<< . .

( 38)

. . >>

making law. In the first edition of The Bramble Bush (1930), he declared

Frank, Modern Mind, 315.
Vaihinger, Philosophy, quoted in Frank, Modern Mind, 315.
Frank, Modern Mind, 320.
Hessel E. Yntema, “The Hornbook Method and the Conflict of Laws,” Yale Law Journal
37 (February 1928): 480.
Herman Oliphant, “Stare Decisis“Continued,” American Bar Association Journal 14
(March 1928): 159.
Savigny, Jhering, Duguit, Holmes, Gray, Frank 197
quite simply: “This doing of something about disputes, this doing of it
reasonably, is the business of law. And the people who have the doing in
charge, whether they be judges or sheriffs or clerks or jailers or lawyers,
are officials of the law. What these officials do about disputes is, to my
mind, the law itself.”28 In the next edition he described these “unhappy
words” as “at best a very partial statement of the whole truth” because
they failed to take “proper account” of law as “an instrument of con-
scious shaping.”29 But how this recantation fits into his general view of
the law remained unclear.
All the Realists drew not only on the pragmatism of John Dewey, but
also on the logical positivism of Bertrand Russell, Ludwig Wittgenstein,
and the Vienna Circle when they were still trying to discover the pure
sensory core of knowledge. But the strictest juristic disciples of logical
positivism were the Scandinavian Realists, most notably Anders Lund-
stedt, Karl Olivecrona, and Alf Ross, all of whom were disciples of Axel
¨ ¨
Hagerstrom. They addressed themselves to disclosing the meaninglessness
of all legal notions, even that of legal validity, along with the logical
distinction between “is” and “ought.” If a word cannot be shown to refer
to some act or sensation, they held, it is metaphysical and meaningless.
Legal concepts have no common factual core to them. The concept of
marriage, for instance, Olivecrona argued, describes an imaginary condi-
tion because the life of every married couple is different. Even the idea of
cohabitation signifies nothing because all husbands do not live with their
wives. Legal concepts are therefore to be condemned and abjured because
they compound superstition, myth, magic, and confusion. Similarly, any
talk of the “ends” of law is nonsense because “ends” cannot be known
scientifically and are merely irrational declarations of feeling: “We like an
end or we don™t. We strive for it. We are uninterested. We strive against it.
That is all.”30 Therefore, the proper study of law has to do with relations
between cause and effect, just as in natural science, for this is the only
object of rational discourse. A practical implication of Olivecrona™s view
was drawn in a book with the ironic title of On Law and Justice (1958), by
Alf Ross, who advocates that jurists should concern themselves with

Karl N. Llewellyn, The Bramble Bush: Some Lectures on Law and Its Study (New York:
Columbia University School of Law, 1930), 3.
Karl N. Llewellyn, The Bramble Bush: On Our Law and Its Study, 2nd ed. (New York:
Oceana, 1951), 3.
Paul Sayre, ed., Interpretations of Modern Legal Philosophies (New York: Oxford
University Press, 1947), 543; quoted in Geoffrey Marshall, “Law in a Cold Climate,” The
Juridical Review (Edinburgh: Green, 1956), 261.
198 The significance of rules

“legal politics,” which would study not the ends of law, but how to make
law correspond to changing ideological conditions.31
Whatever variations they suggested, all Realists were agreed on denying
the objectivity of the law. Some spoke of law as mere acts of will, while
others went so far as to reduce law to apparent acts of will really deter-
mined by chemical or physiological processes. That the Realists drew
attention to the constant change in law constituted part of their appeal.
Even their insistence that this aspect is the whole of the law gained
credence because it seemed to do justice to what some of their predeces-
sors appear to neglect “ the element of contingency in the law. This
contribution led jurists who were not of the “Realist” persuasion, such
as Benjamin Cardozo and Roscoe Pound, to praise Realism for voicing a
useful protest against the dogma that the law consists of rules applied
with mechanical certainty.
Yet for all their insistence on the value of uncertainty, the Realists
smuggled a new kind of monarch onto the throne they affected to despise.
In one fashion or another, all Realists advocated that everything done in
the name of “law” should be ruled by “real social needs.” These needs,
they assumed, as Jhering said explicitly, can be known with certainty
whether by direct intuition or through scientific expertise. From decisions
based on this new myth about “social needs,” the Realists, unlike their
timid ancient and medieval predecessors, promised consequences whose
desirability was certain.
The most curious thing about the Realists is their belief that they were
destroying an ancient myth fostered by medieval theologians. As we have
seen, the recognition that there is an irreducible uncertainty about how to
determine what rules mean in particular circumstances is one of the
dominant themes in the ancient and medieval discussion of law. But the
ancient and medieval writers did not conclude from this that it is impos-
sible to have fixed rules of law. They distinguished, as the Realists did not,
between the different kinds of fixity possible at different levels of abstrac-
tion, and between theoretical and practical reasoning. The true object of
the Realist attack was a modern myth promoted by Kant, that there is
only one kind of rational activity, consisting of demonstrative reasoning
from universally valid laws, which yields indisputable conclusions. This
claim implies that all reasoning has the character of scientific reasoning,
and it permits only two conclusions about the law: that either the law is
absolutely certain, or else it is irrational and wholly uncertain. Having

Alf Ross, On Law and Justice (London: Stevens, 1958), 327; See also H. L. A. Hart,
“Scandinavian Realism,” Cambridge Law Journal (November 1959): 233“40.
Savigny, Jhering, Duguit, Holmes, Gray, Frank 199
discovered, as ancient and medieval philosophers had before them, that
rules of law cannot yield absolutely certain decisions, the Realists felt
obliged to deny that legal rules determine legal outcomes. They accord-
ingly rejected the traditional idea that the law consists of fixed rules that
protect its subjects against arbitrary decisions, and they dismissed the
objectivity of legal decisions as an illusion. In effect, the Realists reduced
law to a miscellaneous source of materials on which judges draw in no
regular fashion. Thus, in the1930s when the Realists™ influence was at its
height, the nature of adjudication became the central preoccupation of the
philosophy of law.
11 The defense of rules: Edward Levi, Hans
Kelsen, H. L. A. Hart

Edward H. Levi
Although ancient and medieval theorists of law took it for granted, as we
have seen, that human law cannot produce certainty, only recently has the
relationship between the logic of the law and its capacity to combine
stability with change and uncertainty been spelled out. In the classic
modern work on the subject, An Introduction to Legal Reasoning (1948),
Edward Levi says bluntly that the law operates under a pretense: “The
pretence is that the law is a system of known rules applied by a judge.”1
He goes on to explain that it is a pretense because no rule of law can
absolutely specify a decision. This gap between general rules and particu-
lar decisions means that rules of law are always ambiguous. The ambigu-
ity is inescapable because of the logical relationship between any general
proposition and a more particular one. Any given particular can be fitted
into a variety of general statements, and the particulars that could be
implied by a general statement cannot be exhaustively stated. Thus, the
character of the logical relationship between a general rule and a particu-
lar instance makes it impossible for any general rule, however clear, to
yield only one correct decision.
Only with general statements of a perfectly abstract nature, as with
figures the sum of whose angles is 180 degrees, can the particulars fitting
under it be unambiguously identified. In the law, the rules are general but
not abstract. To identify the particular instances that fit under a law, we
must always abstract from a cluster of attributes. And while we can and
do make rules for a proper manner of abstracting, any attempt to elimin-
ate all uncertainty would lead us into an infinite regress of rules. But this
ambiguity in law, Levi points out, also contributes to the stability of the
law. Where laws are made by an assembly of men or must be acceptable
to more than one person and at more than one instance, a degree of

Edward H. Levi, An Introduction to Legal Reasoning (Chicago: University of Chicago
Press, 1970), 1.

Edward Levi, Hans Kelsen, H. L. A. Hart 201
ambiguity makes agreement and acceptance easier. Differences of view, at
least within certain areas, can be reconciled under ambiguous words, and
the ambiguity can then be left for the court to resolve as each case arises.
Such ambiguity is indispensable for civil peace.
Moreover, the ambiguity in laws makes it possible for the law to remain
stable while adapting to changing circumstances. Not only do new situ-
ations arise, but peoples™ wants also change: “The categories used in the
legal process must be left ambiguous in order to permit the infusion of
new ideas. And this is true even where legislation or a constitution is
involved. The words used by the legislature or the constitutional conven-
tion must come to have new meanings. Furthermore, agreement on any
other basis would be impossible.”2
What makes this flexibility possible without destroying the stability of
the law is reasoning by example, or analogical reasoning. A legal decision
can be formulated in a syllogism, and the syllogism is not a deception, but
the premises are never self-evident. The problems in legal reasoning arise
out of the difficulty in discovering the premises of the syllogism that are
designed to conclude in a decision. The major premise is in question until
it is decided what rule of law is relevant. If there is agreement about that,
as there may be from the outset in some cases, the court still has to
establish whether the action in dispute is of the sort designated by the
rule or how this case compares with other cases that have come under
the rule.
Even a simple speed law can spawn many awkward questions, and
a law forbidding “combinations in restraint of trade” provides an inex-
haustible subject for argument. Moreover, once a number of cases have
been brought under a statute, the judge is faced with the problem of
precedent. He must decide whether the case before him most resembles
this or that case in the past if he is to see the law as a consistent whole. In
deciding such questions, the judge and the lawyers arguing before him
must proceed by analogical reasoning in order to establish the minor
premise of the syllogism. The question constantly being answered is:
When is it reasonable to treat different cases as though they were the
same? This problem is not, as Bentham said, peculiar to the common law.
Nor can it be resolved by any definition of precedent. A precedent is
binding on a subsequent case because the court finds the cases to be
similar. But as no two cases are ever identical, what constitutes a prece-
dent and how a rule applies to this case are not known until the court has
pronounced. This is what makes the Realists sound plausible when they
say that the judge cannot “find” the law.

Ibid., 4.
202 The significance of rules

According to Levi, however, it does not follow that the judge™s decision
is arbitrary or to be explained only by some irrational cause, as the
Realists suppose. The judge arrives at his decision by reasoning; his
decision has reasons, not causes. And the system of law dictates that
only certain kinds of reasons are appropriate. The judge ought not to
ask himself: “Which of the parties is handsomer, more virtuous, has
suffered most, will pay me more?” Or: “What decision will strengthen
the government?” Or: “Whom can I make happier?” The questions ap-
propriate for a judge to consider are of the order of: “What rule applies to
this case? How does this case compare with other cases like it? Which of
the analogies presented before me is more persuasive?” In answering such
questions, a judge will be concerned only with applying the law as it exists
in statutes and precedents, and he does this by considering whether his
decision is consistent with past decisions. His concern with preserving the
law will inhibit him from doing justice in this case at the cost of destroying
the consistency of the law.
It is therefore just as misleading to think that the judge “makes” the law
as that he “finds” it. He does not make the law in the sense of giving effect
to his will or to arbitrarily selected requirements. What issues from the
court is a decision, not a personal fiat. The judge supports his decision
with a legal argument; he may not say, “I think him so because I think
him so.”
Those who insist that these arguments are spurious and look for some
“real” cause for judicial decisions, as Realist jurists do, are exhibiting a
fallacy that is endemic today, the fallacy of supposing that there are only
three alternatives: certainty, science, and irrationality. Either we must
arrive at an indubitable answer of the sort possible within Euclidean
geometry, or justify our answer in terms of an established scientific
theory, or else we must, in this fallacious view, fall into irrationality. If
the reasoning of the judge is neither certainly infallible nor scientifically
valid, it must be condemned as illusory.
But rational activity is not confined to deductive reasoning from
indubitable premises such as mathematical constructs or from universally
acceptable premises such as scientific theories. Analogical reasoning
cannot, however, produce certainty or even the degree of consensus
generally found among scientists in stable areas of research; it can only
persuade a given audience that one conclusion is more “reasonable”
than another. What is at issue in any case, even if settled easily, is
deciding what constitutes the “facts.” An appropriate criticism will
offer reasons for considering one set of “facts” more reasonable or
relevant than another possible construction. But no construction can be
Edward Levi, Hans Kelsen, H. L. A. Hart 203
Because analogical reasoning cannot reach a conclusion from which no
man could rationally dissent, this uncertainty is often confused with
irrationality, thereby giving credence to Realist theories of law. But the
uncertainty in analogical reasoning does not render its rationality spuri-
ous. A decision is rational because it has reasoned antecedents, not social,
psychological, or physiological causes. There is an order appropriate to
the reasoning and impersonal criteria for judging it. It is therefore rational
though not certainly correct. What emotions the judge may have experi-
enced before or during the trial, or in what order the various ideas in his
opinion occurred to him should worry his biographer, not the jurist. What
matters for the law is that other reasonable men conversant with the
statutes, decisions, and procedures of the law should find his decision
and reasoning plausible even if they themselves might have reached
different conclusions.
The dependence of law on analogical reasoning also explains, Levi
points out, what the Realists assert to be desirable and possible, though
they cannot show how it comes about “ the relation between the law
and the moral convictions of the governed. For when a judge finds
some analogies more reasonable than others, however objectively he tries
to see analogies, he must see them with his own eyes. And a judge, like all
other people, has inherited beliefs, acquired convictions, an outlook on
life, in short, a “mental background.” His judgment always incorporates
assumptions about what constitutes a reasonable interpretation of certain
words or what constitutes reasonable conduct in a reasonable man. His
decision necessarily incorporates judgments about whether conduct is
negligent or responsible, malicious or well-intentioned, reasonable or
unreasonable. In other words, the judge decides what accords with the
law not in a moral vacuum, but in accordance with his understanding of
moral conduct, in terms of which he necessarily interprets the words of
the law and understands the arguments presented to him. Though the
judge is not meant to be and should not be a one man public opinion poll
or a student of public opinion, he is obliged to exercise his discretion in
the manner of a reasonable man of his time, and this implies that he
reflects unselfconsciously about the morality of his time. But he does so in
the course of attempting to apply the law as it is. The notion that the
judge “finds” the law, while incorrect as a description of the law, is a
useful practical maxim that directs the judge to keep his attention fixed on
the appropriate considerations.
Once the character of analogical reasoning is recognized, it becomes
obvious, Levi says, that a controversy with the Realists about “whether
the law is certain, unchanging, and expressed in rules, or uncertain,
changing, and only a technique for deciding specific cases misses the
204 The significance of rules

point. It is both. Nor is it helpful to dispose of the process as a wonderful
mystery possibly reflecting a higher law, by which the law can remain the
same and yet change. The law forum is the most explicit demonstration of
the mechanism required for a moving classification system.”3 The Realists
denied this because they equated rationality with certainty and demon-
strative reasoning and failed to grasp the importance of analogical
reasoning for maintaining a legal system.

Hans Kelsen
A different kind of response to the preoccupations of the Realists came
from Hans Kelsen and H. L. A. Hart, who are known as “Positivists” and
who focused on revising the theories of Kant and Austin. Unlike the
Realists, they retained, indeed emphasized, the traditional view of law
as a set of stable rules that provide the basis for objective decisions by
judges. Kelsen and Hart offered new ways of understanding the character
of a legal system, which they believed could explain the uncertainty
and change in the law so as to establish that the objectivity of the law is
not an illusion.
Kelsen regarded legal Realism as a variety of sociological jurisprudence
that denied the fundamental difference between the natural and the
human world by being concerned only with facts and causes, rather than
with norms and the “meanings of acts of will.” Sociological jurists ac-
cused his “Pure Theory” of law of being “ideological,” Kelsen explains,
because it concerns itself with legal norms rather than with legal acts and
the causes of legal acts. Those who regard that as “ideological” assume
that whatever is not natural reality or a description of it is “ideology.”
If, however, ideology is used correctly to refer to a “nonobjective presen-
tation of the subject influenced by subjective value judgements” that
disfigure “the subject of cognition,” then the Pure Theory of law can be
acquitted of being ideological, for it is uncontaminated by any illicit
considerations of justice. It aims only to discover the essence of law by
analyzing its structure: “The Pure Theory desires to present the law as it
is, not as it ought to be; it seeks to know the real and possible, not the
˜ideal,™ the ˜right™ law. In this sense, the Pure Theory is a radical realistic
theory of law, that is, a theory of legal positivism.”4
Kelsen was no more enamored with traditional jurisprudence than the
Realists were, but he criticized traditional jurisprudence for another

Ibid., 4.
Hans Kelsen, Pure Theory of Law, trans. Max Knight (Berkeley and Los Angeles:
University of California Press, 1967), 102“06.
Edward Levi, Hans Kelsen, H. L. A. Hart 205
defect, i.e., for unconsciously serving as an ideology to justify or oppose
an existing social order. All ideology, he said, is rooted in “Wishing”
rather than “Knowing,” that is, in “subjective values” and interests other
than the pursuit of truth.5 It is therefore concerned with either preserving
or attacking reality, and in order to achieve its end it will either glorify or
denigrate reality, but will never present it undistorted. The Pure Theory of
law, Kelsen is confident, escapes any such danger by denying all “value”
to positive law. He believes it to be a true science of jurisprudence with the
same object as the social science of ethics. Both, he explains, are norma-
tive sciences, not because they prescribe norms for human behavior and
thereby command, authorize, or positively permit a certain kind of con-
duct, but because they describe certain man-made norms and the rela-
tionship between men that is thereby created.6 By describing norms,
Kelsen means analyzing the logical relationship between the rules
governing men without any reference to the purposes that they might be
designed to serve, other than peace or survival, which are biologically
given, not man-made, ends. This exclusion of any relation between law
and the purpose that it might be expected to serve removes any possibility
that positive law as it appears in Pure Theory can either conform to or
violate an ideal law. Therefore, Kelsen argues, neither supporters nor
critics of the existing order could find a weapon in Pure Theory.
The objectivity of Pure Theory comes by recognizing, Kelsen says, that
justice is merely an irrational ideal. It may be an ideal that men need in
order to will and to act, but it is not a subject of “rational cognition.” The
latter knows only “interests” and “conflicts of interests.” These may be
satisfied by sacrificing some to others, but that one alternative is “just”
cannot be established by rational cognition. If, however, justice is thought
of as neutrality, it might be admissible. It would then require only that
the government remain indifferent to the various conflicting interests
and aim to bring about a compromise between them. Such neutrality,
Kelsen argues, has the objective justification that “only a legal order. . .
which brings about such a compromise between opposing interests as
to minimize the possible frictions has expectation of relatively enduring
existence. . . [and can] secure social peace on a relatively permanent
But when justice is used in the ordinary sense, it signifies nothing
but a rationalization of personal preferences pretending to a spurious
universality, or what Kelsen calls “subjective judgment of value.” A norm

5 6
Ibid., 106. Ibid., 86.
Hans Kelsen, General Theory of Law and State, trans. Anders Wedberg (Cambridge, MA:
Harvard University Press, 1945), 13“14, 49.
206 The significance of rules

can be “just” or “unjust” only for those who do or do not desire what the
norm prescribes, although whoever pronounces a social institution to be
just or unjust may be unaware that he or she is expressing a merely private
interest. This is especially obvious in appeals to “natural law,” which
pretend that the norm of justice is immanent in nature, thus displaying
a typical illusion due to an “objectivization of subjective interests.”8 Even
the claim that men have a natural right to be free or treated equally is
really a self-deception “or “ what amounts to much the same thing “ an
ideology,” because it pretends that a subjective judgment of value pro-
ceeds from some indubitable source such as “nature.”9
The only correct sense of justice in Kelsen™s terms is “legality.” It is
therefore “just” for a general rule to be actually applied in all cases where,
according to its content, this rule should be applied.”10 Used in this way,
justice is related not to the “content of a positive order, but to its
application,” and it is compatible with any legal order, capitalist or
communist, democratic or autocratic. A statement that the behavior of

<< . .

( 38)

. . >>

Copyright Design by: Sunlight webdesign