A SOCIAL TiECRY OF aCLTURAL CHARIE - ~~~~~~John Howland Rove, University of California, Berkeley Introductory Note This paper was written for the First Annual Meeting of the Kroeber Anthropological Society, held in Berkeley on My 18, 1957. Its publication has been delayed because I wanted time to do some more thinking about the subject and discuss it further with friends interested in cultural change. The. thought and. discussion of five years has had the effect of increasing my confidence in the value of the argument, at least as a productive stimulus to discussion. It would be tempting to undertake a revision based on wider read- ing, but.such an undertaking would require a further delay in publication. The paper is therefore presented substantially as it was read in 1957 with only minor editorial changes. The relationship of imitation to social hierarchy which is explored here was the subject of two earlier papers which I learned about only after mine was. written (Fallers, 1954, and Srinrvas, 1956). The second was kindly called to MY attention by Edward B. Harper. Eugene A. Hammel, who was partly responsible for the formulation of the argument presented here, has done a field investigtion of some of its implications which carries the matter a step further (Hammel, 1962). - J.H.R. * * * One of the most interesting theoretical developments in modern archae- oloW is the general acceptance by archaeologists of the' assumption that all cultures change. I am not aware that this assumption has been stated by any- one. in so many words, but it is obvious enough in archaeological practice. When an archaeolQgist. today starts work in a previously unknown area he looks for and expects.to find a "sequence of cultures," no matter where he is work- ing. All his methods of dating depend ultimately on the determination of changes.in culture'. Changes in culture have now been found in.so many parts. of. the world that it seems to be a reasonable expectation that they w11 even-. tual.ly be found everywhere. The assumption that all cultures change does not, of course, imply any corol-lary that the rate of change is constant or that all cultures change at the same rate. We now have numerous dated examples in which no change can be traced in the archaeological record over a period of several centuries or even millenia. Even in such cases, however, some change ultimately takes place. The general rule seems to be that the more abundant and varied the archaeological record is the more frequent the changes that can be traced. It therefore seems likely that the apparent lack of change over long periods in some cultural traditions is as much a reflection of the poverty of the surviving cultural inventory as an indication of real cultural stability. The assumption that all cultures change poses some interesting theoreti- cal problems. Without indulging in the reductionism of seeking a psychological explanation for this phenomenon9 we can legitimtely look for reasons on a social and cultural level which may provide at least a partial explanation for what seems to be a general characteristic of human behavioro The problem of explaining cultural change recommended itself for discussion in a seminar on culture change and stability which I offered at the University of California in the fall semester of 1956. The discussion group based on this seminar consisted of Frank B. Bessac, Gordon Lo Qrosscup, Eugene A. Hamel, Dorothy Menzel, Ann M. Norsworthy, Octavio Romano, Robert JO Squier, and myself. This group succeeded in formulating several proiising ideas, one 6f which is the subject of this papero This particular idea grew out of a discussion of the role- of social prestige in culture change. I am especially indebted to Eugene A. Hammel and Ann Mo Norsworthy for help in expanding and refining ito In his Les lois de l'imitation (1890, 1895, 1903) Gabriel Tarde repeatedly makes the point that, in a stable social hierarchy, the direction of influence is from the top down, people of less prestige imitating those with more prestige. He comres the movement of influence to a cascade (1895, ppo 92, 396-397). Tarde does not rule out the possibility of a trickle of influence in the other direction, but he is confident that the usual direction of influence is down the scale. In its simplest form, Tarde's generaliztion may be restated as follows: In a social hierarchy, people of less prestige Imitate the behavior of those with more prestige oftener than people of more prestige imitate the behavior of 2eople with lesso For convenience of reference, I am proposing to refer to this proposition here- after as "Tarde's Law." Tarde's discussion of the "Law" recognizes its obvious corollary: Since imitation necessarily involves the movement of traits, beaLvior traits move downward in a soca hierrchy more often than t move upward. It is the downward movement of traits-which Tarde compares to a cascade. In fact, it is not hard to find-examples of behavior traits which must have moved down the social scale because they were formerly characteristic of an upper class group and are now found only on the lowest social levels. The poncho, for instance, was introduced into highland Peru in the first half of the eighteenth century as a garment worn by upper class Spanish horsemen. It is now the distinctive dress of the lowest social levels of the Inca-speaking population. On the other hand, it is much harder to find examples of behavior traits which have moved upward. There are some qualifications to this statement which will be discussed later. Tarde' s Law and its corollary describe one type of cultural change: change involving the redistribution of traits already present in the cultural system of a given social hierarchy. But we can now go further than Tarde did. In the first place, there are some limitations to the operation of Tarde's Law. Sore bebavior traits are more easily imtated by people on a lo'tr 3oCivul level than others. If, for example, social prestige Is correlated with wealth, behavior traits like the ownership of large tracts of land are not likely to move downward in the hierarchy because people of lower status are not econmical- ly in any position to imitate this aspect of the behavior of the high prestige 76 group. On the other hand, patterns of clothing, manners, linguistic peculiari- ties, songs, dances, and stories can be copied with relative ease by anyone. It is behavior of the more easily imitated type which comes most often within the scope of Tarde's law. Another limitation on Tarde's Law is suggested by the generalization that, In a social hierarchy, some features of customary bebavior are identi- fied with social---osition and serve to symbolize it. It is these particular features which are most likely to move down the social scale through prestige imitation. It may be noted that it is more convenient for a high prestige group to have, as behavior symbolic of its status, traits which are relatively difficult to imitate rather than ones which are relative- ly easy. This factor may explain in part the widespread emphasis on lineage and property in social hierarchies around the world, Behavior traits which symbolize primarily local or professional groups, age grades, sex differences, factions, and so forth, may also serve indirectly as symbols of social position, for the groups they distinguish generally form social hierarchies. Thus, age differences are the basis of one scale of pres- tige and sex differences of another. Prestige imitation takes place within each scale according to Tarde's Law. Thus, in the United States, women can imitate men's dress but men cannot imitate women's dress, Closely analogous to- the specialized type of prestige imitation just dis- cussed is imitation between communities. Certain settlements have more prestige than others. The fact that a settlement is noted for some special activity, product, or historical association may confer prestige, usually in a limited sphere. More general prestige seems to be correlated with the gross size of the settlement, except when a settlement has undergone a sudden increase or decrease of population and its prestige status has not had time to adjust to the new sizeO Thus, a town has more prestige than a village and a city has more prestige than a town. One reason for the greater prestige of larger set- tlements may be that the larger the settlement the greater the range of the local social hierarchy occupying it and the higher the social position of the upper levels. In most of Latin America, at any rate, "high society" groups are found only in the larger cities, and the professional people who generally form the next social level are more numerous and better off in the cities than in the towns. The most respected doctors and lawyers, for example, are found in the cities, though there may be some doctors and lawyers in small towns as well. - Tardets Law works between settlements as well as between individuals or social groups. As the urban geographers have noted, every city has a "hinter- land,"" an area in which its cultural influence dominates all towns and smaller settlements. Southern Colombia, for example, ias part of the cultural hinter- land of Quito in the sixteenth and seventeenth centuries. Street names, churches, festivals, and many other features were copied from Quito by the southern Colombian towns during this period. These old borrowings are conspic- uous because the prestige relations have been substantially reversed in more modern times, 77 Prestige, and hence imitation, relationships may exist between nations or societies which are otherwise independent of one another. It is a commonplace observation that France dominated Europe in prestige and as a model for imita- tion in the late seventeenth and early eighteenth centuries, for example. At an earlier period the Byzantine Empire exercised a similar cultural domination beyond its borders in southeast Europe and the eastern Mediterranean. Tarde, and a number of living anthropologists, have noted that traits once.characteristic of prestigious countries are preserved in ones of less prestige, and traits which began in influential cities can be found. surviving in remote villages. This situation parallels the one we found operating within a social hierarchy of individuals. So far, we have explored the implications of Tarde's Law for the redistri- bution of existing traits within a prestige system and gone on to extend the principle beyond its effect in a single social hierarchy. If two or more social hierarchies are in contact and differential prestige relations are set up between them, we can expect that cultural borrowing will take place and that the direc- tion of influence will be from the unit of greater prestige to that of less pres- tige. Since it is generally the people on the higher social levels who bave the best opportunity to travel and establish foreign contacts, much of the borrowing is likely to be on a high social level. Prestige, then, in all systems of human relations, is acknowledged by imitation and borrowing of traits of behavior. The very existence of prestige difference, therefore, seems to be a sufficient condition of cultural change. We can go one step further. In any given social hierarchy social position is symbolized by behavior, as we have said. Now, if the behavior of the people at the top of the hierarchy did not change while people lower down on the scale imitated it, the symbolic advantage of the people at the top would be progres- sively reduced. To maintain their symbolic advantage, people at the top of the hiearchy must adopt new behavior traits different from those of the people immediately below them. They can get the new behavior traits by outright inven- tion, by borrowing them from another social hierarchy, by reviving obsolete traits, or even by taking up traits characteristic of a social level so far be- low them that they feel that it offers no danger to their position. They must, however, get new traits from some source in order to maintain the system. Thus, a prestige system operates as a mechanism for stimulating invention as well as imitation. There are two questions raised by the foregoing discussion which may re- quire comment. One is the relationship of the prestige theory outlined to the social situation in the United States, and the other is the effect of revolu- tionary movements. I interpret United States society as consisting not of a single social hierarchy but of many. In fact, our society is quite deliberately organized to encourage and reward as many individuals as possible by providing many different kinds of prestige which are not correlated with one another. A bank clerk with no hope of social advancement in his profession may become a fine golf player or reach a position of honor in his lodge. In this type of social system the 78 pattern of prestige and imitation is naturally very much more complicated than it is in a simple class system in which there is only one prestige scale. Within each social hierarchy the patterns which we have described will be found to hold, however. Social revolution can be defined very neatly in terms of prestige theory as a situation in which indiv'iduals of relatively low social status who are not entitled to high prestige positions under the existing rules nevertheless succeed in occupying them. A coup d'etat, in which no change in the social hierarchy occurs, is not a revolution in this sense. A social revolution must be carried out at the expense of the people already occupying the high prestige positions, so these people are the natural enemies of the revolutionists. In order to consolidate their position, the revolutionists must set up a new social hierarchy and a new prestige scale in which they can impose a subordin- ate position on their late rivals. It is natural for them to seek new behavior traits as prestige symbols, and hence all revolutions which qualify under my definition are occasions for wholesale innovation. If the revolutionary move- ment has strong support from one of the lower social levels, it may be conven- ient for the revolutionary leaders to try to give social prestige to certain behavior traits chracteristic of the social position of their supporters, thus creating an exception to the orderly operation of Tarde's Law. It may be noted that "revolution" in this sense does not need to be a political movement. In his own particular social hierarchy, Elvis Presley was a revolutionary. Our exploration of the effects of social prestige on cultural patterns has thus provided us with exactly the explanation that we needed for the phen- omenon of constant cultural change. Wherever we find a social hierarchy we must expect change to occur. It remains to be determined whether all human societies consist of hierarchies in this sense, but a very large number of them certainly do. It is also worth noting that this theory explains only the fact of change. It does not require that the change be in any particular direction or order and is thus not an evolutionary theory in the usual sense. The question of order in culture change is another problem altogether. BIBLIOGRAPHY Fallers, Lloyd A. 1954 A note on the "trickle effect.' Public Opinion Quarterly, vol. 18, no. 3, Fall, pp. 3114-321. Princeton. Hammel, Eugene A. 1962 Wealth, authority and prestige in the Ica Valley, Peru. Univer- sity of New Mexico Publications in Anthropology, no. 10. Albuquerque. Srinivas, Mysore Narasimhachar 1956 A note on Snskritization and Westernization. Far Eastern Quarterly, vol. XV, no. 4, August, pp. 481-496. Ann Arbor. 79 Tarde, Gabriel 1890 Les lois de l'imitation; etude sociologique. Felix Alcan, Editeur, Paris. 1895 Les lois de l'imitation; etude sociologique. Seconde edition, revue et augmentee. Felix Alcan, Editeur, Paris. 1903 The laws of imitation. Translated from the second French edition by Elsie Clews Parsons, with an introduction by Franklin H. Giddings. Henry Holt and Company, New York. 80