Академический Документы
Профессиональный Документы
Культура Документы
Robert K. Merton: Consensus and Controversy. by Jon Clark; Celia Modgil; Sohan Modgil;
Puritanism and the Rise of Modern Science: The Merton Thesis. by I. Bernard Cohen; K. E.
Duffin; Stuart Strickland; "After Merton": Protestant and Catholic Science in Seventeenth-
Century Europe. by Rivka Feldhay; Yehuda Elkana; The Focused Interview: A Manual of
Procedures. by Robert K. Merton; Marjorie Fiske; Patricia L. Kendall; L'Ope ...
Review by: Raymond Boudon
Contemporary Sociology, Vol. 20, No. 4 (Jul., 1991), pp. 519-522
Published by: American Sociological Association
Stable URL: http://www.jstor.org/stable/2071781 .
Accessed: 06/09/2013 12:28
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 support@jstor.org.
American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to
Contemporary Sociology.
http://www.jstor.org
What Middle-RangeTheoriesAre
RAYMOND BOUDON
Universitede Paris-Sorbonne
Sociologists oftenuse the word "theory" in Merton had in mind when he coined his
an idiosyncraticfashion. For many of them, famous notion of the "middle-rangetheory"
"theorizing" means, for instance, identifying (MRT). I remember Paul Lazarsfeld once
a given type of social process or of tellingme: "an importantnotion, but I don't
independentvariable as the main one. Thus know how to defineit." I may be wrong,but
"the theory of conflict" advocates in a I thinkthatMRT means two things,which are
nutshell the idea that conflictsrepresentthe both of utmostimportanceand very simple to
most importantaspect of social life. That grasp. A negative one: it means it is hopeless
conflictsare importantis trivial.That theyare and quixotic to tryto determinethe overarch-
the most importantaspect of social processes ing independentvariable that would operate
is at best an empty statement. For other in all social processes, or to determinethe
sociologists, social classes-alternatively, the essential featureof the social structure,or to
"dominant class"-represent the only really find out the two, three, or four couples of
crucial groups in societies. Some sociologists concepts (e.g., Gesellschaft/Gemeinschaftl)
have even spent their lives developing the that would be sufficientto analyze all social
view that any observable social phenomenon phenomena. In other words, I see first in
should be interpretedas the manifestationof Merton's MRT notionthe polite expressionof
the hidden hand of the dominantclass. Let us a severe doubt about the usefulness of a
call this type of theorizingBT (for "broad "theoretical" activitywhich was widespread
range" or-ad libitum-"bad" theorizing). when Mertoncoined it, and which appears as
a permanentfeatureof our sociological field.
Middle-Range Theory
i That such concepts are importantis one thing; that
It has sometimes seemed hard to see what they constitutea key to all social phenomena is another.
A penetrating British methodologist said with and explain all kinds of heterogeneous
recentlythe same thingin more directterms: phenomena. It is present,forexample, behind
what sociologists call "theory" is oftenwhat Hirschman's "tunnel effect" (1980), which
philosophers would normally consider bad suggeststhatthe sensitivityof social actorsto
philosophy(Pawson 1989). But the notion of the evolution of theirown situationdepends
MRT also has a positive side, which I would on the evolution of other social categories
enunciateplainly: sociologists would be better close to them. Thanks to this model,
off and would gain both in efficiencyand Hirschman explains puzzling observations
respectabilityin the external world if they about the toleranceto inequalitiesin develop-
took the word "theory" in the same sense as ing countries: when two lines of cars stay
do the othersciences, and if they"theorized" bumperto bumperin a tunnel,as soon as the
in the same fashion as do otherscientists. line, say, on the left starts moving, this
That the notionof MRT attemptsto fighta creates expectations in the other line. Those
certain epistemological obscurantismtypical on the rightexpect to move as well. If they
of sociology and thatBT is seen by Mertonas don't, or if theymove less quickly than they
a major obstacle to the advancement of expected, theytendto become more impatient
sociology is clear from what he writes: thanwhen theydid not move at all. The same
"sociology will advance in the degree thatits happens with inequalities, says Hirschman.
major concern is with developing theoriesof Because of this RG mechanism, they can be
the middle range" (Merton 1957, p. 9). In feltmore stronglywhen the absolute situation
other words, substitutingMRT for "theory" of the actors gets better.
(BT) is the main way to give sociology the Obviously, the "tunnel effect" can be
cognitive power a scientific discipline nor- easily related, throughthe mediation of the
mally should aim at. So much forthenegative RGT, to Stouffer'sfamous "relative depriva-
criticalside of the MRT notion. tion" effect,the canonical illustrationof the
As to the positive side: talking about the RGT. The same effect is illustratedby the
"referencegroup theory,"Mertonsays thatit famous analysis in Tocqueville's Ancien
is "one of these theoriesof the middle range Regime, where he explains why discontent
which consolidate otherwise segregated hy- grew in the years before the French Revolu-
potheses and empiricalregularities"(1957, p. tion, at a time where people were becoming
280). In other words, MRT describes effec- betteroff (Boudon 1982).
tively what the other sciences call simply I have elsewhere tried to show that
"theory." As we all know fromour studies in educational inequalities could not be easily
the philosophy and history of the natural understood without introducingRG effects
sciences, a "scientific theory" is a set of (1977, 1990). Educational inequalities are
statementsthat organize a set of hypotheses large and remain so. According to empirical
and relate themto segregatedobservations.If data, theyresultfromthe factthatyoungsters
a "theory" is valid, it "explains" and in other from lower-class families are cognitively
words "consolidates" and federatesempirical handicapped. But they result also from
regularitieswhich on theirside would other- motivational effects. Other things (achieve-
wise appear segregated.This amountsalso to ment, age, and the like) being equal, a
sayingthatmere empiricismis of littleworth. youngster from a lower class feels less
attractedthan a higher-class studentby the
idea of reaching the next stage in the
Reference Group Theory educational race. This motivationalfactor is
by far much more important than the
Reference group theory(RGT) is a good "cultural" one. The effects of the cultural
classical example of a theory in the sense handicap dwindle over the life of a school
Merton advocates. The general idea behind cohortbecause of the overselectionof young-
the RGT is thatmany attitudesand beliefs get stersfromthe lower class, while the motiva-
installedin the minds of social actorsby their tional factor enters repeatedly into action,
taking some persons or groups as a natural generating exponential effects. Now this
reference,given the situation and questions motivation factor is most convincingly ex-
the actors are exposed to. This theoretical plained by the fact that the youngstersand
idea is presentbehindseveral models thatdeal their families judge educational and social