Author's personal copy
Provided for non-commercial research and educational use only.
Not for reproduction, distribution or commercial use.
This article was originally published in the International Encyclopedia of the Social
& Behavioral Sciences, 2nd edition, published by Elsevier, and the attached copy
is provided by Elsevier for the author’s benefit and for the benefit of the
author’s institution, for non-commercial research and educational use including
without limitation use in instruction at your institution, sending it to specific
colleagues who you know, and providing a copy to your institution’s administrator.
All other uses, reproduction and distribution, including
without limitation commercial reprints, selling or
licensing copies or access, or posting on open
internet sites, your personal or institution’s website or
repository, are prohibited. For exceptions, permission
may be sought for such use through Elsevier’s
permissions site at:
http://www.elsevier.com/locate/permissionusematerial
From Wolters, G., 2015. Hempel, Carl Gustav (1905–97). In: James D. Wright
(editor-in-chief), International Encyclopedia of the Social & Behavioral Sciences,
2nd edition, Vol 10. Oxford: Elsevier. pp. 773–775.
ISBN: 9780080970868
Copyright © 2015 Elsevier Ltd. unless otherwise stated. All rights reserved.
Elsevier
Author's personal copy
Hempel, Carl Gustav (1905–97)
Gereon Wolters, Universität Konstanz, Konstanz, Germany
Ó 2015 Elsevier Ltd. All rights reserved.
Abstract
Carl Gustav Hempel was a leading figure of the logicoempiricist movement, originating in the 1920s in Vienna and Berlin. He
is particularly famous for his work on scientific explanation that led to the ‘Hempel–Oppenheim model’ of scientific
explanation and his work on concept formation in empirical science. Hempel, furthermore, connected his methodological
ideas with the thesis of the unity of science, according to which there holds the same methodology both in the natural and in
the social sciences. In his later work, Hempel opened up to T. S. Kuhn’s historic-pragmatic conception of science.
Carl Gustav Hempel was a leading figure of the logicoempiricist
movement that dominated much of American philosophy in
the second third of the twentieth century and has been a principal source of present day analytical philosophy. In the spirit
of logical empiricism, Hempel aimed at a scientific philosophy
with the principal objective to arrive at a purely logical methodology of the empirical sciences, including the social sciences.
His most important contribution in this respect is what can be
regarded as the first theory of scientific explanation (Salmon,
1989). This theory is referred to as ‘the Hempel–Oppenheim
model,’ the ‘covering-law model,’ or the ‘standard model of
scientific explanation.’ In his later years, Hempel gave up the
idea of a purely logical (i.e., syntactical and/or semantical)
‘rational reconstruction’ or ‘explication’ of basic methodological concepts like explanation, confirmation, law of nature,
probability, type, etc. Both the internal failure of this program
and the influence of Thomas S. Kuhn caused Hempel to
acknowledge for a satisfying reconstruction of methodological
concepts also ‘pragmatic’ aspects of science in the sense of
social and historical circumstances of its development.
found a post of instructor at City College, New York, in 1939.
The academic year 1940–41 saw him at Queens College, New
York, from where he went as associate professor to Yale in
1948. In 1955, Hempel became Stuart Professor of Philosophy
at Princeton. After retirement at Princeton in 1973, he worked
as university professor at Pittsburgh University from 1977 until
1985. He died at Princeton on 9 November 1997.
Major Contributions
Hempel has given important contributions to (1) the rational
reconstruction of methodological concepts, particularly the
concept of scientific explanation; (2) the ‘universality thesis,’
i.e., the thesis that the social sciences do not, and should not,
use methods that are essentially different from the methods of
empirical science; and (3) concept formation and taxonomy in
empirical science.
Scientific Explanation
Biography
Hempel was born on 8 January 1905, in Oranienburg (near
Berlin, Germany). He studied mathematics, physics, and
philosophy at the universities of Göttingen, Heidelberg, and
Berlin from 1923 to 1929. After graduation (Staatsexamen) that
qualified him to teach in secondary schools (Gymnasien),
Hempel went to Vienna to study with Moritz Schlick and,
particularly, Rudolf Carnap. Both, together with the mathematician Hans Hahn, the economist Otto Neurath, and others,
had founded in 1923 the ‘Vienna Circle’ (Wiener Kreis), which
was to become the nucleus of logical empiricism. After his
return to Berlin in 1930, Hempel entered teacher’s training and
teaching and worked at the same time with Hans Reichenbach,
head of the ‘Berlin Group’ of logical empiricism, on a philosophical dissertation on the concept of probability. After
graduation in 1934, Hempel preferred to leave Nazi Germany
and went to Brussels to assist Paul Oppenheim, a private
scholar, in carrying out philosophical research that related
principally to concept formation in empirical science. In
1937–38, Hempel worked for a year as research associate with
Carnap in Chicago and again went back to Brussels until he
International Encyclopedia of the Social & Behavioral Sciences, 2nd edition, Volume 10
Hempel reconstructs scientific explanations of concrete events
in nature, history, and society as arguments, i.e., as logical
deductions containing as premises (explanans) at least one
(general) scientific law L1, as well as initial and boundary
conditions Cr and as conclusion (explanandum) the statement
E describing the event in question. We thus obtain the
following model of ‘deductive-nomological (D-N) explanation’:
L1, L2, .,Ln
C1, C2,.,Cr
the explanans
[entails logically]
E
the explanandum
In case the explanandum contains essentially a statistical law
(e.g., for all x: the probability of G, if F is given, equals r and as
boundary condition that F is given for some b), the explanatory
argument does not deliver a deduction of the explanandum
statement E but rather gives inductive support for E, i.e., the
explanans shows that E was to be expected with a certain degree
r of probability. In simplified form (one statistical law, one
http://dx.doi.org/10.1016/B978-0-08-097086-8.61050-4
International Encyclopedia of the Social & Behavioral Sciences, Second Edition, 2015, 773–775
773
Author's personal copy
774
Hempel, Carl Gustav (1905–97)
boundary condition), we thus obtain the following model of
‘inductive-statistical (I-S) explanation’:
p(G, F) ¼ r
F(b)
the explanans
[lends inductive support of
probability r to]
G(b)
the explanandum
In the case of I-S explanations, one must further assume that
the explanans is maximally specified, i.e., the explanans must
contain all available information that is possibly relevant for
the explanation of the explanandum E. In other words, the
explanandum must contain all pertinent statistical laws and
those special facts that can be connected with the E event by
statistical laws. I-S explanations thus relate to the body of
scientific knowledge at a given time, i.e., they become ‘epistemically relative.’ The postulate of maximum specificity avoids
the ‘explanatory ambiguity,’ i.e., the possibility to give acceptable I-S explanations based on true premises for mutually
exclusive explananda.
In case one deduces as explanandum a general or a statistical
law of lesser generality than the one(s) in the explanandum one
obtains the reduction of a law or theory in the first case (e.g.,
explaining the law of free fall by means of the law of gravitation), and a deductive-statistical or D-S explanations of a statistical law by a more comprehensive statistical law in the second.
Originally, Hempel claimed the structural identity of explanation and prediction. Later he had to admit that, although every
adequate explanation is (under pragmatically changed circumstances) also a prediction, not every adequate prediction yields
also an adequate explanation, for symptoms are often adequate
for a prediction but not for the corresponding explanation. So is,
for example, a sudden fall of the barometer (together with suitable
laws) sufficient for predicting a thunderstorm, but it does not
explain it, since the fall of the barometer is not the cause of the
thunderstorm, but rather a symptom of its arrival (Hempel, 1965).
Unity of Science
Hempel claims the methodological unity of empirical science.
With respect to scientific explanation this amounts to
a universality thesis of scientific explanation: There is no
essential difference between explanations in the natural
sciences and those in psychology, the social sciences, or history.
Explanations in these fields likewise aim at showing that the
event in question was to be expected on the basis of antecedent
and boundary conditions and general laws. In history, those
general laws are usually taken from psychology, sociology,
economics, the natural sciences, etc. Hempel leaves the question open whether genuine historical laws exist. Normally,
explanations in history and sociology fail to include an explicit
statement of the laws they presuppose. This failure is due to the
fact that those laws are part of folk psychology and seem to be
tacitly taken for granted. Apart from that, it is difficult to
formulate in a sufficiently exact way the underlying general
assumptions about, e.g., the outbreak and course of revolutions. Furthermore, most of the regularities in the social
sciences are statistical. Therefore, one can expect only explanation sketches in these fields.
The method of empathic understanding that is often
claimed to distinguish the social from the natural sciences is
neither sufficient nor necessary for explanations in these fields.
It is not sufficient, because the individual act of understanding is
only a heuristic device in order to suggest general psychological
hypotheses that might subsequently serve as explanatory
principles in the case under consideration. It is not necessary,
because a historian might well be able to explain the deeds of
a paranoiac historic personality by reference to the principles of
abnormal psychology without being able to arrive at an
emphatic understanding of such a personality.
Concept Formation in Empirical Science
Hempel’s work on concept formation in empirical science
originates in his early cooperation with Paul Oppenheim and
consists basically in the application of principles of logical
empiricism to the formation of scientific concepts (Hempel and
Oppenheim, 1936). Hempel distinguishes (1) classificatory, (2)
comparative, and (3) quantitative concepts. These distinctions
normally characterize at the same time the historical development of a science from its initial descriptive ‘natural history’
stages to its mature theoretical stages. Classification divides
a given set or class of objects (e.g., human beings) into subclasses
(e.g., beings that suffer from certain mental disorders). Each of
these subclasses is defined by means of a certain concept (e.g.,
schizophrenia) that represents the complex of characteristics
essential for the membership in that subclass. Hempel speaks
also of ‘classificatory types.’ The elements of these subclasses
are those individuals to whom the respective concept applies
(e.g., schizophrenics). The characteristics that form the defining
concept are ascertainable fairly directly by observation. Classificatory concepts as well as scientific concepts in general have to
fulfill two requirements. They have to (1) be objective and (2)
have systematic import. Objectivity is attained first by using
operational criteria and (often only partial) operational definitions in a large sense that includes observation as operational
and second by eliminating criteria with valuation overtones.
Scientific concepts have systematic import if they lend themselves to the formulation of general laws or theoretical principles. Classificatory concepts are a yes-or-no affair, i.e., a certain
individual belongs to some subclass or not, whereas comparative concepts admit of more or less with respect to a certain trait
(e.g., x is warmer than y with respect to cold–hot). They lead to
(quasi) linear orderings as soon as one includes the relation of
coincidence (e.g., x is equally warm as y). Hempel calls the
extremes of such orderings ‘ordering types.’ The next step would
consist in giving these linear orderings a metric and thus arriving at quantitative concepts (e.g., temperature).
For Hempel the ‘ideal types’ that have become popular in
the social sciences since Max Weber are neither classificatory
nor ordering types. Ideal types are interpretative or explanatory
schemas that contain a set of empirical generalizations that
establish subjectively meaningful connections between
different aspects of some kind of phenomenon, e.g., purely
rational economic behavior, a capitalist society, and a religious
sect. Accordingly, ideal types must be reconstructed not as
concepts but as theoretical systems that are intended to
provide explanations and therefore must contain testable
hypotheses. Despite their lack of clarity and precision, there
International Encyclopedia of the Social & Behavioral Sciences, Second Edition, 2015, 773–775
Author's personal copy
Hempel, Carl Gustav (1905–97)
thus are also with respect to ideal types no essential methodological differences between psychology or the social sciences
and the corresponding methods in the natural sciences
(Hempel, 1952, 1967).
Influence, Impact, and Current Significance
Hempel’s ideas constitute the mature form of logicoempiricist
thinking (Fetzer, 2001). His work on scientific explanation
‘stands out as the benchmark to which all subsequent studies of
this topic must be referred’ (Salmon, 1998, p. vii). His reconstruction of concept formation in empirical science has become
a sort of standard for objective procedures of concept formation.
It has greatly influenced subsequent editions of the Diagnostic
and Statistical Manual of the American Psychiatric Association
(Houts, 2001). His later work reflects the openness of logical
empiricism to new developments in the philosophy of science
(particularly T.S. Kuhn’s work) that put a strong emphasis on
the historical and sociological (pragmatic) factors of the development of science (Hempel, 1979, 2000; Wolters, 2003).
See also: Empiricism, History of; Explanation: Conceptions in
the Social Sciences; Idealization, Abstraction, and Ideal Types;
Kuhn, Thomas S. (1922–96); Logical Positivism and Logical
Empiricism; Mathematical Models in Philosophy of Science.
775
Bibliography
Fetzer, J.H. (Ed.), 2001. The Philosophy of Carl G. Hempel: Studies in Science,
Explanation, and Rationality. Oxford University Press, Oxford.
Hempel, C.G., 1952. Fundamentals of Concept Formation in Empirical Science.
University of Chicago Press, Chicago (International Encyclopedia of Unified Science.
Vol. II, no. 7).
Hempel, C.G., 1965. Aspects of Scientific Explanation and Other Essays in the
Philosophy of Science. The Free Press, New York.
Hempel, C.G., 1967. Philosophy of Natural Science. Prentice-Hall, Englewood
Cliffs, NJ.
Hempel, C.G., 1979. Scientific rationality: analytic vs. pragmatic perspectives.
In: Geraets, T.F. (Ed.), Rationality To-day – La Rationalité Aujourd’hui. University of
Ottawa Press, Ottawa, pp. 46–66.
Hempel, C.G., 2000. Selected Philosophical Essays. In: Jeffrey, R. (Ed.). Cambridge
University Press, Cambridge.
Hempel, C.G., Oppenheim, P., 1936. Der Typusbegriff im Lichte der Neuen Logik.
Wissenschaftstheoretische Untersuchungen zur Konstitutionsforschung und
Psychologie. A. W. Sijthoff, Leiden.
Houts, A.C., 2001. The Diagnostic and Statistical Manual’s new white coat and
circularity of plausible dysfunctions: response to Wakefield. Behaviour Research
and Therapy 38, 315–345.
Salmon, W.C., 1989. Four Decades of Scientific Explanation. University of Minnesota
Press, Minneapolis, MN.
Salmon, W.C., 1998. Causality and Explanation. Oxford University Press, Oxford.
Wolters, G., 2003. Carl Gustav Hempel – pragmatic empiricist. In: Parrini, P.,
Salmon, M.H., Salmon, W.C. (Eds.), Analytical and Continental Aspects of Logical
Empiricism. University of Pittsburgh Press, Pittsburgh, pp. 109–122.
International Encyclopedia of the Social & Behavioral Sciences, Second Edition, 2015, 773–775