It was in the mid 1980s. I was at the Australian National University, working hard for my Ph. D., when I met Taro Go, a young Japanese anthropology student who, a few years later, died during his first fieldwork on the island of Sumba, and who would have been among the contributors to this collection of essays, had it not been for this tragedy. Anyway — on a certain hot night in Canberra, a few Japanese students got together at Taro’s room, drinking wine and engaging in a nonsensical talkand in useless discussion. One of us challenged us all whether anybody could provide a reasonable explanation why one is prohibited to urinate on an earthworm (as is a Japanese custom).
“An earthworm helps farmers by digging up soils, so farmers think, I guess, that it is disgraceful to urinate on such a helpful creature” said one. Another student (who, incidentally, was working for his medical degree) suggested “Urinating on an earthworm means that you are, at the time, fiddling with soil; that is, you’ve got dirty hands. The prohibition works such that dirty hands never touch a sensitive organ.” These two, one might say, are functional explanations. I myself suggested a third — a symbolic, structuralist interpretation, referring to the metaphorical connection (a morphological similarity) between a penis and an earthworm.
After a while, Taro gave us a definite explanation — “Just Because!”
I now think that Taro’s was the closest to a native explanation (that is, no explanation).1 People often answer, with a tinge of boredom because of the repeated anthropologist’s questions of “Why do you do that ritual in this way?”, “Just because! There is no reason. That, you see, has been our way since our ancestors”.
What is the point of the story? The point is that we, anthropologists, have to admit that “no explanation is a good explanation” — at least, “from the native’s point of view”. The same point has been stated repeatedly in this volume (by Sugishima and Nakamura, among others).
What, then, is the anthropologist’s task if she wants to write ethnography from the native’s point of view — merely describing whatever is the liturgical sequence of the ritual in question without explaining anything? Exotic as it may be, that kind of ethnography would be, almost certainly, of no theoretical significance. Taro’s answer might have been meant to be a challenge, a provocation: “How else can the anthropologist analyse a ritual?”
This paper attempts to show yet another way of an anthropological analysis of ritual — neither a functional, nor a structural, nor a simply descriptive one.2
To make my contention clear, let us imagine we are asked to analyse the game of chess. There can be conceivably an infinite number of ways of analysing chess. De Saussure distinguishes two types of analyses of chess: (A1) external analyses such as a study of historical expansion of chess from India to Middle East, to Europe, to China, or to other countries such as Thailand, Japan, etc, or an aesthetic study of the various chess pieces and (A2) internal analyses such as a study of the interrelatedness of various rules of the game. The second kind (A2) is defined, by de Saussure, as the structural analysis. Yet, apart from the two types mentioned by de Saussure, we can surely make our analyses in yet another way, namely, once we focus on an actual match of chess. This way can also be divided into two types, following the Saussurian approach — (B1) external, such as a study of a historical importance of a certain game (Bobby Fisher becoming the first American champion etc.) or a micro-psychological study of the players in a certain match and (B2) a study of strategy and tactics, like those descriptions one find in common chess literature. I would like to make the (B2) kind of analysis in this paper, with a certain theoretical twist.
Turn back to the actual rituals. In this paper, I will deal with a sequence of ceremonial customs revolving around an occurrence of death, among the Endenese of central Flores. As the aim of the game of chess is, as any player would agree, clear, that is, “to win”, so is that of mortuary ceremony and is agreed upon by almost any player — “to separate the dead from the living”. The entire verbal framework (Fox 1989) is usually presented to consist of different stages, each contributing to that aim. I would like to focus upon smaller segments (than a verbally framed stage) to see how people “do things,” — that is, re-establish their social world — especially “with words.”
In his influential book, How to do things with words, J. L. Austin (1975) (Austin 1975 (1962)) begins his argument by focusing his attention on a special kind of speech acts, which he calls “explicit performatives”, and then proceeds to distinguish between three kinds of speech act (discernible in every speech, not only in explicit performatives), that is, (1) locutionary, (2) illocutionary and (3) perlocutionary acts. In other words, every speech act, according to Austin, performs the three kinds of acts mentioned above simultaneously.
Let us focus on the explicit performatives, whose illocutionary force is conspicuous even to an ordinary speaker, that is, a naive non-philosopher.
The salient fact about this kind of speech genre is that it directly
works on the world. When you say, “Out”, as an umpire (in a baseball
game), then the batter is out; when you say “I name this ship Queen
Elizabeth the second” (as a person entitled to do so in the proper
situation), then the ship is named “Queen Elizabeth the second” and so
on.
In short, in this kind of situation, one can exercise a direct influence
upon the world by the mere utterance of words. The man, who was safe
(“not out”) before, is now “out”; the ship, which bore no name before,
is now named “Queen Elizabeth the second”. This is the way “explicit
performatives” work on the world.
D. Schneider (1984) (Schneider 1984) once emphasised a more dynamic aspect of kinship than that of merely one’s “being” in a certain kinship relation with another person, namely one’s “doing” something towards another. In his words, kinship is not only being but also doing. What I want to establish here is that kinship for the Endenese is not merely being, nor only doing, but also “saying” — saying something to influence the actual world of kinship relations. And a ritual provides the appropriate setting for “saying”. The Endenese make maximum use of the illocutionary force to adjust their kinship world.
Let me introduce you to the proper ethnographical background. The Endenese is an Austronesian speaking population, inhabiting the central part of Flores, to the west of the Lionese (reported, in this volume, by Sugishima as well as Aoki) and to the east of the Nage Keo people(s) (see the contribution by Forth).
Native exegeses of rituals among the Endenese can be said to be composed of two parts — (1) a cosmological and (2) a sociological one. Those two adjectives are conventional labels, employed here to refer to sets of idioms, interrelated to each other, each set somewhat closed on itself — a set I have called a “language system” (see, for example, (Satoshi Nakagawa 1992)). The cosmological language system contains such words as “witch” (’ata porho), “ancestral spirits” (’ata mata, ’embu kajo), “spirit” (nitu), “to attack” (tau) etc. The sociological language system consists solely of idioms of kinship, (see presently).3
The exegeses of agricultural rituals (nggua tana watu) are rich in cosmological interpretations, both official (and unanimously agreed upon) and spontaneous (those of a personal nature, not necessarily with unanimous agreement), but they are lacking sociological interpretations; whereas rites de passage (there is no special native category denoting this genre of nggua) are given both kinds of interpretations.
Be it noted that I am not saying that nggua tana watu have
no “sociological” relevance; they do. For example, a specific
clan may want to emphasise its political power by conducting a nggua
tana watu in an unusual time, thus forcing the rest of the society
to participate in the ritual.4
In this way, people do comment on a nggua tana watu,
or on the way of performing a certain, specific occurrence of
nggua tana watu. However, when asked to formulate the meaning
of a ritual in general (say, a ritual held at the time of
planting, kaki), without referring to any actual performance,
people do not attach any sociological meaning to kaki.
I would like to use the word “political” rather than “sociological” for this genre of exegeses. Political kind of speech5 is the one which mentions the “actor’s intentions”. About this genre, a short theoretical comment, in relation to the overall conclusion, will be given at the end of this paper,
In contrast to the nggua tana watu, rites de passage are rich in both cosmological and sociological interpretations. This paper focuses on one of the rites de passage of the Endenese, i.e., the funeral ceremony.
The Endenese regard an occurrence of death as a potential threat to, as well as an occasion for the re-establishing of, their current (presumably stable) social world. Such a conceptualisation is salient in the indigenous cosmological and sociological exegeses.
The main theme of the ritual is thus the removal of the ill effects of the death from the society and the reconfirmation of the peacefulness of the living world. This cosmological theme is repeatedly confirmed by the informants. They refer to every specific ritual observance by saying “this is done in order to separate the dead from the living” as well as by invoking the idea of witches, who are supposed to devour the dead body. 6 In other words, such exegeses are not hidden deep in the symbolic structure of the ritual, waiting to be found and analysed by the anthropologist, but are explicitly formulated by those involved.
On the fourth night, for example, after a period marked by prohibitions, at dawn, the bereaved go to bathe themselves at a certain place, “in order to cleanse themselves and rid themselves of the deceased”. When they return from the bathing, an ’ata marhi (a medicine man), waiting for them at the threshold of the village, sprinkles bhonggi (enchanted rice grains) over the bereaved, “in order to separate them from the deceased”. Everything is, thus, clear to the natives, except for anthropological questions such as “why the fourth night?”
Sociological exegeses also revolve around the same kind of theme. Yet, it is not until we know enough about the Endenese kinship idioms (alias “sociological language system”) that we can fully understand and analyse the oratory of the Endenese in the context of a mortuary ceremony.
The social world of an Endenese is conceptualised as consisting of
tripartite parts — his/her agnates (’ari ka’E),7
his/her wife-givers (ka’E ’embu) and his/her wife-takers
(weta ’anE). These different types of relation
accrue from the two ordering principles of (1) agnation and (2) the
asymmetric alliance (rules about matrilateral cross cousin marriage),
these being not uncommon in the societies in eastern Indonesia. Often, a
third principle, namely, that of ceremonial gift exchange, as forceful
as the first two, is to be exploited for the ordering of the social
world.
We may call the first two principles (agnation and the asymmetric complex) “regular (kinship) principles” as they are of kinship pure and simple, and the third one (prestation) an “irregular principle.”
The third principle refers to the bridewealth transaction, or more
precisely, the ngawu transaction.8 The
dogma has it that the wife-taking group should give a certain amount of
valuables (such as elephant tusks, gold items and, nowadays, cash) to
the wife-giving group in contracting a marriage.
Actually, a bridegroom and his closest agnates should accumulate the
valuables before contracting the marriage by sending for their own
wife-taking groups. On the receiving end, the valuables are not simply
received by the bride”s group (the wife-giver); they should be further
divided among the wife-giver’s wife-giver and so on.
The principle of prestation does not, in effect, seem to be directly involved in the ordering of the society; it only seems to be a by-product of the kinship principles (agnation and asymmetric complex), such that things (ngawu) flow in a direction already defined in kinship terms. We might say that “being (kinship) determines doing (prestation)”; thus a prestation is not, one might continue, one of the principles of kinship to be placed along with agnation and with the asymmetric alliance.
What makes the prestation the third principle, irregular as it may be, is the fact that the principle of prestation employs, when working as an ordering principle, a reversal of the directionality of this dogmatic explanation (“being determines doing”)—“doing determines being”.
The reversal can be done only when the original idiom (in this case, “being determines doing”) is articulated as a dogma, or an indigenous linguistic rule to be followed.9
New by me A dogma (“being determines doing”) can be sometimes reversed such that one can say that “doing determines being.”
For example, all Endenese assume as a rule that one should marry a girl from his wife-giving group. This is a dogma. The reversal applies too — that is, that the group, from where the girl comes, is wife-giving group. By marriage, an Endenese thus can recruit non-kin into his/her social world articulated in the idiom of affinity.
A mere statistical tendency, the statement of which may look like a rule, does not function in this way (“doing determining being”).
A couple tends to adopt the husband’s sister’s son when they do not have a son themselves for some time after their marriage. But no Endenese has ever mentioned that to me as a rule.10 Thus, even if one adopts a child of a person, other than one’s sister or sister’s husband, that person will never be considered as one’s sister or sister’s husband.
Ngawu transactions are like the marriage rules, since ngawu prestations are transferred from wife-taker to wife-giver, so if ngawu goes from A to B, A is regarded as the wife-taker and B as the wife-giver. In this way, “doing” (paying ngawu) makes “being” (kinship).
Let me give one striking example of how “doing determines being” (“prestation generates kinship”), namely, the institution of what I call “bridewealth-linked brother and sister”.
Suppose there are a pair of a man and a woman, between whom there is no kinship relation, and that you are a rich man, without any kinship relationship to either of them. You are the one who is in charge of the woman ("a la your own daughter) and, thus, when she marries, you receive the bridewealth (ngawu) paid for her. Now the man is poor, so poor that he cannot marry since he has no ngawu. Say, you propose to pay bridewealth for his marriage and he assents. You use, for the promised payment, the ngawu you received for the woman’s marriage. The man and the woman are now in the relation of what I term “bridewealth-linked brother and sister”
Platenkamp’s suggestion If a man (a) has custody over a woman (b) as if she were his own daughter, then he is entitled to the ngawu prestation transferred by the wife-taker upon the woman’s marriage. If the man (a) then agrees to transfer their prestation to the wife-giver of another man (c) who wishes to marry a woman but cannot afford to pay the bridewealth, then the latter man (c) and the woman (b) — for whom the bridewealth was originally received — are now in a relation of what I term “bridewealth-linked brother and sister.”
In Endenese, this situation is described as wa’u se ’imu // nai se ’imu, “one (woman) getting off // one getting on”. The first part refers to the woman (b) who marries out and the payment for whom one uses for the poor man’s (c) bridewealth and the latter part refers to the man’s (c) wife who enters (“getting on”)your family now. This relationship between the man (c) and the woman (b) is conceptualised as one of brother and sister (even though there is no actual kinship relation between them) and the connection between them is considered much stronger than an actual genealogical link between brother and sister (without any bridewealth transaction), in the sense that the man’s (c) daughter must marry the woman’s (b) son.
Let us turn back to the kinship idioms. The main (proper) principles (agnation and asymmetric complex) ordering the Endenese social world consist conceptually of three categories of kin: (1) agnates, (2) wife-givers and (3) wife-takers. This seemingly orderly social world of an Endenese can be disturbed by events of a personal nature, such as ro’i (a purposive rupture of the friendly relationship supposed to be maintained between affines) or a wrong type of marriage (rarha sarha, “a wrong path”). Ro’i turns two affines into non-relatives of each other, whereas rarha sarha reverses the wife-giver/wife-taker relationship. One, thus, can never simply say so-and-so is, say, a wife-taker, for the person referred to might also be categorised as a wife-giver via the different application of the three principles.
When a person traces such multiple types of relationship to someone with whom that person has daily interactions, the two persons have many occasions to discuss which type of relationship should prevail when this question arises. But with a person in a fringe of one’s social world (such as a person in a distant village), one may not have the chance to decide which relationship one should exploit. Thus, when one day during my recent field work, I asked my Endenese friend about his kinship relationship with a person who just passed by his house, he said “rhatu mesa” (Ind. “Ada saja”) — that is, “anything goes,” implying the person could be counted as his wife-giver, agnate or even wife-taker, depending on how he traces a history of kinship.
The real world can be sometimes too muddled for the two main Endenese kinship principles (agantion and asymmetric alliance) to classify exclusively. NEW And it is in such contexts that the principe of “saying” comes to the fore, to articlute the world.
First, an episode from my field notes:
☆
A feud broke out between a mountain village and a coastal village over the ownership of a certain parcel of land which lies midway between the two villages. One day, early in the morning, a group of coastal people came up to the mountain village, armed with bows and arrows. They demonstrated their willingness to do whatever they had to do by brandishing bows and arrows; after a few minutes, seeing that nobody came out of the houses, they went away.
A few days later, a coastal man, who had been on the scene as a member of the warring party, came up to the mountain village, selling fish. He said, “I was songga (”asked for help”)11 the other day. I could not refuse my co-villager’s request, could I? This time, I assure you, I’ve come only to sell my fish. I’m your friend (’orho ’imu).”
This is one example of “making up the world by saying”, frequently employed by the Endenese. A mortuary ceremony employs many of such devices, since a funeral provides the Endenese people with a convenient occasion for a reconciliation with people with whom one had an awkward relationship (such as ro’i) and for a more precise articulation of hitherto unclear relationship.
Ro’i is an institutionalised way of breaking up the affinal relation12 and there is, in turn, an institutionalised way of undoing the rupture.13 Here I rather with to focus upon the Endenese ways of solving less formalised crises, namely, those which make themselves felt on the occasion of a mortuary ceremony.
As often as not, a funeral speech consists of addressing the deceased’s wrong doings such that they, the bereaved, forget the bad influences caused by the deceased and try to create a better social world.
☆
“Something has been going wrong with you and us. We’ve had an awkward relation up until now. Look here. All of it was caused by him, the one lying here in front of us. He is, thank goodness, dead now. He was a witch (’ata porho). He was the one who caused all ill feelings and animosity between us. Now let us forget about him, his doings and the past. And let us start our new relation from now on.”
The past episodes are now set in a context different from the previous one. The past is now reinterpreted such that the only bad person is the one who is dead.
More effective than the mere re-interpretations of the social world as discussed above is the assignment of three roles (the agnates, wife-givers and wife-takers) by means of saying.
Immediately after a death has occurred, co-villagers come to the deceased’s house as ’ari ka’E, or more precisely, as ’ari ka’E ’onE nua, “’ari ka’E in the village”.
☆
In the village where I did my field work, a man died. A co-villager, who was the deceased’s main wife-giver (i.e., his mother’s brother’s son), came to the house and performed many tasks for the preparation of the coming funerary ceremony (ghena). The main wife-giver is supposed to perform only ritually significant actions, such as the first digging of the grave and putting a twig over the grave.14 I was rather amazed by his generosity. He said to the people in the house, “I have come here as a co-villager now. Later, I’ll come as a wife-giver. And, then, treat me as such.”
To quote another episode from my field notes.
☆
A boy, who was a member of the deceased’s wife-takers, was helping women around the hearth. Somebody of the in-group said, “you brought such a small amount of ngawu.” The boy replied that he had not come fully as a wife-taker yet.
In this context, a special importance is attached to the process of sodho, or “notifying”, carried out at an early stage of the ceremony. On the death, a messenger is to be sent to every conceivable relative, to notify (sodho) them of the death and ask them to come to the funeral (ghena). The messenger does not only notify them of the death, but also of their assignment to attend to the mortuary ceremony as “agnates”, “wife-givers” or “wife-takers”. In doing so, people can re-order their actual world, at least, in the current context.
Failing to assign a role (whether appropriate or not) may provide the notified party with an excuse for being absent from the — altogether expensive — ceremony.
☆
After a certain funeral, I talked with an old man, whom I knew as the deceased’s wife-taker, and who had not attended the funeral. I asked him why he had not attended the funeral and if the messenger had not arrived. He said that the messenger had come but had not said that he should come as a wife-taker. “That was not a proper procedure. That’s why I did not attend the ceremony,” he concluded.
It is true that, as somebody else told me later, the incident of not notifying properly might not be a reason for his absence; not the whole reason, anyway — he may have had some political (as defined above) reasons. The point is, rather, that this excuse is regarded as proper. One might, “legitimately” as it were, accuse the failing party in this way.
We have seen that the Endenese people make frequent use of “saying” to influence the world (“being”) mainly articulated in a kinship idiom, agnation and the asymmetric alliance. They do so especially when their world is in crisis such as on the death of a constituting member. What is lacking in the analysis, however, is that we don’t know what we mean, exactly, by “changing the world by saying”. How can mere saying alter the world?
In How to do things with words, Austin argues that the essence of the magic of “doing things with words” is a “tradition” or “institution”, an idea which later is made clearer by Searle (1969) (Searle 1969). He recapitulates it as a “constitutive rule” (see Sugishima’s paper) — corresponding more or less to what in this paper I labelled a dogma.
But still the question remains, this time in a form slightly different from the original one, that is, “How can a speech act, based upon a constitutive rule, act on the world?”
The question can be answered only by analysing the relations which words have with the world. Let me, for the sake of the argument, state the conclusion first. There are two ways of relations words can have with the world: (1) words “fit” the world and (2) words “organise” the world.15
“CAN to describe X as Y really be the same as to call X Y? Or again the same as to state that X is Y?” writes Austin(Austin 1979a). This rather out-of-the-worldly, seemingly useless (that is, philosophical) question can lead to a really fascinating conclusion. I will not go into the details of Austin’s discussion here; instead, I will quote his telltale examples of “describing” and “calling” used by Austin.
☆
You describe it as crimson? But look, it has a lot of blue in it. It is not really like crimson at all.
He describes me as a dictator, whereas in fact, as he must have known, I have always acted only on the advice of Parliament. (Austin 1979a: 148)
He then continues, “If many such examples are studied, the watershed between calling and describing appears to take shape.” Note the important fact that this paper is meant to be a sequel to the paper entitled “Truth” (Austin 1979b). The contrast between “calling X Y” and “describing X as Y” is supposed to shed light on the notion of “truth”. Actually, by that contrast, Austin is now leading us to see the distinction between two kinds of “truth”: the truth of “calling X Y” and that of “describing X as Y”.
The difference lies, according to Austin’s phraseology, in the “directions of the fit”. In my terms, “calling X Y” is a case of “words organising the world” and “describing X as Y” is one of “words fitting the world.” “Calling X Y” can succeed or fail — that is, succeed or fail to “organise the world.” If it succeeds, then the statement is true (because the world is so made); otherwise, it is false. This is a clear case. “Describing X as Y”, on the contrary, is not. The statement can be closer to the world but can never “mirror the world” (Austin 1979b). In this context, it is of interest to note that, in (Austin 1979b), Austin contrasts the two usages of the English word, “true”: (1) “true of” and (2) “true to”. A record of proceedings can be “true of” the proceedings of which it is a record; but a replica or a photograph cannot be “true of” the original — it can only be “accurate or lifelike (true to)” ((Austin 1979b: 126)).
Now let me summarise the above discussion. When words “organise” the world, the statement can be “true of” the world because the world is so organised. The statement really changes the world. When words merely “fit” the world, the statement can only be “true to” the world. The statement is, so to speak, to be in accordance with the world. Constitutive rules are rules for the words to organise the world, thus enabling them to change the world. My cases of the Endenese people changing their kinship world by “saying” are apparently those of words organising the world, making use of the relevant constitutive rules. What, then, is the theoretical importance of the second kind of statement?
To answer this question, let us turn our attention to the community in question, the community in which the said statement is regarded as true or false (or, more true or less true). If, in the case of “calling” (as in sodho), the statement is true, all the members should agree with the statement — a constitutive rule has such power. In contrast, in the case of “describing” (as in a funeral speech), even if the statement is, more or less, true (“true to the world”), there still can be some who do not agree with it – witness the examples by Austin. The statement is, to use a vogue anthropological idiom, an “interpretation” of the world and it can never be said to be true of the world. The community in question has been called an “interpretive community” whose border is rather blurred and fuzzy, just like the truth value of the statement itself. What I termed “political” speech is the case in point now. Political speech, such as guessing at the unstated intention of the players and/or actors, is a kind of interpretation in this sense. It can alter the world as well, but not in the way “calling” or sodho can change the world. A political statement can increasingly pervade the community in question and can be a major opinion, when it becomes “truer to” the world. It changes people’s conceptualisation of the world gradually. That is the functioning of such political interpretations as cited in the paper. The man who made a speech about the wrong doings of the deceased can, thus, modify the (kinship-)world view of the people around him.
And, this is the end of my, somewhat belated, funeral speech.