Journal of Philosophy, Inc.

The Function of General Laws in History

Author(s): Carl G. Hempel
Source: The Journal of Philosophy, Vol. 39, No. 2 (Jan. 15, 1942), pp. 35-48
Published by: Journal of Philosophy, Inc.
Stable URL: http://www.jstor.org/stable/2017635 .
Accessed: 10/11/2014 16:51

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
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].

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

http://www.jstor.org

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 35

vestigation might lead us. It is, however, pertinent to say that

much more in the way of positive results has already been attained
than is indicated anywhere in this article.
JOHN DEWEY.
COLUMBIA UNIVERSITY.

THE FUNCTION OF GENERAL LAWS IN HISTORY

1. It is a rather widely held opinion that history, in contra-
distinction to the so-called physical sciences, is concerned with the
description of particular events of the past rather than with the
search for general laws which might govern those events. As a
characterization of the type of problem in which some historians
are mainly interested, this view probably can not be denied; as a
statement of the theoretical function of general laws in scientific
historical research, it is certainly unacceptable. The following con-
siderations are an attempt to substantiate this point by showing in
some detail that general laws have quite analogous functions in
history and in the natural sciences, that they form an indispensable
instrument of historical research, and that they even constitute the
common basis of various procedures which are often considered as
characteristic of the social in contradistinction to the natural
sciences.
By a general law, we shall here understand a statement of uni-
versal conditional form which is capable of being confirmed or
disconfirmed by suitable empirical findings. The term "law" sug-
gests the idea that the statement in question is actually well con-
firmed by the relevant evidence available; as this qualification is,
in many cases, irrelevant for our purpose, we shall frequently use
the term "hypothesis of universal form " or briefly "universal
hypothesis" instead of "general law," and state the condition of
satisfactory confirmation separately, if necessary. In the context
of this paper, a universal hypothesis may be assumed to assert a
regularity of the following type: In every case where an event of
a specified kind C occurs at a certain place and time, an event of a
specified kind E will occur at a place and time which is related in
a specified manner to the place and time of the occurrence of the
first event. (The symbols "C" and "E" have been chosen to sug-
gest the terms "cause" and "effect," which are often, though by
no means always, applied to events related by a law of the above
kind.)
2.1 The main function of general laws in the natural sciences is
to connect events in patterns which are usually referred to as ex-
planation and prediction.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
36 THE JOURNAL OF PHILOSOPHY

The explanation of the occurrence of an event of some specific

kind E at a certain place and time consists, as it is usually expressed,
in indicating the causes or determining factors of E. Now the as-
sertion that a set of events-say, of the kinds C1, C., . . ., C.-
have caused the event to be explained, amounts to the statement
that, according to certain general laws, a set of events of the kinds
rmentionedis regularly accompanied by an event of kind E. Thus,
the scientific explanation of the event in question consists of
(1) a set of statements asserting the occurrence of certain events
C1, . . . C, at certain times and places,
(2) a set of universal hypotheses, such that
(a) the statements of both groups are reasonably well con-
firmed by empirical evidence,
(b) from the two groups of statements the sentence asserting
the occurrence of event E can be logically deduced.
In a physical explanation, group (1) would describe the initial
and boundary conditions for the occurrence of the final event;
generally, we shall say that group (1) states the determining con-
ditions for the event to be explained, while group (2) contains the
general laws on which the explanation is based; they imply the
statement that whenever events of the kind described in the first
group occur, an event of the kind to be explained will take place.
Illustration: Let the event to be explained consist in the cracking of an
automobile radiator during a cold night. The sentences of group (1) may
state the following initial and boundary conditions: The car was left in the
street all night. Its radiator, which consists of iron, was completely filled
with water, and the lid was screwed on tightly. The temperature during the
night dropped from 390 F. in the evening to 250 P. in the morning; the air
pressure was normal. The bursting pressure of the radiator material is so
and so much.-Group (2) would contain empirical laws such as the following:
Below 32? F., under normal atmospheric pressure, water freezes. Below 39.2?
F., the pressure of a mass of water increases with decreasing temperature, if
the volume remains constant or decreases; when the water freezes, the pressure
again increases. Finally, this group would have to include a quantitative law
concerning the change of pressure of water as a function of its temperature
and volume.
From statements of these two kinds, the conclusion that the radiator
cracked during the night can be deduced by logical reasoning; an explanation
of the considered event has been established.

2.2 It is important to bear in mind that the symbols "E," "C,"

"tCe,''"C2,y" etc., which were used above, stand for kinds or prop-
erties of events, not for what is sometimes called individual events.
For the object of description and explanation in every branch of
empirical science is always the occurrence of an event of a certain
kind (such as a drop in temperature by 140 F., an eclipse of the
moon, a cell-division, an earthquake, an increase in employment, a

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 37

political assassination) at a given place and time, or in a given

empirical object (such as the radiator of a certain car, the planetary
system, a specified historical personality, etc.) at a certain time.
What is sometimes called the complete description of an indi-
vidual event (such as the earthquake of San Francisco in 1906 or
the assassination of Julius Caesar) would require a statement of
all the properties exhibited by the spatial region or the individual
object involved, for the period of time occupied by the event in
question. Such a task can never be completely accomplished.
A fortiori, it is impossible to explain an individual event in the
sense of accounting for all its characteristics by means of universal
hypotheses, although the explanation of what happened at a speci-
fied place and time may gradually be made more and more specific
and comprehensive.
But there is no difference, in this respect, between history and
the natural sciences: both can give an account of their subject-
matter only in terms of general concepts, and history can "grasp
the unique individuality" of its objects of study no more and no
less than can physics or chemistry.
3. The following points result more or less directly from the
above study of scientific explanation and are of special importance
for the questions here to be discussed.
3.1 A set of events can be said to have caused the event to be
explained only if general laws can be inidicated which connect
"(auses" and "effect" in the manner characterized above.
3.2 No matter whether the cause-effect terminology is used or
not, a scientific explanation has been achieved only if empirical laws
of the kind mentioned under (2) in 2.1 have been applied.1
3.3 The use of universal empirical hypotheses as explanatory
principles distinguishes genuine from pseudo-explanation, such as,
say, the attempt to account for certain features of organic behavior
by reference to an entelechy, for whose functioning no laws are of-
fered, or the explanation of the achievements of a given person in
terms of his "mission in history," his "predestined fate," or simi-
1 Maurice Mandelbaum, in his generally very clarifying analysis of rele-
vance and causation in history (The Problem of Historical Knowledge, New
York, 1938, Chs. 7, 8) seems to hold that there is a difference between the
'causal analysis ' or ' causal explanation ' of an event and the establish-
ment of scientific laws governing it in the sense stated above. He argues
that "scientific laws can only be formulated on the basis of causal analysis,"
but that "they are not substitutes for full causal explanations" (l.c., p. 238).
For the reasons outlined above, this distinction does not appear to be justified:
every "'causal explanation"I is an ''explanation by scientific laws" ; for in
no other way than by reference to empirical laws can the assertion of a causal
connection between certain events be scientifically substantiated.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
38 THE JOURNAL OF PHILOSOPHY

lar notions. Accounts of this type are based on metaphors rather

than laws; they convey pictorial and emotional appeals instead of
insight into factual connections; they substitute vague analogies and
intuitive "plausibility" for deduction from testable statements and
are therefore unacceptable as scientific explanations.
Any explanation of scientific character is amenable to objective
checks; these include
(a) an empirical test of the sentences which state the determin-
ing conditions;
(b) an empirical test of the universal hypotheses on which the
explanation rests;
(c) an investigation of whether the explanation is logically con-
clusive in the sense that the sentence describing the event to
be explained follows from the statements of groups (1) and
(2).
4. The function of general laws in scientific prediction can now
be stated very briefly. Quite generally, prediction in empirical
science consists in deriving a statement about a certain future
event (for example, the relative position of the planets to the sun,
at a future date) from (1) statements describing certain known
(past or present) conditions (for example, the positions and mo-
menta of the planets at a past or present moment), and (2) suitable
general laws (for example, the laws of celestial mechanics). Thus,
the logical structure of a scientific prediction is the same as that of
a scientific explanation, which has been described in 2.1. In par-
ticular, prediction no less than explanation throughout empirical
science involves reference to universal empirical hypotheses.
The customary distinction between explanation and prediction
rests mainly on a pragmatical difference between the two: While
in the case of an explanation, the final event is known to have hap-
pened, and its determining conditions have to be sought, the situa-
tion is reversed in the case of a prediction: here, the initial condi-
tions are given, and their "effect" -which, in the typical case, has
not yet taken place-is to be determined.
In view of the structural equality of explanation and prediction,
it may be said that an explanation as characterized in 2.1 is not
complete unless it might as well have functioned as a prediction:
If the final event can be derived from the initial conditions and
universal hypotheses stated in the explanation, then it might as well
have been predicted, before it actually happened, on the basis of
a knowledge of the initial conditions and the general laws. Thus,
e.g., those initial conditions and general laws which the astronomer
would adduce in explanation of a certain eclipse of the sun are

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 39

such that they might also have served as a sufficient basis for a
forecast of the eclipse before it took place.
However, only rarely, if ever, are explanations stated so com-
pletely as to exhibit this predictive character (which the test re-
ferred to under (c) in 3.3 would serve to reveal). Quite com-
monly, the explanation offered for the occurrence of an event is
incomplete. Thus, we may hear the explanation that a barn burnt
down "because" a burning cigarette was dropped in the hay, or
that a certain political movement has spectacular success "because"
it takes advantage of widespread racial prejudices. Similarly, in
the case of the broken radiator, the customary way of formulating
an explanation would be restricted to pointing out that the car was
left in the cold, and the radiator was filled with water.-In ex-
planatory statements like these, the general laws which confer upon
the stated conditions the character of "causes" or "determining
factors" are completely omitted (sometimes, perhaps, as a "matter
of course"), and, furthermore, the enumeration of the determining
conditions of group (1) is incomplete; this is illustrated by the
preceding examples, but even by the earlier analysis of the broken-
radiator case: as a closer examination would reveal, even that much
more detailed statement of determining conditions and universal
hypotheses would require amplification in order to serve as a suffi-
cient basis for the deduction of the conclusion that the radiator
broke during the night.
In some instances, the incompleteniessof a given explanation may
be considered as inessential. Thus, e.g., we may feel that the ex-
planation referred to in the last example could be made complete
if we so desired; for we have reasons to assume that we know the
kind of determining conditions and of general laws which are rele-
vant in this context.
Very frequently, however, we encounter "explanations" whose
incompleteness can not simply be dismissed as inessential. The
methodological consequences of this situation will be discussed later
(especially in 5.3 and 5.4).
5.1 The preceding considerations apply to explanation in history
as well as in any other branch of empirical science. Historical ex-
planation, too, aims at showing that the event in question was not
"a matter of chance," but was to be expected in view of certain
antecedent or simultaneous conditions. The expectation referred
to is not prophecy or divination, but rational scientific anticipation
which rests on the assumption of general laws.
If this view is correct, it would seem strange that while most
historians do suggest explanations of historical evenits, many of
them deny the possibility of resorting to any general laws in history.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
40 THE JOURNAL OF PHILOSOPHY

It is, however, possible to account for this situation by a closer

study of explanation in history, as may become clear in the course
of the following analysis.
5.2 In some cases, the uniiversal hypotheses underlying a his-
torical explanation are rather explicitly stated, as is illustrated by
the italicized passages in the following attempt to explain the
tendency of government agencies to perpetuate themselves and to
expand (italics the author's):
As the activities of the government are enlarged, more people develop a
vested interest in the continuation and expansion of governmental functiolls.
People who have jobs do not like to lose them; those who are habituated to
certain skills do not welcome change; those who have become accustomaed to
the exercise of a certain kind of power do not like to relinquish their control-
if anything, they want to develop greater power and correspondingly greater
prestige.
Thus, government offices and bureaus, once created, in turn institute drives,
not only to fortify themselves against assault, but to enlarge the scope of their
operations.2

Most explanations offered in history or sociology, however, fail

to include an explicit statement of the general regularities they
presuppose; and there seem to be at least two reasons which account
for this:
First, the universal hypotheses in question frequently relate to
individual or social psychology, which somehow is supposed to be
familiar to everybody through his everyday experience; thus, they
are tacitly taken for granted. This is a situation quite similar to
that characterized in 4.
Second, it would often be very difficult to formulate the under-
lying assumptions explicitly with sufficient precision and at the
same time in such a way that they are in agreement with all the
relevant empirical evidence available. It is highly instructive, in
examining the adequacy of a suggested explanation, to attempt a
reconstruction of the universal hypotheses on which it rests. Par-
ticularly, such terms as "hence, "therefore," "consequently,"
"because," "naturally," "obviously," etc., are often indicative of
the tacit presupposition of some general law: they are used to tie
up the initial conditions with the event to be explained; but that
the latter was "naturally " to be expected as "a consequence " of
the stated conditions follows only if suitable general laws are pre-
supposed. Consider, for example, the statement that the Dust
Bowl farmers migrate to California "because" continual drought
and sandstorms render their existence increasingly precarious,
2 Donald W. McConnell, Economic Behavior; New York, 1939; pp. 894-
895.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 41

and because California seems to them to offer so much better liv-

ing conditions. This explanation rests on some such universal
hypothesis as that populations will tend to migrate to regions which
offer better living conditions. But it would obviously be difficult
accurately to state this hypothesis in the form of a general law
which is reasonably well confirmed by all the relevant evidence
available. Similarly, if a particular revolution is explained by
reference to the growing discontent, on the part of a large part of
the population, with certain prevailing conditions, it is clear that
a general regularity is assumed in this explanation, but we are
hardly in a position to state just what extent and what specific
form the discontenit has to assume, and what the environmental
conditions have to be, to bring about a revolution. Analogous re-
marks apply to all historical explanations in terms of class strug-
gle, economic or geographic conditions, vested interests of certain
groups, tendency to conspicuous consumption, etc.: All of them rest
on the assumption of universal hypotheses 3 which connect certain
characteristics of individual or group life with others; but in many
cases, the content of the hypotheses which are tacitly assumed in a
given explanation can be reconstructed only quite approximately.
5.3 It might be argued that the phenomena covered by the type
of explanation just mentioned are of a statistical character, and
that therefore only probability hypotheses need to be assumed in
their explanation, so that the question as to the "underlying gen-
eral laws" would be based oln a false premise. And indeed, it
seems possible and justifiable to construe certain explanations
offered in history as based on the assumptioln of probability hy-
potheses rather than of general "deterministic" laws, i.e., laws in
the form of universal conditionals. This claim may be extended
to many of the explanations offered in other fields of empirical
science as well. Thus, e.g., if Tommy comes down with the measles
two weeks after his brother, alnd if he has not been in the com-
pany of other persons having the measles, we accept the explana-
tion that he caught the disease from his brother. Now, there is a
general hypothesis underlying this explanation; but it can hardly
be said to be a general law to the effect that any person who has
not had the measles before will get them without fail if he stays in
the company of somebody else who has the measles; that a con-
tagioni will occur can be asserted only with a high probability.
3 What is sometimes, misleadingly, called an explanation by means of a
certain concept is, in empirical science, actually an explanation in terms of
universal hypotheses containing that concept. "Explanations" involving con-
cepts which do not function in empirically testable hypotheses-such as
"entelechy" in biology, "historic destination of a race" or "self-unfolding
of absolute reason" in history-are mere metaphors without cognitive content.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
42 THE JOURNAL OF PHILOSOPHY

Many an explanation offered in history seems to admit of an

analysis of this kind: if fully and explicitly formulated, it would
state certain initial conditions, and certain probability hypotheses,4
such that the occurrence of the event to be explained is made highly
probable by the initial conditions in view of the probability hy-
potheses. But no matter whether explanations in history be con-
strued as "causal" or as "probabilistic" in character, it remains
true that in general the initial conditions and especially the uni-
versal hypotheses involved are not clearly indicated, and can not
unambiguously be supplemented. (In the case of probability
hypotheses, for example, the probability values involved will at
best be known quite roughly.)
5.4 What the explanatory analyses of historical events offer is,
then, in most cases not an explanation in one of the meanings de-
veloped above, but something that might be called an explanation
sketch. Such a sketch consists of a more or less vague indication
of the laws and initial conditions considered as relevant, and it
needs "filling out" in order to turn into a full-fledged explana-
tion. This filling-out requires further empirical research, for
which the sketch suggests the direction. (Explanation sketches
are common also outside of history; many explanations in psycho-
analysis, for instance, illustrate this point.)
Obviously, an explanation sketch does not admit of an empirical
test to the same extent as does a complete explanation; and yet, there
is a difference between a scientifically acceptable explanation sketch
and a pseudo-explanation (or a pseudo-explanation sketch). A
scientifically acceptable explanation sketch needs to be filled out
by more specific statements; but it points into the direction where
these statements are to be found; and concrete research may tend to
confirm or to infirm those indications; i.e., it may show that the kind
of initial conditions suggested are actually relevant; or it may
reveal that factors of a quite different nature have to be taken into
account in order to arrive at a satisfactory explanation.-The filling-
out process required by an explanation sketch will, in general, as-
sume the form of a gradually increasing precision of the formula-
tions involved; but at any stage of this process, those formulations
will have some empirical import: it will be possible to indicate, at
least roughly, what kind of evidence would be relevant in testing
4 E. Zilsel, in a very stimulating paper on ''Physics and the Problem of
Historico-Sociological Laws " (Philosophy of Science, Vol. 8, 1941, pp. 567-
579), suggests that all specifically historical laws are of a statistical char-
acter similar to that of the " macro-laws " in physics. The above remarks,
however, are inot restricted to specifically historical laws since explanation in
history rests to a large extent on non-historical laws (cf. section 8 of this
paper).

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 43

them, and what findings would tend to confirm them. In the case
of non-empirical explanations or explanation sketches, on the other
hand-say, by reference to the historical destination of a certain
race, or to a principle of historical justice-the use of empirically
meaningless terms makes it impossible even roughly to indicate the
type of investigation that would have a bearing upon those formula-
tions, and that might lead to evidence either confirming or infirming
the suggested explanation.
5.5 In trying to appraise the soundness of a given explanation,
one will first have to attempt to reconstruct as completely as possible
the argument constituting the explanation or the explanation sketch.
In particular, it is important to realize what the underlying explain-
ing hypotheses are, and to judge of their scope and empirical foun-
dation. A resuscitation of the assumptions buried under the grave-
stones "hence," therefore," because," and the like will often re-
veal that the explanation offered is poorly founded or downright
unacceptable. In many cases, this procedure will bring to light the
fallacy of claiming that a large number of details of an event have
been explained when, even on a very liberal interpretation, only
some broad characteristics of it have been accounted for. Thus, for
example, the geographic or economic conditions under which a
group lives may account for certain general features of, say, its
art or its moral codes; but to grant this does not mean that the
artistic achievements of the group or its system of morals has thus
been explained in detail; for this would imply that from a descrip-
tion of the prevalent geographic or economic conditions alone, a
detailed account of certain aspects of the cultural life of the group
can be deduced by means of specifiable general laws.
A related error consists in singling out one of several important
groups of factors which would have to be stated in the initial con-
ditions, and then claiming that the phenomenon in question is "de-
termined" by and thus can be explained in terms of that one group
of factors.
Occasionally, the adherents of some particular school of ex-
planation or interpretation in history will adduce, as evidence in
favor of their approach, a successful historical prediction which was
made by a representative of their school. But though the predictive
success of a theory is certainly relevant evidence of its soundness,
it is important to make sure that the successful prediction is in fact
obtainable by means of the theory in question. It happens some-
times that the prediction is actually an ingenious guess which may
have been influenced by the theoretical outlook of its author, but
which can not be arrived at by means of his theory alone. Thus, an
adherent of a quite metaphysical "theory" of history may have a

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
44 THE JOURNAL OF PHILOSOPHY

sound feeling for historical developments and may be able to make

correct predictions, which he will even couch in the terminology of
his theory, though they could not have been attained by means of
it. To guard against such pseudo-confirming cases would be one
of the functions of test (c) in 3.3.
6. We have tried to show that in history no less than in any
other branch of empirical inquiry, scientific explanation can be
achieved only by means of suitable general hypotheses, or by the-
ories, which are bodies of systematically related hypotheses. This
thesis is clearly in contrast with the familiar view that genuine
explanation in history is obtained by a method which character-
istically distinguishes the social from the natural sciences, namely,
the method of empathetic understanding: The historian, we are
told, imagines himself in the place of the persons involved in the
events which he wants to explain; he tries to realize as completely
as possible the circumstances under which they acted, and the mo-
tives which influenced their actions; and by this imaginary self-
identification with his heroes, he arrives at an understanding and
thus at an adequate explanation of the events with which he is
concerned.
This method of empathy is, no doubt, frequently applied by lay-
men and by experts in history. But it does not in itself constitute
an explanation; it rather is essentially a heuristic device; its func-
tion is to suggest certain psychological hypotheses which might
serve as explanatory principles in the case under consideration.
Stated in crude terms, the idea underlying this function is the fol-
lowing: The historian tries to realize how he himself would act
under the given conditions, and under the particular motivations
of his heroes; he tentatively generalizes his findings into a general
rule and uses the latter as an explanatory principle in accounting
for the actions of the persons involved. Now, this procedure may
sometimes prove heuristically helpful; but its use does not guarantee
the soundness of the historical explanation to which it leads. The
latter rather depends upon the factual correctness of the empirical
generalizations which the method of understanding may have sug-
gested.
Nor is the use of this method indispensable for historical ex-
planation. A historian may, for example, be incapable of feeling
himself into the role of a paranoiac historic personality, and yet he
may well be able to explain certain of his actions; notably by ref-
erence to the principles of abnormal psychology. Thus, whether
the historian is or is not in a position to identify himself with his
historical hero, is irrelevant for the correctness of his explanation;
what counts, is the soundness of the general hypotheses involved, no

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 45

matter whether they were suggested by empathy or by a strictly

behavioristic procedure. Much of the appeal of the "method of
understanding" seems to be due to the fact that it tends to present
the phenomena in question as somehow "plausible" or "natural"
to us; 5 this is often done by means of attractively worded meta-
phors. But the kind of "understanding" thus conveyed must
clearly be separated from scientific understanding. In history as
anywhere else in empirical science, the explanation of a phenomenon
consists in subsuming it under general empirical laws; and the
criterion of its soundness is not whether it appeals to our imagina-
tion, whether it is presented in suggestive analogies, or is otherwise
made to appear plausible-all this may occur in pseudo-explanations
as well-but exclusively whether it rests on empirically well con-
firmed assumptions concerning initial conditions and general laws.
7.1 So far, we have discussed the importance of general laws for
explanation and prediction, and for so-called understanding in his-
tory. Let us now survey more briefly some other procedures of
historical research which involve the assumption of universal
hypotheses.
Closely related to explanation and understandinig is the so-
called interpretation of historical phenomena in terms of some par-
ticular approach or theory. The interpretations which are actually
offered in history consist either in subsuming the phenomena in
question under a scientific explanation or explanation sketch; or
in an attempt to subsume them under some general idea which is
not amenable to any empirical test. In the former case, interpreta-
tion clearly is explanation by means of universal hypotheses; in the
latter, it amounts to a pseudo-explanation which may have emotive
appeal and evoke vivid pictorial associations, but which does not
further our theoretical understanding of the phenomena under
consideration.
7.2 Analogous remarks apply to the procedure of aseertaining,
the "meaning" of given historical events; its scientific import con-
sists in determining what other events are relevantly connected
with the event in question, be it as "causes," or as "effects"; and
the statement of the relevant connections assumes, again, the form
of explanations or explanation sketches which involve universal
hypotheses; this will be seen more clearly in the subsequent section.
7.3 In the historical explanation of some social institutions great
emphasis is laid upon an analysis of the development of the institu-
tion up to the stage under consideration. Critics of this approach
5 For a criticism of this kind of plausibility, cf. Zilsel, I.c., pp. 577-578,
and sections 7 and 8 in the same author 's "'Problems of Empiricism,'" in
International Encyclopedia of Unified Science, Vol. II, 8.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
46 THE JOURNAL OF PHILOSOPHY

have objected that a mere description of this kind is not a genuine

explanation. This argument may be given a slightly different
aspect in terms of the preceding reflections: A description of the
development of an institution is obviously not simply a statement
of all the events which temporally preceded it; only those events
are meant to be included which are "relevant" to the formation of
that institution. And whether an event is relevant to that develop-
ment is not a question of the value attitude of the historian, but an
objective question depending upon what is sometimes called a causal
analysis of the rise of that institution.6 Now, the causal analysis
of an event consists in establishing an explanation for it, and since
this requires reference to general hypotheses, so do assumptions
about relevance, and, consequently, so does the adequate analysis of
the historical development of an institution.
7.4 Similarly, the use of the notions of determination and of
dependence in the empirical sciences, including history, involves
reference to general laws.7 Thus, e.g., we may say that the pres-
sure of a gas depends upon its temperature and volume, or that tem-
perature and volume determine the pressure, in virtue of Boyle's
law. But unless the underlying laws are stated explicitly, the as-
sertion of a relation of dependence or of determination between
certain magnitudes or characteristics amounts at best to claiming
that they are connected by some unspecified empirical law; and
that is a very meager assertion indeed: If, for example, we know
only that there is some empirical law connecting two metrical mag-
nitudes (such as length and temperature of a metal bar), we can
not even be sure that a change of one of the two will be accompanied
by a change of the other (for the law may connect the same value
of the "dependent" or "determined " magnitude with different
values of the other), but only that with any specific value of one
6 See the detailed and clear exposition of this point in M. Mandelbaum's

book; l.c., Chs. 6-8.

7 According to Mandelbaum, history, in contradistinction to the physical
sciences, consists "not in the formulation of laws of which the particular case
is an instance, but in the description of the events in their actual determining
relationships to each other; in seeing events as the products and producers of
change" (l.c., pp. 13-14). This is essentially a view whose untenability has
been pointed out already by Hume; it is the belief that a careful examination
of two specific events alone, without any reference to similar cases and to
general regularities, can reveal that one of the events produces or determines
the other. This thesis does not only run counter to the scientific meaning of
the concept of determination which clearly rests on that of general law, but
it even fails to provide any objective criteria which would be indicative of the
intended relationship of determination or production. Thus, to speak of em-
pirical determination independently of any reference to general laws means
to use a metaphor without cognitive content.

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
 FUNCTION OF GENERAL LAWS IN HISTORY 47

of the variables, there will always be associated one and the same
value of the other; and this is obviously much less than most authors
mean to assert when they speak of determination or dependence in
historical analysis.
Therefore, the sweeping assertion that economic (or geographic,
or any other kind of) conditions "determine" the development and
change of all other aspects of human society, has explanatory value
only in so far as it can be substantiated by explicit laws which state
just what kind of change in human culture will regularly follow
upon specific changes in the economic (geographic, etc.) conditions.
Only the establishment of concrete laws can fill the general thesis
with scientific content, make it amenable to empirical tests, and
confer upon it an explanatory function. The elaboration of such
laws with as much precision as possible seems clearly to be the
direction in which progress in scientific explanation and under-
standing has to be sought.
8. The considerations developed in this paper are entirely neu-
tral with respect to the problem of "specifically historical laws":
neither do they presuppose a particular way of distinguishing
historical from sociological and other laws, nor do they imply or
deny the assumption that empirical laws can be found which are
historical in some specific sense, and which are well confirmed by
empirical evidence.
But it may be worth mentioning here that those universal hy-
potheses to which historians explicitly or tacitly refer in offering
explanations, predictions, interpretations, judgments of relevance,
etc., are taken from variousg fields of scientific research, in so far
as they are not pre-scientific generalizations of everyday experi-
ences. Many of the universal hypotheses underlying historical ex-
planation, for instance, would commonly be classified as psycho-
logical, economical, sociological, and partly perhaps as historical
laws; in addition, historical research has frequently to resort to
general laws established in physics, chemistry, and biology. Thus,
e.g., the explanation of the defeat of an army by reference to lack
of food, adverse weather conditions, disease, and the like, is based
on a-usually tacit-assumption of such laws. The use of tree
rings in dating events in history rests on the applicatioll of certain
biological regularities. Various methods of testing the authentic-
ity of documents, paintings, coins, etc., make use of physical and
chemical theories.
The last two examples illustrate another poilnt which is relevant
in this context: Even if a historian should propose to restrict his
research to a "pure description" of the past, without any attempt
at offering explanations, statements about relevance and determina-

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions
48 THE JOURNAL OF PHILOSOPHY

tion, etc., he would continually have to make use of general laws.

For the object of his studies would be the past-forever inaccessible
to his direct examination. He would have to establish his knowl-
edge by indirect methods: by the use of universal hypotheses
which connect his present data with those past events. This fact
has been obscured partly because some of the regularities involved
are so familiar that they are not considered worth mentioning at all;
and partly because of the habit of relegating the various hypoth-
eses and theories which are used to ascertain knowledge about past
events, to the " auxiliary sciences" of history. Quite probably,
some of the historians who tend to minimize, if not to deny, the
importance of general laws for history, are actuated by the feeling
that only " genuinely historical laws " would be of interest for
history. But once it is realized that the discovery of historical
laws (in some specified sense of this very vague notion) would not
make history methodologically autonomous and independent of the
other branches of scientific research, it would seem that the prob-
lem of the existence of historical laws ought to lose some of its
weight.
The remarks made in this section are but special illustrations
of two broader principles of the theory of science: first, the separa-
tion of "pure description" and "hypothetical generalization and
theory-construction" in empirical science is unwarranted; in the
building of scientific knowledge the two are inseparably linked.
And, second, it is similarly unwarranted and futile to attempt the
demarcation of sharp boundary lines between the different fields
of scientific research, and an autonomous development of each of
the fields. The necessity, in historical inquiry, to make extensive
use of universal hypotheses of which at least the overwhelming ma-
jority come from fields of research traditionally distinguished
from history is just one of the aspects of what may be called the
methodological unity of empirical science.
CARL G. HEMPEL.
QUEENS COLLEGE,
NEW YORK.

BOOK REVIEW
A Dialectic of Morals. Towards the Foundations of Political Phi-
losophy. MORTIMER J. ADLER. Notre Dame, Indiana: The Re-
view of Politics. 1941. x + 117 pp. $1.80.
The author calls this volume a "distillation of actual arguments
which President Hutchins and I have had with students in courses
devoted to reading the great works in Ethics and Politics.'" There

This content downloaded from 146.83.195.38 on Mon, 10 Nov 2014 16:51:56 PM

All use subject to JSTOR Terms and Conditions