE JOURNAL OF PHILOSOPHY

of a summum bonum to which every duty is s

basic duty to actualize supreme values. Kantorowicz does not
indicate what these values are. Perhaps he is merely doing
formal analysis and cannot be expected to treat here the area in
which legal philosophy and ethics merge. But it would be impor-
tant to know how close he is to a natural-law position and how
consistent it would be with other things he has to say. Dogmas
are rules of which the validity is guaranteed by other rules whose
validity is already recognized. They are the rules which are the
output of legal dogmatics, the core of legal science. Kantorowicz
states that the relationship between a dogma and its justifying
rule(s) is one of logical implication, but he does not analyze this
any further although he recognizes that syllogistic logic is inade-
quate for the process. This touches upon one of the most compli-
cated aspects of the juristic process, legal reasoning. Logical impli-
cation in its ordinary senses seems inappropriate here, and the
topic merits the attention of logicians. There is no recourse to a
basic rule here, he says, for we are dealing with a system which
"relies for its validity on itself." The justification of the dog-
matic process appears to him to stem from a duty which rational
beings recognize to submit the law to a process of rationalization.
A suggestive analogy is drawn between legal and theological
dogmatics.
The function of rules of whatever type is to prescribe. Kan-
torowicz maintains that permissions can be expressed as prescrip-
tions and rights can be reduced to duties. This is a rather com-
plicated business, as the author realizes. Campbell indicates doubt
on Kantorowicz's view in the Preface.
Rules of law prescribe external conduct only, and this distin-
guishes them from ethical rules. While legal rules may be con-
cerned with good faith, they never prescribe it. Kantorowicz
views ethical rules as consisting in prescriptions of "internal con-
duct" (which is certainly an odd notion). Justice is assigned to
the sphere of "quasi-morality." A man who conforms to a just
law but has bad motives is said to be acting in a just manner.
Law, even just law, only demands external conformity. Many
will undoubtedly regard Kantorowicz's internalistic conception of
ethics as much too narrow. And the separation of morals and
law, while important for some purposes, may be unfortunate for a
history of legal science. The total inclusion of morals in this area
of research would be foolhardy, of course. But in some ancient
legal systems ethics and law are too closely bound up for easy sepa-
ration. One could speak of the legal aspects of such systems in
contrast with their ethical aspects, but this might rob these bodies

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
 BOOK REVIEWS 713

of their systematic character, to display which is a task of legal

science.
Kantorowicz believes that it is useful not to classify social
custom (e.g., rules of etiquette, forms of greeting, etc.) as law.
The chief difficulty in drawing a distinction is that both law and
social custom consist of rules prescribing external conduct. The
author exhibits his impressive erudition in dealing with this issue.
Various criteria-content, state origin, enforceability, and enforce-
ability by courts-are rejected. In connection with the last he
makes the interesting comment that one cannot define "courts of
law" without defining "law" first. "The law is not what the
courts enforce; the courts are one of the institutions which enforce
the law." This he regards as a telling blow against the views of
the American legal realists. His own definition utilizes the more
neutral notion of "judicial organ," and avoids, he thinks, any
vicious circularity. A judicial organ is any definite authority
concerned with the application of principles to individual cases of
conflict between parties. We may speak of judicial organs "wher-
ever rules of any kind are applied." The idea of rule-application
preserves what is valid in the untenable requirement of enforce-
ment; for there are many instances of legal rules which are applied,
although the judgments falling under them cannot be enforced.
Two more items complete the definition of "law." The first is a
"subjective ingredient," namely, that rules of law are rules which
are considered fit to be applied by judicial organs. Kantorowicz
is careful to point out that "considered fit" does not refer to the
opinions and prejudices of legal and social philosophers, but
rather is used in regard to those who actually apply the rules or
wish them to be applied. This is all that he says on this important
part of his definition. This requirement can be understood in a
number of ways, so that it is difflcult to comment upon it, even to
say that "considered fit by those who wish them to be applied"
might be opening the door too wide for a history of legal science.
The final item is that there must be some more or less definite
procedure in the application of the rule by the judicial organ.
This serves to distinguish law from the rules of social custom.
Kantorowicz admits that "considered fit" and "some kind of pro-
cedure" are vague enough to leave room for theoretical doubts and
practical uncertainty. So much for the last two words of his
definition: "considered justiciable." It must be kept in mind
that "justiciable" does not refer to disputes which might be recog-
nized as suitable for adjudication, but to the rules themselves.
They are considered fit to be applied by a judicial organ according
to some definite procedure.

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms
714 THE JOURNAL OF PHILOSOPHY

Three brief comments are in order. First, one may question

whether Kantorowicz has completely avoided the circularity of
which the legal realists are accused. Can "recognized authority"
be defined without reference to law? The same question may be
raised about "definite procedure." Second, the restriction of the
concern of judicial organs to the application of principles to in-
dividual cases of conflict between parties appears too narrowing.
Where, necessarily, is the conflict when a confessor applies a rule
of penance to the "internal conduct" of his co-religionist? Third,
it would have been interesting to have had a detailed exposition of
the qualities which make a rule justiciable, as Goodhart point out.
Kantorowicz argues that his definition applies to primitive
societies, and that judicial organs may be found in such societies.
Finally, he recognizes that there are bodies of rules which are not
ordinarily made an object of legal science, but which fall under
his definition, e.g., rules of the chivalrous duel applied and en-
forced by courts of honor, rules of sports and games applied by
umpires, etc. He convincingly argues that this raises no difficulty
for his definition, that the analogy between such bodies of rules
and what is ordinarily called "law" is sufficiently close for all of
them to be classified under the same heading. The only exception
is rules of gangs of bandits, for they seem not to be applied with
the requisite definiteness of procedure.
In sum, Kantorowicz's little book is packed with interesting
ideas and deserves to be read by those interested in the philosophy
of law. It might also prove profitable to methodologists interested
in seeing how a person actively engaged in a discipline has gone
about staking out its area.
M. P. GOLDING
COLUMBIA UNIVERSITY

NEW BOOKS

BARTH:, KARL: Protestant Thought: From Rousseau to Ritschl.

Being the Translation of Eleven Chapters of Die protestantische
Theologie im 19. Jahrhundert. New York: Harper [1959].
435 p. $7.00.
BEVAN, EDWYN: Stoics and Sceptics. Four Lectures Delivered in
Oxford during Hilary Term 1913 for the Common University
Fund. New York: Barnes & Noble [1959]. 152 p. $4.50.
BRETON, STANISLAS: Approches phe'nomenologiques de l'idee d'e'tre.
Paris-Lyon: Emmanuel Vitte, 1959. 254 p. (Probl'emes et
doctrines, XVI.) 2100 fr.

This content downloaded from 37.111.1.85 on Sun, 22 Oct 2023 16:41:37 +00:00
All use subject to https://about.jstor.org/terms