Nomothetic vs. idiographic
The distinction between nomothetic and idiographic methods was raised on an
internet list for statistics educators. Since this distinction is so
important in Brunswikian research, I asked Ken Hammond to comment on the
message. The original message and Ken Hammond's response are reproduced
The message from Kubovy:
Date: Thu, 05 Jan 1995 09:05:32 -0500 (EST)
Subject: Re: nomological and nomothetic
To: Multiple recipients of list email@example.com.EDU;
In Article 1995Jan4.firstname.lastname@example.org,
>I am looking for some information on the terms nomothetic and
>nomological. I can not find a standard reference that defines them.
For the first, see Nagel, E. (1961). The structure of science: Problems in
the logic of scientific explanation. London: Routledge & Kegan Paul, p
547-8, thinks that Aristotle is source of the distinction between
_nonomthetic_ sciences, "which seek to establish abstract general laws for
indefinitely repeatable events and processes; and the _ideographic_, which
aims to understand the unique and nonrecurrent" (p 547). He attributes the
terms to Windelband, W. (1915). Geschichte und Naturwissenschaft, In
Praeludien (Vol. 2, pp. 136-160). Tuebingen. See also Bunge, M., & Ardila,
R. (1987). Philosophy of psychology. New York: Springer, pp. 223-224.
I can't give you a reference to the latter. I believe the term comes from
the debate in psychology between operationalism and more liberal conceptions
of the meaning of scientific terms. Is it Paul Meehl who coined the term
"nomological network" to describe the structure within which many
theoretical terms are embedded, without individually receiving operational
|\ /| / Michael Kubovy (email@example.com)
| \/ |/ Dept. of Psychology, Univ. of Virginia, Gilmer Hall
| |\ Charlottesville, VA 29903-2477
| | \ (804)982-4729, FAX 982-4766
I agree with Kubovy that these terms are important; they are also
highly relevant to Brunswikian approach because Brunswik was
virtually the only experimental psychologist from the 1940s onward
who was drawing the distinction and coming down on the idio side.
The distinction is described in the glossary (Anderson, B. F., Deane, D.
H., Hammond, K. R., McClelland, G. H., & Shanteau, J. C. (1981).
Concepts in judgment and decision research Definitions, sources,
interrelations, comments. New York: Praeger, p. 166) and in
Hammond, K. R., McClelland, G. H., & Mumpower, J. (1980). Human
judgment and decision making: Theories, methods, and procedures.
New York: Hemisphere/Praeger in which there are 14 references in
the index to each of these terms, including definitions. It is no
accident that there is this much material in the latter inasmuch as
one of the authors was a student of Brunswik. Neither these terms
nor nomological net appear in Meehl's 1954 book. Brunswik's 1956
book contains 2 references to "nomothetic laws" at least in my index
(the original contains no index): One reference is to 2 pages of text in
which Brunswik's general theory of behavior is described and is
contrasted with "nomothetic-reductionism" as part of his summary of
the book. His 1943 Psychological Review paper also treats this topic.
Kurt Lewin should not be forgotten in this regard; he was scornful of
averages of population scores, noting that they generally represented
the behavior of no single person. (See his "Dynamic theory of
personality," 1935, pp. 1-26.) Brunswik showed his agreement with
Lewin's idiographic position in this way: "I agree with Lewin when
he makes it clear that there is no place for statistics in a strictly [note
"Strictly") nomothetic . . . discipline." (Example: The law of the lever
or the gravitational constant were not produced by statistics.) "In
fact, not even averages from a large number of cases [of subjects] . . .
are in order. Indeed, those psychologists who have accepted the
ideology of accumulated observation have already deviated from the
strictly nomothetic path. If all the relevant conditions are known, or
rather if all disturbing influences are eliminated, only one
observation is needed to ascertain a general law [which is what
nomotheticism is about once and forever. Lewin . . . refers to Galileo's
study of falling bodies as an example" (Brunswik, E. (1943).
Organismic achievement and environmental probability.
Psychological Review, 50, pp. 265-266). Brunswik then stated his
own view: There must be recognition of the fact that there can be no
truly molar psychology dealing with the physical relationships of the
organism with its environment unless it gives up the nomothetic
ideal in favor of a thoroughly statistical conception" (p.270) (of the
organism environment relationship). I think that the relinquishment
of averages will be the big accomplishment of the next 50 years,
roughly a century after Brunswik's and Lewin's attempts.
I also think that the nature of Brunswik's emphasis on
organismic environmental relationships makes him the first modern
evolutionary psychologist. I try to follow up on this aspect of his
work in my book.
I do not think that the nomo v. idio distinction has anything to do
with operationalism arguments. Brunswik made a point of saying
that idio terms were just as conducive to operational definition as
nomo terms, as, of course, they are. The distinction is one of
theoretical direction, that is, defining the task of psychology, and
methodological arguments over generalization. (See Hammond, K. R.,
Hamm, R. M., & Grassia, J. (1986). Generalizing over conditions by
combining the multitrait-multimethod matrix and the representative
design of experiments. Psychological Bulletin, 100, 257-269 which
says quite a bit about this and offers a current example of how the
nomo ideal fails us.)
I am glad Kubovy brought this matter to our attention and I hope
he pursues his interests in it.
Kenneth R. Hammond