Law and the State as Pure Ideas: Critical Notes on the Basic Concepts of Kelsen's Legal

Philosophy
Author(s): Kurt Wilk
Source: Ethics, Vol. 51, No. 2 (Jan., 1941), pp. 158-184
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/2989062
Accessed: 23/07/2010 01:40

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ucpress.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

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 support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Ethics.

http://www.jstor.org
 LAW AND THE STATE AS PURE IDEAS: CRITI-
CAL NOTES ON THE BASIC CONCEPTS OF
KELSEN'S LEGAL PHILOSOPHY
KURT WILK

I
AT of society is
A timewhen a profoundtransformation
reflected in widespread departure from established
legal values and a quest for new fundamentalrights
problemsinevitably come to the fore which touch upon the na-
ture of law itself. The contemporaryrevolutionariesof Conti-
nental Europeclaim to createlaw of an entirelynew orderafter
destroyingobsolete legal systems. Of the underlyinglegal and
constitutional concepts of the Anglo-Americanworld, mean-
while, many have been subject to a processof reinterpretation.
Now, finally,the violent clash betweennationalsocietiesorgan-
ized in differentdirectionsand stages of transitionhas resulted
in a series of breaches and pretendedredefinitionsof interna-
tional law. All this raises questionsof the law's relation to the
world of changing facts to which it is to apply. Can law be
changed,can it, indeed,be made and unmade,by alteredfacts
or by the responseto alteredfacts in creativecourt decisionsor
politicalfiat? How far and in what sense can the law itself gov-
ern the facts of changingsocial relationships? Can it be con-
ceived of as a self-centeredregulativesystem, and, if so, are the
contents of its rulesadjustableto changesin factual conditions?
What, in particular,is the role of the state in sustaining,adjust-
ing, or creating and destroyinglaw?
Questionssuch as these revive an old antagonismin legal and
politicalthought. Its developmenthas been markedby the con-
flicting ideas of those to whom law, in the last analysis, is the
pronouncementof a sovereignwill backed by political power
i58
 LAWAND THE STATEAS PUREIDEAS I59

and those who ascribeto the law a validity of its own, independ-
ent of factual forcesand ultimately foundedon ideal principles
intelligible to human reason. The progressof thinking along
these two distinctlines is indicatedby the namesof Machiavelli,
Hobbes, Austin, Bergbohm,and the school of legal positivism,
on the one hand, and of the monarchomachs,Althusius,Grotius,
Coke, Locke, and Bentham, on the other, with influencesfrom
both sides mergingin conceptssuch as Rousseau'svolontegene-
rale, Savigny'sVolksgeist,and Hegel'sWeltgeist.The controver-
sy has not been resolvedbut raised to anotherlevel by the ad-
vent of Kant's critical epistemologyand later of sociology and
social psychology: the formerhas subjected the basic assump-
tions of scientificthinkingto a critiquethat may be extendedto
the social sciences,while the latter are broadeningand deepen-
ing our empiricalknowledgeof social phenomena. Under this
doubleimpactthe conflictingideasof law as sovereigncommand
and as commonreasonhave evolvedinto a contrastin emphasis,
first, on the more complexfactual influences-economic, social,
and psychologicalas well as strictly political-that may shape,
and in turn be shapedby, the law; and, second,on the natureof
law as a system of regulatoryideas that may as such be dis-
tinguished from the social and individual mental processes
which may underlie them. The first outlook has manifested
itself in economic interpretationsof law of the Marxian and
othervarieties,in the rise of sociologyof the law, of legal plural-
ism and "realism";the second has found expressionin the neo-
Kantian "puretheory" of law, the phenomenologicaltheory of
legal concepts a priori,and their variations.
Nor have endeavorsbeen lacking to bridgethe apparentgulf
between these distinct aspects of the law and to visualizelaw as
both regulativeof and adapted to the needs of society. Such a
concept of the law underlies,as a workinghypothesis, the rea-
soning of a Holmesand a Cardozoas well as the methodsof in-
terpretationelaboratedby GermanInteressenjurisprudenz. But
emphasishas so far been placed here on applying such a hy-
i6o ETHICS

pothesisto the practicalinterpretationand developmentof laws

rather than on cementingits theoreticalfoundations.
The basic assumptionsof ideal regulatoryvalues, on which
both rationaljudicialinterpretationand theoreticalinsight into
the law as a distinct order depend, have been forcefully chal-
lengedby the recrudescenceof the olderconceptof law as sover-
eign power in the legal doctrineof GermanNational Socialism.
That doctrineopenly questionsthe very possibilityof objective
reason from which legal values could be derived, whether
through the mediation of political power or otherwise. On its
own part, to be sure,it relieson a disregardof criticalanalysisof
power as a political phenomenonand accepts the acts of the
power that be as expressionsof a presumedracial group mind,
naively personifiedin the politicalleader. This attack adds im-
portanceto the increasingattention of Americanthinkersto the
theoreticalbases of law as an instrumentalityof social regula-
tion, including the critical review of contemporaryEuropean
thought on the subject. Any such re-examinationof the role of
law in society and its relation to power and reason must first
take stock of those ideas which altogetherdeny the power ele-
ment in law-ideas which start from the conclusionsof critical
epistemologyand deducetherefroma radicaldissociationof law
fromall relationshipsof politicaland socialfact.
The boldest elaborationof such a purely ideal concept of the
law as a body of rulesof social significanceyet wholly divorced
as such fromsocialcausesand effectsis still that of Hans Kelsen.
His theory firstbrokeaway fromthe establishedpositivisticand
the emergingsociologicalapproachesto law in Germanyby re-
fining and applying the conclusionsof Kant's critical idealism.
It has since stimulatedlegal and politicalthought in many Con-
tinental countries,both throughevokingfurtherrefinementand
modificationand through challengingcritical opposition,'and
it has more recentlyattracted also just attention in GreatBrit-
I See the bibliographyby R. A. MOtall,appendedto H. Kelsen, Reine Rechtslehre

(Leipzig-Wien,I934), pp. I55-222.

 LAW AND THE STATE AS PURE IDEAS i6i

ain and America. ProfessorKelsen himself has presented the

English-readingpublic with an abbreviatedversion in the Law
QuarterlyReviewof Oxford,2of his magisterialsummaryof the
"pure theory of law,"3which his previous writings had elabo-
rated in greaterdetail and fortifiedwith penetratingcriticism.4
Some other papers of his in English present his conclusionson
particularquestionsof legal philosophyor of practicallegal and
political concern.5 In addition, there are now a number of
studiesby British and Americanscholarssetting forth the prin-
cipalpoints of his so-called"Viennaschool"of legal thoughtand
commentingcritically on some of its major aspects and on its
place in modernsocial philosophy.6There is no need to add to
2
"The Pure Theory of Law," Law QuarterlyReview,L (I934), 474-98; LI (i935)
5I7-35.
3 ReineRechtslehre.
4 In particular:Hauptprobleme der Staatsrechtslehre(2d ed.; Tflbingen,I923); Das
ProblemderSouverdnitdt unddie Theoriedes Volkerrechts (2d ed.; Tubingen,I928); Der
soziologische (2d ed.; Ttlbingen,I928); Allgemeine
undderjuristischeStaatsbegriff Staats-
lehre(Berlin,I925); Die philosophischen Grundlagen derNaturrechtslehre unddesRechts-
positivismus(Charlottenburg,I928); "Th6orieg6ndraledu droit internationalpublic:
probldmeschoisis," Acaddmiedu droit international,recueildes cours, XLII (I932),
Part IV, I I 6-3 5 I.
5 TheLegalProcessand InternationalOrder(London,I934); "The Party-Dictator-
ship,"Politica,II, No. 5 (I936), I8-32; "Centralizationand Decentralization,"in Au-
thorityand the Individual("HarvardTercentenaryPublication"[Cambridge,I937]),
pp. 2io-39; Legal Techniquein InternationalLaw: A TextualCritiqueof the League
Covenant(Geneva, I939).
6 E. Bodenheimer, "Power and Law," Ethics, L (ig40), I27-43, at I30-33; I. Husik,
"TheLegalPhilosophyof Hans Kelsen,"Journalof SocialPhilosophy,III (I938), 297-
324; H. Janzen, "Kelsen'sTheoryof Law," AmericanPoliticalScienceReview,XXXI
(I937), 205-26; J. W. Jones, "The 'Pure'Theoryof InternationalLaw,"British Year
Bookof InternationalLaw(I935), pp. 5-i9; J. L. Kunz,"Onthe TheoreticalBasisof the
Law of Nations,"GrotiusSocietyTransactions, X (1925), II5-42; "The'ViennaSchool'
and InternationalLaw,"New YorkUniversityLaw QuarterlyReview,XI (I934), 370-
42I; H. Lauterpacht,"Kelsen'sPure Scienceof Law," in W. I. Jennings,ModernThe-
and
oriesof Law (Oxford,I933), pp. I05-38; I. Mattern,Conceptsof State,Sovereignty,
InternationalLaw (Baltimore,I928), pp. I2I-39; C. W. Stern,"Kelsen'sTheoryof In-
ternationalLaw," AmericanPolitical ScienceReview,XXX (1936), 736-4i; E. Voe-
gelin, "Kelsen's Pure Theory of Law," Political Science Quarterly,XLII (I927), 268-80;
C. H. Wilson,"The Basis of Kelsen'sTheoryof Law,"Politica,I, No. I (1934), 54-82.
An Englishedition has been announcedof the comprehensivebook by W. Ebenstein,
(Prague,I938); cf. reviewby A.
Schuleder reinenRechtslehre
Die rechtsphilosophische
Brecht,AmericanPoliticalScienceReview,XXXII (I938), II73-75.
162 ETHICS

these presentationsof Kelsen's theory to the Americanreader.

This paperis directedratherat a criticalanalysisof his twofold
centralconcept,fromwhichthe many ramificationsof his theory
spring in unusual logical consistency. At the core of Kelsen's
thought is a peculiarconcept of law as a compulsoryregulative
orderand of the state as identical with the law. The meaning
and implicationsof that concept will be outlined here only as
far as necessaryto obtain standardsof critique, and from the
application of these standards some conclusionswill be ven-
tured that may bear upon the nature of the law and the state.
II
Kelsen's views on the essence of law and of the state have
been expressedin clear-cutdefinitions. The law, accordingto
him, is, generically,a system of rules or norms. Specifically,it
is a system of such norms as institute compulsion. Both the
generic characterand the specific differenceof the law in this
definitionare understoodby Kelsen in a distinct and peculiar
sense.7
A norm in that sense is by no means a link in the causalcon-
nection of naturalevents. It is differentnot only fromphysical
7 Kelsen'sconceptsare all coloredby the peculiarmethod of his philosophicalap-
proachto law and the state. He proposesto eliminatefromjurisprudenceall elements
that are not whollyand exclusivelylegal. In this he impliesthat anythingthat can be
logicallyabstractedfroman object of knowledge,such as the law, is thereforeessen-
tially foreignto its nature. Thus, he followswhat he calls a "pure"method. He also
impliesthat any particularobjectof knowledgeand the approachto it are of necessity
determinedby each other,so that the methodof researchdependson the limitationof
its object,while it defines,in turn,what the limits and the characterof that objectare
to be. To that extent he followswhat may be called "limitativemethod." Finally, he
impliesthat scienceswhich are based on differentmethods,correspondingto and de-
terminativeof essentiallydifferentdelimitationsof theirsubjectmatter,lack a common
groundof knowledge.Indeed,he maintainsan irreconcilable contrastbetweenhis legal
methodandthe methodormethodsof certainnonlegalsciences.Tothatextenthefollows
what may be calleda dualisticor possiblypluralisticmethod. Withinits limitedrealm,
on the otherhand, any such methodhas a distinctlymonistictendencyto referall its
conceptsback to the same imitative assumptionson its basic subject matter. While
oppositionto Kelsen'stheoryhas frequentlybeenbasedon criticismof his method,this
paperconcentratesratheron analyzingthe substantiveelementsof his concepts. For a
fine summaryof the methodicalproblemsinvolved,see Lauterpacht,op. cit.
 LAW AND THE STATE AS PURE IDEAS i63

processesbut also frommental processes. In particular,it must

be distinguishedfrom those mental acts throughwhich it is per-
ceivedby individuals. The normis the objectivecontent of per-
ceptions,but it is no perceptionitself, nor is it the effect of per-
ception. It furnishesa standard by which the significanceof
humanbehaviormay be judged,but in itself it is neitherhuman
behaviornor the cause of any behavior. In short, the norm is
idea and not at all naturalreality. Its contents may and do re-
fer to those of other norms, but they have no connectionwith
acts of natural causality. They embody no commandor any
other act of interferencewith naturalfacts but ratherconstitute
a logical judgment. Indeed, each norm is a hypothetical judg-
ment of specific character,viz., the logical connection of two
circumstancesby an imputation. That imputationis a matter
of the "Ought,"which, therefore,is the specificlogical category
by whichnormsare to be understood. "If there is A thereought
to be B" means that the circumstanceB is imputed to, or con-
ceived as the consequenceconditionedby, the circumstanceA.8
A system of norms,accordingly,can rest only on theirmutual
connection,which consists in the referenceof any such imputa-
tion to contents of anothernorm. For, if the contents of norms,
while by no means arising out of facts as such, are to be con-
ceived at all as meaningfuland significant,they must refer to
some standardof meaning,which then can be foundonly in an-
other norm. Thus, there exists of necessity a systematic co-
herenceof normsas a sequenceof references,step by step, of the
contents of each normto those of another. It leads back to one
ultimate norm,which itself cannot be referredback any further
but must be assumedex hypothesi to give meaningto the entire
system. This hypothetical norm, then, is the fundamentalor
original norm that is required to support the systematic co-
herence of norms. Not only does each single norm contain a
8 Law QuarterlyReview, L 477-8I, 484-85; Reine Rechtslehre,pp. 4-7, 20-24;
(I934),
also Problem der Souverdnitdt,pp. i-8; Staatsbegriff, pp. 75-8I; Allgemeine Staatslehre,
pp. i6-2I.
i64 ETHICS

hypotheticaljudgment,but it is part of a system which in turn

is purely hypotheticalin character
Again, institution of compulsion, which in Kelsen's view
marksthe differencebetweenlaw and all other norms,has noth-
ing to do with the actualbringingabout of force as the sanction
of a command. That would be outside of the purely significa-
tive sphereof norms. It rathermeans the insertionof compul-
sion as one link in the logical chain of imputation. In other
words,in the legal normit is a compulsoryact that is imputed,
as a consequence, to some conditional circumstance. This
specificcharacterof the law is quite independentof what are the
particularcontents of any legal rule. For it admits of institut-
ing various consequentcompulsoryacts and of assumingany
conditionalcircumstanceswhatever. So law appearsas a social
technique that can serve any social purpose. It is not only a
hypotheticalsystem but, in its specificessence, a truly neutral
system. Furthermore,there being no particular contents of
norms that could be called essentiallylegal, the systematic co-
herenceof law cannotbe soughtin its contents. It may be found
only in a seriesof commonformalcriteriaof the makingof law of
differentscope, that is, criteriadeterminingwhether there has
come into being, say, a constitution,treaty, statute, ordinance,
customary rule, contract, judicial decision, or administrative
order. In this sense the law is a formalsystem.'0
Such is Kelsen'sbasic conceptionof the law, out of which the
other major notions of his legal theory are evolved with great
consistency. It is only consistentfor him, for instance,to define
the validity of the law as the specifichypothetical existenceof
legal normsoutside of the sphereof naturalfacts;"'or to define
positive law as a legal system the fundamentalnorm of which
9 Law Quarterly Review, L (I934), 479, 497; Reine Rechtslehre, p. 66; also Staats-
begriff, pp. 93-95; "Centralization and Decentralization," pp. 213-14.
IO Law Quarterly Review,L (I934), 486, 488; ReineRechtslehre, pp. 25, 28-33, 62-89;
also Staatsbegriff,p. 82; Die philosophisckenGrundlagen,pp. 30-4i; "Centralization and
Decentralization,"p. 212.
Il Staatsbegriff,
p. 95.
 LAW AND THE STATE AS PURE IDEAS A65

cannot be derivedfrom another, that is a nonlegal,norm'12 It

is also consistentfor him to concedelegal significanceto natural
facts only in so far as they are referredto in legal rules as mere
elementsof theirnormativecontents-there being no true con-
nection betweenthe actual facts and the contents of normsbut
merely a certainparallelism.'3So, again, his premiseslead him
to deny any substantivedistinctionof normswithin a system of
law. No distinguishingline can be drawn between public and
privatelaw or betweenobjectivelaw and subjectiveright, for all
law is conceivableonly as a formallycoherent,hence logically
uniform,system. Neither can a line be drawnbetween natural
and artificiallegal persons,since the personis merelyan expres-
sion of the unity of some particularbundle of norms. Nor can
any be drawnbetweenpersonaland real obligationsand duties,
since individual duties of any kind are merely legal norms
viewed from different special angles; or between lawful acts
and wrongs,since torts or criminaloffensesare simply one kind,
among others, of conditions of legal consequencesand cannot
thereforeproperlybe conceivedas unlawfulacts.14 All of which
leads us back, for a critical understandingof Kelsen's whole
legal theory, to a closer analysis of his fundamentalconcept of
the law.
We may take up the elements of Kelsen's definition in his
order,proceedingfromthe genericto the specificaspects of law.
In his view law is normativein kind; and the norm is a kind of
objective and ideal meaning,quite distinct as such, as we have
seen, from all natural facts. By this token the law is sharply
distinguishedfromthe individuals'experiencesof what is legally
ordainedor enforcedand also from the outer facts of social life
envisaged in legal relationshipsand institutions. To be sure,
in that sharpdistinctionwhich is at the bottom of Kelsen'sthe-
I2 Ibid., pp. 87-88.
I3 Ibid., pp. 93, 94, 96; AligemeineStaatslehre,
pp. i8-i9.
'4 Law QuarterlyReview,L (I934), 487, 492, 494-97; ReineRechtslehre,
pp. 26-27,
40-4I, 46-59, I09-14; also Ailgemeine Staatslehre, pp. 47-70, 80-94.
i66 ETHICS

ory, it is not ignoredthat such mental and social facts somehow

correspondto the abstract ideas of law. It is not claimed that
the law is something existing a priori and independentof its
everbeingconceivedor recognizedor established. It is admitted
that the objective meaningof law, like all ideal meaning,is at-
tached to, though it must not be confoundedwith, subjective
acts of humanminds.'5Nor does Kelsen'sdistinctionabsolutely
oppose law, as idea, to reality. Far from regardingthe law as
unreal,he attributes to it a reality of its own, namely, ideal as
opposed to natural reality.'6
This basic thesis of Kelsen'stheoryis, indeed,of transcendent
value. True, as he has not failed to point out,'7 it is not a
thought that belongs originallyand exclusivelyto his pure the-
ory of law. He has, rather, carriedon, refined,and adapted to
jurisprudencea line of thought prevalentin Germanphilosophy
since Kant-the distinction between act and significance,be-
tween psychophysicalnature and ideal contents, and, ultimate-
ly, between causal determinationand moral freedom.'8 Yet
within that multifariousphilosophicaltrend and its ramifica-
tions in GermanjurisprudenceKelsen's theory is of outstand-
ing significance:first, because of the cogent criticismwhich he
has incessantlydirectedfrom the basis of that fundamentaldis-
tinction against the confusionof causal and psychologicalwith
ideal and no-ologicalconceptionsin jurisprudenceand political
science and which should help to dispel much loose thinkingin
those fields,and, second,becausehe was the first to build a com-
prehensiveframeworkof juristicthought upon that distinction.
While critical objections may be and often have been raised
'5 p. 93.
Staatsbegriff,
I6 LawQuarterlyReview,L (I934), 48i; Staatsbegriff,
pp. 76-78.
17 E.g., ReineRechtslehre,
Vorrede.
A8 Germanlegal thoughthas been influenced
by this distinctionin differentversions;
notablythe neo-Kantian(Marburgschool: H. Cohen,SystemderPhilosophie,Logikder
reinenErkenntnis,Vol. II: EthikdesreinenWillens[Berlin,I902-4]) and the phenome-
nological(Freiburgschool: E. Husserl,LogischeUntersuchungen [Halle, igoo], esp. I,
228-45).
 LAW AND THE STATE AS PURE IDEAS i67

againstthis theoreticalframework,they must neverlose sight of

its underlyingtruth that the contents and meaning of an act,
and in particularof a mental act, are not identicalwith that act
itself and that this applies to the law no less than to any other
sphere of thought. Recognition of this common foundation
would help to eliminatemuch theorizingabout the law that is
frankly or confusedly psychologicaland naturalistic. For the
rest, it should reducethe controversiesbetween Kelsenites and
anti-Kelsenitesto justerproportions.
Yet Kelsen has not been, and cannot be, content merely to
recognizethat law belongs to the realm of ideas and meanings.
He must determinemoreclosely the ideal naturenot only of ob-
jective meaningin generalbut of such meaningas may be con-
tained in law. What is the more particularkind of ideal con-
tents to which the law, along with possibly other regulatory
ideas, belongs? The whole conceptof the law dependson this its
genusproximum.Its definitionis the crucialproblemof Kelsen's
theory and, indeed, of any legal theory which shares the basic
assumptionof law as idea.
The problemmay be divided into three separate questions.
First, what is the characterof normativemeaningsuch as may
be containedin law, comparedwith other complexesof mean-
ing? Second, what is its relationshipto the extra-idealworld?
And, third, what interconnects meaning of this characterso
that we can think of it as a coherentwhole called the law or
some similar comprehensivesystem? Kelsen's theory under-
takes to answerthese three questions. But a criticalanalysisof
his answerssuggeststhat they are beset by certain seriouslogi-
cal difficulties.
First: To characterizemore especially the normative con-
tents of the law, as we have seen, Kelsen definesthe norms as
the logical formsof values, expressedin "Ought"clauses. The
evaluating judgments or norms, in this definition, are but a
particularkind of hypotheticaljudgments,which in turn are a
class of logical judgmentsor connectionsmade betweenpercep-
i68 ETHICS

tions througha priori categoriesof thought. A norm is distin-

guishedfromother hypotheticaljudgmentsby the specialcate-
gory of thought which it uses for connectingits elements;it is
this specialnormativecategorythat Kelsen calls "imputation."
But does this really distinguishthe normsfrom all other logical
judgments? To be sure, there is a distinction of hypothetical
from other, especially categorical,judgments with which stu-
dents of logic are familiar.'9But no satisfactorydistinctionhas
yet been drawnbetweenthe imputationand other hypothetical
connections. Kelsen'sdefinitionof the imputationas the logical
connection between a conditioned and a conditional circum-
stance is simply a definitionof the hypothetical connectionas
such. In other words, Kelsen does not establish a positive dis-
tinction betweenhypothetic "Ought"judgments("if there is A
there ought to be B") and hypothetic "Is" judgments("if there
is A thereis B"): in both of them conditionsare connectedwith
consequences. His only distinction is a negative statement.
Startingfromthe fundamentaldisparityof causalexistenceand
normativevalues, he deems the normshypothetical judgments
the particularlogical category of which is not causality. But
such a negativedistinctiondoes not sufficeto explainwhat is the
peculiar characteristicof the values, the "Ought,"which con-
stitutes the presupposeddisparityof causaland normativejudg-
ments. Indeed,it presentsa meretautology. The supposeddis-
tinction of normativefrom other hypotheticaljudgmentsis im-
plicit in the underlyingassumptionof a disparity between the
"Is" and the "Ought"and fails to add anything to it.
Thus, Kelsen cannotbe said to have succeededin distinguish-
ing the normsfromother meaningmerelyby a formalcriterion,
viz., a logical categoryof their own which would explain them
as a specialkind of logical judgments. It is submittedthat any
such distinctionis bound to be unsuccessful. If, in accordance
with Kent, the differentkinds of logical judgmentsare defined
'9 Cf. Kant's distinction of categorical,hypothetical, and disjunctivejudgments
(KritikderreinenVernunft[Akademieed.; Berlin],pp. 87-89).
 LAW AND THE STATE AS PURE IDEAS i69

and distinguishedby their inherent categories,then the hypo-

thetical judgmentwill be found to be determinedby the cate-
gory of causality, that is, the specificconnectionbetween con-
dition and consequence.20 This generaland formalcriterionof a
kind of logical judgments, however, is applicableto empirical
objects of any kind. It does not divide the empiricalworldinto
two spheres that would be separately cognizable and would
thereforehave to be consideredas objectivelydistinct. Causal-
ity as a transcendentallogical category is not confinedto judg-
ments regardingphenomenaof the naturalworld. It is turned
into the particularcausality of naturalsciencesonly when it re-
fers to those phenomenawhich appearin space and time.
Moreover, even if a noncausal hypothetical category were
conceivable,it would fail to explainthe norm as an evaluating
judgment. For if the norm were a hypothetical judgment, it
could not but be neutral toward the circumstanceswhich con-
stitute the elements of such a judgment. The elements of any
hypothetical judgment, taken by themselves, are not assertive
but problematicaljudgments.21 That is to say that in a state-
ment that "if there is A there ought to be (or is) B" we put a
qualificationor evaluation neither on the circumstanceA as
such nor on the circumstanceB as such but only on the connec-
tion between the two. For attributingvalues to circumstances
does not mean to keep them in a state of logical suspense;rather
does it mean to affirmor negative, to select or reject,them. Im-
puting a circumstanceas a consequenceto some other condi-
tional circumstanceoffersno indicationas to whetheror not the
consequenceor the conditionthemselvesought to be. It follows
that a hypotheticaljudgmentas such does not at all expressan
evaluationand, therefore,does not constitute a norm. An eval-
uation may be contained only in a categoricaljudgment or in
such anotherjudgment-say, a hypotheticone-as refersto and
20
In fact, Kant himselfregardscausalityanddependency(orcauseandeffect)as the
categoryspecificto hypotheticaljudgments(ibid.,pp. 86-87, 88-89, 92-95).
2I Accordingto Kant's terminology(ibid.,pp. 89-90).
170 ETHICS

dependsupon a categoricaljudgment. The specific nature of

normativejudgments,then, is to be explainednot by any spe-
cial logical characterof theirs, as assumedin Kelsen's theory,
but by the particularmode of perceptionapplied to their ob-
jects. The objects to which normativejudgmentsreferareper-
ceived, not as phenomenain space and time, but ratheras mere
possibilitiesin space and time which are contemplatedsubject
to some presupposedstandard of values. In other words, the
normis an imperativejudgment. That is to say, it is, of course,
not a commandin the psychologicalsense of a causative act, as
Kelsen has convincinglyshown;it is an imperativeas objective
meaning.
Second: In determiningthe relationsbetween the normsand
the nonnormativeworld, Kelsen starts from the assumption
that normsand naturemust be logicallyisolatedif eitheris to be
understoodat all. But at the same time he assumes that the
norms,too, are real and are not entirely divorcedfrom the nat-
ural acts underlyingthem. Thus, he is faced with the problem
of how to explainthe connectionof the two essentiallydisparate
spheresof psychophysicalacts and normativevalues. Yet this
problemis insolubleif the objectivedisparityof naturalacts and
values is soughtin an incompatibilityof ways of cognition,as in
Kelsen's theory. For if, as he assumes,the disparity of norms
and naturerests on the disparityof cognitionsof the normative
and natural sciences and this in turn rests on the formal dif-
ferenceof two incompatiblelogical categories,then it is possible
only to understandeither isolated spherebut not an objective
connectionof both. The two sphereshave nothing in common
that could constitute such a connection. If we assume two in-
compatiblesorts of logical judgmentswe are necessarilyled to
two separategroupsof cognitionsand cannot arriveat one cog-
nition that would connect the two under an idea common to
both. The radicaldualismof the causal "Is" and the normative
"Ought,"then, to whichKelsenis led by his rigorousseparation
of two pure scientificmethods,may at best help to explain the
 LAWAND THE STATEAS PUREIDEAS I7I

ideal contents of values. But it can in no way make clear how

values affectthe worldof socialrelationshipsand behaviors. An
essentialpart of normativereality thus remainsunexplained.
Kelsen himself clearly recognizesthe existence of some con-
nection between norms and causality without, however, being
able to explainit fromhis basicassumptions.This is revealedby
his definitionof the relationof legal normsin particularto hu-
man behavior. In accordancewith his strict separationof the
normativeand causal sphereshe regardsthe validity and posi-
tiveness of the law merelyas attributionsof the abstractsystem
of objective normative meaning.22Consequently,he ascribes
actual effectivenessnot to the law itself but solely to what indi-
viduals think of it and are thus influencedby-that is to say, to
the law's ideology.23The relevancyof law within the world of
facts, then, is to him merelya certainrelationbetween the nor-
mative validity of law and the actual effects of legal ideology.24
Apart fromthis relation,legal normsin his view are exactly like
any other sort of ideal meaning,such as mathematicalor logical
statements. That is to say, they correspondto mental acts, but
they are neither dependent on them as a condition per quctm
nor realized by them.25Yet in what this special relation be-
tween the validity and the effects of law consists Kelsen does
not, and indeedcannot,say. For if he reallywereto refernature
and norms to each other he would to that extent join them by
common judgments; and this would imply just that unitary
cognition of all phenomenawhich his assumptionof an irrec-
oncilabledualismof an ideal and a natural world excludes.
Indeed, Kelsen merely registers the empiricalphenomenon
that law and actual behaviorare in part parallel. He does not
explainit as an elementof legal reality but rathertakes it for a
by-productof an economy in legal thought. In his view social
22 Staatsbegrifi, pp. 87-88, 95.
23 Ibid., pp. 87, 92, 93; ReineRechtslehre,pp. 125-27.
24
Staatsbegriff,pp. 87, 93; Allgemeine Staatslehre,pp. i8-i9.
25 5taatsbegriff, PP. 79, 93.
I72 ETHICS

relevancy does not essentially inhere in law in the sense that

sociallyirrelevantpretendedlaw wouldsimplynot be law. It is,
rather, a matter of expediencyto select one of the numberless
conceivablefundamentalnorms and resultinglegal systems as
the hypothesisthat is best suited to give meaningto actual hu-
man behavior.26Thus Kelsen renouncesany attempt to give an
explanationof the law's social significance.Resigninghimselfto
describingit as a phenomenon,he arrivesonly at the somewhat
commonplaceobservationthat there is neither strict identity
nor completedivergenceof lawfuland actual conductbut rather
a certain tension between these maximumand minimumlim-
its.27 So Kelsen's treatment of the relation between law and
facts only serves to confirmthe conclusionthat the problemof
the factual relevancyof normsand, in particular,of legal rules
cannotbe solvedon the basis of a strictly dualistictheoryof cog-
nition regardingthe "Is" and the CCOught.1128
Third: Closely connectedwith the foregoingare certain ob-
jections that may be raised against Kelsen's view of the sys-
tematic unity of the law. Since he isolates the law from the
legally constituted human society, he cannot rest the unity of
a legal system on any relevancy it may have for one and the
same social group. He can deriveunity of the system only from
the abstractrelationof its normsto one another,based on their
commonlogical dependencyupon the same fundamentalnorm.
In this way it is, indeed,possibleto set forth a coherentsystem
of law, providedonly that all norms ascribedto such a system
26 Ibid., pp. 95-96.
27 Ibid., p. 96; Allgemeine Staatslehre, pp. i8-i9.
28Kelsenvirtuallyadmitsthat he posesan insolubleproblemwhenhe acknowledges
"the apparentlyinevitableantinomybetweena dualismof the "Is" and the "Ought,"
whichmust be presupposed,and the undeniablerecognitionof an objectiverelationship
connectingthe two systems that are presupposedto be unrelatedto each other"(All-
gemeineStaatslehre,p. ig). This inherentcontradictionhas been frequentlynoted by
criticsof Kelsen'stheory(see, e.g., Wilson,op. cit., pp. 67-82; H. Heller,Die Souverani-
tat (Berlin, I927), pp. 53-55; J. Mo6r, "Reine Rechtslehre," Zeitschriftfiir dffentliches
Recht,XV (I935), 330-43, at 338-43; H. Morgenthau,La Rgalite des normes(Paris,
I 934).
 LAW AND THE STATE AS PURE IDEAS I73

are so interpretedas to fit into its coherence. In other words,

each normmust be interpretedchieflywith a view to upholding
the presupposedlogicalcoherenceof the law at largeratherthan
with regardto the immediatepracticalintents and purposesof
the particularrule. Wheneverthere is a conflict between the
two viewpoints,the unity of the system can be maintainedonly
at the price of artificial or impracticableinterpretations. In
such cases a legal rule will be interpretedby readinginto it cer-
tain contents for the sake of theirlogical consistencywith other
rules, whatever its more immediateand obvious meaningmay
be.
Such misinterpretationsappearto be inevitable if the unity
of a legal system is sought in the abstract coherenceof all its
norms. For such coherence,beinga matterof ideal meaning,ex-
ists in thought only, whereas in fact legal rules have often
emergedand may still emerge as the unreflectedand isolated
standardsof variousphasesof humanbehaviorwithin the same
community. These rules which apply in the same community
but do not necessarilyrefer to or agree with one another must
under Kelsen's theory all be fitted into normative coherence;
that is, they must be relatedthroughother intermediarynorms
to a commonhypothetic fundamentalnorm. Thus, in Kelsen's
view unconstitutionalstatutes and also ordinancesor regula-
tions ultra viresmust be theoreticallyconstruedas subordinate
legal norms,whichare temporarilyand conditionallyauthorized
by the superiornorm even though they are at variancewith it.
That is to say, these unconstitutionalor illegal provisionsmust
be accepted in legal theory as valid norms unless and until set
aside by legally established procedure,when they lose their
normativevalidity becauseof their declaredinconsistencywith
superiornorms. The same appliesto judicialdecisionswhichdo
not correctlystate the law. They must be acceptedin legal the-
ory as delegatednormsof limitedvalidity, whichwouldend only
if and when they are reversedor overruledby later decisions.29
29 pp. 229-6i;
AllgemeineStaatslehre, pp. 73-89.
ReineRechtslehre,
I74 ETHICS

Furthermore,carriedto its logical conclusion,this theory can

conceiveof normsas legal only inasmuchas they are parts of a
comprehensive,rationally coherent system of law. And, since
there are no logical limits to such a system or any cogent cri-
terion to distinguishseparatebodies of law on a purely logical
basis, the same theory must assume, as the most satisfactory
hypothesis, a unitary system of all law. It must thereforeaim
to construeevery legal ruleor decisionas an elementof that all-
inclusive body of law. Thus, for instance, the distinct rules of
internationaland of municipallaw or of civil and ecclesiastical
law or of various autonomouslaws must all be construed as
logically coherent within one common legal system,30even
thoughtheircontentsmay makeno referenceto one anotherand
may be intentionallyseparateand even incompatible. Similar-
ly, for the sakeof the presupposedobjectiveunity of the law, the
unreflectedcoexistencein more primitivelaws of separatelegal
maximsfor differentphasesof humanbehaviorwouldhave to be
ignored. By the same token the frequent inconsistenciesbe-
tween legal doctrinesthat have developedfor differentcases in
customarylaws or in judge-madelaw would have to be neglect-
ed, since all doctrinesmust be consideredas derivingfrom the
supposedlyuniformsystem of law.
Yet the thought that all law is consistent and harmoniousis
itself an afterthoughtof the legal scholarreflectingon existing
laws. It is the productof his effortsto co-ordinatethe individual
legal ruleswith whichhe is confronted. Kelsen,on the contrary,
makesit the basic assumptionwhich is to determinewhetheror
not a given rule is legally valid and how a rule claimingsuch
validity is to be construedand interpreted. Thus, he must close
his eyes to the historical developmentof law, which has pro-
gressedby gradualsteps only to the concept of a legal system
embracingall the diversestatutory and judicial rules and doc-
trines. Likewise,he must ignoreany trends growingout of the
increasingcomplexityof modernsociety towardlegal pluralism,
30 Allgemeine Staatslehre,
pp. ii5-33; ReineRechtslehre,
pp. i34-36.
 LAWAND THE STATEAS PUREIDEAS I75

that is, the developmentof voluntary agreementsamong large

groups of people into autonomousbodies of law, such as rules
arrivedat by collectivebargainingor by voluntaryco-operation
of businessfirms and associationswith governmentalagencies.
A legal theory that is to understandthese aspects of law will to
that extent deviate from Kelsen's view. It will considersys-
tematic coherencenot as an essentialcriterionof law as suchbut
rather as a feature of a particularstage in the developmentof
laws. Also, it will seek the unity of a concretebody of law not in
the systematic coherenceof all meaningit may contain but in
the identity and unity of the particularsociety whichit governs.
From these aspects of the generic characterof the law as a
system of regulatoryideas we may turn brieflyto its specificcri-
terionin Kelsen'stheory,whichis the institutionof compulsion.
We may agree, indeed, that the institution of compulsionis
specificallylegal or, in other words,that the ideas of regulation
whichare containedin legal rulesareideasof regulationby com-
pulsion-at least, if law is taken to include rules of social cus-
tom. But we may well questionwhetherthe institutionof com-
pulsion marksthe only distinctionof law fromother regulatory
ideas. Kelsenassumesthat it does; consequently,he ascribesto
the law a neutral and completely formal character. He is led
to this view by his strict isolationof normsfromfacts. He can-
not accept compulsionas an elementin the realizationof law in
human society, since that would provide a common concept
combininglegal ideas and social relationships. Thus he is con-
strainedto definelegal compulsionas a merepart of the abstract
meaningof legal rules. Conversely,his definitionwould appear
cogent only if the strict isolation of normative meaning from
naturalfacts could be accepted.
Looking back over our analysis of Kelsen's concept of the
law's generic characteras a regulatorysystem, we recall that
its normativemeaningcannot be explainedwithout contradic-
tion in termsof a logicaljudgmentopposedto causality;nor can
the relevancy of law and its unity as a body of diverse legal
176 ETHICS

rules be understoodby divorcinglaw from the social facts to

whichit is to apply. In otherwords,for an understandingof the
law as a body of regulatoryideaswe have to departfromassump-
tions of "pure"causality and "pure"normality. Thereby,we
abandonthe basis for a conceptionof legal compulsionas a mere
logical link in a noncausaljudgment-a link by which alone the
law could be distinguishedfrom other systems of purely logical
meaning. From this point of view there is, then, no need to as-
sumethat law, being characterizedby compulsion,shouldother-
wise be susceptive of any conceivable contents. There is no
need to considerall its preceptsas purely accidental,unrelated
to any extralegalvalues, and dependentfor their validity solely
on their complying with prescribedprocessesof lawmaking.3'
III
The state, accordingto Kelsen, is nothing but the law. This
identificationof the state with the law as understoodby him
carriestwo majorimplications-one negative and one positive.
First, like the rulesof law the state does not belongto the world
of facts.32 Second, with all law being regardedas a single co-
herent system, each individual state is identical with a partial
legal orderof a certainkind.33
To supportthe negativepart of his definitionof the state Kel-
sen argues that it is impossibleto explain any social organiza-
tion as a factual associationof men. For in purely factual, that
is, psychophysical,termswe cannotarriveat any understanding
of such featuresof social organizationas objectivity, constancy,
unity, impersonality,and spatial extension. Yet these are the
very featuresthat distinguishorganizationslike the state from
3I The purificationof law fromall extralegalvalueshas beenanotherprincipalpoint
of criticismagainst Kelsen's theory (see, e.g., Lauterpacht,op. cit., pp. I3I-38; E.
Kaufmann,Kritik der neukantischen Rechtsphilosophie(Berlin, I921), pp. 20-32; L.
Pitamic, "ReineRechtslehre,"ZeitschriftfuirOffentlichesRecht,XV (I935), 4IO-I5, at
4I3-15.
32 Staatsbegriff,pp. 4-74; Aligemeine Staatslehre, pp. 6-I3.
3 Staatsbegriff, 87-9I; Ailgemeine Staatslehre, pp. I3-I8; Reine Rechtslehre,pp. II7-
19, I29-36; "Centralization
and Decentralization,"pp. 2I5-I6.
 LAW AND THE STATE AS PURE IDEAS I77

passing and fluctuatinghuman relationshipssuch as a friend-

ship or an occasionallygatheringmultitude.,4 It follows that
the distinctive natureof the state can be explainedonly by ref-
erenceto meaningfulideas, which, as we have seen, may be un-
derstood as the objective contents of mental acts but must be
clearly distinguishedfrom the natural processesand effects of
those acts.
Examiningthe positive aspects of the state as idea, Kelsen
finds that it has the same characteristicsas those attributedto
the law. It is, to him, a normativesystem of values; and it is a
system instituting compulsion.35From this conformityof their
characteristicshe concludesthat the state and the law are iden-
tical. In otherwords,fromthe point of view of any state or of its
municipallaw the two coincide so fully that one cannot con-
ceive of any state acts outsideof the law or of any legal concepts
and institutions beyond the reach of the state."6 To the ob-
server who does not confine himself to the law of a particular
state but looks at law as a whole every state appearsas identical
with a well-definedpart of the legal order. For, if law can be
conceivedof only as a singlecoherentsystem, it must be deemed
in the last analysisto comprisethe laws of all states as parts of
the total system. Also, it must be deemed to include certain
rules that serve to set the limits between those partial systems
without belongingto any one of them. In this total view of the
law, then, the law of nations, existing outside of the municipal
law of any state, constitutes the supposedsupremebody of dis-
tributive and delimitativenorms,while each municipallaw and
the state identifiedwith it form one of the limited component
bodies of norms.37
Kelsen's whole political theory is a bundle of conclusions
34 Staatsbegriff, pp. I4-I5, 4I, 44-45.
35 Ibid., pp. 43-45, 75-84; Allgemeine Staatslehre, pp. i4-i6.
36 Staatsbegriff, 87-89; Allgemeine Staatslelre, pp. i6-i8.
37 Problemder Souveranitdt,
pp. 204-68; Staatsbegriff,
pp. I04-5; Allgemeine Staats-
lekre, pp. II5-28; Reine Recktslelre, pp. I29-36.
I78 ETHICS

drawnfrom this basic identificationof law and the state which

culminatesin the conceptof a universallegalorderheld together
by the law of nations. He turns what is commonly taken for
specific attributions of the state into purely juridical notions
which concernthe validity and the making of the limited legal
system that in his view constitutes the state. Thus, the people,
territory,and political authority-often referredto as the ele-
ments of the state -are deemed mere externaldemarcationsof
the reach of the rules of a limited legal system. Similarly,the
variousformsof governmentaldecentralizationand union,such
as local governmentsand administrativedistricts, federations
and monarchicalunions, are all taken only to indicate different
lines of definitionof the scope of certain rules within the same
limited legal system.38 The governmentalpowersof legislation,
administration,and adjudicationare defined,in the same sense,
as mere steps or degrees in the making of law; governmental
authoritiesand agenciesare definedas the instrumentalitiesof
lawmakingat these different steps; and the various forms of
governmentare consideredas so many alternative methods of
lawmaking.39It is true that Kelsen on occasion goes beyond
these purely legal definitionsof what pertains to the state and
dealswith ideasand programsof actualpolitics, like democracy,
socialism,dictatorship,and parliamentaryrule. But he is inter-
ested primarilyin explainingtheirlegal aspects and fitting these
into the frameworkof his legalistic theory of the state and, for
the rest, revealingand describingthem as ideologiesoperating
on the psychophysicalplane of causesand effects.40
A critiqueof Kelsen'sconceptof the state must beginwith an
analysis of the objectionswhich he raises against any view of
the state as a factual phenomenon. He predicatesthese objec-
PartII:
Staatslekre,
38 Allgemeine "Centralization pp. 2I6-39.
and Decentralization,"
Part III.
Staatslehre,
39 Allgemeine

SozialismusundStaat(2d ed.;Leipzig,I923);
40 orMWesenundWertder Demokratie
(2d ed.; Tubingen,I929); Das Problemdes Parlamentarismus (Wien-Leipzig,I925);
Staatsformund Weltanschauung (Tubingen,I933); "The Party-Dictatorship,"op. cit.
 LAW AND THE STATE AS PURE IDEAS I79

tions on the assumption that the only consistent attempt at

explainingthe state in factual terms is a psychologicaltheory
of the state a theory, that is to say, which would explainthe
state as some sort of associationor combinationof individual
humanminds. He proceedsto show that any such explanation
of the state in termsof socialpsychologyis boundto fail. Point-
ing out that psychologyas such is concernedwith the individual
mind and its processes,he doubts if a psychologicalapproach
can suffice to explain interconnectionsbetween these individ-
uals, not confined to the mental processeswithin each mind,
whichwe may call social.4 But, even if subjectivementalproc-
esses could be conceived as interacting and somehow binding
individualmindsto one another,he maintainsthis could at best
serve to explaina generalfeatureof society as a whole, a char-
acteristiccommonto all associationsof individuals. It can fur-
nish no clue to the understandingof what constitutes a social
body of a particularkind, such as the state, and distinguishes
it as such from other social groupings.42
These argumentsmay, indeed, be said to contain in a nut-
shell a refutation of psychologicaltheories of the state. This
is true even if we considerthat the individualmind is endowed
with social predispositionsand abilities. The social endowment
of the individualmay serve to explainhow it is at all possible
for him to enter into relationshipswith his fellows, that is to
say, how his mindis able to influenceand react upon their men-
tal processes. Still it does not explainwhat are the grouprela-
tionships within which these influencesand reactions actually
take place. In other words, from the angle of the individual
mind, we may at best understandthe possibility and the pre-
conditionsof association. But we cannot therebydeterminethe
objective characterof any associationor the objective distinc-
tion of one associationlike the state from others of different
purport, such as a tribe, a church, a national or international
political party. Kelsen'sconclusionthat the state as such can-
pp. I2-I5.
4'Staatsbegriff, 42 Ibid., p. 8.
i8o ETHICS

not be definedin psychologicalterms is sound, whether or not

we agreewith his generalphilosophy.
Yet Kelsen's refutation of the psychological approach to
politicalthought does not disposeof any conceptionof the state
that recognizesits aspects as a factual phenomenon. It does
refute the reductionof the state to a bundle of purely causal,
that is, psychophysical,relationshipsbetween individuals. It
also disposesof the merejuxtapositionof the social community
and the objectiveidea of the state in a compounddefinition,for
instance, by territory,population, and ruling authority. Such
a definition,as Kelsen reveals, rests on the uncriticalconfusion
of divergentconcepts, using as it does both social concepts of
causal, psychophysicalcharacterand ideal concepts of regula-
tory characterA3 But his objectionsdo not reach a more com-
prehensiveview that takes social life as a total unity including
both factualacts and their idealcontents as complementaryand
mutually conditional elements. To this broader view social
reality consists in the very complementationand mutual de-
pendency of subjective acts and their objective significance,in
the very interrelationbetween the psychophysicaland the spir-
itual. Both spheresmay be distinguished,as Kelsen has shown
they must be if the nature of either is to be understood. But
they cannot be kept separate in social theory lest the way be
barredto an understandingof how the meaningof any individ-
ual's thought can provoke reactions in the thinking of others
and therebyestablishsocialrelationships. It may be suggested,
indeed, that it is the fundamentalproblemof social theory to
show, or at least to adumbrate,the coherenceand interrelation
between the causationsof individualactions and mental proc-
esses within a groupof people and the particularideas that are
common to or interlockingin their thoughts, beliefs, and emo-
43 Representativeof this "two-aspectstheory"of the state against which Kelsen's
criticismis directedis G. Jellinek,AligemeineStaatslelre(5th ed.; Berlin, I929). For
an Americanversionof the same theorysee W. W. Willoughby,FundamentalConcepts
of PublicLaw (New York, I924).
 LAW AND THE STATE AS PURE IDEAS i8i

tions and thus constitute them a social group. The state, like
any other organizationor associationof men, would then have
to be consideredas a part of that comprehensivesocial reality-
a part possessedof its own distinctivegroupideas and reactions
which politicaltheory is to identify and understand.44Kelsen's
denial of the state's factual existenceturns away from this cru-
cial problem. Far frombeingan inescapablecorollaryof his con-
clusive critiqueof psychologicaldefinitionsof the state, it truly
parallelstheir failureto interpretthe social reality of the state.
Turningto the positive elementsin Kelsen'sdefinitionof the
state, we need not add much to our commenton the definition
he gives of the law. Forin his view of the attributionsof the state
are those of law. They carry the same implications,and, con-
sequently,it is submitted that they invite the same objections.
Whetherit is called a legal system or the orderof a state, the
concept of a self-containedand logically coherentbody of com-
pulsory rules is beset with the difficultyof explainingessential
aspects of such rules. Neither their peculiar logical structure
nor their bearingupon the world of facts in which compulsion
applies nor their referenceto one another and their relation to
noncompulsoryvalues can be satisfactorilyexplainedin purely
normativeterms. These terms,then, cannot serve to explainan
effective orderof humanbehavior.
But, if the state cannot be defined at all as a body of pure
norms,thereis no point in comparingit with what also purports
to be such a body of pure norms. We have, then, no groundon
which to identify the state and the law as what is supposedto
be the same body of compulsory rules under two different
names. Nor do we have a groundon which to speculateupon
relationsof this supposedbody of rules called a state or, identi-
44See, e.g., H. Freyer,Soziologieals Wirkliclkeitswissenschaft
(Leipzig,I930), pp.
20, 79-8i; H. Heller, op. cit., pp. 8i, I02-8; Staatslehre (Leiden, I934), pp. 37-48, 69-
I25; T. Litt, Individuum und Gemeinsciaft (Leipzig, 1922), pp. I44, I69-88 (3d ed.,
I926), pp. 234-74; R. Smend, Verfassungund Verfassungsrecht
(Miinchen-Leipzig,
I928), pp. 5-2I; K. Wilk,Die Staatsformbestimmung
derWeimarer Reichsverfassung
(Ber-
lin, I932), pp. 34-35, 38-40, 43-
i82 ETHICS

cally, its legal order with the bodies of rules that may be
ascribedto other states or with those rules that are called the
law of nations. If the state cannot be conceivedas purelya sys-
tem of legal rules, it cannot, of course,be consideredas a mere
part of a comprehensivesystem of all law that would be delim-
ited by the provisionsof the law of nations.45
Thereremainsthe questionwhetherthe state and the law can
be identifiedat all, wholly or in part, even if their identity can-
not be of a purelynormativenature. Obviously,any such iden-
tificationwouldbe bound considerablyto narrowthe conceptof
the state. Many acts and processesin social life that receive
their meaningand purportfromtheirreferenceto the state can-
not be consideredas pertainingto the state's legal order. Some
of these have nothingwhateverto do with the law. Witness,for
instance, political manifestoes and ceremonialacts of govern-
ments, the use of governmentalsymbols,the providingof public
service facilities, and the exerciseof purely political discretion
by governmentalagents-not to speakof groupaction morere-
motely referringto the state though essential to the workingof
its government,such as party activities beyond the scope of
legislativeregulation,politicalnegotiations,or the expressionof
views contributingto the formationof publicopinion. Witness,
also, governmentalaction which has become fixed in more or
less definite usages of the constitution-usages not commonly
consideredlaw althoughthey might perhapsbe consideredlegal
in a wider theoreticalsense, like the president'scabinet or the
nominationof presidentialand vice-presidentialcandidatesby
national party conventionsin the United States, essential ele-
ments of the parliamentarycabinet system in Britain, and the
indissolubilityof the French Chamberunder the Third Repub-
45This does not, of course, dispose of the question as to whether and to what extent
the states' municipal laws as such are subordinate to the law of nations. Critical re-
marks on Kelsen's stand on that point may be found, e.g., in A. Verdross, Die Einheit
desrechtlichen Weltbildes(TUbingen, 1923), pp. 76-86 (from the viewpoint of a substan-
tive international legal monism), and in the papers by Jones, Kunz, and Stern (cf.
above, n. 6).
 LAWAND THE STATEAS PUREIDEAS i83

lic. Otheracts of extremeimportancefor the state are outright

unlawful,includingcoups d'etat and revolutions. It would be
pointless to stretch the concept of the law so far that all such
acts would be construed in theory as legal. For a law thus
stretchedor strainedcouldnot serve as a governingstandardof
values by which alone those acts of utmost importanceto the
body politic could be tested.
If, then, the conceptof the state were to be narrowedso as to
coincidewith the law as a governingorderof social values, how
could those legally irrelevant or unlawful acts of the state's
agents and citizens be explainedin theory? Kelsen's answer,
quite consistently,is that they can be explainedonly as purely
factual events, that is, as individualactions of the personscon-
cerned, without any referenceto the objective order of the
state.46 Yet this means that as far as those acts are concerned
the state, which provides their frameworkof reference, dis-
appearsas an objective whole transcendingthe fluctuatingpsy-
chophysicalrelationshipsbetween individuals. To this extent
the state is reduced to a bundle of such casual relationships
between individuals;which, on a largerscale, is the legitimate
object of Kelsen's own scathing criticism against the psycho-
logical theoriesof the state. It may be concludedthat an ade-
quate conceptionof the state as a comprehensivephenomenon
of social life, as the commonframeworkof acts of political and
governmentalsignificance,cannot at all be gained by identify-
ing the state and its legal order.
In any event the question of the state's relation to law, if
definedin other terms than those of purely normativeidentity
within a single and universallegal system, leads us beyond the
boundsof Kelsen'spure legal theory. This theory does, indeed,
provide a viewpoint of permanent heuristic value by distin-
guishingbetween the objective meaningof the legal orderand
its social relevancyand operationand isolatingthe two for pur-
poses of investigation. It runs into difficultiesonly when it ele-
46S taatsbegriff,pp. 96-97.
i84 ETHICS

vates those heuristicprinciplesinto absolute verities. It gives

due emphasisand freshilluminationto the independenceof the
contents of legal values and standards, as such, from their
underlyingfacts, which include above all the relationshipsof
power. Its bold technique of identifying or coordinating all
regulatory ideas with one another while divorcing them from
social facts does not resolvethe problemof how those objective
values are relatedto socialpowerand organization.But it helps
more forcefullyand clearly to restate that problem.
BROOKLYNCOLLEGE