This paper sets out to examine some of the notions of “value” which are current in anthropology, to relate them, and to consider the problems they set for the empirical identification of values in historical cultures. The treatment aims not at an exhaustive presentation but at highlighting a number of issues somewhat neglected in anthropology. I shall use some economic concepts which, while useful to clarify thought, suggest that the field is scattered with traps which we are unable, at the moment, to avoid.

The most systematic and influential exponent of a theory of value in anthropology today is undoubtedly Kluckhohn, who has gone so far as to state that values constitute systems which should be a subject of inquiry just as are cultures and social struc­tures.’ In most cases, however, this degree of systematic abstraction is not achieved, and values are thought of as elements of culture or of social structure.

1 C. K. M. Kluckhohn, “Value and Value-Orien­tations in the Theory of Action,” in Toward a Gen­eral Theory of Action, ed. Talcott Parsons and E. A. Sbus (Cambridge, Mass.: Harvard Univer­sity Press, 1951), p. 395.

In Kluckhohn’s thought two separable ideas are brought together. In this presentation they are distinguished, and a third is added which, however, Kluckhohn rejects.

The essence of the first, the Type A approach, as expressed with great clarity by Nadel,² is that values are ideas about worth­whileness. As Nadel states, the conceptual index may vary (using scales such as “good—bad,” “desired—not desired”). Such values are significant because of a relationship to action or potential action. Kiuckhohn holds that a value is a conception relating to a code or standard. It implies the desirable—but just any kind of desire will not do: it must be justified morally, by reasoning, by aesthetic judgment, or by some combination of these.⁸ Kluckhohn’s Type A notion of value is thus more restricted than Nadel’s; it covers a narrower range of judgments and preferences. But Kluckhohn does explicitly recognize that value implies choice (“selection”).

Now Type A values are largely, as Firth

 

²S.   F. Nadel, Foundations of Social Anthropol­ogy (London: Cohen & West, 1951), p. 264.  

3 Op.cit.

 

556                               THE AMERICAN JOURNAL OF SOCIOLOGY

 

puts it, concerned with ideals or expressions about the dictates of moral obligation.⁴ The expressive feature is often uppermost in  Kluckhohn’s thinking, and he is quite explicit about the means available for identifying them. It must, he says, be possible to conceptualize the value and express it verbally.⁵ The expression will be made either by the actor himself or by the anthropologist who gains the actor’s acceptance or rejection of the formulation. In other words, values are abstract qualities attaching to verbal statements. 

Although Type A values may range widely (including aesthetic judgment, for example), they specifically include, and are sometimes considered to center upon, ethical with approval statements about value in-or moral judgment. Despite this, it is interesting to note that the two most exhaustive recent accounts of the ethical systems and moral codes of a preliterate people make only passing reference to values, and, if reference to values were removed from their argument, the argument would not suffer. (This suggests at least the possibility that we could do without the Type A notion of value, since its reference is covered by other terms.) Brandt holds that common usage demands that values should be standards of permanence and goodness or desirability.° Dictionary definitions do not support his idea of ordinary usage, and it is not at all clear as to how desirability, worthwhileness, and goodness are to be distinguished and related. Ladd, following Kluckhohn, regards moral values as a part of values in general.⁷ Both Brandt and Ladd are interested in systems of ideas. application of his analysis to social systems 

Common usage, however, as found in the  Shorter Oxford English Dictionary, speaks of ethical value in terms of “esteem for its own sake” and “intrinsic worth.” Here at least the social scientist must reject common usage, since ethical judgment involves preferences and the relegation of some alternatives to an inferior position of esteem. So  that even ethical values, as all values of Type A, must relate to judgments about degree and scale. The Type B approach to values is normally present in the same argument that uses Type A. Firth draws attention to Type B values as “social imperatives. . . the basic assumptions of a society.”⁸ Brandt talks of relative permanence,⁹ and Ladd quotes with approval statements about value interests as “wider and more perduring.”¹° Kluckhohn, writing primarily of Type A values in relation to a theory of action, points out their relevance for a Type B analysis. He conceives of Type A values organizing “a system of action,” shows that individual values must adjust to goals, the maintenance requirement, and the interests of the actor, of other individuals, and of the sociocultural system, and adds that “values add an element of predictability to social life.”  Nadel puts the Type B method in a nutshell: “When we take the respective forms of behaviour to be instances of a ‘value,’ we understand that here such-and-such an idea pf worthwhileness is consistently applied  to the various occasions of acting.”¹²  If we include ideal statements or reaction to projective tests as behavior, Kluckhohn’s application of his analysis to social systems falls within this category. For he is concerned with finding consistencies in behavior, relating them to ideas about worthwhileness, and deducing valuational themes which apply to whole cultures or subcultures. (There i of course an obvious danger of circularity here.  If values are identified as persistent themes of worthwhileness, one

 ⁴ Raymond Firth, The Study of Values by Social Anthropologists (The Marett Lecture [London, 1953]). Cf. Man, CCXXXI (1953), 1—8.

 ⁵Qp cit p. 397                                        

  ⁶R. B. Brandt, Hopi Ethics: A Theoreticalysis (Chicago: University of Chicago Press, 1954), pp. 38—39.

 7 John Ladd, The Structure of a Moral Code: A   Philosophical Analysis of Ethical Discourse Applied   to the Ethics of the Navaho Indians (Cambridge, Mass.: Harvard University Press, 1957), pp. 62—63.

⁸op. cit.                            ^{9 Op.} cit.           ¹⁰ Op. cit.   ¹¹ Op. cit.   ¹² Op. cit. (Italics mine.)

cannot explain persistent themes by reference to them, since they are the same thing.)

The method, with its emphasis upon ideas and concepts, and upon the anthropologist’s interpretations of consistency, is a sophisticated and highly technical advance upon the older approaches to themes in culture, to ethos, and to cultural patterns. A few minutes spent with Mead and Métraux,¹³ Bateson,¹⁴ Honigmann,¹⁵ and Benedict¹⁶ reveal that, although there are many important and subtle differences of emphasis, their gross concepts can be translated into Kluckhohn’s framework. A major characteristic common to all these investigators is their search for unifying or organizing gross cultural categories built upon psychocultural premises. Sometimes the new phrasing of the Type B approach seems to move very far indeed from the Type A requirement of verbalization recognizable to the actor. Kiuckhohn’s themes, such as “determinate—indeterminate,” “unitary-pluralistic,” and “autonomy-dependance,” are highly abstract and seem to be interpreted directly from behavior without the intermediation of Type A values or indirectly from behavior through Type A values.¹⁷

In short, Type B is recognizably related to values when, as Firth¹⁸ says of Ruth Benedict, there is “an explanation of dominant traits of civilization in terms of cultural choice.” But when choice and scale are left out, and we have dominant theme alone, we

 

^13Margaret Mead and Rhoda Métraux, Themes in French Culture (Stanford, Calif.: Stanford Uni­versity Press, 1954).

14 Gregory Bateson, Naven (Cambridge: Cam­bridge University Press, 1936).

15John J. Honigmann, Culture and Ethos of  Kaska Society (“Yale University Publications in Anthropology,” No. 40 [New Haven, Conn.: Yale  University Press, 19491).

10 Ruth Benedict, Patterns of Culture (Boston: Houghton Muffin Co., 1934).

¹⁹ C. K. M. Kluckhohn, “Toward a Comparison of Value-Emphases in Different Cultures,” in The State of the Social Sciences, ed. L. D. White (Chicago: University of Chicago Press, 1956).  

have left values far behind, and the reader must perform extra acts of interpretation to think of them. Theme implies values, but the act of thematic model-building removes the variable of choice from the anthropolo­gist’s purview. This is an important way of contributing to theory, but it should not be allowed to conceal the fact that the identi­fication of value should be related to evidence about choice and scale. If this were not so, and we had theme alone, we would not need the concept of value but could rely solely on concepts such as ethos or theme.

The third, or Type C, approach begins with the analysis of individual preferences as expressed through behavior (whether or not verbalized by the actor). Firth’s review of values in anthropology suggests its importance alongside the others.¹⁹ Kluckhohn rejects it and, in rejecting it, describes some of its elements very well:

  A cathexis is ordinarily a shortterm and narrow response, whereas value implies a broader and longterm view. A cathexis is an impulse; a value or values restrain or canalize impulses in terms of wider and more perduring goals. A football player wants desperately to get drunk after his first big game, but this impulse conflicts with his values of personal achievement and loyalty to teammates, coach and university.²⁰

  The Type C approach would regard this as a conflict of possible objectives, reflecting the presence of inconsistent values, each with a different valence and potentiality for being translated into action. The actual choice in the light of behavior would be a reflection of the dominant value subscribed to by the actor, given the specific circum­stances surrounding the action. The approach is neutral between “more perduring and less perduring,” since the emphasis upon more or less in this matter is itself a matter of valuation. If the football player gets drunk despite the opposition of coach, teammates, and university, this would, according to Type C, represent the valuation at that

 

¹⁹ Ibid.

²⁰ “Value and Value-Orientations in the Theory of Action,” op. cit., p. 399.  

time and place and would show that, in spite of the importance the player may attach to these others, there are elements in the situation which cause him to pay the price, such as it might be.

These three approaches to values have been set out to emphasize the importance of scale and degree, on the one hand, and the relationship between worthwhileness and preference, on the other. The concept here advanced is close to common usage and to technical usage in fields such as music, painting, and economics. Value implies worth, and worth cannot be thought of intransitively; we ask, “Worth what?” In its simple form this implies a conception of equivalence, which in turn implies measurement. Measurement by direct equivalence is impossible in many contexts; where this is so, rank order takes its place^.21 This point of view is essential to the legitimate use of the concept in such a way that it does not conflict with other concepts already estab­lished in technical anthropological usage.

We may now turn from this threefold classification to examine the conventional materials available for anthropologists in their attempts to identify the values of individuals or cultures, to place the values in the scale of preferences, and thus to assign an importance and a weight to them.

One set of data which Kluckhohn would emphasize consists of individual statements about worthwhileness of conduct and goals, related to a conception of the world around the speaker cast in a moral frame of reference. In the first place, the field worker can elicit statements of theory (e.g., descriptions of ideal conduct), and he can record comments which imply judgments of worthwhileness in response to normal situations. An example would be a mother scolding her child, or an old man commenting on the be­havior of a neighbor’s son. He can also use projective tests and like techniques to imply a basis of personality, and therefore a frame­work of moral judgment, without the subject being aware of what he is revealing about himself.

This approach is necessary in the search for values, but by itself it does not identify them. Verbal statements can be interpreted as rationalizations justifying one’s course of action despite knowledge that the action runs counter to the mores. They may be used to discredit or approve of another person’s conduct, not on the moral grounds stated in the text, but for some social reason not stated in the text. Nevertheless, precisely because they are rationalizations, they are so formulated as to be likely to appeal to the moral views of at least some of the hearers. This implies that the speaker is reflecting to some extent what he believes to be the current mores of the community (or, in carelessly set up field situations, of the anthropologist). Does this mean that the rationalizations describe values? Certainly, with appropriate interpretation, they reflect something which may roughly be described as the “value tone” of the community; but this is only one side of the medal.

Projective tests circumvent objections based upon rationalization but do not avoid others. Quite apart from the effectiveness or otherwise of the tests in obtaining defined psychological information (a question which I am not competent to discuss), projective tests and verbal statements have certain common merits and disadvantages. They do reveal some potential goals, principles of conceptualization, and ideal ordering among ends. They do not put the actor to the test of his actions, making him responsible for actual choice, with all its consequences visited upon him. Although there may be a connection between ideal values and psychological characteristics, on the one hand, and the valuations of living persons in historical context, on the other, it is not a simple correspondence. Prediction of human choice on the basis of ideal statement or psychological characteristic is, for several reasons, likely only under very limited circumstances. First, each situation being complex and unique, the analyst cannot foresee the effective variables; this observation affects social time

  prediction in general but cannot be allowed to prevent the formulation of theory. More serious is the second reason: both ideal statement and analysis of projective test are abstractions which do not take into account such matters as discounting the pain of future situations and, for this reason, do not with certainty anticipate the likely concrete judgment from the point of view of the actor himself. There is a very close similarity here to parties of political opposition (including those in colonial territories), which, when faced with responsibility in government, may develop markedly different corporate personalities.

A second analysis, which can build upon the former and which is particularly related to Type B values, is to construct a “system of values” for a culture. The statistically or morally normal values of the culture are here based on statements about individuals. Precisely the same reservations need to be made as for the first type of analysis, but there are other difficulties.

In societies with few authoritarian, theological, or theoretical devices, what weight is to be given to differences in ideas and viewpoints? Sometimes it is possible to state that authority leans to one view and that there are sanctions against others. Again, it is possible to introduce deliberate imprecision, indicating that variation of certain degrees is tolerated. Sometimes, as with Gluckman, conflict and disagreement are shown to have functional implications.²² But sometimes the concepts of national character, basic personality, and world view—all of which concepts imply systems of values for cultures—push the exploration much further, seeking dominant principles, even in complex societies. Such examinations are worthwhile, particularly as theoretical models to be argued about, but they are clearly oversimplifications if they are meant to refer to historical societies. The trend toward the concept of modal personality, permitting a spread of characteristics, has not been sufficiently used as an analogy for the treatment of national character, world view, ethos, values, and like concepts which may be distributed throughout culture according to structural, idiosyncratic, or other principles.

A third approach is concerned with an abstraction of the working principles of a culture or the relation of cultural processes to an ideal model. On the one hand, verbal statements or observed behavior are treated abstractly to provide, for example, a structural frame of reference. On the other, verbal statements or observed behavior are related to an a priori frame of reference which describes the normal operation of a society or culture in relation to assumed goals of maintenance, equilibrium, mental health, and the like. In either case an external standard of value may be applied (just as when “more perduring” was used in a different context as a criterion of value). We can speak of an institution as “important” because not only do people adhere to it but it fits in with a social purpose. This teleological functional approach has had its significance for our society. It has countered the thoughtless reformer by providing a basis for the analysis of a presumed purpose in apparently irrational custom and conduct. But it has also led to an arid relativism which seems to assume that all behavior gives evidence of equal value. But the aridity is more apparent than real. It is based on a weak handling of social models. Merton’s concept of dysfunction²³ and Leach’s of a normal disequilibrium,²⁴ though they are by no means ideal correctives, offer promise of new models which at least put the question: When does value exist, and how much? It is necessary to ask this question, for now in social theory (as distinct from points of view about social work) we have to ask ourselves whether this or that institution has a negative function in the workings of society or whether it represents a force opposed to others. But even this approach does not offer a technique for assessing the raw material upon which it is

 

²³ Robert Merton, Social Theory and Social Struc­ture (Glencoe, Ill.: Free Press, 1949).

²⁴ E. R. Leach, Political Systems of Highland Burma (London: Bell, 1954).

based or for identifying particular values to place them in the operational context of the model in use.

A fourth scheme is possible. It focuses primarily upon the identification of particular values (rather than value in general), so that their significance for social and cultural processes may be deduced on a firm base of data. This will be a disappointing scheme, since there are probably as many pitfalls as in any other. It offers one advantage over others, and one only: it endeavors to trace the steps needed in order to state categorically that a specific cultural group places a known value upon a given item. If my conclusion must be that in strict terms we cannot provide the evidence in an operational form, we may nevertheless endeavor to approximate it and improve both analysis and technique. Above all, the statement of complexities may do something to reduce a sometimes overconfident optimism.

The approaches so far considered have centered upon characteristics of individuals or cultures. Sometimes they are expressed verbally, sometimes they are inferred from casual or manipulated behavior, and sometimes they are inferred from or stimulated by custom. One cannot conceive of a value without an expression of the goal envisaged. Thus a moral precept, for example, is important because it expresses a goal; if it did not relate to behavior in this way, it would have no significance as a value. Thus a significant element in the anthropological treatment of value so far is that it purports to elucidate goals. Whether an author sets out to present a “value system” or to analyze the customs, culture, or social structure of a particular people, the end result is a statement of goals. In the case of a value system, more weight, it is true, may be given to the description of an ethical or philosophical set of ideas to which people relate their goals; but, unless the ethics or philosophy do come down to earth in this fashion, we have ideas only and not values.

Insofar as goals are verbalized or inferred from non-situational data (e.g., by projective tests or from dreams, myths, or statements of ethical theory), they are only potentially achievable. The costs of achievement, and the bearing of competitive goals, weighted for value, cannot be assessed accurately from such data. There is therefore one set of goals which are not achieved by means of current behavior. For these one may adopt the standard economic term, “potential demand,” implying that what persons say or think they want, or what the outsider infers they want from unconscious motivation, represents a potentiality of action only which might, in fact, never be achieved. Potential demand is largely represented by hopes, desires, moral judgments, and unconscious motivations.

It is clear that forecasting the manner in which potential demand is or is not translated into reality is an extremely tricky business and is not likely to be accurate or possible save in smallscale units of shortterm analysis, in cultures known to be static (and seldom do we really know this), or in parts of culture for which reward and cost are readily quantifiable and are known and relevant to the actors as well as the observ­ers. Cost is of the utmost importance to the analysis, just as is the competition of values for prepotence, and most treatments of value in anthropology completely fail to take cost into account as a variable of significance.

The role played by cost can be seen even more if we examine the bearing of descriptive field-work studies upon the identification of value. It might be assumed that, since such studies describe what people actually do, that is, the effective demand of a culture, they would reflect current wants and preferences and hence the operation of valuation in a specific cost situation. In other words, field-work studies should describe values for a given place and time.

Whether field-work studies can do this precisely is doubtful, but some progress can be made by recognizing limitations in the current approach. The first requirement in empirical value study is weighting by quantity; it would be tempting to say "measurement" Clearly we cannot measure such  matters as religious satisfaction directly, but sometimes we can determine how frequently persons are involved in initiations or in magical rites, and it may be possible from that to make suitable, though limited, deductions. Fortunately, such detail is becoming more and more a criterion of good field work; but, unfortunately, many writers who deal in values or in cultural themes turn their backs on this kind of weighting as if it were unimportant. The idea, the principle, is the datum to be obtained by the most direct approach, counting smacks of statistics, economics, and materialism.

I do not contend that counting or measurement imply objectivity or that counting and measurement are the essential methods of science. Some problems are beyond our reach; for other problems an undue stress on counting and measurement may be unnecessary or dangerously misleading. And the study of values is the study of some quantitative aspects of cultural qualities. Value implies worthwhileness, which implies degree, which implies scale, which is compounded of quantity and measurement. Furthermore, statements about the characteristics of a population are essentially quantitative, for they imply that the appropriate proportion of the population shares the characteristics. Such quantitative judgments to the required degree of precision may be achieved without actual counting or measurement; but, unless we know and state our margins of error, we are left with subjective inference. Again, this may be useful; however, there is always the danger that subjectivity creeps upon authors unawares.

But measurement, whether exact or inexact, is only another step toward assessing valuation. It involves primary difficulties which would have to be removed before the objective could be achieved. The first is to construct a scale. When economists speak of market value, they indicate measurement according to price (which may be manipulated or constructed in such a way as to correct for inflation and similar distortions in the monetary unit of measurement). For cultural aspects as a whole, such a scale is hardly possible. Physical measurement involves using an artificial device on which the symbols of the criterion are marked; the criterion may be a parallel characteristic (say, length) which applies equally to the artificial device and to the object measured, or it may be responsive to forces resulting from a process set in motion (say, as measured in a voltameter or balance). Economic measure­ment is in a sense responsive, for it occurs as a result of processes; where value is concerned, it is indirect. It is assumed that, the more worthwhile an object (to the market), the more people are prepared to meet the cost per unit. This is precisely the element of analysis which is lacking in anthropology.

In point of fact, this extension of analysis does not go far enough. Market value is a particular subvariant of our genus; it reflects certain valuations, but it does not represent the whole range of valuations, nor is it valuation per se. And we cannot accept money or any other tangible good as the relevant element in price if we are to compare values throughout a culture. There is only one resource which is given up every time an action is undertaken, that is, which can be regarded as a universal cost element in valuation: time. And, paradoxically enough, although economists accept the no­tion that time is a resource, it enters infrequently into anthropological discussions of resources.

But it is not possible, without a great deal of difficult calculation, to convert every element in cost into a time equivalent. Such a computation would mean, for example, stating that an act of religious ceremony involved hours of attendance and performance from the participants, plus the sacrifice of foods which themselves possessed a certain time-price, plus the use of certain ceremonial objects, depreciation of which involved a certain time-price. Clearly, the arithmetic involved would be extremely difficult, but not quite impossible, and at least the steps are available for verbal analysis. It should be pointed out as an aside that this is getting close to a labor theory of value, which has been rejected by most economic theorists; but the analogy is superficial, since the labor theory of value is concerned with the explanation of prices, supply, and demand, whereas our cultural theory of value is an attempt to state what things and modes of behavior are regarded as worthwhile in a culture as a whole and not merely in the market.

There is an obvious objection to the argument so far: no one-to-one relationship exists between the quantity of time given to achieve a goal and its value. The criticism applies to any other form of assessing cost as well and is the reason why market value cannot be taken as equivalent to cultural value. A possible way around this difficulty is suggested by the proposition that an index of value, or worthwhileness, is the degree to which persons are prepared to give up an objective, or to consume more of it, because cost conditions (translated into time-price) have changed. This proposition leads to the notion of elasticity of demand, which draws attention to the responsiveness of demand, or goal achievement, to variations in costs. When translated into cultural terms, it suggests that for every goal there is a ratio between a movement in the quantity of the end product achieved and a movement in the total balance of advantage-disadvantage appropriate in the relevant situations. If an actor wants something desperately, he will be prepared to achieve it in at least the same quantity as before, even though the costs of achievement rise, and this is a reflection of the value of the goal to him; he does this by reducing the actual achievement of other objectives, even though their cost of achieve­ment may have fallen. By following through ratios in these kinds of ways, the anthropologist would lay bare the ramifications of choice and would assess value by showing how much persons are prepared to hold onto the goals they want or to give them up in response to changing conditions. An empirical exercise of this kind would be difficult and would of necessity involve diachronic studies.²⁵

Clearly, the practical difficulties of using the tool to provide neatly measured quanti­ties are insuperable. We simply do not have available the techniques which would enable us to measure units of cost and consumption in such fields as religion and aesthetics, let alone responsiveness to changes in time-price.

Since this is so, what are the conclusions? In straightforward conventional field-work analysis we imply values but do not describe them. Normally, two related but quite separable things are described: effective demand across the total range of institutions and the ideational content of potential demand.

This does not mean to say that we are unable to approach the objective a little further. With a slightly different orientation of field work, we could present new data bearing upon the problems mentioned above, though still without the precision of agreed measurement. In the future, field work should pay much more attention than hither­to to costs of achievement of existing goals and to responsiveness on the part of groups to alterations in costs, for these are essential elements in a concept of value.  

UNIVERSITY OF BRITISH COLUMBIA

 

25 C. S. Belshaw, “Revaluation of Time in a Papuan Community,” South Pacific (Sydney), VI (1952), 466—72, and Changing Melanesia (Mel­bourne: Oxford University Press, 1954), pp. 133— 52.  

 

NOW   check out the elaboration of these and linked ideas in my 

Conditions of Social Performance: an exploratory theory. Routledge and Kegan Paul 1969