We are left with the critical question, which I have so far evaded, but which must at length be faced: why piety? or in language which may sound a bit old fashioned to-day, what is the function of piety in the context of the kinship and religious institutions I have described?

Piety is a word packed with ambiguity, for us, and not altogether free of derogatory associations. If, as the Oxford English Dictionary tells us, it commonly stands for 'habitual reverence and obedience to God (or the gods)' and 'faithfulness to duties naturally owed to parents and relatives, etc.', it also carries overtones of hypocrisy and to quote the same dictionary, of 'fraud and the like practices for the sake of religion'. This is no doubt an understandable ambiguity in a culture which deems it specially meritorious for outward forms of manners and conduct to match inner states of sentiment and belief. Prevarication and hypocrisy are not, however, confined to our civilization. People of good repute among the Tallensi speak of such practices with contempt. At the same time there is little or no questioning of the sincerity of outward conduct. Morality is what is seen in a person's conduct and actions and these are deemed to be expressions of genuine intention, feeling or belief. This is assumed in Tale religious custom and ritual practices, as it seems to have been in the cultures I shall presently refer to, those of ancient Rome, and of traditional China.

To avoid the flavour of unctuous conformity carried by the English word, I venture to use its Latin ancestor, pietas, in this lecture, though not without hesitation. Is it not redolent for many of us of long hours of wrestling with the tedious affairs of the Pious Aeneas? ' We are wearied', confesses even the great Virgilian scholar, John Conington (1872, vol. II, p. 11), writing in more pious times than ours, 'by being constantly reminded of his piety.' Significantly though, he adds that this 'may be partly owing to our misapprehension of the epithet', for Aeneas's piety 'is not merely nominal; it shows itself in his whole feeling and conduct to the gods, his father and his son'. What adds to our understanding of this 'luckless epithet', as another commentator calls it (Irvine 1924, p. 103 n.) is the observation that it is never applied to Aeneas in the Fourth Book while he is Dido's lover but is restored to him when, to quote this commentator, 'pietas has conquered self' and he leaves her to inspect his fleet. ⁴ This marks what Warde Fowler (1911 lect. xviii) describes as the 'taming of his individualism' in the interests of the state; for, he continues, 'pietas is Virgil's word for religion and religion (not knowledge, or reason or pleasure) is the one sanction of Aeneas's conduct'.

The Tallensi do not have a concept for that complex of reverent regard, moral norms, ritual observance and material duty in the relationship between parent and child, more particularly of son to father, both during the lifetime and after the death of the parent, which I am calling pietas. But they would readily understand the Roman ideal alluded to in Conington's apology for Aeneas. A modern authority contrasts pietas as 'concerned with the circles of family and kinship' with 'fides' which 'pertains to the extra-familial', that is the political, side of Roman life. Originally, we are told, it meant 'the conscientious fulfilment of all the duties which the di parentes of the kin group demand'. Later it meant both the dutiful discharge of cult obligations to the divine 'members of the kin group' and 'reverence and consideration towards the living, human members' (Pauly 1950, s.v. Pietas). An apocryphal story, recorded by various writers of antiquity as an example of superlative piety, and quoted by most commentators, tells of the daughter who kept her aged father alive imprison with milk from her breasts (Roscher 1902, s.v. Pietas). This story would strike Tallensi as bizarre but not fantastic. They would applaud the praise implied in the 'luckless epithet' for Aeneas's filial love and would comprehend he definition of pietas as 'dutiful conduct towards the gods, one's parents, relatives, benefactors and country' (Lewis & Short 1880). And they would agree with the significant implication that it belongs to the realm of parenthood not to that of marriage and sexual life.

This need not surprise us. Considering only Fustel's account and overlooking the differences due to the more complex civilization of the Romans, we can see that their patrilineal descent and patriarchal family system, closely bound up with an ancestral cult, has close parallels with those of African peoples like the Tallensi, the Yoruba, the Thonga, and many others who also have segmentary patrilineal lineage systems and patriarchal family structures inextricably tied to ancestor cults. For the nuclear element in all these systems is one and the same phenomenon, the ambivalent interdependence of father and son in the nexus of final authority versus subordination, identification by descent versus division by filiation, transient possession of paternal status versus its inevitable and obligatory supersession by the filial successor.

However, by all accounts ⁵ the patria potestas of the Roman father in the domestic domain was far more absolute than is that of a Tale, or any other patrilineal African father. Age made no difference, as the example of Anchises in the Aeneid shows. Fowler (1911, p. 414) notes that he is the 'typical Roman father' maintaining his authority to the end of his life, 'to whom even the grown up son, himself a father, owes reverence and obedience.' ⁶ I do now know if Roman sons were compelled to practise avoidances towards their father, but if Cicero's attitude in De Officiis is typical it suggests considerable shyness and formality in their personal relations. ⁷ It is difficult to believe that there was not a good deal of latent antagonism in the relations of sons to fathers. The savage punishment traditionally prescribed for parricide and stories of fathers who sacrificed sons for the public weal ⁸ lend colour to this influence and fit in with the rigorous jural sanctions that supported paternal authority. It is understandable, therefore, why the right of filial succession was not only very strictly entrenched, but was in part enforced as an inescapable obligation to the familia in law and to the ancestral deities in ritual. One can easily imagine the legal constraint and moral compulsion that was required to bind a son in loyalty to his father, until the time when, as Maine puts it, the paternal power was extinguished by his death. For not till then was the son jurally adult and autonomous, which meant ipso facto having the capacity to officiate in religious rites, as Aeneas does in AeneidV, in commemoration of his father's death. Nor is it without significance that this should be in his first important religious act.

In the same way, Tale norms of pietas resemble those of Confucian ethics in China, ⁹ if allowance is made for the refinement and elaboration added by literary transmission through the agency of specialists and scholars. Tallensi would accept the Confucian ideal of pietas as consisting in 'serving one's parents when alive according to propriety; in burying them when dead according to propriety and in sacrificing to them according to propriety', to follow Douglas's translation (1911, p. 119) or 'according to ritual' in Waley's version of the Analects (1938, p. 89). For propriety and ritual are overlapping categories among the Tallensi, too. We have already seen that Namoos define the avoidances of first-borns as taboos, whereas Talis say they observe them out of propriety; an this is typical of many Tale religious ideas and practices.

The common ground lies again in the basic similarities between Tale patrilineal descent group and family organization, with its religious projection in ancestor worship, and that of the Chinese. Granted the differences in range and scale due to the greater complexity of a literate, economically and social differentiated, technologically sophisticated, culturally wealthy and historically oriented civilization, these similarities are noteworthy. Classical treatises on ethics and ceremonial (Douglas 1911; Waley 1938; Grube 1910; De Groot 1910) ¹⁰ are echoed in the field observations of sociologists and anthropologists of to-day. All testify to the Chinese veneration of pietas (hsiao) as the supreme virtue in the relationship of children to parents, especially of sons to fathers, both in life and after the elevation of the father to ancestorhood.

Filial piety, in the modern Chinese community described by Hsu (1949, p. 207; and see Freedman 1958) is said to be the 'foundation stone' of its social organization and is given exactly the same sense as in my quotation from the Analects. But what is more to the point is the weight given by recent students, also in accord with classical treaties, to the patri-filial nexus in the Chinese family and descent system. 'The basis of kinship is patriliny,' states Hsu (1949, p. 58), 'and the most important relationship is that of father and son. The father has authority of life and death over a son, and the son has to reverence and support his parents. Mourning and worship after the death of the parents are integral parts of the son's responsibility.' Tale fathers do not have unrestricted authority of life and death over their children, but with qualifications this formula would be acceptable to Tallensi.

Furthermore, it appears that in China, too, the first-born son has a unique place in the sequence of the generations, though it is not, apparently, marked by avoidances. Thus it seems that in former times 'the eldest son being the direct propagator of his father's line, had the sole right to make sacrifices to deceased parents. This was associated with pre-eminent rights of inheritance in regard to property and ancestors' (Sing Ging Su 1922, p. 37). Judging from the references to primogeniture in recent literature, this rule still prevails (cf. Freedman 1958, p. 82). We learn from Dr Lin's fascinating story (1948, chap. xii) of the fortunes of a Chinese lineage that the first-born has a 'legal right to an extra portion of the joint property as a special recognition of his primogeniture', and, incidentally, that this may give rise to serious conflicts in the family.

The general impression one forms is that fathers treat their sons with affection and indulgence during infancy, but with increasing authority and formality as they grow to manhood. The emancipation of jural majority, economic independence and the ritual autonomy demonstrated in the right to perform sacrifices to the ancestors comes (as in ancient Rome) at length only after the death of the father, as Hsu specifically states (1949, p. 209). This coincides with the dead father's establishment as an ancestor; and it is noteworthy that tablets dedicated to ancestors are so arranged that those of fathers and sons are on opposite sides of the ancestral hall, successive generations being thus kept apart after death, as they were divided by degree in life, and alternate generations being grouped together in accordance with the well-known principle of the merging of alternate generations ¹¹ (Granet 1951, pp. 86 seq.).

I have digressed from Africa to look briefly - and I fear too superficially to satisfy the experts - at the two civilizations which are most renowned for the exalted place accorded to the rule of pietas in their schemes of moral and religious values. These are the paradigmatic cases, often discussed by scholars. ¹² They have the advantage that more or less formulated doctrines can be examined to see what is meant by the concept of pietas. But what is most instructive in comparing these paradigmatic institutions with their relatively amorphous Tale - and I believe, more generally, African - counterparts is to consider the reasons for their indubitable efficacy amongst all these peoples, irrespective of how much explicit doctrine there is. It is unnecessary to go farther afield in order to propose some hypotheses. Indeed even if I wished to explore the African data more fully there would be little profit in it. For I know of only one modern study of an African religious system in which the observance or neglect of pietas has received particular attention and that is Dr John Middleton's impressive work on Lugbara ancestor worship (Middleton 1960).

Among the Lugbara, as among the Tallensi, the Romans, the Chinese, and all other ancestor-worshipping peoples, a man becomes an ancestor when he dies not because he is dead but because he leaves a son, or more accurately, a legitimate filial successor, ¹³ and he remains an ancestor only so long as his legitimate lineal successors survive. This goes with the rule that ancestors have mystical power only with respect to their descendants and not, for example, with respect to collateral kin. On the other side, a man has no jural authority in his family and lineage, whatever his standing may be in wealth or influence or prestige, if he has no ancestors and until he acquires the status which permits him to officiate in the cult of his ancestors. For, as Dr Middleton demonstrates at length, authority comes not by delegation from those over whom it is exercised but by transmission and assumed devolution from ancestors. That is why jural authority can be acquired only by succession, in these systems.

But let us consider the paradoxes in these requirements. To become an ancestor a man must have sons; hence the inordinate value attached to male offspring; Tallensi say that a man who dies sonless has wasted his life, and the Chinese, according to Hsu (1949, p. 77), compare him to a tree without roots. We might well ask what deeper motives underly this profound desire for sons, but that would be an unwarranted digression. All we have to note is that sons are desired and needed so that the apotheosis of ancestorhood may be attained; but it can be attained only by so cherishing sons that they eventually supplant one. On the other side, legitimate status in family, lineage, and community can be acquired only by being legitimately fathered; but jural autonomy, which is the source of authority and power even in the domestic domain, can be achieved only upon the death of the father any by assuming his mantle - quite literally so among the Namoos.

But that is not the end. Purely descriptively considered, as Dr Middleton acutely observes, both as father in his family and as elder in his lineage, the man who holds authority in the name of his ancestors is, by that token, subject to their authority and, in the last resort only to that authority. But is is not the same kind of authority as that of a living person over his dependants. It is imputed to account for things that happen to and amongst his descendants.

So we have the paradox that a man may desire to be allotted responsibility towards the ancestors for ills that befall himself and his family, since this is evidence for all to see that he is directly subject to ancestral authority and by that token jurally autonomous, hence entitled to exercise secular authority in family and lineage affairs as well as to officiate in ritual service of the ancestors.

I must resist the temptation to expatiate on this point, and return to the question why pietas? and I shall begin with a proposition that must here be stated dogmatically, though there is ample evidence in the literature I have cited to justify it. In the type of social system we are discussing, at any rate within the domains of kinship and descent groups, jural autonomy and authority are highly prized, indeed the most highly prized capacities a man can aspire to, since he cannot reach full adulthood without them. They are, in a sense, scarce goods, since they are attached to exclusive genealogical positions in a descent group. But jural authority is also indispensable to the organization of society; indeed it is the very heart of the social system. And this must be the reason why authority never dies - must never be allowed to die - though its holders of a given time have to, by the laws of nature. I argue that jural autonomy and authority are attributes to fatherhood. Indeed they issue solely from paternal status, in this type social system. We see this at all levels of social structure, for lineage eldership presupposes paternal status is not only the kingpin of the social structure in patrilineal systems; it is, among the Tallensi, deeply embedded in each person's life experience by upbringing and through daily cognizance of its existence.

Thus when we say that jural authority never dies and must not be allowed to die we can translate this to mean that fatherhood never dies and must not be allowed to die though fathers in the flesh have to die. There is, of course, a very tangible sense in which fatherhood never dies where patrilineal descent is the governing principle of social structure. For as long as a man's descendants last, their place is in society and their social relations to one another and to the rest of society are ordered by reference to his paternal career. They are himself, replicated by social selection as well as by physical continuity. But I am here concerned with the moral and religious representations of this fact. From this position, ancestorhood is fatherhood made immortal, in despite of the death of real fathers; that is to say, it is paternal authority, above all, that is made immortal and impregnable in despite of the transience of its holders. But fatherhood which confers the capacity for authority is worthless without its antithesis of sonship; and sonship is meaningless without the right to attain the coveted status of fatherhood. So we see father and son bound to each other in ineluctable mutual dependence, one might even day in tacit collusion, to maintain this precious value, yet inescapably pitted against each other for its eventual possession, as Tallensi recognize.

In this relationship the sons are at a disadvantage, being under the authority they must support, and what is more, restrained by the premise of kinship amity which outlaws strife between kin (Fortes 1949). Nor must we forget that fathers do love and cherish their sons, and at heart most of all their first-borns, as is apparent from the grief of a father whose son pre-deceases him. Both among the Chinese and among the Tallensi, they do so openly in their sons' formative years, and still, among the Tallensi, behind the façade of the avoidances, in later years, when their own life situation, and Tale beliefs about human nature, prompt recourse to the defence of thrusting first sons out. Can sons do other than strive to reciprocate their fathers' devotion even while, perhaps only inarticulately coveting their status?

But the problem remains how to reconcile the moral imperative of kinship amity with the rivalry of interests between the generations or, to put it from the angle of the individual's life experience and motivation, how to preserve the trust and affection engendered by the life-long reciprocity of parental solicitude and filial dependence against the pull of the underlying mutual antagonism generated in this very relationship of upbringing. It is, in short a question of resolving the ambivalence that is built into the relations of successive generations in the unilineally-organized descent systems we are considering. We should bear in mind that there is no means of total escape from the family and lineage structure other than by complete severance of all ties with kin and community and the consequential abandonment of all rights and claims to sources of livelihood , jural status, ritual insurance, and political protection. Traditionally, in a society like that of the Tallensi, one could not live in a community except as either a legitimate member of a lineage and a family, or a kinless and rightless slave attached to a lineage and a family and able to survive only by virtue of being accorded quasi-kinship status. There are good structural reasons, therefore, for institutional devices and cultural values that will serve to regulate the potentiality of schism between successive generations.

Ancestor worship provides the medium through which this end is attained. It represents not only the apotheosis of parental authority but its immortalization by incorporation in the universal and everlasting dominion of the lineage and clan ancestors. How subtly the beliefs and ritual practices of ancestor worship lend themselves to regulating the opposition between successive generations can be appreciated. from the manner in which Tallensi rationalize it by recourse to the concept of Destiny (cf. Fortes 1959). This enables them to externalize the latent conflict in symbolic guise, and thus to acknowledge it, without destroying the relationship to which it belongs. But the inequality of power and authority is not eliminated. To accept this more is needed than symbolic cognizance of its character. And here pietas comes into play.

Pietas is rooted in the relations of living parents and children, as I have already emphasized. It enjoins obedience and respect towards parents, submission of personal will and desires to their discipline, economic service to them, and acquiescence in jural minority. The tangible reward for keeping the rules is the gratification of parents and kin, and the diffuse approval of society. There is also a moral reward in that pietas towards the living is eo ipso pietas towards the ancestors and is deemed to conduce to their benevolence. We might therefore regard pietas as the temporary renunciation of self-interest in order to maintain indispensable social relationships. But I would rather avoid such conjectures and merely say that conformity to these norms is an avowal of contentment with parental authority and power.

The upshot is that sons who might be tempted to rebel and fathers whose patience is exhausted are both kept in check. But this does not wholly rule out the chances of acrimony or discord between them. Tallensi would say that human nature is like that. There are people who resent authority, or evade customary duties, or flout religious precepts. Sanctions are necessary but must not be expected to work unfailingly. That is how it is even with such emotionally and institutionally compelling rules as the avoidances of first-borns. Tallensi certainly perceive how these observances segregate the spheres of father and son and enable them to put an interpretation on their relationship which reduces friction and suppresses open rivalry especially in the all-important matter of rights over persons and property. But though breach of first-born taboos is unheard of, Tallensi say that faithful adherence to them is a matter of the kind of propriety, and the kind of morality which I have called pietas, not of blind fear. And where pietas is wanting sons may turn against their parents and fathers repudiate their sons, in defiance of both sentiment and religion. I have recorded instances elsewhere (Fortes 1949) and, as we might anticipate, it is usually a mature first-born of an ageing father who rebels and breaks away. But I have not heard of a case in which the ultimate sanction of pietas did not eventually prevail. When the father dies the son must - and in my experience always does - return to perform the funeral and assume his inheritance. Kologo's tragic fate was widely cited as an object lesson. He quarrelled with his father and departed to farm abroad. But when messengers came to tell him that his father had died he hurried home to supervise the funeral. He had barely taken possession of his patrimony when he fell ill and died. The general belief was that this was retribution for failure to make up his quarrel with his father. When he came home from the funeral he had made submission to the lineage elders and they had persuaded his father's sister to revoke his father's curse. But this was not enough, as his death proved. The diviner revealed that his father, now among the ancestors, still grieved and angered by his desertion, complained to them of his impiety and so they had slain him.

It is pietas then, which makes living authority acceptable. Transposed into ritual form it becomes the pietas towards the ancestors which is the essence of their worship. This corresponds to the continuity between the living and the ancestors that is embodied in the descent group. But there is tangible foundation for this transposition from mundane custom to religious practices and belief. Among the Tallensi, as among the Chinese but perhaps more conspicuously and familiarly, the ancestors, far from being remote divinities, are part and parcel of the everyday life of their descendants. Their shrines stud the homesteads, their graves are close by, their names are constantly cited in social transactions. It is often impossible to tell, when Tallensi speak of a father or a grand-father, whether they are referring to a living person or to an ancestor. In a large expanded family not a week passes without some sacrifice or libation to the ancestors. And the attitude of the officiant in domestic rites of this kind is but a more reverent version of his relations with a living parent.

I have often taken part with the head of the family in rites that seem so informal as hardly to merit the title of religious worship. On the eve of the sowing season, for example, every family head goes round his homestead and his home farm pouring a libation of millet flour mixed in water upon each of his ancestor shrines in turn in order to inform the ancestors of the tasks that lie ahead. He addresses them with deference and pleads for their protection against accidents and for health, fertility, and well-being for all. But his matter-of-fact manner and conventional words might easily mislead an onlooker to see no religious meaning in his actions. Characteristically, when I found one of my friends supervising the sowing of his home farm he explained that he was late in starting because, as he put it, 'I had to tell my father first.' As his own father was still alive, I asked if he meant he had to inform him first. 'Yes of course I had to tell him,' he said, 'one can't do anything so important without telling one's father. But I don't mean him. I mean my father who became my Destiny, my ancestor.' Pietas towards the ancestors consists primarily in ritual tendance and services in the form of libations, sacrifices and observances whenever they are demanded. The parallels with pietas towards the living is seen, among the Tallensi, in their description of sacrifice as giving food and drink to the ancestors, though they make it clear that this is not meant in the material sense. in return, they say, ancestors 'back up' (dol) their descendants.

However, the aspect of ancestor worship which I wish to dwell upon is its value in resolving the opposition, structural and inter-personal, of successive generations. Granted the premises of belief and value, death palpably removes fathers; but it is not assumed to extinguish fatherhood. On the contrary, it furnishes the conditions for elevating fatherhood above mundane claims and commitments. What is more, it provides the occasion for society to compel sons to accept their triumph as a moral necessity and to make up for it by undergoing the ritual exigencies that metamorphose fathers into ancestors. It is reassuring for a son to know that it is by his pious submission to ritual that his father is established among the ancestors for ever. He sees it as the continuation of submission to the authority that was vested in his father before his death.

And let me interpolate that we must not be deceived into assuming that funeral rites are necessary in order to turn to a dead person into an ancestor for what are vulgarly thought of as superstitious reasons. Similar rites are performed on behalf of living men in order to confer office and status. Ancestorhood is a status in a descent structure as Van Gennep (1909) showed and as such students of African religions as one of our former Presidents the late Dr Edwin Smith often emphasized (e.g. Smith 1952). The ritual establishment of ancestorhood defines the realm of events and social relations within which the power and authority of ancestors are believed to be displayed.

To go back to what I have been saying, it should be recollected that death is legitimated as the doing of the ancestors. It s they themselves, the fountain-head of authority and the final sanction of pietas, who remove fathers and open the way for sons to succeed. That they cut down fathers in just retribution for conduct which they are believed to regard as impious is consistent with their status. Is there a more effective way of asserting power and authority than by imputing and punishing disobedience?

It can be seen that in these systems a person never escapes from authority. The jural authority of the living father is metamorphosed into the mystical authority of the ancestor father, backed by the whole hierarchy of the ancestors and the more formidable for that reason. Thus a father's status is held by grace of the ancestors. For all its rewards, it is not an easy office, for it carries not only material responsibilities for dependants but the more onerous ritual responsibilities to the ancestors. But to succeed him need to be interpreted as supplanting the father, but rather as taking over and continuing the office that was temporarily vested in him. It is submission to duty and this divests it of guilt, the more readily so since the opposition between successive generations is not ended but merely transposed to a new level. And it is in some ways more acute, for misfortunes, disease, and death are the lot of mankind ad quite unforeseeable. These are interpreted, in Tale philosophy, as manifestations of dissatisfaction on the part of ancestors. The man who holds paternal status is constantly faced with unforeseen demands from the ancestors. The right to officiate in sacrifice to them gives jural and economic power and authority over living dependants. But it also imposes the burden of responsibility for the proper tendance and service of the ancestors. And one can never be sure that one is fulfilling these obligations satisfactorily, as one can be with one's duties to living parents. (Cf. Fortes 1959.)

The saving grace is pietas. If one conducts one's life to the best of one's lights, in accordance with the dictates of pietas, one can have faith in the justice of the ancestors. What is more, one can accept what comes from them without remorse and in a spirit of submission to authority that cannot be questioned. Hope remains; for expiation and reconciliation are always open to one. This is no more than admitting that one has failed in pietas, a very human failing, and the institutional means are there for reinstatement. To give the ancestors what they demand in sacrifice, service and observance is to submit to their discipline and so to recover pietas. I do not think that I have embroidered extravagantly on the ethnographic facts. As far as the Tallensi are concerned, a man's ancestor-father and forefathers are, as I have said, believed to be in his vicinity all the time, ritually accessible to him at the shrines dedicated to them. This is not a superstitious fiction. Their authority, protective no less than disciplinary, in intimately felt to be ever present in normal life, just as living parents are. I was reminded of this on an occasion when the Tongraana was discoursing to his elders about a dispute between two clan heads at which he had been asked to give evidence as an authority on native custom. He explained how he had felt obliged to refute the claims of one of the parties, even though he was a kinsman.

'He was lying', said the Tongraana, 'and lies only get you into trouble. I hate deceit. I will not tell lies, cos what is may. If a man has been properly brought up by his father he will not be a liar. When I was a small boy my father used to beat me and beat me if I deceived him or told lies, that is why I do not speak lies. My father will not permit this.'

To take another instance. Teezien was patiently trying to make me grasp the point of the last fruits rites at the external boghar at the end of the dry season.

'We provide for them (i.e. the ancestors) he said, and beg crops. We gave him (that is, the boghar personified as the collectivity of all the ancestors) food so that he may eat and on his part grant us something. If we deny him he will not provide for us, he will not give to us, neither wife nor child. It is he who rules over us so that we may live. Supposing you are cultivating your farm and the crops spoil, will you not say that it is your father who let this happen? If you are breeding livestock and they all die, won't you say your father permitted this? If you gave him nothing will he give you anything? He is the master of everything. We brew beer for him and sacrifice fowls so that he may eat to satisfaction and then he will secure guinea corn and millet for us.'

[Ancestors, I protested, are dead; how can they eat and do such material things as making crops thrive?] 'It is exactly as with living people.' he answered imperturbably, 'If you have a son and you are bringing him up and he refuses to farm, you upbraid him. You say you fathered him with tribulation and here he is refusing to farm what then are you to eat? If he doesn't farm will he ever get himself a wife, will he achieve children? Now if someone does you a favour wouldn't you go and thank him? And if you do someone a favour and he comes to thank you with however small a token would you not do him a favour again?'

We can see that fathers are held in mind as if they had never died. And the image in which they are cast is one that accentuates the authority and discipline which they exercised. They are recalled with pious gratitude for the moral scruples they inculcated and the obedience they exacted, but also with affection for the benevolence they showed to loyal sons. This is more revealing since Tale fathers, in reality, very rarely have recourse to corporal punishment and normally have easy-going and tolerant relationships with their growing sons, as I have previously noted.

In this way the concept of ancestorhood and the religious institutions in which it is ritually and socially embodied serve as the medium that enables the individual to keep up his relationship with his father, even after his death, as if he were a part of himself. The father who controlled his conduct during life turns into an internal censor of his conduct when he becomes an ancestor. And this is effective because he is, at the same time, externally available both for the imputation of absolute power and authority, and for acts of appeasement. This provides grounds that seem rational and objective for the rituals of solicitude by means of which a man may hope to control, or at least to influence and certainly to negotiate the changing fortunes of life. Pietas is the bridge between the internal presence and the external sanctity of paternal authority and power.

Go to Next section | Return to Previous section | Return to the beginning