The Competitiveness of Nations in a Global Knowledge-Based Economy
Robert K. Merton
The Unanticipated Consequences of
Purposive Social Action
American Sociological Review, 1 (6
Dec. 1936, 894-904.
Content
IN SOME ONE of its numerous forms, the problem of the unanticipated
consequences of purposive action has been treated by virtually every substantial
contributor to the long history of social thought. [1] The diversity of context
[2] and variety of terms [3] by which
this problem has been known, however, have tended to obscure the definite
continuity in its consideration. In
fact, this diversity of context - ranging from theology to technology - has been
so pronounced that not only has the substantial identity of the problem been
overlooked, but no systematic, scientific analysis of it has as yet been
effected. The failure to subject
this problem to such thorough-going investigation has perhaps been due in part
to its having been linked historically with transcendental and ethical
considerations. Obviously, the
ready solution provided by ascribing uncontemplated consequences of action to
the inscrutable will of God or
Although the phrase, unanticipated
consequences of purposive social action, is in a measure self-explanatory, the
setting of the prob-
1. Some of the modern theorists,
though their contributions are by no means of equal importance, are:
Machiavelli, Vico, Adam Smith (and some later classical economists), Marx,
Engels, Wundt, Pareto, Max Weber, Graham Wallas, Cooley, Sorokin, Gini, Chapin,
von Schelting.
2.
This problem has been related to such
heterogeneous subjects as: the problem of evil (theodicy), moral responsibility,
free will, predestination, deism, teleology, fatalism, logical, illogical and
non-logical behavior, social prediction, planning and control, social cycles,
the pleasure- and reality principles and historical
“accidents.”
3. Some of the terms by which the
whole or certain aspects of this process have been known are: Providence
(immanent or transcendental), Moira, Paradoxie der Folgen, Schicksal,
social forces, heterogony of ends, immanent causation, dialectical movement,
principle of emergence and creative synthesis. The present writer hopes to devote a
monograph now in preparation to the history and analysis of this problem. The vast scope and manifold implications
of the problem necessitate my being elliptical at times in the present brief
exposition. For the same reason of
limitation of space, I have had to eliminate most of the concrete material upon
which the discussion is based.
lem demands further specification.
In the first place, the greater
part of this paper deals with isolated purposive acts rather than with their
integration into a coherent system of action (though some reference will be made
to the latter). This limitation is
prescribed by expediency, for a treatment of systems of action would introduce
further complications. Furthermore,
unforeseen consequences should not be identified with consequences which
are necessarily undesirable (from the standpoint of the actor). For though these results are unintended,
they are not upon their occurrence always deemed axiologically negative. In short, undesired effects are not
always undesirable effects. The
intended and anticipated outcomes of purposive action, however, are always, in
the very nature of the case, relatively desirable to the actor, though they may
seem axiologically negative to an outside observer. This is true even in the polar instance
where the intended result is “the lesser of two evils” or in such cases as
suicide, ascetic mortification and self torture which, in given situations, are
deemed desirable relative to other possible alternatives.
Rigorously speaking, the
consequences of purposive action are limited to those elements in the
resulting situation which are exclusively the outcome of the action, i.e., those
elements which would not have occurred had the action not taken place. Concretely, however, the consequences
result from the interplay of the action and the objective situation, the
conditions of action. [4] We will be primarily concerned with the sum-total
results of action under certain conditions. This still involves the problem of causal
imputation (of which more later) though to be a less pressing degree than
consequences in the rigorous sense. These sum-total or concrete consequences
may be differentiated into (a) consequences to the actor (s), (b) consequences
to other persons mediated through (1) the social structure, (2)
the culture and (3) the civilization. [5]
In considering purposive
action, we are concerned with “conduct” as distinct from “behavior,” that
is, with action which involves motives and consequently a choice between various
alternatives. [6] For
4. Cf. Frank H. Knight, Risk,
Uncertainty and Profit,
5. For the distinction between
society, culture and civilization, see Alfred Weber, “Prinzipielles zur
Kultursoziologie: Gesellschaftsprozess, Civilisationsprozess und
Kulturbewegung,” Archiv für Sozialwissenschaft und Sozialpolitik, 47, 1920, 3-49; R. K. Merton, “Civilization and
Culture,” Sociology and Social Research 21, 1936, 103-113. (Cf. foregoing article, definition
19. Ed.)
6. Knight, op. cit., p.
52.
895
the time being, we will take
purposes as given, so that any theories which “reduce” purpose to conditioned
reflexes or tropisms, which assert that motives are simply compounded of
instinctual drives and the experiential shaping of these drives, may be
considered as irrelevant. Psychological considerations of the
source or origin of motives, though they are undoubtedly important for a more
complete understanding of the mechanisms involved in the development of
unexpected consequences of conduct, will thus be ignored.
Moreover, it is not assumed that in
fact social action always involves clear-cut, explicit purpose. It may well be that such awareness of
purpose is unusual, that the aim of action is more often than not nebulous and
hazy. This is certainly the case
with habitual action which, though it may originally have been induced by
conscious purpose, is characteristically performed without such awareness. The significance of such habitual action
will be discussed later.
Above all, it must not be inferred
that purposive action implies “rationality” of human action (that persons always
use the objectively most adequate means for the attainment of their end).
]7\
In fact, part of the present
analysis is devoted to the determination of those elements which account for
concrete deviations from rationality of action. Moreover, rationality and irrationality
are not to be identified with the success and failure of action, respectively.
For in a situation where the number
of possible actions for attaining a given end is severely limited, one
acts rationally by selecting the means which, on the basis of the available
evidence, has the greatest probability of attaining this goal and yet the goal
may actually not be attained. [8] Contrariwise, an end may be attained by action which, on
the basis of the knowledge available to the actor, is irrational (as in the case
of “hunches”).
Turning now to action, we may
differentiate this into two types: (a) unorganized and (b) formally organized.
The first refers to actions of
individuals considered distributively out of which may grow the second when
like-minded individuals form an association in order to achieve a common
purpose. Unanticipated consequences
may, of course, follow both types of action, though the second type would seem
to afford a better opportunity for sociological analysis since the very process
of formal organization ordinarily involves an explicit statement of purpose and
procedure.
7.
Max Weber, Wirtschaft und
Gesellschaft, Tübingen, J. C. B.
Mohr, 1925, pp. 3 ff.
8 See
J. Bertrand, Calcul des probabilités,
Paris, 1889, pp. 90 ff.; J. M. Keynes, A Treatise on Probability,
London, The Macmillan Co., 1921,
Chap. XXVI.
Before turning to the actual
analysis of the problem it is advisable to indicate two methodological pitfalls
which are, moreover, common to all sociological investigations of purposive
action. The first involves the
problem of causal imputation, the problem of ascertaining the extent to which
“consequences” may justifiably be attributed to certain actions. For example, to what extent has the
recent increase in economic production in this country been due to governmental
measures? To what extent may the
spread of organized crime be attributed to prohibition? This ever-present difficulty of causal
imputation must be solved for every empirical case which is
studied.
The second problem is that of
ascertaining the actual purposes of a given action. There is the difficulty, for instance, of
discriminating between rationalization and truth in those cases where apparently
unintended consequences are post facto declared to have been intended.
[9] Rationalizations may occur in connection with
nation-wide social planning just as in the classical instance of the horseman
who, on being thrown from his steed, declared that he was “simply dismounting.”
This difficulty, though not
completely obviated, is significantly reduced in cases of organized group action
since the circumstance of organized action customarily demands explicit (though
not always “true”) statements of goal and procedure. Furthermore, it is easily possible to
exaggerate this difficulty since in many, if indeed not in most, cases, the
observer’s own experience and knowledge of the situation enables him to arrive
at a ready solution. Ultimately,
the final test is this: does the juxtaposition of the overt action, our general
knowledge of the actor(s) and the specific situation and the inferred or avowed
purpose “make sense,” is there between these, as Weber puts it, a “verständliche
Sinnzusammenhang?” If the analyst self-consciously subjects these elements to
such probing, there is substantial probability that his conclusion in respect to
purpose is not too far afield in the majority of instances. The evidence available will vary in
different cases and the probable error of the imputation of purpose will
likewise vary.
9. This introduces the problem of
“chance,” which will be treated in another connection. It should be realized that the aim of
an action and the circumstances which
actually ensue may coincide without the latter being a consequence of the
action. Moreover, the longer the
interval of time between the action and the circumstances in view, the greater
the probability (in the absence of contrary evidence) that these circumstances
have happened “by chance.” Lastly,
if this interval is greatly extended, the probability that the desired
circumstances will occur fortuitously may increase until virtually the point of
certainty. This reasoning is
perhaps applicable to the case of governmental action “restoring prosperity.”
Compare V. Pareto, Traité ole sociologic
générale, Paris, Payot, 1917, II,
par. 1977.
897
It must be freely admitted at this
junction that these problems have not been further treated in the ensuing
discussion, but despite the absence of any further explicit treatment, the
limitations set by these methodological difficulties are implicitly recognized
throughout.
Lastly, it may be urged that a
frequent source of misunderstanding will be eliminated at the outset if it is
realized that the factors involved in unanticipated consequences are -
precisely, factors, and that none of these serves by itself to explain any
concrete case.
The most obvious limitation to a
correct anticipation of consequences of action is provided by the existing state
of knowledge. The extent of this
limitation may be best appreciated by assuming the simplest case where this lack
of adequate knowledge is the sole barrier to a correct anticipation.
[10]
Obviously,
a very large number of concrete reasons for inadequate knowledge may be found,
but it is also possible to summarize several classes of factors which are most
important.
The first class derives from the
type of knowledge - usually, perhaps exclusively - attained in the sciences of
human behavior. Properly speaking,
the social scientist almost invariably finds stochastic (conjectural)
associations and not, as in most fields of the physical sciences, functional
associations. [11] This is to say, in the study of human behavior, there is
found a set of different values of one variable associated with each value of
the other variable(s), or in less formal language, the set of consequences of
any repeated act is not
10.
Most previous discussions of unanticipated
consequences limit the explanation of unanticipated consequences to this one
factor of ignorance. Such a view
either reduces itself to a sheer tautology or exaggerates the ro1e of but one of
many factors. In the first
instance, the argument runs in this fashion: “if we had only known enough, we
could have anticipated the consequences which, as it happens, were unforeseen.”
The apparent fallacy in this
post mortem argument rests in the word “enough” which is implicitly taken
to mean “enough knowledge to foresee” the consequences of our action. It is then no difficult matter to uphold
the contention which then reads in effect: “if we had known, we would have
known.” This viewpoint is basic to
several schools of educational theory, just as it was to Comte’s dictum,
savoir pour prevoir, prevoir pour pouvoir. This intellectualist stand has gained
credence partly because of its implicit optimism and because of the indubitable
fact that sheer ignorance does actually account for the occurrence of some
unforeseen consequences in some cases.
11.
Cf. A. A. Tschuprow, Grundbegriffe und
Grundprobleme der Korrelationstheorie,
constant but there is a range of
consequences, any one of which may follow the act in any given case. In some instances, we may have
sufficient knowledge of the limits of the range of possible consequences, and
even adequate knowledge for ascertaining the statistical (empirical)
probabilities of the various possible sets of consequences, but it is impossible
to predict with certainty the results in any particular case. Our classifications of acts and
situations never involve completely homogeneous categories nor even categories
whose approximate degree of homogeneity is sufficient for the prediction of
particular events. [12] We have here the paradox that whereas past experience
[13] is the sole guide to our expectations on the assumption
that certain past, present and future acts are sufficiently alike to be grouped
in the same category, these experiences are in fact different. To the extent that these differences are
pertinent to the outcome of the action and appropriate corrections for these
differences are not adopted, the actual results will differ from the expected.
As Poincare has put it, “...
small differences in the initial conditions produce very great ones in the final
phenomena... Prediction becomes impossible, and we have the fortuitous
phenomenon.” [14]
However, deviations from the usual
consequences of an act may be anticipated by the actor who recognizes in the
given situation some differences from previous similar situations. But, insofar as these differences can
themselves not be subsumed under general rules, the direction and extent of
these deviations cannot be anticipated. [15] It is clear, then, that the partial knowledge in the
light of which action is commonly carried on permits a varying range of
unexpected outcomes of conduct.
Although no formula for the exact
amount of knowledge necessary for foreknowledge is presented, one may say
in general that consequences are fortuitous when an exact knowledge of many
details and facts (as distinct from general principles) is needed for even a
highly approximate prediction. In
other words, “chance cones-
12.
A classification into completely homogeneous
categories would, of course, lead to functional associations and would hence
permit of perfectly successful prediction, but the aspects of social action
which are of practical importance are too varied and numerous to permit such
homogeneous classification.
13.
A priori calculations of probability are manifestly irrelevant to
specific social acts.
14. Henri
Poincaré, Calcul des probabilite’s,
15.
The actor’s awareness of his ignorance and
its implications is perhaps most acute in the type of conduct which Thomas and
Znaniecki attribute to the wish for “new experience.” This is the case where unforeseen
consequences actually constitute the purpose of action, but there is always the
tacit assumption that these consequences will be desirable. The nebulous purpose in this class of
action is satisfaction.
899
quences” are those which are
occasioned by the interplay of forces and circumstances which are so complex and
numerous that prediction of them is quite beyond our reach. This area of consequences should perhaps
be distinguished from that of “ignorance,” since it is related not to the
knowledge actually in hand but to certain knowledge which can conceivably be
obtained. [16
The importance of ignorance as a
factor is enhanced by the fact that the exigencies of practical life frequently
compel us to act with some confidence even though it is manifest that the
information on which we base our action is not complete. We usually act, as Knight has properly
observed, not on the basis of scientific knowledge, but opinion and estimate.
Thus, situations which demand (or
what is for our purposes tantamount to the same thing, appear to the actor to
demand) immediate action of some sort, will usually involve ignorance of certain
aspects of the situation and will bring about unexpected
results.
Moreover, even when immediate action
is not exacted, there is the economic problem of distributing our
fundamental resources, time and energy. Time and energy are scarce means and
economic behavior is concerned with the rational allocation of these means among
alternative wants, only one of which is the anticipation of consequences of
action. [17] In our present economic order, it is manifestly
uneconomic behavior to concern ourselves with attempts to obtain knowledge for
predicting the outcomes of action to such an extent that we have practically no
time or energy for other pursuits. An economy of social engineers is no more
conceivable or practicable than an economy of laundrymen. It is the fault of the extreme antinoetic
activists who promote the idea of action above all else to exaggerate this limit
and to claim (in effect) that virtually no time or energy be devoted to the
acquisition of knowledge. On the
other hand, the grain of truth in the anti-intellectualist position is, as was
just observed, that there are not only decided economic limits to
the
16.
Cf. Keynes,
op. cit., p. 295. This distinction corresponds to that made by Keynes
between “subjective chance” (broadly, ignorance) and “objective chance” (where
even additional wide knowledge of general principles would not suffice for
foreseeing the consequences of a particular act). Much the same distinction appears in the
works of Poincaré and Venn, among others.
17.
Cf. Knight, op. cit., p. 348.
The reasoning is also applicable to
cases where the occupation of certain individuals (e.g., social engineers and
scientists) is devoted solely to such efforts, since then it is simply a
question of the distribution of the resources of society. Furthermore, there is the practical
problem of the communicability of knowledge so obtained, since it may be of a
very complex order and the effort of persons other than social engineers to
assimilate such knowledge leads us back to the same problem of distribution of
our resources.
advisability of not acting until all
or as much as possible uncertainty is eliminated, but also psychological limits
since excessive “forethought” of this kind precludes any action at
all.
A second major factor of unexpected
consequences of conduct, which is perhaps as pervasive as ignorance, is error.
Error may intrude itself, of
course, in any phase of purposive action: we may err in our appraisal of the
present situation, in our inference from this to the future objective situation,
in our selection of a course of action, or finally in the execution of the
action chosen. A common fallacy is
frequently involved in the too-ready assumption that actions which have in the
past led to the desired outcome will continue to do so. This assumption is often fixed in the
mechanism of habit and it there finds pragmatic justification, for habitual
action does in fact often, even usually, meet with success. But precisely because habit is a mode of
activity which has previously led to the attainment of certain ends, it tends to
become automatic and undeliberative through continued repetition so that the
actor fails to recognize that procedures which have been successful in
certain circumstances need not be so under any and all conditions.
[18]
Just as rigidities in social organization often balk and
block the satisfaction of new wants, so rigidities in individual behavior may
block the satisfaction of old wants in a changing social
environment.
Error may also be involved in
instances where the actor attends to only one or some of the pertinent aspects
of the situation which influence the outcome of the action. This may range from the case of simple
neglect (lack of systematic thoroughness in examining the situation) to
pathological obsession where there is a determined refusal or inability to
consider certain elements of the problem. This last type has been extensively dealt
with in the psychiatric literature. In cases of wish-fulfilment, emotional
involvements lead to a distortion of the objective situation and of the probable
future course of events; such action predicated upon “imaginary” conditions must
inevitably evoke unexpected consequences.
The third general type of factor,
the “imperious immediacy of interest,” refers to instances where the actor’s
paramount concern with the foreseen immediate consequences excludes the
consideration of further or other consequences of the same act. The most prom-
18.
Similar fallacies in the field of thought
have been variously designated as “the philosophical fallacy” (Dewey), the
“principle of limits” (Sorokin, Bridgman) and, with a somewhat different
emphasis, “the fallacy of misplaced concreteness”
(Whitehead).
901
inent elements in such immediacy of
interest may range from physiological needs to basic cultural values. Thus, Vico’s imaginative example of the
“origin of the family” which derived from the practice of men carrying their
mates into caves to satisfy their sex drive out of the sight of God might serve
as a somewhat fantastic illustration of the first. The doctrine of classical economics
according to which the individual endeavoring to employ his capital where most
profitable to him and thus tending to render the annual revenue of society as
great as possible is, to quote Adam Smith, led “by an invisible hand to promote
an end which was no part of his intention,” may serve as an example of economic
interest leading to this sequence.
However, after the acute analysis by
Max Weber, it goes without saying that action motivated by interest is not
antithetical to an exhaustive investigation of the conditions and means of
successful action. On the contrary,
it would seem that interest, if it is to be satisfied, demands such objective
analysis of situation and instrumentality, as is assumed to be characteristic of
hominis oeconomici. But it
is equally undeniable that intense interest does in fact often tend to preclude
such analysis precisely because strong concern with the satisfaction of the
immediate interest is a psychological generator of emotional bias, with
consequent lopsidedness or failure to engage in the required calculations. It is as much a fallacious assumption to
hold that interested action in fact necessarily entails a rational calculation
of the elements in the situation [19] as to deny rationality any and all
influence over such conduct. Moreover, action in which this element of
immediacy of interest is involved may be rational in terms of the values basic
to that interest but irrational in terms of the life organization of the
individual. Rational, in the sense
that it is an action which may be expected to lead to the attainment of the
specific goal; irrational, in the sense that it may defeat the pursuit or
attainment of other values which are not, at the moment, paramount but which
none the less form an integral part of the individual’s scale of values. Thus, precisely because a particular
action is not carried out in a psychological or social vacuum, its effects will
ramify into other spheres of value and interest. For example, the practice of birth
control for “economic reasons” influences the age-compo-
19.
This assumption is tenable only in a
normative sense. It is indubitable
that such calculation, within the limits specified in our previous discussion,
should be made if the probability of satisfying the interest is to be at
a maximum. The error lies in
confusing norm with actuality.
sition and size of sibships with
profound consequences of a psychological and social
character.
Superficially similar to the factor
of immediacy of interest, but differing from it in a highly significant
theoretical sense, is that of basic values. This refers to instances where there is
no consideration of further consequences because of the felt necessity of
certain action enjoined by certain fundamental values. The classical analysis of the influence
of this factor is Weber’s study of the Protestant ethic and the spirit of
capitalism. He has properly
generalized this case, saying that active asceticism paradoxically leads to its
own decline through the accumulation of wealth and possessions entailed by
decreased consumption and intense productive activity.
This process contributes much to the
dynamic of social and cultural change, as has been recognized with varying
degrees of accuracy and cogency, by Hegel, Marx, Wundt and many others. The empirical observation is
incontestable: activities oriented toward certain values release processes which
so react as to change the very scale of values which precipitated them. This process may in part be due to the
fact that when a system of basic values enjoins certain specific actions,
adherents are not concerned with the objective consequences of these actions but
only with the subjective satisfaction of duty well performed. Or, action in accordance with a dominant
set of values tends to be focussed upon that particular value-area. But with the complex interaction which
constitutes society, action ramifies, its consequences are not restricted to the
specific area in which they were initially intended to center, they occur in
interrelated fields explicitly ignored at the time of action. Yet it is because these fields are in
fact interrelated that the further consequences in adjacent areas tend to
react upon the fundamental value-system. It is this usually unlooked-for reaction
which constitutes a most important element in the process of secularization, of
the transformation or breakdown of basic value-systems. Here is the essential paradox of social
action - the “realization” of values may lead to their renunciation. We may paraphrase Goethe and speak of
“Die Kraft, die stets das Gute will, und stets das Böse
schafft.”
There is one other circumstance,
peculiar to human conduct, which stands in the way of successful social
prediction and planning. Public
predictions of future social developments are frequently not sustained precisely
because the prediction has become a new element in the concrete situation, thus
tending to change the initial course
903
of developments. This is not true of prediction in fields
which do not pertain to human conduct. Thus, the prediction of the return of
Halley’s comet does not in any way influence the orbit of that comet; but,
to take a concrete social example, Marx’s prediction of the progressive
concentration of wealth and increasing misery of the masses did influence the
very process predicted. For at
least one of the consequences of socialist preaching in the nineteenth century
was the spread of organization of labor, which, made conscious of its
unfavorable bargaining position in cases of individual contract, organized to
enjoy the advantages of collective bargaining, thus slowing up, if not
eliminating, the developments which Marx had predicted.
[20]
Thus, to the extent that the
predictions of social scientists are made public and action proceeds with full
cognizance of these predictions, the “other-things-being-equal” condition
tacitly assumed in all forecasting is not fulfilled. Other things will not be equal just
because the scientist has introduced a new “other thing” - his prediction. This contingency may often account for
social movements developing in utterly unanticipated directions and it hence
assumes considerable importance for social planning.
The foregoing discussion represents
no more than the briefest exposition of the major elements involved in one
fundamental social process. It
would take us too far afield, and certainly beyond the compass of this paper, to
examine exhaustively the implications of this analysis for social prediction,
control and planning. We may
maintain, however, even at this preliminary juncture, that no blanket statement
categorically affirming or denying the practical feasibility of all
social planning is warranted. Before we may indulge in such
generalizations, we must examine and classify the types of social action
and organization with reference to the elements here discussed and then refer
our generalizations to these essentially different types. If the present analysis has served to set
the problem, if only in its most paramount aspects, and to direct attention
toward the need for a systematic and objective study of the elements involved in
the development of unanticipated consequences of purposive social action, the
treatment of which has for much too long been consigned to the realm of theology
and speculative philosophy, then it has achieved its avowed
purpose.
20.
Corrado Gini, Prime linee di
patologia economica,
904