INTRODUCTION⁽¹⁰¹⁾

The Ga⁽¹⁰²⁾ theory of the nature of man has received little or no attention from philosophers. This may be because it has been assumed to be virtually identical with the often-discussed theory of the neighboring Akan. Thus Debrunner refers to "the Gas, whose psychological concepts are almost identical with the Twi terms," and adds in parenthesis: "It is remarkable that the Twis, Gas and even the Ewes use the same word kra, kla, klama--there is a great temptation to link this with the Egyptian concept of the ka."⁽¹⁰³⁾ Nor have the Ga beliefs about immortality received any philosophical discussion.

These two topics are obviously closely related. In this paper I shall explore them and attempt to elucidate the nature of the relationship: that is, in what way, and how successfully, the theory of man's nature provides a framework in terms of which his survival of death may be described and explained.

The major study of these matters from an anthropological point of view are the two chapters "The Dogma of Human Personality" and "The Cult of the Dead" in Margaret Field's classic Religion and Medicine of the Ga People.⁽¹⁰⁴⁾ For the purposes of this study I have supplemented the data in Field with some original field-work. In the enterprise of deriving generalizations about Ga thought from beliefs stated by informants and recorded in Field, two constraints presented themselves.

Firstly, the influence of Christianity in the area has been deep and widespread. When a statement is made which seems Biblical in content or expression, it is occasionally difficult to know whether this is a case of independent concurrence between Ga traditional thought and Christianity, or of the influence of the latter on the former. Instances of this will be indicated in footnotes.

Secondly, on some subjects informants differed from each other, or from the views recorded in Field. In this situation there are two possible stances on the part of the investigator. If he accepts the characterization (by Horton, etc.) of a traditional culture as one in which a single world-view has a monopoly on the intellectual assent of its members, then he will assume that one or another of the views in question is more representative of the traditional point of view and hence to be regarded as more authoritative. This approach, however, has often been contested, most recently by Gyekye, who stresses that traditional thought must have been formed by individual thinkers, and that therefore a monolithic orthodoxy is not to be expected. The members of a traditional culture, operating with concepts suggested by their common language, may arrive at and promulgate differing views, which in turn are accepted or rejected by others in accordance with the results of their individual reflection. In this way, competing views come to be at large in the community, and though one or another of them may predominate, all of them have a title to be regarded by the investigator as representative of the philosophy of the culture.⁽¹⁰⁵⁾

For practical purposes, I shall adopt a modified version of the second stance. Where competing views exist, they will be indicated, and it will not necessarily be assumed that one of them is the Ga view. But this is without prejudice to the possibility that further consultation of custodians of traditional culture might yield grounds for regarding one view as more authentic than the other. Further, where external influence would seem to be a potentially viable explanation of a discrepancy among views, as is the case with concepts of the kla, I shall indicate as much.

The Ga have a tripartite theory of man. (The expression "tripartite" will later to be found to be misleading, but it will serve for the moment.) Within, or otherwise associated with, the body are two unseen entities, the susuma and the kla. There is no consensus on the proper translation of these terms. In my view "spirit" is a reasonably apt translation of susuma, but "soul" does not correspond to kla.⁽¹⁰⁶⁾ Further, "spirit" is required also for the translation of mum.⁽¹⁰⁷⁾ In order, therefore, to avoid erroneous associations and to preserve necessary distinctions, it seems best to transliterate both terms.

The three "parts" of man will respectively form the principal subject-matter of the three sections of this paper. Among the questions to be discussed will be the nature of the kla and the susuma, the grounds for postulating their existence, and their relation to the body and to each other. The Ga conceive personal immortality to take two main forms: survival of death in a disembodied state, and renewed life in a different material body. For reasons of space, however, the topic of reincarnation will have to be omitted in this paper. Since disembodied existence is precisely the independent existence of the susuma, it will be discussed in the course of the section on the susuma.

THE BODY

The Ga call the body of a man gbmtso, the same word being used for the body of animals. This is a compound word consisting of two elements, gbm and tso. Gbm, pl. gbmi, is the word for man or woman, person; m, pl. mi, also have this sense. (The latter may be applied to animals, but when it is, there seems to be always a degree of personification.) "Human being" is adesa, pl. adesai, or gbm adesa. Tso, pl. tsei, is used of plants with an upright and particularly a woody stem, as distinct from creepers and soft-stemmed plants, which are bai. Thus all trees are tsei. Tso may also be used of a piece of wood, a plank, stick, pole or staff, both in isolation and in compounds:

obonu k tso - drum and stick

nyimtso - walking-stick

denghimtso - lit. "a stick for holding in the

hand", such as an officer's staff

Tso also means the material, "wood". When it is used in this sense in compounds, the name of the object made out of wood normally follows the tso element:

tsotsu - wooden house

tsoshinaa - wooden door

tsosamfl - wooden window

tsogbm - wooden (statue of a ) human being

but:

saatso - bedstead (saa = bed)

In the last compound, tso occupies the same position as in gbmtso. The relationship between the two elements, saa and tso, is, as in the first list, tso-for-X, or, possibly, tso-of-X, and not, as in the second, S-from-tso. We may therefore conclude that gbmtso is "tso for, or of, gbm".

Given the meanings of each element of the compound in isolation and in other contexts, and given the suggestions carried by the structure of the compound, what conception of the body is conveyed by the word gbmtso? Two possibilities come to mind.

Firstly, is tso perhaps being used in an extended sense to mean material in general, and not just wood, so that gbmtso means "stuff of man"? Anyone acquainted with the ancient philosophy of the West is irresistibly reminded of Aristotle's choice of the Greek word hyle, meaning "timber", as the technical term for his new concept of matter. But the fact that tso seems not to be used in other contexts to mean 'stuff" makes this impossible.

Secondly, does the word have reference rather to the shape of the human body, with a central bigger part (the "trunk") and smaller parts going off it? This seems altogether more natural and likely. Three other considerations work in its favor. Firstly, the phrase tso kng, whose literal meaning is "shoulder of tree", is used to mean the crook of a brand or the branch itself. The analogy between the shape of a human body and a tree is thus registered elsewhere in the language. Secondly, tso is used in three other compounds referring to parts of the body: vitso ("head"), nkutso ("elbow") and nakutso ("knee"). In at least the two latter instances, the jointed form of a tree probably prompted the expression. Thirdly, the semantically equivalent Akan word onipadua (Twi) or nyimpaua (Fante), where onipa/nyimpa = "man" and dua = "tree", is normally taken by Akan scholars to carry a reference to the shape of the body. Christaller, for example, gives its basic meaning as "the figure, form, shape of the body".⁽¹⁰⁸⁾

It is relevant to recall at this point that, as in most West African traditional thought systems, trees, and plants generally, are not regarded as inanimate objects. The Ga make the generalization that tsi f l, wji ji am: "all the plants are wji". Wng (pl. wji) may be translated according to context as "(minor) god", "spirit", "power", "juju": the central idea is that of an invisible thing which has real, though limited, power. The seriousness of the belief that plants are endowed with some kind of sentient spirit is shown in the practice of offering prayer before roots are gathered. Certain trees, notably the silk-cotton tree and the baobab tree, are regarded with especial awe, and rarely cut down.

The body then is conceived as tree-like in shape; and the word gbmtso may carry associations of power, dignity, even sacrosanctity. But the Ga conception of the body is as of a combination of the sublime with the ridiculous. For there is a saying that the body is a mask; and the word used for "mask" in the saying, kakamotobi, denotes a comic or grotesque mask, usually home-made, which is donned by young men who go round in groups especially at Christmas-time to amuse the adults and frighten the children. It is said that when we enter life, we choose the mask which we will wear. The implication of describing the body as a mask is that what you see when you look at a human being does not give you his real nature. Man is something other than his body, something more enduring than it. The next two sections will deal with what the Ga believe goes to make up a human being apart from his body.

THE SUSUMA

Akan Controversies

With regard to the possibly parallel Akan theory of mind, Gyekye has maintained that the tripartite superficies conceals a dualistic kernel. The sunsum is a part of the okra:

insofar as things asserted of the okra are not assertable of the sunsum, the two cannot logically be identified. However, although they are logically distinct, they are not ontologically distinct. That is to say, they are not independent existents held together in an accidental way by an external bond. . . . The sunsum may, more accurately, be characterized as a part--the active part--of the kra (soul).⁽¹⁰⁹⁾

All earlier interpreters had held the view that the kra and the sunsum were logically distinct. But there were radically differing views to their respective natures. Danquah speaks of the sunsum as the "material mechanism" and says that "sunsum . . . is, in fact, the matter or the physical basis of the ultimate ideal of which Okara is the form and the spiritual or mental basis".⁽¹¹⁰⁾ Wiredu, on the other hand, holds that the kra, while not a straightforwardly physical object, has some quasi-physical properties. For example, it can be seen by medicine-men or those whose normal powers of perception have been extended by medicinal means or those gifted with extra-sensory perception.⁽¹¹¹⁾ Thus Danquah and Wiredu each hold that one of the two non-bodily "parts" of man in Akan thought is physical or quasi-physical, but they differ as to which of the two is such. Gyekye, on the other hand, does not accept the attribution of physical properties to either the sunsum or the kra. He holds that the Akan position is basically a Cartesian one: the kra, an immaterial entity, inhabits the body during life, and leaves it at death. The kra and the sunsum survive death as a "spiritual unity", and it is on this basis that the Akan hold man to be immortal.⁽¹¹²⁾

I believe it will provide a useful orientation for the ensuing discussion of the Ga beliefs if I list in advance the positions which I shall take up vis-a-vis these points of controversy which have emerged with reference to the Akan beliefs. (This is not, however, to imply anything as to the validity of the parallel.)

(1) The susuma and the kla are ontologically distinct

(2) The kla is non-physical.

(3) The susuma has some physical properties.

(4) Kla and susuma do not survive death as a unity.

(5) Personal immortality in a disembodied form

consists in the continued existence of the

susuma.

It will thus be apparent that the position to be advocated in this paper bears more resemblance to Danquah's than to either Wiredu's or Gyekye's. Although these scholars were not dealing directly with the beliefs of the Ga, I shall at various points have to deal with arguments in their writings which are relevant to this subject-matter.

Towards Defining Susuma

The Ga say that when God created man, he breathed into clay, and activated it. That breath of God which gives life to the clay is man's susuma.⁽¹¹³⁾ As already indicated, the word for breath is mumo. The susuma, therefore, is mum. The kla is also regarded as mum. The inter-relations between these concepts are at first glance puzzling. But I now believe they are explicable in terms of the following two theses.

Firstly, mum is a generic term applicable to anything that is conceived (a) as immaterial, i.e., not composed of gross matter like the body, and (b) as personal or quasi-personal. Susuma and kla are thus both species of mum. Other species of mum are wng (referred to in Section I) and gbeshi. Both of these are entities which do not form part of the normal constitution of a human being but which (because they are not material as the body is material) can superimpose themselves upon a human being and occupy the same space as his body or part of his body, and (because they are personal) can utilize that body to produce expressions, either vocal or motor, in which pattern or purpose is discernible. In the case of a wng, such expressions often involve displays of supernatural strength or stamina, such as frenzied dancing or running. They are therefore both limited in duration (not usually more than five or six hours) and easy to recognize. Possession by a gbeshi, on the other hand, does not give rise to such conspicuous manifestations (cf. Section III), and because its manifestations do not involve supernormal capacities, they may be of much longer duration than those characteristic of possession by a wng. Both wng and gbeshi, then, are thought capable of possessing a human being; and both are species of mum alongside susuma and kla.

Secondly, apart from being the genus comprising these four species, mum also has a narrower use: for as soon as any question of distinction or degree among the various spiritual entities enters in, there is a tendency for mum to slide up the scale in the direction of those seen as "higher", more divine, or less akin to matter. Mum is normally used in the Ga Bible, for example, to translate pneuma ("spirit"), which is seen as being higher than psyche ("soul"), on the grounds that God is pneuma.⁽¹¹⁴⁾ Again, there is, as we shall see, a traditional doctrine to the effect that when a man dies his susuma goes to the World of the Dead, but his kla goes to God; and an alternative way of expressing this is to say that while his susuma goes to the World of the Dead, his mum goes to God.

These two and any other similar usages might lead to an objection that mum is not, as I have maintained above, the genus to which susuma belongs, for it is sometimes contradistinguished from it. To this I would answer that mum is used in two senses; that mum in its wider sense is the genus of susuma; and that it is when mum is being used in its narrower sense that it is contradistinguished from susuma.⁽¹¹⁵⁾

Within the class of immaterial personal entities, what distinguishes susuma from the others? Like kla, but unlike gbeshi and wng, it is an integral and not an adventitious "part" of a human being. How does it differ from kla? One difference is that it can leave the body without causing death; whereas the kla is associated with the body right up to death. The detachability of the susuma from the body will be very prominent as we consider the susuma in relation to consciousness.

Susuma, Mind and Consciousness

If one enquiries of an exponent of Ga traditional thought about the nature of man, one will invariably be taught about the body, the susuma and the kla; the mind will not be mentioned. Yet the word for mind, jwengm, exists; it is used for example in the phrase eb jwengm, literally "he hasn"t got a mind", i.e., "he has no brains". This situation prompts two questions. Why is mind scarcely mentioned in the account of human nature? And what is the relation between this Cindarella and the more important and interesting components?

It would be wrong to say that the mind, for the Ga, is part of the body. For a distinction is certainly drawn between the mind and the brain (ans). Nevertheless it would seem that the mind is regarded as a function of the brain. It is difficult to get definite statements on this issue; and the reason for this seems to be that thought was not traditionally regarded as posing philosophical problems or standing in need of explanation in terms of occult entities, as are some other phenomena associated with man. Wiredu has made a comparable comment in connection with the Akan okra:

The kra is postulated in Akan thought to account for the fact of life and destiny but not of thought. The soul, on the other hand, seems in much Western philosophy to be intended to account, not just for life but also for thought. Indeed, in Cartesian philosophy, the sole purpose of introducing the soul is to account for the phenomenon of thinking.⁽¹¹⁶⁾

Gyekye has taken up the reference to Descartes to cast doubt on the suggested contrast between Akan and Western philosophy. He points out that the Cartesian cogitatio is wider than thought in the sense of conscious ratiocination, which he believes (although it is not quite clear why) is the sense in which Wiredu is using the word; and

Any living being must have consciousness. This being the case, consciousness, which is equivalent to the soul or mind in Descartes, can be a translation of okra.⁽¹¹⁷⁾

It seems to me that the contrast can be defended against this criticism in the following way. The fact that a living being is a conscious being does not go to show that the okra is postulated to account for consciousness rather than to account for life itself. If, on the other hand, we do take thought in the narrow sense of ratiocination, it is possible to make the contrast quite sharply: Descartes certainly held that thought in that sense was a function of the soul; but the Akan do not connect it particularly with the kra, of which Gyekye's own preferred summary characterization is "the principle of life of a person and the embodiment and transmitter of his or her destiny (nkrabea)".⁽¹¹⁸⁾ The fact that Descartes also attributed other conscious states or activities (e.g., perception and volition) to the soul does not affect the point that the Akan do not specifically attribute ratiocination to the kra.

We return therefore to the Ga concept of mind strengthened in the belief that conscious thought (ratiocination, daydreaming, etc.) was not considered to constitute a problem, and that this is probably the reason why so little appears to be said about the mind in their theory of human nature. The relative dearth of evidence constitutes an obstacle in ascertaining what the theory does maintain about the nature of the mind. But the impression I have formed is that the mind is regarded as a faculty of the brain just as hearing is the faculty of the ears, seeing of the eyes, etc. If this is correct, and the mind is a function of part of the body, then it is not surprising if the mind is not mentioned along with the body, the susuma and the kla in the theory of human nature. It is considered to be subsumed under the first of these components. We have thus a ready, though perhaps disappointingly simple answer to the first of the two questions with which we began, and to the first third of the second question, that is, the relation of the mind to the body.

We have now to consider the relation between the mind and the kla and susuma respectively. But the relationship between the mind and the kla forms an aspect of the relationship between kla and susuma-plus-body, and can usefully be considered in Section III, where the nature of the kla is discussed. The relationship between mind and susuma is a complex and interesting problem. Here we encounter a further sharp difference between this and Western theories of mind. For those of the latter which postulate, behind the succession of thoughts, perceptions and volitions, a self whose thoughts, perceptions and volitions these are, have usually made this single entity the subject of both conscious and unconscious thoughts. But on the Ga view conscious thoughts alone are attributable to the mind; unconscious thoughts are experiences of the susuma. This is not said in so many words, but I believe it is a generalization which may validly be made on the basis of scrutiny of the types of activities which are attributed to the susuma. These include the following activities, with reference to all of which the chief point stressed is that the susuma is separated from the body: (i) witchcraft activities; (ii) out-of-the-body experiences in terminal illness; (iii) dreaming/sleeping. Beliefs about the part played by the susuma in these activities will now be described.

With regard to witchcraft activities, I shall not have very much to say. The term is intended to cover diverse activities which witches are believed to perform out of the body by night, such as travelling to a meeting-place, taking part in a discussion or feast, and procuring food by a spiritual attack on a victim.⁽¹¹⁹⁾ These activities, the reality of which is very widely believed in, are said to be performed by the susumai of witches which leave their bodies by night. This is the only instance among the activities we are reviewing where the separation of the susuma from the body is subject to volition.

It sometimes happens in a terminal illness that a patient who has appeared to be asleep reports, on awakening, that he has left his bed and travelled either to a familiar or to an unknown place, mixing sometimes with the living and sometimes only with the dead. This intermittent "travelling" may go on for days or weeks (rarely, months) before death. Here again it is believed that the susuma has actually left the body, and that the reported experiences are genuine experiences of the susuma in this independent state. When a person lies unconscious in a coma, it is supposed that his susuma has gone to Azizanya, the transition point into the World of the Dead, where he is being judged; if he were guilty, he would not come back.

In a very similar way, dreams are held to be veridical experiences of the susuma. It is believed that when we sleep, the susuma leaves the body,⁽¹²⁰⁾ visits other places, and interacts with the susumai of other people. This explanation of dreaming (common to many African peoples⁽¹²¹⁾), which is liable to seem very implausible to the outsider, Gas sometimes defend by two arguments.

Before stating the first argument, it must be mentioned that the susma is believed to be capable of travel in time as well as in space. So when we dream about a past experience, our susuma is said to leave the present and go back to the time at which the experience occurred. The fact that we sometimes dream about genuine past experiences is not thought to constitute any evidence for the veridical nature of dreams, or for the departure of the susuma to an earlier time-segment than that in which the body exists when the dream takes place. For in waking life we may remember the experience, so that a re-awareness of the experience is patently compatible with the temporal copresence of the susuma with the body.⁽¹²²⁾ The occurrence of premonitory dreams, however, is thought to constitute such evidence. The argument may be formulated as follows. Since some dreams "come true", the content of the dream must have been a real existent or occurrent, which, since it was not in the same time-segment as the existence of the body of the dreamer, the susuma must have travelled in time in order to be acquainted with it.

The second argument used is as follows. It sometimes happens that X expresses a belief that Y was dreaming about him on a particular night, when it is in fact true that Y was dreaming about X on that night. What could underlie X's possession of this true belief except some actual mutual encounter? And since the encounter did not involve the bodies of X and Y, it must have involved their susumai.

One is not obliged to accept these two arguments, but they are interesting as showing that empirical evidence is deemed relevant to establishing the existence and determining the nature of the susuma. The two arguments are each based on a fact: the first, that some dreams "come true", and the second, that two people sometimes dream about each other on the same night. One may think that the hypothesis of the existence and capacity for "travel" of the susuma is not the most economical way of explaining these two facts. But, at the least, the arguments show a willingness to relate facts to theories. Such an attitude does not fall under the heading of "superstition" as defined by Wiredu:

By "superstition" I mean a rationally unsupported belief in entities of any sort. The attribute of being superstitious attached not to the content of a belief but to its mode of entertainment. . . . When, however, we come to the traditional African belief in ancestor spirits--and this, I would contend, applies to spiritualistic beliefs everywhere--the position is different. That our departed ancestors continue to hover around in some rarefied form ready now and then to take a sip of the ceremonial schnapps is a proposition that I have never known to be rationally defended.⁽¹²³⁾

The brief compass of our discussion of the susuma so far indicates that this judgment may be too severe. For we have already noted two other cases apart from dreaming where empirical evidence is brought to bear on establishing the existence and functions of the susuma: the alleged "travelling" of the susuma of a sick person as death approaches, and the alleged "travelling" of the susuma of a witch in pursuance of witchly purposes. Both of these rest on testimonial evidence: the patient's report and the witch's confession, respectively.

One may feel that adequately stringent criteria for assessing such testimonies are not always applied; but to stigmatize the "spiritualistic belief" based on them as "superstitious" must amount to either (a) discounting the relevance of human experiences or alleged experiences to the theory of human nature, which would be a remarkable and certainly untenable approach; or (b) a judgment that none of the particular experiences in question could be authentic, i.e., that all out-of-body experiences claimed by terminal patients or witches are either sincerely or mistakenly claimed--a judgment which could, it would seem, only stem from the perception of an incompatibility between the critic's theory of human nature and that to which the alleged experiences seem to point. If this were the case, the blanket repudiation of the evidence would be a kind of petitio principii, as dogmatic in its own way as the "superstition" on which it is a comment.

The approaches requisite in the context would seem to be rather (c) a careful assessment of the testimonies in question in terms of the generally accepted criteria for judging the worth of testimonial evidence; (d) consideration of whether the facts as claimed warrant the theory as propounded. If (d) yielded a negative result, then, as far as the evaluation of this theory was concerned (though not perhaps that of some alternative possible theory), (c) would be necessary. The adoption of approach (b) above would suggest that the critic's conduct of task (d) has yielded a positive result. Paradoxically, therefore, the blanket repudiation of the testimonial evidence could constitute an indication of approval, however reluctant, of the theory in another respect: the validity of its derivation from those particular alleged facts.

After this excursus into the grounds for belief in the susuma, I return to the topic of the relation of the mind to the susuma. I hope enough has been said to justify the statement that activities which in the West are attributed to the unconscious or subconscious mind, in Ga thought are attributed not to the mind, but to the susuma. Of this, indeed, the chief and perhaps the only example we have had is dreaming; for Western philosophy has scarcely thought it worth taking account of alleged out-of-body experiences such as the other two types of case consist in, nor, as far as I know, do parapsychologists or popular thought ascribe them to the unconscious. At this point may be mentioned the idiom Esusuma ke le wie, "His susuma spoke to him". This is a comment made when someone stops short of taking a disastrous step. Since it implies that one's susuma does not speak to one all or most of the time, it perhaps supports the view that the susuma is an unconscious or subconscious element of a person.

We cannot, however, say that the susuma is in effect the unconscious mind, and that the theory could be amended so that the susuma is past or unrecognized thoughts, wishes, fears, etc. of which the proper subject is the jwngm. We can see this from the way the theory interprets the experiences of the terminally ill patient. The patient characteristically expresses a desire for death during the period in question. This fact might be used to connect the supposed "travelling" of the susuma with Western concepts of the unconscious as a repository of wishes and fears--a connection which could probably be made unobjectionably in the case of dreams. But in the case of the patient, the desire for death is a fully conscious one; and the role of the susuma is as a would-be implementer of wishes to which the conscious mind cannot, by mere volition, give effect. The susuma has, as it were, a mind and a will of its own.

Three observations concerning the relation between the mind and the susuma now suggest themselves. Firstly, if there were a complete separation between the mind and the susuma during dreaming, we should suddenly receive a pack of dreams each time the susuma returns to us. But dreaming is a progressive experience, as can be seen from physical reactions on the part of the dreamer. Presumably, therefore, the mind is residually active during sleep, and is able to register the impression of the absent susuma in such a way as to produce the physical reactions appropriate to the dream. What, then, is the nature of the link between them, and how does it differ from that which obtains during waking life? This is left mysterious. There is supposed to be an "invisible thread" between the dreamer and his susuma, but as far as I know it is not further described.

Secondly, the out-of-body experiences of the sick person near to death seem to be accompanied by a clarity and sense of reality superior to that which characterizes dreams. How is this difference to be explained if they are both alike experiences of the separated susuma? The theory appears to be silent on this point.

Thirdly, there is a difference between the sleeper's dream and the patient's "journey" when each is looked at in the manner broached earlier, as an exercise in wish-fulfillment. For the desires which the susuma is supposed to execute in dreams are sometimes not desires which are given countenance to by its owner; indeed they may have been censored from admission to consciousness. This is not so however, of the desire for death in the other case. Moreover, that desire has the peculiarity that it is precisely a desire for a condition of the susuma (according to the Ga view whereby after death the susuma exists in separation from the body). But the desires which the susuma is supposed to execute in dreams are desires for a condition of the whole person. (It is conceivable that someone might hold, within the framework of the theory, that such desires are desires of the susuma for a condition of itself in relation to other susumai and not, after all, desires of the person in relation to other persons. But that would be to posit a much greater degree of independence between the mind and the susuma.)

From the second and third observations, we can see that there are differences between dreaming and the alleged "travelling" before death in regard to both their felt quality and the character of the purposiveness which they exhibit; and in view of these differences, the verdict seems inescapable that the identical explanation of them is too bare. Either dreaming should be eliminated here, or further details should be supplied which make the differences between the two types of cases comprehensible. Without such elaboration, and an account of the link between susuma and body such as was desiderated in the first observation, the complex question of the relation between the mind and the susuma cannot be fully resolved. Perhaps it can be said that the account of dreaming does not sit very comfortably with the rest of the theory.

The Susuma After Death

The susma leaves the body at death. Esusuma eshi l--"His susuma has left him"--means "He is dead". The departure of the susuma alone however, is not sufficient to cause death. It is when the susuma and the kla both leave the body that death occurs.

The question whether animals have a susuma, and whether, accordingly, the death of an animal either consists in, or is accompanied by, the departure of its susuma, typically does not meet with a very ready or assured answer. But some Gas at least hold that animals do not have a susuma, and that that is why (except for those traditionally regarded as gods, such as the hyena, the python, etc.) they are permitted to be killed. This position however, would seem inconsistent with the role ascribed to the susuma in human dreaming, in view of the fact that animals obviously do dream. Furthermore, how is the death of animals to be explained if they do not have a susuma? For animals apparently do not have a kla (at least in one sense of kla), and one presumably wants to explain their death in a manner parallel to that of human beings. Thus a positive answer to the question whether animals have a susuma would be more consistent with beliefs about the susuma of humans in relation to both dreams and death. The unwillingness of discussants to commit themselves in answer to this question is perhaps due (if my earlier suggestion concerning the empirical basis of the theory is correct) to the fact that in the case of animals no reported experiences are available.

In the remainder of this subsection I shall give a mainly descriptive account of those Ga beliefs concerning the fate of man after death which seem relevant to determining the nature of the susuma.

When the susuma leaves the body at death it travels very quickly⁽¹²⁴⁾ and reaches a river which it must cross. Before that, if very rapid action is taken, the susuma may sometimes be brought back. If the person's name is shouted three times at the nearest crossroads, and the person responds, then the fleeing susuma is said to have been recalled. Alternatively, the body may be besmeared with pepper or pepper may be burnt in the room. The susuma, which is believed not to like the smell of pepper may sneeze, and all is well.

The susuma of a person who has died in an accident or by violence (otfo) does not travel from the place of death until after pacification has been performed. An otfo is angry and may haunt passers-by in a rough and frightening manner until it is pacified and its spirit transferred.

The river which must be crossed is not identified with any geographical river (Christians tend to call it Yordon Faa, "the River Jordan"), but the arrival-place of the newly-dead is known by the name of a geographical town, Azizanya, which is sited where the River Volta flows into the sea. This is a picturesque expression of the belief that we are all one with eternity which we are eventually going to join. Money is put into the coffin as the fare to the further bank of the river. At azizanya the nose is said to be broken, for ghosts are reputed to speak nasally. The dead person thus irrevocably enters gbohiiajeng, "the World of the Dead" (from gbohii, "dead people", and jeng, "world"), also called sisaiajeng, "the World of Ghosts".

But on earth his ghost may be seen for up to about three weeks after death. This is about the period it takes for someone's susuma to become impotent and lose contact with people in the physical world. The ghost may be seen in different places, sometimes far apart, usually by people who knew the deceased well. It is most commonly seen in the first three days after death. It can enter a room through even the smallest hole. It may be seen by one or more of the people in a room without being seen by all. Its presence can be detected by a characteristic fragrance of krb⁽¹²⁵⁾ or by a sensation of cold even when it is not seen, or seen by only some of those present. To see a ghost is always a frightening experience. To sit on a chair upon which a ghost has been or is sitting is widely believed to result in impotence, infertility or even death. For sisa nii l, atasa he, "one does not touch the things belonging to a ghost". For this reason upright chairs are often faced about and tipped against the wall after use so that ghosts will not sit on them. Animals are also believed to be capable of seeing ghosts. When a dog starts barking furiously for no apparent reason, the explanation is liable to be given that he has seen a ghost.

The ghost is universally identified with the susuma: susuma l ji sisa ni ak le, "what we call the ghost is the susuma". To convey the meaning "I have seen a ghost", mina susuma is often used instead of mina sisa (as being less frightening). The Ga affirm that we receive a new body at death. But questioning always elicits that this new body is none other than the susuma. It is a new body in the sense that it has not previously been the person's outer garment, as it were; he is now unencumbered by the visible body, which he has laid aside. His new body can not only move faster but also see more than the old one; it is said picturesquely that ghosts have four eyes.

While the susuma goes to the world of the Dead, the kla, which is believed to be the presence of God in us, goes to God. What happens to it is not known. But some say that it loses its individuality. For example, one informant said "When you die you are in two different places. The spirit of God which is in me goes back to God. But what makes me Tettey is my susuma. That goes to gbohiiajeng. That is what has the scent of krb. It doesn't lose its individuality as the other does." The matter of the kla's loss of individuality belongs to the section below on the kla, where I shall attempt to cast some light on the doctrine, at first sight puzzling, of the dual destination of the dead. Here it may merely be noted that it appears to be a unanimously held doctrine, and that there also appears to be general agreement on the point that it is the susuma which sustains the individuality of the person. The susuma is the person who has died, but it is less than the person. God has taken his own power away; the body is in the grave; what remains is the susuma, and this still actively works.

Four kinds of powers are attributed to it in relation to the physical world.

Firstly, it may hover around and become visible, or otherwise perceptible, to the living as sisa, just described.

Secondly, it may possess a living person, usually a medium, as wng. There exist professional mediums who are reputed able to contact virtually any dead person. They make use of a ssi, a big wooden bowl containing water, herbs, etc. It is believed that the reflection of a ghost in a mirror, in water, etc., is sometimes visible when the ghost itself cannot be seen. After invocation the medium and the client see the dead person in the water; the medium may speak in a voice which the client identifies as that of the dead person.

Thirdly, it may materialize, assuming the appearance, voice, etc., which the person had while living, so as to deceive anyone who sees it into thinking that they are looking at a normal living person. This usually only happens only in an hour of need of a child or grandchild of the deceased. He appears not to them, but to an intermediary, sometimes a total stranger or a distant acquaintance who has not heard of the death, saying, for example, "Take this money and give it to Akeley. Tell her Auntie Akoshia sent it for her." Messages, for example, instructions as to the disposal of property, are believed to be sent in the same way. But the dead are shy of being seen and recognised by the living, except when they have a special purpose such as this in view. Anecdotes of people who have casually caught sight of an acquaintance looking just as he did in life are numerous. This is not considered a frightening experience. Indeed, the acquaintance may be addressed almost like a naughty child, to the effect that "I have spotted you". He is typically unwilling to engage in conversation, and gets out of sight as soon as possible.

Fourthly, the dead are believed to be able to influence events on earth by means not ordinarily perceptible, as they are in the other three cases. They are, therefore, not specified but the reality is firmly believed in to the extent that prayers to the dead for peace, prosperity and other blessings are a normal feature of traditional life. A considerable body of belief, into which I need not enter here, centers on the occasions and exigencies that stir the dead to exercise their powers of intervention in the world of the living.

With regard to the nature and quality of life within the World of the Dead, it is said that the dead who possess mediums do not reveal this. But some at least of the dead are thought to be more pure and holy than men on the earth, and their life to be more sublime.

As to whether judgment is a feature of the World of the Dead, there appears to be no belief in a general judgment, though it is left open that individuals might be rewarded or punished. The common phrase Nyngm baawo bo nym, "God will punish you", is often taken to refer to this life or to a subsequent life on earth, rather than to anything which is to happen in the World of the Dead; but the ancestors are believed to administer judgment upon the individual there.

The location of the World of the Dead is indeterminate. Gas see the world as composed of three main levels. Ngwei is "sky", "heaven"; it can also function adverbially as "on high", "upwards". Whatever is above us, like the moon, stars, airplanes, is at ngwi. It is the word used for the Christian concept of heaven, the place where God is, often conceived of as existing above us. Shikpng is the earth on which we tread; anything below it is at shishi, "underneath", "bottom part". Within this scheme the World of the Dead is assigned no definite place. It is not specifically held to be below the earth, as the Igbo believe; nor is it above us, as the Ewe believe.⁽¹²⁶⁾ The implications of the absence of any definite location for the World of the Dead will be considered below.

Much more might be said concerning beliefs about the after-life. For instance, the interesting Ga beliefs about the "sky family" have not been touched upon. But the above are the main beliefs relevant to establishing that for the Ga, personal survival of death consists in the continued existence of the person's susuma, and (in conjunction with the beliefs about the susuma of a living person described in the previous section on the susuma after death) to supplying a basis for analysis of the concept of the susuma, a task to which I now turn.

The Nature of the Susuma

The susuma cannot be understood behaviouristically as a set of dispositions belonging to a person. It is itself a substance or owner of properties. But is it an immaterial or non-physical substance, in effect a Cartesian ego as Gyekye has maintained the Akan susuma to be? I shall argue that the susuma does not answer to this description. For while one cannot say straightforwardly that the susuma is a physical thing, it yet seems to have some physical properties. I shall, firstly, indicate what these are; secondly, attempt to answer an argument brought by Gyekye against Wiredu's characterization of the Akan kra as quasi-physical which would apply equally to my characterization of the susuma; and, thirdly, show why (if we have to choose) it is better to say that the susuma is a physical thing than to say that it is non-physical. I shall then suggest a different characterization of the susuma.

The susuma is plainly not gross matter like the flesh and bones of the body. At the same time, it seems to have some of the properties of a physical object. In the first place, it exhibits movement through space. As we saw in the section above regarding the susuma after death, the susuma moves from the body in dreams and may also do so when the person is approaching death. Neither of these movements are subject to volition; but the susuma of a witch can move through space at will. And the susuma, as we saw, can move in time as well as in space. In either case its movement is extremely rapid. Perhaps we can compare the Western belief that the whole life of a drowning man passes before him in a flash. Now if the susuma can move through space (and time), this presumably means that it occupies space (and time). And this is surely the defining characteristic of a physical object. The same applies to the movements of the susuma after death which were discussed in the previous section. Whether manifesting itself as sisa, possessing a medium as wng, or materializing to appear just like a living person, it is present in a specific location, and, occupying space, must be physical in nature. If the word "occupies" is thought to be inappropriate, we could alternatively say that it "occurs in" space (and time). Then while it might not be appropriate to say that it is a physical object, yet still we would surely have to accept that insofar as it moves through space the characterization "physical" must be applied, just as, for example, a wave is physical, although not a physical object.

What about the collection of susumai which is gbohiiajeng? Because it is not assigned any specific place in the three levels of the physical world, should we draw the conclusion that it is not located anywhere and therefore (as Professor Gyekye has claimed of the seemingly parallel Akan asamando)⁽¹²⁷⁾ does not exist in space? This conclusion does not seem reconcilable with beliefs about the local presence of ghosts, etc. It would seem more consistent with these to suppose that the World of the Dead exists somewhere in the three levels, but we do not know where: perhaps in several "departments", some of which may be superimposed upon our level. How else could one accommodate beliefs about ghosts haunting places on earth? Thus the absence of any specific assigned location for the World of the Dead should not upset our conclusion that its inhabitants move in space and have spatial and temporal locations.

Secondly, the susuma has electromagnetic properties. When it becomes visible as a ghost, it presumably emits photons. The same applies to the "witch-light", a rapidly oscillating glow which the susuma of a witch is supposed to give off as it travels through the air. The susuma in its ghostly form is supposed also to be perceptible to the other distance senses, hearing and smell. In parenthesis, we note here again the prominence of observation as grounding for statements about the susuma: what counts as evidence of the presence of a ghost is visual, auditory and olfactory sensations.

Wiredu suggested that it is a reason for characterizing the Akan kra as quasi-physical, that medicine-men or people with ESP or medicinally-heightened perception are said to be capable of seeing the kra. This is in essence the same argument as the one I have just used about the susuma. Gyekye has objected that:

It must be noted, however, that these phenomena do not take place in the ordinary physical world; otherwise anyone would be able to see or communicate with the kra. This must mean that what those with special abilities see or communicate with is something non-spatial. Thus, the fact that the okra can be seen by such people does not make it physical or quasi-physical (whatever that expression means), since this act or mode of seeing is not at the physical or spatial level.⁽¹²⁸⁾

But if the kra really is seen at a particular spot, we surely cannot rebuke the inference that it exists in space. At least, it would not seem possible to do so by means of the argument here used, that the object seen is not physical because the act of seeing is not physical, otherwise everyone would be able to perform it. For by the same token one might argue that dogs do not physically hear the high notes of a dog whistle, otherwise humans would be able to hear them.

How does Gyekye account for the mobility and perceptibility of the susuma on his Cartesian interpretation of it, or, more precisely, of the okra of which he takes the sunsum to form a part? The answer is contained in the following words:

[i]t cannot be inferred that they [spiritual beings] are physical or have permanent physical properties. It means that a spiritual being can, when it so desires, take on physical properties. That is, even though a spiritual being is nonspatial in essence, it can, by the sheer operation of its power, assume spatial properties.⁽¹²⁹⁾

Gyekye here envisages a temporary assumption of physical properties by a spiritual being. Since he goes on to quote with approval Mbiti's statement "Spirits are invisible, but they make themselves visible", this is tantamount to an agreement that, when a ghost is seen, it is seen physically and in space. (Since ghosts are often not seen by everyone in the room, this admission would seem incompatible with his earlier position.) So the matter appears to devolve upon the following question: what is the nature of that being which, when it is physically observed, is a ghost, and again when it is not so observed, is a ghost? Are we to say it is physical or non-physical?

The theory wisely avoids committing itself here, and no doubt it would be better if we did the same. Once the contents of beliefs are agreed upon, and their implications drawn out, there is little point in pressing them for a decision on questions couched in terminology to which they have nothing to correspond. (There appears to be no word for "matter", "material" or "physical" in Ga.) But if we must make a choice, it would seem better to describe the susuma as "physical", rather than as "non-physical but possessing a capacity for materialization through the sheer exercise of its power"-- i.e., presumably, by thought. For the description of it as physical licenses a reason why it is now in a particular place which is familiar and comprehensible to us ("it has moved"). The other approach offers a reason ("it has thought") which is not comprehensible, or at least not familiar. And I think we may get some confirmation for the preference for the former alternative from the statement that we receive a new body at death. As was pointed out in II (d), this new body is agreed to be the susuma. If the susuma can be described as a body, it is presumably conceived as an organized physical entity.

It is appropriate at this point to consider the significance of the fact that the word susuma also means "shadow". Why is the susuma so called, and what, if anything, is it the shadow of? There is no agreement on these points among informants. Two hypotheses are worth considering.

(i) The susuma is the shadow of kla. It is the representation of it, the only thing which enables us to know what the kla is like. Here one must bear in mind the belief that one may see the reflection of something which is not itself seeable (for example, the reflection of a ghost may be seen in water or a mirror when it itself is not seen).

(ii) The meaning "shade", i.e., "ghost", is paramount. The ghost is a shadow of the body: it is not the body, but a reflection or projection of it. Because the in-life component of the person is taken as being identical with the post-death apparition, it is given the name "shadow", which is really or primarily appropriate to the shadowlike--seeable but untouchable--appearance after death.

(i) is supported by some highly knowledgeable informants. Otherwise one would have tended to prefer (ii), which is more straightforward and draws more closely on the literal meaning of "shadow". Moreover, one is familiar from other languages with the word for "shadow" being used for "ghost", e.g., Latin umbra, though it is a special feature of the Ga usage that the same word is used for an in-life component of the person.⁽¹³⁰⁾ Had (ii) been the correct explanation, it would perhaps have provided further support for the preference for the description "physical" over "non-physical" (for the perceptibility of the susuma would then be revealed as so essential a feature of it as to have determined even its name).

But even though this discussion of shadowness has not yielded clear support for our preference, it seems to me that the arguments which preceded it do show clearly that (whatever we may later find to be the case with the kla) we cannot give a Cartesian account of the susuma as pure consciousness devoid of any physical properties, such as is the sunsum of Akan thought on the Gyekyean interpretation of it. Far from being an immaterial entity inhabiting a body, the susuma is itself a body, and shares with the material body the properties of being organized and of possessing a spatial and temporal location, and the powers of movement through space and time and (intermittently) of perceptibility to the distance senses of human beings and animals.

Thus I suggest that the correct account of the susuma is that it is a Strawsonian person, to which both mental and physical predicates are applicable, and not a Cartesian ego, to which only the former category would apply. Strawson in Individuals argued that the fact that we apply predicates ascribing physical characteristics, and predicates ascribing states of consciousness to a person, should not lead us to think a person consists of two things: a body, which is the real subject of the mental predicates, and a mind, which is the real subject of the physical predicates (any more than when we say "The brick is square" and "The brick is red", we should then think that squareness and redness do not both characterize the brick, but independently characterize two distinct things, the brick's shape and the brick's color). Persons just are a kind of things to which both mental and physical predicates apply (just as bricks are a kind of things to which both shape- and color-predicates apply); and the notion of a person is logically prior to that of an individual consciousness. Since, as we have seen, the susuma possesses physical properties as well as mental ones, it falls under the concept of a person as that concept is analyzed by Strawson. It can only be represented as a pure consciousness or immaterial ego at the price of neglecting some beliefs which form a salient part of Ga conceptions of human nature and personal immortality.

If the susuma is a person in this sense, does this mean I am two persons, and not one? If so, (a) how do we reconcile this with our usual intuitions that I am only one? and (b) how are the two persons related? If not, what explanation can we give of why a person plus a person does not yield two persons? To these and other questions which are raised by our account of the susuma as a person in the Strawsonian sense, an answer can be attempted only after the nature of the kla has been considered.

THE KLA

Concepts of the Kla

The variety of the statements made by Gas about the kla is most striking, and creates initial bewilderment. For example, on the one hand the kla is said to be the highest element in man's nature, while on the other it is said that plants too have kla. Each of these statements is corroborated by a number of informants; and such apparent incompatibilities can be multiplied. One very obvious task which an analysis of the concept of the kla must fulfill is to account for these glaring discrepancies. It would seem that a fuller investigation must yield one of four possible conclusions.

(i) The inconsistencies are only apparent and disappear when the complex concept of the kla is understood. (ii) The concept of the kla is hopelessly confused. The best one can do is to document the various beliefs and idioms concerning it. But to articulate them in a coherent manner is an enterprise doomed to failure. (iii) Different people hold different concepts of the kla, and these rival views have taken hold--now here, now there--within the community. (Here one may invoke Gyekye's thesis of the importance of individual reflection in the formation of traditional thought systems.) (iv) With the kla, we are dealing not with one concept but with several, different to the extent that they could with a gain in clarity be expressed in different words. If (iii) were correct, thinkers a, b and c would all hold that we have a kla, but they would differ as to what kind of thing the kla was. If (iv) were correct, one and the same thinker would hold that we have a kla in sense a, a kla in sense b, and a kla in sense c. For (iv) to be correct, indeed it should be the case that many people are prepared to acknowledge, when they are invited to consider the matter, that they do have belief answering to each such statement.

The conclusion to which the evidence seems to me to point is (iv), that the kla is not a single complex concept, nor yet rival versions of the same concept, but more than one as relatively simple concepts which have gotten grouped together under the same name. (Indications that one of these is the original one, and the other or others later comers, will be mentioned as a matter of interest; but this is a historical matter on which I am not really equipped to comment.) As was stated above, a rich and varied collection of beliefs, linguistic idioms and customs has grown up around the kla, and the distinctions we shall draw between senses of kla must be based on and interspersed with short descriptive accounts of some of these.

The kla is said by Gas to be a part of God's nature in man. It is regarded by them as higher than the susuma. If it be asked "higher in what sense?", the answer is threefold.

Firstly, the kla is more powerful than the susuma. It can direct the susuma, whereas the susuma never directs it. The kla dictates a man's destiny, the message he is to bring into the world and the task he is to fulfill in it. We are said to take leave of our kla when we come into the world. It normally continues to guide and protect a man throughout his life. Someone who enjoys good fortune or has had a series of lucky escapes may be described as kla kpakpa ts, "the owner of a good kla". To say of someone that Ekla nyi ese, "His kla is following him", means that he is lucky. The kla is sometimes actually identified with destiny: Okla l,oshadi ni, "Your kla is your destiny".

A creative as well as directive power is sometimes attributed to the kla, for it is even said that Le ebbo, "He is the one who created you". Since it is his kla which makes a person what he is, gratitude to him may be expressed by referring to his kla: "Thanks be to your kla", "May God bless your kla". By the same token, a person's kla may be abused with all the insults which might be heaped upon its owner: "Your roguish kla", "Your kla's foolish face", etc. The kla is strongly associated with physical health and vigor. If you are allergic to something, your kla does not like it. Ob kla, "You haven"t got a kla", is a statement made to someone who is thought to be too passive, someone who puts up with things against which he should react or rebel.

Secondly, the kla is morally perfect. It is tarnished when its owner commits a serious misdeed, and because of its association with health, its displeasure may be manifested in sickness. Concerning this there is some evidence dating from the eighteenth century. The Moravian historian Oldendorp writes:

The priests of the Akkran [i.e., the Ga] have a theory about diseases. They see in them a result from a misunderstanding between spirit and soul. In their opinion as long as the two live together in peace and concord like husband and wife, man is healthy, but if one of the two commits a fault, the harmony is disturbed: the pure part wants to separate itself from the impure one, hence arise inner trouble and sickness of the body.⁽¹³¹⁾

It will be noted that Oldendorp records a belief that either the kla may be alienated from the impure susuma, or vice versa. But I have not found any informant who accepts the latter situation as a possible one, nor is there any evidence to that effect recorded in Field. The kla always seems to be regarded as essentially pure; any taint which falls upon it from the action of its owner should be washed off by means of a prescribed ceremony.

Physical illness is not the only possible outcome of offending the kla. It may withdraw its protection and offer no more moral or spiritual guidance. The consequences of this are supposed to be dire. Ekla eje es, "His kla has left him" (the opposite of Ekla nyi es) is a statement made when madness, alcoholism, etc., has befallen a person. The withdrawal of the kla's protection may be followed by the intrusion into the personality of an alien element known as a gbeshi. As mentioned in II (b), this seems to be regarded as a kind of spirit. Any form of socially unacceptable behavior which does not occur in a man's immediate family, and thus cannot be attributed to heredity, is liable to be attributed to a gbeshi. It is regarded as a disruptive force which interferes with the links binding the kla and the susuma, and prevents the victim from fulfilling his destiny. Some medicine-men are believed to be able to perform ceremonies which will rid a person of gbeshi.

Thirdly, the kla is regarded as more "honorable" than the susuma. After childbirth, recovery from a serious illness, survival of an accident, or a signal success of any kind, the kla is "washed" and thanked at a special ceremony. Some other ceremonies in connection with the kla have been referred to already; yet others are described in Field.⁽¹³²⁾ The susuma however, receives no such veneration.

A fourth point, more controversial than the preceding ones, may be added. This is that the kla is never seen; thus it is either immaterial, or further removed from the ordinary material world than the susuma which, as we have seen, is intermittently perceptible, either itself or through a reflection. However, some say that medicine-men can see the kla. But this is denied by others, who explain that what the client wants to be assured of is that the practitioner has really been in communication with the kla: a medicine-man might use the terminology of vision in order to satisfy him of that; but he does not, properly speaking, see the kla. The question stands in need of closer investigation.

To the question: How do we know that we have a kla? What is thought to make the postulation of it necessary? the answer is obscure. There is, or at least is considered to be, a lot of empirical evidence, known to the ordinary person, to support the theory of the nature and activities of the susuma. But the theory of the kla seems to enjoy no such backing. It is possible that the pronouncements of medicine-men, who, as just remarked are believed able to communicate with the kla, have formed an important source of the beliefs relating to it.

What conception of the kla is suggested by these beliefs, idioms and observances? I believe it is one which figured largely in the popular thought of the Western world in earlier times: that of an individual of a higher-than-human order of being, who determines one's destiny and watches over one's welfare: an attendant personal spirit like the Greek personal daimon, the Roman genius, or the Christian guardian angel. I have found that when this interpretation of the kla is put to informants, it is readily accepted. If correct, it explains among other thing the honors paid to the Kla after successes and deliverances from danger; why your kla may be said to abandon you or to follow you; why (a matter not touched upon above) prayers to the dead can continue even when a reincarnation of them is believed to be alive; and why Gas sometimes feel perplexity as to whether the kla is oneself or is outside oneself. Since a spirit of the kind in question is a more divine being than a man this interpretation of the kla also accounts for why the kla is said to be God's power in man.

A person's kla is sometimes said to be related to the day of the week on which he was born. Miss Field writes as follows:

The "day-name" which is given according to the day of

the week on which a person is born, if often known as

the "kla-name". It is said that "all people born on the

same day have the same kla". . . . The day-name may be used

by medicine-men in killing, and furthermore the killing

may reach the wrong person as well as the right one

through their common name. For instance, if you want to

kill a man named Kwaku (the name means "born on Wednes-

day") you call in a bad medicine-man and he prepares a

medicine using the name of Kwaku, lays it on the ground

and arranges that when Kwaku walks over it he will sick-

en and die. However, another Kwaku may walk over it and

die instead of the first.⁽¹³³⁾

Any ceremony involving the kla is held on the day of the week on which one was born; and because of the belief that all those born on the same day have the same kla, the presence of people born on that day is considered especially appropriate. The belief that those born on the same day have the same kla is also found among the Akans. Eva Mayerowitz reported that the Akan assign each day of the week to the rulership of the deity of a particular planet who protects those born on that day and whose influence is responsible for the common traits of character supposedly possessed by them.⁽¹³⁴⁾

It is difficult to see how this particular astrological conception of the kla, whereby there are only seven forms of kla, can plausibly by combined with a conception of the kla as the guardian spirit of an individual. What would happen when the interests of two Kwakus conflicted? But the Greeks and Romans also used to worship their daimn or genius on their birthday (once a year, however). The connection between astrology and the concept of a guardian spirit is a time-honored one, and therefore the difficulties in understanding any version of the connection, let alone the present simple one, should not necessarily lead us to feel that the indications that the kla is to some extent astrologically determined cast doubt on our interpretation of the kla as a guardian spirit. The belief that each person has an individual guardian spirit, and the belief that the character and destiny accruing to one from such a spirit are determined by the disposition of the planets at one's birth have, however mysteriously, often been conjoined in popular thought over the centuries.

Miss Field also records another view linking the kla with names. This concerns the Ga naming system whereby names determined by order of birth, one set for male children and another for female, recur in alternate generations of a family. According to Miss Field, it is a postulate of this system that a child (a) possesses the same kla as. and (b) is a reincarnation of, the grandfather, grandmother, great-uncle or great-aunt whose name he bears.⁽¹³⁵⁾ If (a) alone were believed, then one would be able to infer either (i) that possession of the same kla is not, for the Ga, constitutive of personal identity, or (ii) that personal identity is not, for the Ga, a one-one relation. For an eldest son, for example, who had several sons of his own, some of whom in turn had a son, would have the same kla as each of his eldest grandsons. If he is not regarded as the same person as them, then possession of the same kla does not constitute personal identity; if he is, then a single person can be identical with more than one person--not merely after his death, but during his lifetime. But Miss Field's explicit addition of (b) to (a) at first sight seems to necessitate (ii) (since a reincarnate is presumably the same person as he of whom he is a reincarnation). The only way of avoiding (ii) would be to infer (iii) that, for the Ga, "Y is a reincarnation of X" does not entail "Y is the same person as X".

If, then, these beliefs were held by the Ga, one would have to conclude that they had either a very unusual notion of personal identity, whereby a person could be identical with one or more of his contemporaries, or a very weak notion of reincarnation, whereby X may be a reincarnation of Y without being the same person as Y. It is also worth noting that there are obvious difficulties in combining the beliefs linking the kla with lineage-names with the beliefs referred to in connection with day-names; for two possessors of the same lineage-name might not be born on the same day, so that by the one criterion they had the same kla, but by the other they did not.

My inquiries however, yielded different results, which would not involve these difficulties; for my informants all denied that possessors of the same lineage-names necessarily have the same kla. They had, indeed, never heard of more than one person being supposed to reincarnate the same person at the same time, and evinced hesitation and unease at the question whether it was possible for this to happen. My impression was that they found the question conceptually odd. The explanation of this which most naturally suggests itself is that they look on personal identity as inherently a one-one relation. It is certainly often believed that children take on the characteristics of those whose names they bear. But this appears to be most usually attributed to a rather vague "influence" which comes short of full-scale reincarnation. One also hears of members of the grandchild generation who after the death of the member of the grandparent generation whose name they bear, are perceived by those around as becoming more like him. Whatever else this may suggest, it at least supports the conclusion that the junior is not regarded as a reincarnation of the senior either during his lifetime, or merely in virtue of bearing his name, and doubt is cast upon the assertion that namesakes are automatically looked upon as reincarnating their eponyms. For these reasons, I believe we can discount the suggestion that there is a relationship between the kla and lineage-names, and justifiably sidestep the task of deciding what conception of the kla would be involved in such a belief.

A further set of beliefs concerns the kla and witchcraft. Beliefs to the effect that witches operate by eating the kla of their victims are amply documented in Field and Debrunner, and I shall not dwell upon them here. Witches are said to eat the kla limb by limb and organ by organ, either on one night or over a longer period; when the heart is taken, the victim dies. Since each part of the physical body has a kla counterpart, the conception of the kla involved in such beliefs is, as Field describes it, "an invisible body, the perfect double of the physical body".⁽¹³⁶⁾ The kla is said to reside in the blood, and an alternative way in which witches are said to operate is by sucking the blood of their victims. The expression Obe kla, literally "You haven't got a kla", means "You are able to withstand witchcraft".

The conception of the kla involved in this set of beliefs, which for convenience we shall call kla II, has it in common with that of the kla as guardian spirit ("kla I") that in both cases the kla is essential to physical health and vigor. But otherwise they seem rather different from each other. Firstly, kla I is pre-existent--one takes leave of it in coming into the world-- whereas kla II resides in the blood of the physical body, and its existence is therefore presumably contemporaneous with that of the body. Secondly, kla I is a divine and powerful part of man and the susuma is subject to it; whereas kla II is something which can be preyed upon by the susuma of a witch. Thirdly, the results of kla I's abandoning a person are moral weakness, folly or madness and it abandons him as a whole; whereas kla II can be taken from him piecemeal, and the results are not moral or intellectual weakness, but physical weakness. Finally, kla II seems to be an impersonal entity, not the sort of thing which could be thanked for good fortune or in general to which agency could be ascribed, as it freely is to kla I.

Because of all these differences, I believe that kla I and kla II are actually quite different conceptions which have got grouped together under the same name. Kla II may perhaps be described as a "life-force", or what I believe theosophists call the "life-body". An indication of the difference may be found in the fact that the phrase Ebe kla ("He has no kla") has two quite different, and almost opposite meanings. These have already been explained: the phrase can mean "He has no spirit" (in the sense of "spunk") and "He is not vulnerable to witchcraft". In the former sense it speaks of weakness, and in the latter sense, of strength. The fact that identical sentences containing the word can have almost opposite meanings seems strongly to confirm the hypothesis that the word itself has two quite different meanings, and also to suggest that one of them is not original to the language; for languages normally tolerate ambiguous words only to the extent that they do not occur in similar contexts.

There is reason to think that, if one of these meanings is a later comer, it is kla II and not kla I. In the first place, the only area where kla II seems to enter in is attacks by witchcraft. And Miss Field observes that witchcraft may not be indigenous to the Ga, citing two facts. First, the witchcraft practices are less common among them than the neighboring ethnic groups; and secondly, there is no Ga word for a witch, instead, the Fante aye or the Twi Obeyefu is used.⁽¹³⁷⁾ Further, a view of the kla as capable of agency, which we have seen to be characteristic of kla I, is deeply entrenched in both language and custom. It would be tedious to demonstrate this in detail; a review of the idioms and customs referred to will make it sufficiently plain. Çla I, therefore, has a title to be regarded as the original Ga concept, on to which kla II has been grafted--a process perhaps helped by the presumed connection of them both with physical life and health. But kla I is presumed to affect health in ways that kla II is not. For example, it is supposed to be displeased when its owner is a victim of neglect or improper treatment. It may manifest its displeasure, among other ways, in the person's falling prey to a long wasting sickness, which may lead those around to inquire into the cause of the problem and rectify the grievance. Kla II is not credited with motives of this nigh-personal type.

Relation of the Kla to the Susuma and the Body

It has just been suggested that the word kla is used in two quite different senses. In discussing the relations which the kla bears to the susuma and the body, it is obviously very necessary to determine whether we are talking about kla I or kla II in any given case. In practice, however, almost all the evidence bearing on this topic fairly clearly concern kla I. This is not surprising if, firstly, the suggestion that kla II is a latecomer to the Ga conceptual scheme is correct and, secondly, it is borne in mind that kla II seems to be mainly confined to contexts of witchcraft.

The doctrine that the kla leaves the body at death might be thought an exception to this. Might not this be referring to kla II if kla II is a kind of life-force? But a fuller statement of the doctrine runs, "The kla leaves the body at death and goes back to God", and it therefore almost certainly refers to kla I. In what follows then, "kla" will refer to kla I. What is the relation of the kla to the other two "parts" of the human being?

Presence "in" or "with" the body is ascribed to the kla by some Gas. This appears to be the only candidate for a physical attribute of the kla (apart perhaps from its being seen by medicine-men; although this was provisionally rejected). But others think of the connection between kla and susuma plus body (as a shorthand device I shall refer to the latter pair as "person" in the next few paragraphs) as a force binding them together rather than as a compresence.

A relation of possession or ownership holds between a person and his kla. The person is said to be the owner of his kla, and not vice versa. The kla is the kla of that particular person. Can we go further and say that the kla could only be the kla of that particular person? Probably not. There is, as far as I know, no evidence of the belief that your kla is uniquely yours. (Indeed, the belief that persons born on the same day have the same kla seems to be evidence to the contrary. But it is not quite clear whether this is so, and what is common to them is the full kla of each, or whether one of two other possibilities holds: (a) it is not the entire kla that they share, but there is an individual residue; or (b) the kla falls into types and those born on the same day have the same type of kla.)

On the other hand, we can say that the person could only have that particular kla, for "it is your kla that makes you what you are". Thus a causative, creative or productive relation exists between a kla and its owner. How this is implemented is left as mysterious as the nature of the kla itself. Whether the susuma resembles the kla, so that the kla's making you what you are consists at least partially in imparting its nature to you, is also left unstated; although if the view that the susuma is the shadow of the kla is correct, presumably some resemblance must obtain between them.

The role of the kla which most arouses the interest and engages the emotion is that of director of destiny. We must say that the kla directs, not that the kla controls. The person has freedom, has a mind of his own, and the unity between him and his kla is not that between a robot and his program. It will be recalled that a person may displease his kla to the extent that the kla withdraws his guidance and abandons him to an intrusive gbeshi. Again, the mechanics whereby the kla directs the person are not spelled out.

It will be seen that we have got along very nicely referring to the body plus the susuma as "the person". This has not led to any problems. The fact is symptomatic of the kla's being, in a sense, external to the body plus susuma, as they are not to each other. All the relations which hold between the kla and the body plus susuma--compresence (?), possession, causation, direction--are ones which may hold between two quite separate entities. The body and the susuma are connected in a close though not inseparable unity; but the kla cannot really be said to be a constituent of, or element in, the person.

This fact may help to explain why, after death, the identity of the person is sustained by his susuma, while the kla is said to lose its individuality. By this it is perhaps meant, as we can now see, that it loses its association with the individual who has owned it, not is own individuality; on our interpretation of kla I, its individuality and that of its owner are distinct. The import of the belief can now be seen to be as follows: that upon a person's death, the connection between him and his guardian spirit is dissolved; that the guardian spirit returns to God; and that the person, in the form of his susuma, enters the World of the Dead. The unity of the kla with the susuma does not persist through death, any more than does its unity with the body. The kla survives death, but the immortality of the person does not consist in its survival, but in the survival of the susuma, which is supposed to retain a considerable range of the characteristics of the living person. The separation of kla and susuma after death is one of the strongest indications that the susuma is not a part of the kla, as, according to Gyekye, the susma is of the okra, but is ontologically as well as logically distinct from it.

Personal Identity, Unity and Immortality

It was shown in II (d) that a person's survival of death, the identity of the person who exists after death with the person who existed before it, can be said to consist on the Ga view in the continued existence of his susuma, which is conceived as a substance or owner of properties. This sounds like a Cartesian account of survival, but the analysis of the concept of susuma in the above section on susuma after death will have made it clear that the susuma is vastly different. For the susuma is itself a body which exists in time and space, so that its movements should in principle be able to be mapped continuously in both dimensions for any beings with the means of monitoring and recording them. This view of personal survival would thus not face some of the logical difficulties which critics of Cartesianism have shown to attach to the notion of personal identity in the absence of spatiotemporal continuity. But at least two other problems arise from it. Firstly, since human beings do not meet the specifications of the beings by whom such continuity would be checkable, and hence usable as a criterion, what grounds can the theory proffer for maintaining that the susuma continues to exist and to retain its identity?

An answer to this question has to be elicited by the interpreter from the nexus of received ideas and arguments; he will not find the answer ready-made. I would hazard the following account as the kind of justification which is suggested by the views recorded in the above section on the Susuma after death. Reports of instances in which the dead have appeared to the living in any one of the three (or four) recognized ways--ghostly apparition, possession, materialization (and reflection)--are taken to license the statement "Some of the dead have been perceived by the living on at least one occasion". To the question "How do you know that the person exists before and after such a manifestation--or between such manifestations if there are more than one?" it might be answered that the assumption that he does so exist is rendered much easier than the hypothesis that he comes into being afresh on each manifestation, by the parallel assumption we make about human beings whom we meet intermittently in life. To the further question "Granted the authenticity of the evidence and granted the last-mentioned assumption-- supposing, that is, that some men survive death--how do we know that all do?" the answer might be to the effect that those who have been perceived cover such a range of variables conceivably relevant to fitness for survival that no such characteristics has yet been found which has not been instantiated both among those who have, and those who have not, appeared to the living. Pending therefore the discovery of some relevant difference between the two groups, the evidence points to the conclusion of the universal human survival of death.

Such might be the shape which a response to the request for a justification of assertions of survival consonant with Ga beliefs would take. It will be clear that the onus falls mainly on the question of the authenticity of the reported sightings, hearings and smellings. Here again, in the area of survival of death, we find the views of the Ga exhibiting the same empirical bias that we discovered in their account of the nature and activities of the susuma of a living man.

The grounds for supposing the susuma to continue in existence which have just been referred to involve positing a continuity of physical appearance (face, stature, voice, etc.) between the person in life and his susuma after his death. Psychological continuity in the form of identity of memories, affections, objectives, etc., is also held to exist. This gives rise to the second problem with which the Ga view of personal survival might seem to be faced. We have argued that the susuma duplicates the structure of the human being in possessing both physical and mental attributes. (This becomes fully evident only when the beliefs about the susuma after death are considered. For after death, the susuma is believed to be intermittently perceptible to waking human beings in a normal state, as it had not been in life--in dreams it may be perceived by the susuma of another person, but not by a non-dreamer or someone not exercising special psychic powers.) How is this duplication compatible with the unity of the person in life?

Again, one can only attempt an answer which is as far as possible consonant with known beliefs. Such an answer might appeal to the fact that in life there is no temporal overlap between the activities of the body and the susuma. They are like workers on different shifts and with different schedules. When the body (which includes the mind) is on duty, it might be said, the susuma is off duty, and neither is able to perform the other's functions. The well-known fact that dreams often occur in series might be cited to show that dream life has a continuity of its own, just as waking life does. Since body and susuma complement one another both temporally and functionally, they are candidates for a holistic union.

However, the more the lack of overlap between the body and the susuma be stressed, the more puzzling it becomes how the susuma alone can represent, or be, the person after death. For we seem almost to have the case imagined by Locke:

Could we suppose two distinct incommunicable consciousnesses acting in the same body, the one constantly by day, the other by night; . . . I ask . . . whether the day and the night man would not be two as distinct persons as Socrates and Plato.⁽¹³⁸⁾

This can be sharpened by introducing some considerations about moral responsibility. I am surely not responsible for what my susuma does in dreams; and it is presumably not responsible for any conscious misdeeds committed by me in my waking life. Now in the World of the Dead, the susuma is punished for wrongs consciously performed by the person in his lifetime. How does it come to bear the responsibility for them? It should be noted that unlike the Akan, who speak of a "good sunsum", meaning a generous disposition,⁽¹³⁹⁾ the Ga do not appear to apply moral predicates to the susuma at all. This seems to be only consistent with the plausible position that it is conscious choices which determine our moral character. But how is it consistent with the belief that it is the susuma which faces reward or punishment hereafter? Is the susuma not being punished for what is not its fault, or rewarded for what is not to its credit? As Locke forcibly argued:

If the same Socrates waking and sleeping does not partake of the same consciousness, Socrates waking and sleeping is not the same person; and to punish waking Socrates for what sleeping Socrates thought, and waking Socrates was never conscious of, would be no more of right then to punish one twin for what his brother-twin did, whereof he knew nothing, because their outsides were so like they could not be distinguished.⁽¹⁴⁰⁾

These considerations suggest that an attempt to account for the unity of the person cannot succeed by setting up a demarcation between our day-time and our night-time consciousness. Another consideration which tends in the same direction is that whereas the susuma can, when it is apart from the body (in dreams) function in the manner characteristic of it as an individual entity, the mind apparently has no such power. It can produce its own characteristic operations (i.e., ratiocination and other conscious thought processes) only when the susuma is present with the body. This leaves it open for thoughts which occur when the susuma is present with the body to be the product of the joint operation of the two of them. This is consistent with the fact that the human susuma is certainly regarded by Gas as being rational, and also with the fact that dreams may have reference to events of our waking life long since forgotten, whereas in our waking life we are unaware of most of our dream-experiences. To express this difference in terms of a distribution of knowledge between the mind and the susuma, we may say that the susuma has access to the mind's information, whereas the mind has access to only a limited amount of the susuma's information. Perhaps it is partly for this reason that Gas say that the susuma is "wiser than the man himself", or "knows more than the man himself". But the question of the unity of the person may be approached from another angle as well.

Adapting Plato, who said that life is a practice for death, Gas might well say that sleep is a practice for death. For, as we have seen, they believe that sleep consists in the existence of the susuma in independence both of the kla and of the body; and this is precisely the state it is in after death. Thus the Ga view of immortality has affinities to Christian descriptions of the pre-resurrection life after death as a sleep. This has to be inferred; it is not stated, because sleep would be an unsuitable metaphor for death from the Ga point of view. In the first place, death (i.e., here the mode of existence of the dead) is not regarded as essentially a halfway house to another mode of existence, for some but not all of the dead are reincarnated, and in the second place it is not seen as preparatory to a fuller existence, for there is no conception of a condition for human beings which would be superior to it.

If both sleep and death consist in the independent existence of the susuma, wherein lies the difference between the two states? One difference presumably stems from the more final separation of the susuma from the body which death involves. Whereas in sleep the susuma is bound to return to the body, and therefore it cannot establish a full life of its own, after death it has no such ties and is able to establish a life on its own account. Its powers of perception, feeling, thought and volition could attain whatever perfection they are capable of when their exercise is not continually being interrupted by the necessity of attention to needs and interests arising from embodied existence. We recall the belief that the susuma only gradually, over a period of about three weeks, loses its attachment to the places and pursuits it had frequented in the body; but then it generally does lose it, except in cases of exceptional physical or emotional trauma.

Another difference between sleep and death might be found in the fact mentioned above, that it is only after death that the susuma is perceptible to a waking human being in a normal state; in dreams the susuma may be perceived by the susumai of other persons, but not by the persons themselves.

The (intermittent) perceptibility of the susuma after death, it might be said, should properly not be taken, as it was in the above section regarding the nature of the susuma, as an indication of what the susuma was like all the time, and used as an argument for the physical nature of the susuma generally, in life as well as after death. The beliefs on which that argument was based do not entail that death is a separation of three things previously combined, one of which, the body, undergoes change from the moment of death onwards, while the other two, the susuma and the kla remain unchanged. The beliefs in question are quite consistent with the possibility that the susuma is itself changed in nature, either (like the body) after death, perhaps in virtue of its entry into the World of the Dead, or at the time of death, or perhaps even before death. The last of these three possibilities might seem appropriately consistent with the supposed "travelling" of the susuma of a living person in a terminal illness. But however this may be, the occurrence of a change in the susuma may be reflected in the statement that we receive a new body after death; although, as we saw, this always refers to the susuma, the description of it as a "new body" perhaps indicates that it now takes on some physical characteristics not possessed by it before.

This point, if accepted, does not undercut the arguments for the physical nature of the susuma in life altogether. We are still left with the spatial mobility of the susuma. But it might very well be argued that the erosion of the claims about perceptibility as regards the living constitutes a serious weakening of the grounds on which the concept of the susuma was claimed to have a similar structure to the concept of the person, i.e., that the susuma is a bearer of physical as well as mental properties. This, it might be said, applies only to the susuma after death; before death, the susuma does not have a sufficient range of physical properties for us properly to call it a person. Once it is admitted that any commentary on the Ga theory of man must allow for the possibility of a change, perhaps a quite radical change, in the nature of the susuma at or around death--a change whereby it takes on a power to manifest itself to the senses of waking human beings in a normal state which it did not previously possess--we are acknowledging the possibility of a discontinuity in the status of the susuma. The susuma is a kind of understudy whose role in the provinces is confined to rehearsals, but when the play opens in the capitol it appears in the leading role. Features of the susuma after death cannot just be lumped together with those before death to yield a composite portrait of the susuma which perhaps is not true of it at any stage of its history.

Acceptance of this point, then, might mitigate the rather stiff problems which the Ga doctrines initially seemed to face in the areas of the unity of the person in life. The considerations about moral responsibility showed that we cannot deal with these problems by setting up sharp barriers between our daytime and our nighttime consciousness. Then considerations about the dependence of the mind's operation on the presence of the susuma, and about the distribution of knowledge between them, further showed that the susuma's consciousness comprises the mind's, but not vice versa. Finally, the discussion of how, on the Ga teachings, sleep can be differentiated from death showed that at or around death may, or perhaps even must, be posited, not merely a change in the susuma's relation to the body, but an intrinsic change in the susuma itself, whereby it acquires, in addition to its mental characteristics and its power of independent motion, the full range of physical characteristics minimally necessary for human personhood. What precedes then, is far from explaining the unity between the susuma and the body; but it is hoped that it serves to indicate that the explanation must lie in one direction rather than another.

CONCLUSION

In this paper, an attempt has been made to clarify the Ga concepts of the kla and the susuma, to explain their relationships to the body and to each other, and to show how they function as a framework for beliefs about one of the two forms of immortality accepted by the Ga, disembodied survival. (The other, reincarnation, could only be touched upon in passing within the bounds of this study.)

The puzzling diversity of beliefs about the kla has been tentatively explained by the hypothesis that the word kla is used in two senses, that of guardian spirit and that of life-force or life-body. The first of these is capable of accommodating the connection of the kla with the day-names; but an attempt to link the kla with lineage-names faces serious problems. The kla in the first sense (the most firmly entrenched and perhaps the original one) stands in a somewhat external relation to the susuma plus body: though each human being has a kla, the kla is not a part of him in the sense of being an element in or constituent of him. For this reason, the kla is not involved in disembodied survival (although it is involved in reincarnation; if X is a reincarnation of Y, X has the same kla as Y). Uniqueness of one's kla is thus not part of what it is to be an individual human being. The source of beliefs about the kla is probably esoteric, and the links between it and the individual were found at certain points to be too mysterious to permit of meaningful discussion.

A Cartesian interpretation of the susuma is rejected, on the grounds of the physical attributes which it possesses: mobility in space and (after death) perceptibility to the distance senses. The susuma of a living person comprises, but is not confined to, the unconscious mind: the susuma overlaps with the jwengm, or (conscious) mind, which appears to be conceived in a materialistic fashion. Some of the problems which these views raise for the unity of the person are considered. Four states in which the susuma is held to exist independently of the body are shown to require a closer characterization or a more explicit causal explanation than the theory seems to provide. The Ga hold both physical and psychological continuity to obtain between a person in life and susuma after his death. Since it is the susuma which after death sustains the individuality of the person, the susuma after death is a Strawsonian person, a bearer of both physical and mental properties. In this way, it is a shadow (susuma) of the person. But in life it has this status only potentially, being not yet possessed of the full range of physical characteristics essential to human personhood. It is clear at many points that the theory of the susuma has an empirical basis, in the sense that it is a means of explaining phenomena whose occurrence its proponents regard as having been adequately established by either observation or testimonial evidence.