Chapter 1 (pt.1)


The Question of Ancestral Authority

A Traditional Social Order

The Evidence for Stability

The Units of Social Organization

I say that the ancestors are dead. True, a Taleng might say, but they have authority over everyone. If we are both right, what does it mean?

The Tale social order is based on tradition, but then, tradition is based on the social order. Thought circular in form, this statement is not just a tautology. Any modicum of social order is the result of a complex interplay of many factors, acting on each other through many paths of influence. These paths are not just artifacts of an analytical model, for changes in quantity on each variable occur in time, affect others in sequences and over varying durations, alter that synthesis which we call social order. Being the interplay of many variables, social order itself is a complex variable. The Tallensi have a great deal. Americans and Englishmen have less.

We are concerned, not with a broad historical generalization that social order prevails but with the analytical problem of identifying the social mechanisms which operate to produce a greater degree of social order than would obtain, if these mechanism were not called into play. (Merton, 1957, p. 371)

The question of what ancestral authority means for society gives this thesis its theme. The structure of its argument is given by the attempt to construct and explicate a model of the rise and fall of social order, the rates of its coming and going and occasionally, as in Taleland, its staying for a very long time. I think that though it may not succeed equally in all it attempts, this thesis does show that social order may be understood as a complex and meaningful socio-historical phenomenon, and not simply as one of a dozen weak synonyms for society.

The question of Ancestral Authority: In traditional Tale social thought all authority is vested in ancestors, fathers of at least two generations remove, usually dead, and significant in being points of genealogical unification and differentiation. Though living persons do have authority in some matters, it is always be assumed devolution from the ancestors and is never absolute. It is also determined by reference to the same genealogical organization as that of the ancestors. The authority of ancestors is thus morally pervasive. It acts not only in the abstract moral ideas of individual men and groups, however. It is concretely manifested in their interactions and indeed in their very identities. All corporate groups in Tale society are defined in the genealogical organization by reference to the ancestor who begot the, that is, the ancestor who has authority over all of them and over no one else. Ancestors are seen as active in the affairs of men, as responsible for the eventualities of fate, but their importance is not limited to 'action' or their explanatory function. It is strongest as an aspect of social organization.

The authority of ancestors is the key to the operation of a traditional social order among the Tallensi for several reasons. First, it fixes moral authority in the past, directly in tradition and in the assumption that the past is similar to the present. Second, it fixes authority in men who by reason of their decease can no longer take direct action except 'after the fact'. In other words, ancestors have all the authority, but no intentional power. They have only the power of fate (fate of course being in part a function of the social order.) This limits the upset which authority could bring to the system and ties it directly to the consensual basis of divination. Third, the monistic (single principle) organizational system by virtue of which the ancestors have authority provides for a high degree of congruency or consistency in Tale culture and society. Since all groups are defined by the same criteria, rules of transformation from one level of corporation to another are implicit and conflicting definitions of situations are minimized. Further, this monism of organization makes for a high degree of multiplexity of social bonds, as it ensures that the same social groups will be relevant for an individual in (what we would see as) different facets of his life. Let us look now at what we mean by a traditional social order, since this is the context in which we are considering ancestral authority.

A Traditional Social Order: The word traditional is used quite straightforwardly in this phrase, but it is important to specify exactly what we mean by social order before the meaning and conceptual context of the phrase as a whole can be made clear. At its crudest, social order as we use it refers to a quantitative variable of the extent to which sociation (that is patterned social interaction based on at least one-directional constructions and anticipations) is orderly, or, in other words, predictable. The extent to which accurate social prediction is possible is at the heart of our conception. Of course, this immediately implies a measure of stability as well. Social order, as implied above, is a function of a number of variables, each influenced by some of the others and, presumably each by traditional influences from outside the model. The eleven variables which make up the model presented formally in Chapter 8 are as follows:

1. homogeneity-heterogeneity of population

2. multiplexity-monoplexity of social bonds

3. solidarity-weakness of social groups

4. inclusivity-discreteness of group structure

5. long-short term of planning

6. assumed adequacy-inadequacy of information by social actors in decision processes.

7. similarity-dissimilarity of definitions of situations among social actors

8. centralization-decentralization of power

9. accessibility-inaccessibility of ends

10. stability-instability of social relations

11. authority-anomie

It should be noted that all of these variables are seen as implicitly including the specification 'among a given population'. How the population is defined is not specified in the model, though clearly the utility of any discussion will be affected by the definition of population employed and the explicitness of criteria involved in the decision. All of this is in accord with Fortes' suggestion that our uncritically assumed notion of societal boundaries should be abandoned, and

we must substitute the concept of society as a socio-geographic region, the social elements of which are more closely knit together among themselves than any of them are knit together with social elements of the same kind outside the region. We must offer a relative and dynamic concept for an absolute and static one. (1945, p. 231)

It is clear how relativity and absoluteness are opposed in this statement; it is not so clear how one concept is dynamic and the other static, except that the former might refer to using different definitions in different situations. Thus, using the "dynamic" (i.e., relative) way of speaking in our everyday discourse we refer sometimes to British society (as opposed to French) and sometimes to European society (as opposed to American and sometimes to Western society (as opposed to Asian).

This usage is very closely in accord with the organization of segmentary societies, particularly their bonds of contraposition. An often-cited Arab proverb sums this up, saying "I against my brothers; I and my brothers against my cousins; I, my brothers and my cousins against the world". It might be argued that such a usage is limited in utility to applications to segmentary societies. I think not. Specifically, I think that the usage as elaborated (ie., as a variable) says something very important about social order, and social order is very closely related to a useful definition of society as a variable (see below). Such a hierarchy of inclusive identifications is quite important to the maintenance of a social order, and not just the Tallensi or other segmentary orders. Such a definition becomes less obvious precisely as social order becomes less. In many large scale but weekly integrated populations only very broad and very personal identities are fixed. In other words, there are nations, multi-national firms, and individuals, but little in between. On the other hand, such a definition is also less obvious in isolated and/or very small and discrete populations, as are characterized of many small hunting and gathering societies which are organized into bands. As Fortes comments (following in part various others; citations not reproduced here):

it does seem that unilineal descent groups are not of significance among peoples who live in small groups, depend on rudimentary technology, and have little durable property. On the other hand, there is evidence that they break down when a modern economic framework with occupational differentiation linked to a wide range of specialized skills, to productive capital and to monetary media of exchange is introduced. Where these groups are most in evidence is in the middle range of relatively homogenous, pre-capitalistic economies in which there is some degree of technological sophistication and value is attached to rights in durable property. (1953, p. 76)

The mention of rights in durable property brings out the fact that 'traditional' here means something more than a nostalgic affection for the past. In Tale society the working of this other aspect - an emphasis on continuity in transmission of property - is evident in such areas as land tenure and redistribution. But tradition is not sheerly a matter of economic benefits, whether intended or unintended. It could equally be seen in terms of religious or political continuities. It is under the former category that ancestral authority has generally been treated, usually described as 'ancestor worship'. I have not followed this practice. I think, in the first place, that worship is an inadequate description of the relation of men to their ancestors. It is perhaps misleading in connotations of Western devotional practices, but more than that it tends to narrow the scope of analysis which is brought to bear on ancestors in the social systems to which they are central. I have chosen rather to treat ancestral authority as simply (or perhaps complexly) social, taking this as the broadest category. Politics, that is the mechanics of power in defined populations or 'publics', is of course important to this analysis. But it is not central. Politics, however defined, cannot provide the key to understanding Tale social order. This must come from the analysis of the multiplex role of ancestors. Indeed, the nature of Tale politics could hardly be understood without reference to the other variables in our model of social order.

In taking a traditional social order as topic of study, I have also eschewed much of the sort of analysis which Goody brought to bear on the neighbouring Lo Daaga (1962). I have attempted wherever possible to limit reference to individual/experiential variables and perspectives. This is not because these are not productive or relevant, nor is it because I believe in any model of an isolate and independent sociology. Rather, it is in order to try and show how the social order itself operates, including the ways in which such a social order may contain inherent tendencies to change and/or implicit lines on which changes will take place. It should be noted that Goody's analysis is much more than psychological, including for example incisive consideration of the reciprocal influences of differing property relations and ritual practices relating to death. Neither death nor its relation to escatology are the foci of our concern, however. Death becomes relevant as the condition for certain manifestations of the ancestral authority system in social action. To wit: some mechanism among the living must operate to allow the ancestors a voice. Clearly escatology influences social action, just as do other ideational or sensate constructions. But equally clearly they are not the same.

The adjective 'traditional' clearly characterizes the social order of which we speak through reference to people's minds. Tradition involves sentiments relating to the past. I contend, however, that traditional social order cannot be understood by primary reference to sentiments, or by any analysis of "efficient causes" (cf. Homans and Schneider, 1955). To advocate an analysis based simply on efficient causes would be to advocate the sort of myopia with which any individual must necessarily view the workings of his own society. Among the arguments in this book, a central one is that social events take place over greatly varying 'objective' durations. That is, put crudely, some things happen within the range of individuals' vision and some things do not. This is true quite irrespective of the very obvious differences in ranges of individual vision. Social institutions, such as unilineal descent, are particularly unlikely to be historically the result of 'asoption' by consciously deciding individuals. This is the argument which Homans and Schneider put forward in opposition to their erroneous characterization of Lévi-Strauss's (1949) position. Their misreading of Lévi-Strauss seems to follow, however, on the conclusions which they wish to draw, the recommendations which they wish to make for the future of comparative sociology and social theory. These recommendations and conclusions are individualist in bias, and geared toward a maximization of ends analysis complete with assumed ability of individuals not only to identify but to achieve their ends. Their stance suggests that both the extent of individuality and the extent of maximization of ends are social constants. Part of the burden of the present treaties is to show that they are not. In particular I suggest that as individualization increases and persons turn their 'maximizing attention' more to their personal ends such long-term social institutions as unilineal descent must be destroyed, or at least removed from their position of social importance. Paradoxically, the very social institutions which Homans and Schneider set out to study have as an essential characteristic the limitation of the individuality on which these authors base their analysis. The stronger the unilineal descent system, the less applicable Homans' and Schneider's argument. The "psychological preoccupation" for which Needham faults them (with devastating criticism, 1962, p. 126) is largely the result of their assumptions - and they are nothing more than assumptions - about the nature of man. Though they wrote nearly two centuries later, they shared the fault for which Marx criticized natural man theorists and utilitation economists alike: the assumption of the state of man in alienated capitalist "civil society" as man's universal and inevitable nature.

The primary topic of our study is, then, the internal workings of what we have called 'social order'. In particular we are concerned with a single example of social order: traditional Tale society. Though the first part of the thesis especially (Chapters 2-4 and most of 5) the writing is directed to detailed explication and analysis of this social organization. Though this part of the work is clearly influenced by and organized in accord with the more theoretical argument of the latter chapters, it is presented without extensive theoretical references. This is done primarily for reasons of clarity. These chapters attempt to set out the information necessary to an understanding of Tale social order, but also to set it out in such a way as to implicitly make some of the points which will be made explicitly in the theoretical chapters. The rest of this introductory chapter will be devoted to setting out such of the overall argument - in capsule - as to make possible the most productive reading of the early empirical chapters. First we shall look at the evidence which exists for the stability of traditional Tale social organization. It should be remembered that when we speak of Tale society, unless otherwise specified, we speak primarily of its operation, as much as we know it, up to including the 1930's. This is not an analysis, however interesting and relevant to our argument that would be, of its change after that time. Some references to such later information as is available are made, but they are supplementary. From what is written about the Tallensi today, it would seem to me that only an analysis of regional and national organization could approach adequacy in dealing with changes in social order. As Fortes long ago emphasized, social change in small-scale societies tends to be not so much a matter of outside influences as of incorporation into larger units (1936a).

The Evidence for Stability. Since at many points in our argument we assume Tale social order to have been relatively unchanging over long periods of time, we should here note some of the evidence for taking this stance. Fortes makes it clear that he found sufficient evidence to satisfy him:

I was impressed, in the field, with the apparent stability of Tale society over the historical time span consciously recognized by the natives, that is, five or six generations, and probably for a much longer period. I have tried to show this is associated with the lineage system. (1945, p.ix)

Fortes found further evidence for this in the absence of a native conception of historical time. Tallensi

do not think of the lapse of time as being associated with cumulative changes in their culture or social structure, but rather as a periodical or cyclical rhythm of eternal repetition. (op. cit., p. xi)

For many of the arguments advanced herein it is sufficient that Tallensi do not plan on the basis of short-term change. In other words, where we are concerned with social action we need to ascertain primarily the constructions which actors have or are likely to have of their social situations. It is thus important that Tallensi assume stability. It contributes for example very directly to long-term planning. Objective stability is also important, however.

It is difficult to do more than comment on the general stability which impressed early observers and the (again general) similarity of their accounts to later ones. The most important of the early accounts were those published by Cardinall (1921) and Rattray 91932), of which the latter was much more comprehensive and accurate. Cardinall's account figured later in Worsley's (1955) attempt to show that the breakdown of Tale kinship structure was already in progress when Fortes did his field work. Fortes himself specifically discounted the veracity and reliability of Cardinall's account (1945, p. 6). Other writers on the region such as Labouret (1931, considered more reliable by Fortes,) offer little material on the Tallensi. Beyond the overall similarity of the structural outline of Tale society through the first third of the twentieth century we must simply observe that the extent of change is debatable and indeterminate. We can note some of the changes, but it is impossible to assess with the substantial degree of accuracy or reliability the extent of their significance.

The single great area of change we know to have been in progress before British colonization was the population movement which results in the "Namoo-Talis split" which Fortes describes. This is essentially a distinction between self-proclaimed (and mutually recognized) indigenes - the Talis - and newcomers - the Namoos. Similar splits occur in other West African groups and within other populations in the immediate vicinity of the Tallensi (see for a related discussion Goody's analysis of the directional tendencies "Lo" and "Dagaa" among the LoWiili and LoDagaba (1962). How much Namoo-Talis is directional and relative or more categorical, as Fortes describes it, is uncertain). The Namoo immigrants were originally closely linked to the Mamprussi and their chief at Tongo was appointed by the Paramount Chief of the Mamprussi. In time this tie became less active, and by the end of the last century

the Namoos were cut off from Mamprussi and lived on a basis of peaceful equality with the indigenes, speaking the same language and co-operating with them in both ritual and secular matters. (Hart, 1974, p. 4)

As Hart goes on to note, it is clear that in terms of relative population, the Namoos were on the increase. Their settlements proliferated widely while the Talis remained a relatively small cluster. Eventually trade associated with the ritually important B?????? of the Tong Hills (controlled by the Talis) gave them considerable economic weight. This way, however, centered in the person of the Galibdaana, and when he died leaving a large fragmented family, as Fortes predicted, the Tong Hills pilgrim trade ceased to be a major source of economic power (Hart,1974, p. 30). Local political struggles have turned largely on national representation with the foci of dispute being Tong and Ba'ari (Hart, op. cit., p. 31). This weakens the significance of Fortes' major distinction between the Tongo Namoos and the Tong Hills Talis. Whether he over- or under-estimated the significance of the split is a matter for further research, debate and speculation. I do not think the issue is central to the present analysis.

The changes which took place in the early twentieth century up to the time of Fortes' field work were primarily associated with British rule. There were very few missionary inroads into Taleland by 1936 (1945, p. 12). More significant was the extensive armed attack made on the Tong Hills in 1911 resulting in massive population displacement. Some three hundred to five hundred families were resettled in low-lying areas by British order (1945 p. 12; Hart, 1974, p. 7). Armed 'self-help' was also eliminated by British rule (see discussion of collective responsibility and liability, Chapters 3 and 5). 'Native chiefs' and headmen were instituted with powers never held by traditional officials. Though this created some of the imbalance which led to greater differentiation in later years. Fortes argues that its economic influence was still kept largely in line by radically redistributive inheritance in the 1930's. As we have seen, some of this influence still persists today. The two remaining influences were that Taleland began to be bypassed by new traderoutes where it had been directly connected to old (Hart, 1974, p. 8), and migrant labour became extensive (Hart, 1969, 1974, 1975). Fortes reports administrative estimates of 10% of the adult male population absent at any one time in 1936-37, and 7% in 1938-39 (1945, p. 11). This was clearly only a beginning. Early migration was often to the cocoa plantations (see Hill, 1963 for a discussion of migrant cocoa-farmers in southern Ghana) and to the Gold mines of Ashanti. In later years migration became more extensive to urban areas. In both cases migrants tended to be cyclical and temporary, never losing their ties with home. By 1968 the absentee rate in villages was over 50% (Hart, 1974, p. 24), and it was obvious that despite the continued "vitality" of Tale patrilineal institutions which Hart argues, the Tallensi were very largely integrated into wide economic ranges of influence. Their 'system' was now 'open'.

It can only be guessed how much this affected Fortes' account. I do not think that his picture was either substantially inaccurate or substantially different from what it would have been a few years earlier. In only one respect would the basic factors in our model of social order have changed radically. This is the polarization of wealth. There were rich men in the 1930's on a scale previously impossible. The influence of this both Fortes and I have noted, as well as the factors which kept this influence from being greater than it was/is. In sum, the Tallensi were stable, but not without change, before the time Fortes studies them. Seeds had been sown for radical changes, however, and the 1930's were the last era in which this sort and extent of stability could be adduced.


The Question of Ancestral Authority - A Traditional Social Order - The Evidence for Stability - The Units of Social Organization

Previous chapter - Return to Contents - Next chapter- Return to Ancestors Page- Go to The Virtual Institute of Mambila Studies -online