The Batsbi?

Other Peoples

Resources & Links



Oracular speech

gvrini & kadagoba


Gvrini (Georgian: გვრინი) are a type of dirge characteristic of Pshavi and Khevsureti, which were sung by men when they were mowing (G. 'tibva', თიბვა). The Khevsurs and Pshavs believed that, besides being a vital activity, the act of mowing (i.e. scything grass to make hay with which to feed cattle during winter) was in close symbolic relation with the land of the dead and with honouring the memory of the souls of the deceased.

Here is an incomplete (alas! and very rare) recording (link) of a gvrini
recorded in Khevsureti by the noted Georgian ethnomusicologist Shalva Aslanishvili (1896-1981).

Georges Charachidzé had this to say concerning gvrini:

[Introductory note: Charachidzé distinguishes those women who lament the passing of the soul of the deceased during the funeral ("les pleureuses par appel" whose mourning is an integral part of normal funerary rites) from those through whom the soul of the deceased itself addresses the living—the "pleureuses par la voix".]

'It is not, however, enough to ensure the right equilibrium between the world of the living and that of the dead; communication between these two completely separate universes must also be established. As is the case in matters of religion, relationships between these two universes can only take place via the medium of possession or rite.
The first category of relationships only involves women, the mourners. One must distinguish three groups of these.

1) The women mourn the deceased. In doing so, they establish a one-way communication, from the earth to the afterlife. They communicate with the soul of the deceased, who hears them but does not answer. Women belonging to the clan of the deceased "mourn out loud" ["pleurent à haute voix"], and "strangers [i.e. those women who do not belong to the clan of the deceased —A.B.] mourn in silence" ["se livrent aux pleurs silencieux"] (Mt. III, III, p.6). It is absolutely forbidden for men to mourn the deceased, both literally and figuratively speaking: they must neither speak to the deceased, nor cry [i.e. wail —A.B.]. This first category of mourners only intervenes during the funerary ceremonies which immediately follow the death. The Khevsurs define their activity in these terms: they "mourn by calling" ["elles s'adonnent aux pleurs par appel"].

2) The "pleureuses par la voix" are possessed [by the soul of the deceased —A.B.]:
"The lamentations of the "pleureuses par la voix" are very different from normal acts of mourning, and not everyone is capable of pronouncing them. For a woman to mourn in this way, her lamentations must burst out, as if the deceased were forcing her to mourn despite herself (...).
"She began to tremble and to shake, tossing her head from side to side, she began to foam at the mouth, to wring her hands, to grind her teeth; as if she did not want to speak or could not do so; sometimes she let herself fall to the ground, exhausted (...). Before the lamentations, she suffered, was nervous and often appeared to want to escape; the other women caught up with her and begged her to let her tongue speak (= to let the deceased speak through her).
"At last, the woman let the voice [of the deceased —A.B.] out. She spoke with the tongue of the deceased, saying for example:—Begin to move, my interpreter (meene, G. მეენე, literally "linguist"), do not sadden me in the Land of the Souls, make me take part in the lamentations, I want you to be my interpreter" (Mt. III, XXI, p.39).

The soul of the deceased then recites [through the aforementioned "pleureuse par la voix" —A.B.] poems of a special kind, called gvrini. These are learnt by the men, who recite them when the time has come to mow [i.e. cut grass for hay —A.B.]. We have devoted an article to this singular poetical genre, to which we refer the reader for more details (G. Charachidzé, "Travail et mort dans la montagne georgienne", in L'éthnographie No. 54, 1960, p. 45-62).

3. The mesultane.


The task of maintaining relationships with the Land of the Dead based upon the mediation of ritual was the exclusive responsibility of the Priest of the Soul (G. sulis khutsesi, სულის ხუცესი):
"The priest of the Soul only served the dead: when offerings were made, he blessed the funerary tables and drinks. Not everyone could carry out the priesthood of the soul. Only a few knew the invocations which had to be pronounced when offerings were made... He who wished to become Priest of the Soul learnt these invocations by heart." (Mt. III, XXIV, p.43)
The sulis khutsi accomplishes all the rites destined for the land of the dead, including the sacrifice of specifically funerary victims (goats, cows) (Vaja Pshavela, VII, pp. 12, 72-75). He also fulfils an important economic function: it is under his supervision that grass is cut for hay. Just as the first day of the harvest is marked by festivities, so the first swing of the scythe in the fields is the culmination of a funerary ceremony:

"It is forbidden to begin the mowing of grass without having celebrated the cult of the most recently deceased beforehand." A funerary ritual therefore inaugurates the gathering of hay. This ritual is carried out by the priest of the soul, the sulis khutsi. On "Hay Tuesday" (G. tibis samshabati, თიბის სამშაბათი), i.e. the first day of the year upon which one begins to mow, all the men of the village go to the house of the most recently deceased, where the "master of mourning" ["maître du deuil"] greets them... A table is laid, covered with cakes [probably what the Georgians call kada, i.e. soft, flat, round breads with a filling of sweetened clarified butter —A.B.], cheese, buttermilk and jugs of brandy. [All these are, according to Charachidzé, specifically funerary foods —A.B.] The Priest of the Soul lights candles and blesses the food. Slightly removed, the women gather and the "pleureuses par la voix" and the mesultane [another woman who was able to communicate with the dead —A.B.] begin to officiate... When everyone has eaten their fill, women and men having sat separately, "the village is allowed to bring the scythe into the grass". The Priest of the dead and the Master of mourning scythe the first grass, followed by all the men... (G. Charachidzé, 1960, p.55-56; Mt. III, XXV, 4, p.47).'

         From Le système religieux de la Georgie païenne—Analyse structurale d'une civilisation
         Paris: Francois Maspero, 1968 (reprinted by La Découverte, 2001), pp.266-268, my translation.

According to Charachidzé's sources, this song is therefore oracular in nature i.e. its text was not written or composed in the normal sense of the word, but was instead transmitted to the singer by the soul of the person most recently deceased before mowing began via an intermediary oracle (a "pleureuse par la voix").


Another example of oracular speech (one of very few) is given in Tinatin Ochiauri's ქართველთა უძველესი სარწმუნოებიის ისტორიიდან (1954). In this case, however, it is not the mere soul of a deceased person which is communicating with the living, but a divinity itself—one of the pre-Christian divinities the Khevsurs and Pshavs call the "children of God" (G. ghvtis-shvilni, ღვთისშვილნი).

The following lines are the words of the Archangel Michael—the main pre-Christian divinity of the Khevsur community of Arkhot'i, north of the main Caucasus range—who is addressing his vassals via the intermediary of his shrine's kadag (G. ქადაგი). A kadag was a man to whom the divinity of a shrine revealed itself, having chosen him to be its mouthpiece on earth. Perhaps akin to the English term "prophet" or "oracle", the kadag could communicate with the divinity of his community at will, and through him the divinity in question was thus able to communicate its commands, demands or advice to its worshippers.

ჰაი! ჰაი! მე ორ მთავარ–ანგელოზი, მე მაქვს ძალიდ’ შაძლებაი! | ჩემთ ყმათ აღარაად გავაჩნიორ! | მხარი მხარს გამისწორესად კისერ კისერს! | მათრევენ სიმურშიად’ რიოშანში! | აღარაადვის გავაჩნიორ, მაგრამ მე აისივ მთავარ–ანგელოზი ორ, მე აისივ ძალიდ’ შაძლებაი მაქვ! | ჩემ ყმათ ურჩევ, რო ჭკვით იყვნან! | მე ნუ გადამაგდებენ თორე ვანან, ჩემს წესს მე არ დავიყრი, რაიც ხთისგან მაქვ მალოცვილი!

Which roughly translates as:

Hai! Hai! I am the Archangel, I have strength and power! | My slaves no longer respect me! | They have levelled the arm to the arm, the neck to the neck [i.e. they consider themselves equal to the divinity, they have lowered me to their level]! | They are dragging me into soiled waters! | I am no longer different from others, but I am nonetheless the Archangel, I nonetheless have strength and power! | I advise my slaves to be wise! | May they not treat me without respect, otherwise I will punish them, I will not renounce my rule, for I hold it from the heavens!

The following is a unique (to my knowledge) photograph of kadagoba i.e. people who have gathered to listen to the very words of their community's divinity, speaking to his vassals through the intermediary of the shrine's kadag:

This photograph was copied from Giorgi Tedoradze's remarkable ხუთი წელი ფშავ–ხევსურეთში ('Five Years in Pshav-Khevsureti', Tbilisi: 1930); the caption reads: 'Pshav women and men listening to the kadag on one of a shrine's festival days. The kadag himself is leaning forwards, and cannot be seen properly.'

The following is my mediocre and ongoing translation of the chapter on the kadag (shrine oracles) of Georges Charachidzé's outstanding Le système religieux de la Géorgie païenne (see above).

1: The calling of the kadag

If one were asked to define Pshav-Khevsur culture as laconically as possible, the answer would have be: "tsatsloba" and "kadagoba". Tsatsloba, the custom of the brother-husband, has been studied in the second section of this work. Kadagoba, "shamanism", is a verbal noun derived from the Georgian substantive kadag, "shaman". We will henceforth use the Georgian term, but to refer to the corresponding action we will use the verb "to shamanize", more practical than the Georgian kadagoba. Georgian shamanism deserves to be studied for several reasons. First, because of its very existence, which European scholars seem to have barely noticed. Secondly, because the set of facts and activities which this term refers to form a particularly clear and well-articulated system. If one were to personify Georgian culture, one would have the impression that this person has meditated on the question of kadagoba more than on any other question, and that the conception and elaboration of this question have been considered as comprehensively as was humanly possible. This is not to say that kadagoba does not contain characteristics which cannot be found anywhere else—on the contrary—but its structure has been drawn with a firm hand. Thirdly, Georgian shamanism offers an important advantage: it has remained extremely vibrant. As late as 1946, as we will see, there were still active kadag in Pshav-Khevsureti. As it happens, the only information to have been recorded directly which we have at our disposal dates back to as recently as 1944-1946.

Until 1944, kadagoba had only been barely or badly studied.[*] Most of the authors who have written about Georgian shamanism did so incidentally or indeed abusively. In the works of a first group of these authors, one generally finds a handful of indications on scenes of possession and on how the ritual itself unfolds, or isolated details regarding partial (if not actually downright misleading) aspects of the kadag's functions; all these works limit themselves to a few identical sentences or even to a few words: we will therefore ignore them. A few historians or sociologists have, however, provided some useful information, which we will give an account of throughout this study. The most noteworthy of these are Kovalevski, R. Eristavi and V. Bardavelidze, as well as the two authors who lived in the mountains of Georgia, A. Qazbegi and Vaja Pshavela. [...]

The first complete study to give accounts of fieldwork only appeared in 1954. Under a rather vague title, T. Ochiauri published the result of several years of fieldwork among Georgian mountaineers. For the history of archaic beliefs in Georgia (Tbilisi, 1954) consists of a compilation of myths and legends and—above all—of autochthonous accounts of kadagoba. This material is rendered even more precious by the fact that it was gathered on the eve of the disappearance of shamanism in Georgia—between 1944 and 1946 in the mountains and in 1947 and 1948 in the lowlands. It is very unlikely that a study carried out now would be able to provide similar results. To date, only part of this material has been published. The full text is held in the archives of the Ethnographic Institute in Tbilisi, N°s. 336, 355 and 363. The set of texts published by Ochiauri in 1954 constitutes a description of kadagoba, but it is very difficult to date this description and to determine which period it applies to. The age of those interviewed ranges between thirty and a hundred, with most of them being over fifty years old. Some speak in the present tense, whilst others use the past tense. When a centenarian speaks of an event which took place in the past, this event could obviously just as well have taken place in 1940 as in 1860. Nevertheless, it can be said that this set of texts constitutes an account of Georgian shamanism during the first half of the XXth century. This shamanism would also appear not to have undergone a great deal of change during this period. It is difficult to discern any differences between accounts given on the same subject by young men of thirty and centenarian old men and those given by Vaja Pshavela at the end of the XIXth century. Lastly, an observation confirms the topicality of our materials: as we know, kadagoba was until recently still practiced in the mountains of Georgia.

As for its antiquity, however, we are poorly informed. It is customary among Georgian authors to call upon the text in which Strabo describes a [Caucasian —A.B.] Albanian shrine, during the first century B.C. This description could apply to a session of kadagoba taking place in the XXth century, but is unfortunately quite short. Speaking of a hiera chora, one of the "sacred regions" which were widespread in the Caucasus and in Asia Minor, Strabo mentions a "sancutary of the Moon" on the border with [Caucasian —A.B.] Iberia. 'Its shrine (of the Moon) is close to Iberia; the cult held within is celebrated by a man, the most respected person after the king, at the head of the Hiera Chora, vast and populous, who leads the "hierodules" [OED: 'A slave (of either sex) dwelling in a temple, and dedicated to the service of a god.' —A.B.], many of whom are possessed by the divinity and prophesy.'[*] The text reads enthousiôsi and prophêteuousi; these people therefore prophesy in a state of divine ecstasy. The "possessed" here seem to be included in the more general category of "hierodules". Describing other "sacred regions" of Asia Minor, in Cappadocia, Strabo distinguishes the [? théophorètes] from the Hierodules, and it is to the former that he applies the verbs "to be enthused" and "to prophesy".[*] Be that as it may, this shamanistic and religious organization of the 'sacred regions' seems to have been very widespread in the ancient Caucasus[*]. Strabo even says that certain inhabitants of the Caucasus, living on the shores of the Caspian, prophesy in their sleep: 'There is also situated the town of Anariaka, where some say one can witness people prophecizing in their sleep.'[*] Two facts should be kept in mind: the ancient Caucasus knew and practiced extatic divination, "enthusiasm" and "prophecy"; the custom is attested in connection with most of the Caucasian peoples, Albanians, Iberians, Armenians, &c. But the most important is this: these [? théophorètes] represented a social category, not to say a class, and their functions generally differed from normal religious functions.

Despite the fact that it would of course be an exaggeration to consider contemporary kadag as being the direct descendants of the shamans of the ancient Caucasus, it would equally be an exaggeration to separate the two; it seems reasonable to see in kadagoba a distant inheritance of the shamanism referred to in Strabo's work. A lack of intermediary documents, however, prevents one from pursuing this reasoning. For between the Geography of Strabo and the Georgian geography of Vakhushti in the XVIIIth century, we lose all trace of the shaman-kadag. Indeed, it is in the XVIIIth century that the term 'kadag' first appears in Georgian literature, referring to the shaman. The regent Vakhushti speaks thus in his Description of the Kingdom of Georgia (1745):

'The Pkhovi (= Pshav-Khevsur) are Georgians in religion and language, except for the fact that they entertain a belief concerning the kadag, who appears as unconscious and lost; he cries out loud the name of Saint George and the people believe everything he says and consider his words to be the highest truth.'[*] Until this text was published, one could only come across the word kadag in its Christian sense. The verb kadag-eba, used as an infinitive, means "to publish, preach, proclaim"; as a verbal substantive: "the sermon". kadag-i denotes a herald, the public crier (for "preacher", old Georgian prefers to use the derived noun m-kadag-eb-el-i). The definition Saba Orbeliani gives of a kadag-herald in his XVIIth-century dictionary suits the kadag-shaman particularly well:

'He who proclaims the science out loud.' The origins of the term kadag are therefore Christian, and the word must have been introduced to popular discourse at quite a late period; understood in the wider sense of "spokesperson", it has come to denote the shaman. But we know the ancient term, which still exists: the Khevsurs only use it rarely, for the term belongs to the "language of the crosses", a "secret" language who use was, in principle, the exclusive right of the priests and shamans. Although the word is Georgian, only in Khevsureti is it used to refer to a shaman: kandara. In modern Georgian, kandara denotes a length of wood stuck into a wall or suspended by two lengths of string, and used to hang coats or as a perch for birds. In ancient Georgian, the word kandara was only used to refer to a "perch for birds", as Saba Orbeliani noted. This metaphor is perfectly suited to its subject: the kadag is the divinity's perch. We will see later on that the reality is more complex, and that this single image is not enough to account for the use of the word kandara when referring to a kadag.

The kadag is indeed the divinity's perch, for the khat'i [literally "an image, an icon", but in this context the word also denotes "a divinity" —A.B.] descends from the heavens to settle within the kadag himself. The god then speaks through him, through his mouth. Another Pshav-Khevsur term illustrates this last feature: possessed oracles are referred to as meene, "linguists" (from ena, tongue: me-ene, "linguist", in the same way as: puri, "bread", me-pure, "baker"). meene, like kandara, belongs to the secret language of the priests and shamans, the 'language of the cross', jvart ena. The simplest definition of kadag is therefore: a man possessed who serves as an intermediary between the gods and men, from above to below, that is to say from the divine world to the world of men. At first glance, one can therefore say that a kadag is a man possessed. The Georgian mountain tribes use two words to denote possession: damizezebuli and dacherili. dacherili means: "captured, captive, seized". damizezebuli is more difficult to translate; literally, one would have to translate it as "motivated". The best translation would be "concerned", which both conveys the idea of being a prisoner and of being personally targeted. For dacherili, we will adopt the term: "possessed". For damizezebuli: "concerned", always in inverted commas, since we are giving this term a special meaning. T. Ochiauri tries to distinguish between these two terms and is quite right to note that the word "concerned" is used 'in a wider and more general sense' than "possessed". One should qualify this opinion and add that "possessed" is stronger than "concerned" and, especially, more definitive. For example, one says that a lunatic is dacherili, possessed. But if all is not well in a household, if the animals are ill or if family arguments simply become too frequent, the Khevsur asks himself: 'Might not our household be "concerned" by a khat'i?' Here is an illustration of something one might refer to as a kind of "domestic possession". The following is a popular Pshav song in which a husband complains, in an ironic tone of voice, that his wife is a bad housekeeper:
     'I really have a wife with skilled hands,
     Is it even possible that my clothes become shabby?
     She falls asleep even when she is eating; she is not the one to say "the sun has risen".
     And now she sends me to see the soothsayer (mk'itkhavi),
     It is quite possible that we are "concerned" (mizezit viqot).'

It would therefore appear that this kind of possession is less serious than the other. But this distinction is not rigorously made; the two terms are very often used to denote each other, or are even linked: damizez-dacherili, used in the general sense of 'possessed'.

For the Khevsurs, as we shall see further on, there are two kinds of possession: that which leads the possessed to become a kadag, and that which is symptomatic of illness. This is why we described the kadag earlier on as a kind of possessed, for in his case the possession is a sign which announces that he has been chosen by the khat'is to become a shaman. In most cases, one can obviously only determine whether the possession was a divine sign or symptomatic of illness—i.e. whether the possessed will become a kadag or not—retrospectively. One necessarily comes across this problem of time when dealing with shamanism, which is why it is practically useless to consider individual episodes of the life of a shaman or a possessed. Only by studying his complete biography can this question be answered, precisely because of this retrospective assessment which will deem one possessed to be either ill or a shaman. It is because they omitted this factor that the authors who wrote about kadagoba made so many mistakes (except T. Ochiauri and V. Bardavelidze); this omission also explains why one encounters apparently contradictory information or information which seems incompatible with the declarations of native informers gathered between 1944 and 1948.

V. Romanovski, in an article published in 1909, stated that the kadag is exclusively chosen, elected, called (izbran in Russian) during particular religious ceremonies, notably during the one which marks the New Year: 'The person the khat'i has chosen trembles, loses his mind, screams, foams at the mouth, thus indicating to those present that the khat'i has chosen him to be its servant. This is enough for the worshippers to accept him as a kadag.'[*] If one were to take Romanovski's description literally, one would have the impression that the calling of the kadag is an instantaneous manifestation, a sudden state of ecstasy linked to a religious festival. The author, in fact, misses the main point, of which the scene he describes is merely the outcome: the story of the kadag and of his possession. Ochiauri is right when he says that, 'in reality, Romanovski is dealing with the final stage of the possession, that during which the possessed "releases the tongue" (a Khevsur expression), i.e. the moment when the possessed effectively becomes a kadag, which is revealed by the appearance of speech: from now on, that person will be a "linguist".'

[In the XIXth century], N. Khizanashvili made a similar mistake by also considering isolated episodes of kadagoba. This is why individual pieces of information he gives are absolutely correct in themselves yet false if one considers them as a whole; this mistake also explains why elements of his description seem contradictory: 'Young girls express the wish to become kadag, but people rarely trust them, as they believe these girls lack "understanding" [...] The khat'i possesses the young girls, makes them suffer.' Old men or old women are more capable shamans, they 'answer questions better'.[*] A few lines previously, however, one may read: 'The nomination of the kadag is the khat'i's business, and none may interfere.' If this nomination is indeed dealt with between the possessed and his god, then how can the people disapprove of the nomination of the "possessed" young girls whom the khat'i caused to "suffer"? The Georgian ethnographer has gathered different pieces of information without replacing them in their historical context and without seeking to understand their meaning. In doing so, he made two opposed mistakes: on one hand, he mistook distinct beings for each other, and, on the other, he divided a single being in two. First mistake: he believed that these young girls, possessed but whom the people refuse to accept as kadag, were failed kadag, as if they had been unable to pass a test. Actually, these girls are possessed but do not aspire to practice kadagoba. Having misinterpreted the nature of their possession, Khizanashvili saw in its symptoms the mark of a calling. Basically, he mistook the sick for those who heal them. Conversely: when he says that the god alone chooses his kadag, he is referring to the first "call", to the moment when the khat'i first reveals itself to the person it has elected. When he adds that the people prefer to approve of elderly possessed, he is referring to the fulfilment of the kadag's career. However, he fails to understand that these two men are, in reality, one and the same: in the first case, this person is considered during the beginning of his career, and in the other, during the end. Lastly, Khizanashvili has unknowingly put forward a serious problem which is at the very heart of any kind of shamanism or magic: the relationships which unite the shaman to his audience. For if it is wrong to say: the people approve of or do not approve of the possessed as kadag, one can equally not deny the existence of a sort of profound harmony, a reciprocal relationship between the kadag and those for whom he shamanizes, the need for public consent.

All these facts will become clear through the analysis of information gathered in the field; the declarations of those involved will answer the question of how one becomes a kadag.
The origin of kadagoba is a meeting with the divinity, the khat'i (or the angel, the Cross, the Khtishvili (= child of God), all these terms are synonymous). The khat'i, for various reasons, decides to choose as kadag such and such a person, whom he possesses. Sometimes, there is no clear reason for this choice: the god simply liked the person whom he elected.

'The angel meets someone, chooses this person, he takes him and makes him suffer' (for every informant, we indicate this person's age and the year when the declaration was recorded. We will also specify if the informant was a woman. Here: 60 years old, 1946, O., p. 15. O., p. indicates the aforementioned book by Ochiauri, the number being that of the page). In general, the khat'i possesses a human being when it foresees that it will have need for a kadag:
'The Khtishvil chooses a kadag when it needs a linguist. It chooses him, it marks him; it takes him when he is a young child or it also takes him when he is a teenager' (a seventy-five-year-old woman, 1945, O., p. 14).
Regarding the age as from which the possessed is taken by the khat'i, our information varies, but this is because, as we shall see later on, there are several different ways to accede to kadagoba. It would appear that the most beautiful shamanistic career is that which is open to the kadag who is 'called' from childhood:

'We do not know of possession from the day of birth. For this is not based upon inheritance. It is the person the khat'i chooses, and that's that' (1945, O., p. 17). By this, the informant means: the calling is not hereditary and only depends upon the will of the khat'i. Other Khevsurs explain themselves in more detail:

'An angel meets a human being, and takes him as kadag. The khat'i never chooses an old man. It tends to take a child instead, and the younger the child is when it takes him, the better. It is a very good thing if it takes a very young child. It does not make the child shamanize immediately. The child suffers. It is tortured at night, tortured in dream, it suffers. Khtishvils appear before it' (60 years old, 1945).
'A khat'i sometimes takes a five-year-old child. It falls ill. It alternately loses consciousness and goes mad. The child is taken to see the kadag, who tells it: such and such a khat'i has taken you. The khat'i generally prefers bachelors' (25 years old, 1945).
'The Khtishvil chooses the moment the child is born; it keeps it, accompanies it. When the child is 15-16 years old, the khat'i makes it shamanize. Thus is the law: one must be young and one is chosen beforehand, one is made to suffer' (80 years old, 1946, ibid.).
'It tends to capture it from childhood; the child falls ill, suffers, is listless. It suffers, its heart is tortured' (54 years old, 1945, O., p. 18).

One thing we can be certain of: future kadag are called as early as possible. It is quite normal that informations vary regarding the earliest age at which this takes place: if kadagoba is not hereditary, it therefore only reveals itself through the child's psychic disturbances. But the aforementioned symptoms—nightmares, prostration, strange visions, loss of consciousness, attacks of 'madness'—imply that the subject has already reached a certain age. The age of five years mentioned by one informant is probably exact. In addition to this, all these signs indicate without the shadow of a doubt that the child is an elect of the khat'i, destined for kadagoba. One can therefore reasonably conclude, retrospectively, that this calling does indeed date back to the day of birth, even if it only reveals itself a few years later. This is why the different declarations mentioned above seem to us to be not so much diverging opinions than more or less developed forms of the same opinion.

But we had expressed a reservation: if kadagoba is not hereditary. Strictly speaking, it is not. The Khevsurs state this, and one must believe them. More precisely, it is not directly and automatically transmissible, it does not rigorously follow the lineage. But some families are predisposed to possession. This is what the Khevsurs mean when they say: 'kadagoba preferably falls to men of a same father. In Amgha (a village), for example, kadagoba tends to fall to the Tsolik'auri clan, and khutsoba (= priesthood) to the Nislauri' (a seventy-four-year-old woman, 1944, O., p. 25). 'Kadagoba tends to fall more readily to the men of a same father. It is the people of a same lineage who are led to shamanize, and the same is valid for the priesthood' (40 years old, 1946, O., p. 25). 'Kadagoba falls more readily to the people of a same lineage, kadagoba and khutsoba as well. When a kadag appears in a clan, his son will in turn also begin to shamanize or become a priest, whichever it may be.' Another informant provides us with very interesting information on this subject: 'kadagoba and khutsoba tend more often to fall to the men of a same lineage. This phenomenon is called madeniloba. madeniloba refers to those people to whom these sorts of things happen. If, for example, people have an uncle who is a kadag, it happens to them. My father was khutsi, his elder brother was kadag. My grandfather was kadag—beyond that, I don't know' (59 years old, 1946, O., p. 25).
The uncle referred to here is the father's brother. It would seem that this property transmits itself exlusively down the paternal line. One has the impression that the transmission of kadagoba is perceived more as some sort of germ, a contagion, rather than as an inheritance. Or is it what one could call a heredity of acquired characteristics? Let us repeat that this inheritance is not automatic. Generally speaking, shamanic power is not transmitted directly. The phenomenon is interrupted, and resurfaces with the calling of the son or of the nephew. It is discontinuous. It happens in this way: my father is a shaman, my uncle a priest, &c. My life is normal and I remain an ordinary man, unless the sacred manifests itself as an illness again. I fall ill, at any age, young or old, and I will only recover my health by becoming a priest or a shaman. Here is a clear example of the discontinuous process of the transmission of shamanic powers:
'My paternal grandfather was a priest; his father was both priest and kadag. My own father is a priest. When the khat'i possesses him, his spirit is troubled, he goes mad. It happened in this way for my father: he fell ill, seriously ill. His father shamanized and taught him that he would in turn become a priest. That he did, and was cured' (1945, O., p. 123).

We can see that the question of how kadagoba is transmitted has no simple answer, all the more so as it is linked to the transmission of religious power. Indeed, most kadag are also priests; and many priests are also kadag (the proportion of priests among the kadag is higher than that of kadag among the priests). This is why we will leave this question unanswered for the time being, for we will come across it again in the next section, when we will study the relationship which exists between priest and shaman.

In any case, even if kadagoba is transmissible, the calling itself retains its value and is essential to the career of the kadag. This is what the Khevsurs mean when they claim that kadagoba is not hereditary: the fact that the father is a kadag does not necessarily imply that his child will be called by the khat'i. These are two separate elements. The first step is therefore when the calling is revealed to the child. For around ten years, until adolescence and until he becomes a kadag—i.e. the moment when he becomes a 'linguist' and 'speaks' for the first time—the child will 'suffer and learn'.

During this intermediary period between the revelation and the declaration, one can distinguish different phenomena or steps in the symptoms and behaviour of the possessed.

Having described the suffering of the "called" child, a Khevsur adds:
'He who is destined to become a priest learns this from the mouth of the kadag; he who will himself become a kadag is tormented in such a way that he becomes a kadag himself and shamanizes. A child is possessed, but when it is old enough to understand everything and to officiate, then the khat'i causes it to shamanize. Until then, it does not cease to suffer. The khat'i appears to it in dreams, tortures it. The child behaves in such a way that it cannot understand whether it is being tormented by a khat'i or simply ill. It cannot know itself that it is the khat'i who are tormenting him, it cannot tell. The khat'i do not leave him and he cannot speak. They never leave' (1945, O., p. 18).

'I was ill, and suffered from all kinds of illnesses. Either my liver was ill, or my heart. The heart causes prostration, it is unsettled. One can imagine the state one is in! Then one begins to lose one's mind, to say all kinds of things, to go mad. Later (= after the first session of kadagoba) one is no longer mad' (a kadag, 1945, O., p. 18).

'The child goes mad, and shamanizes in a disorderly manner, like a madman. It loses its mind, and is completely lost. The child cries out, but knows not what it cries. It cries what it is told to cry out. In the beginning, it understands nothing of its cries' (54 years old, 1946, O., p. 18).

More often than not, the child manifests symptoms of possession from the age of five, then its periods of illness spread out and temporarily disappear, but reappear in more violent forms between the ages of fourteen and twenty, as the case may be. The ages most frequently mentioned are fifteen, sixteen and seventeen. Symptoms can also appear for the first time at this age; but this delayed crisis gives meaning to the childhood illnesses and it is quite difficult to distinguish these two cases. It is during adolescence that an evolution clearly appears. Two tendencies then manifest themselves: on one hand, the possessed resists the khat'i's grip and tries to fight. This resistance, this inner struggle will last until the final crisis. Moreover, the aim of this period is to attain a degree of language, to be able to communicate. Throughout the texts referred to, one can make out the slow development, the difficult acquisition, the conquest of a final language. One witnesses the adolescent's dramatic efforts to 'release the tongue' (G. ena gamoit'anos). For if he is lost, his madness is no less turned towards verbal expression: he begins with cries, elementary speech. He cries out meaningless things. Then he begins to cry out words and sentences whose meaning is hidden from him. Finally, he 'shamanizes in a disorderly manner, like a madman'. This is a real learning process, in which dreams also play a part. The oneiric agonies of the possessed are not meaningless visions. The khat'i appear before him and thus make him become used to his future dealings with the gods. But the dreams also constitute a practice kadagoba, a prophetic babbling. The Khevsurs say that between the moment when he 'shamanizes like a madman' and the first session of kadagoba, the teenager 'shamanizes in dream':

'Before he begins to speak, before releasing his tongue, he begins to have visions, he shamanizes in dream' (80 years old, 1946, O., p. 19). But, in addition to this solitary learning process and to this struggle with the angel, the possessed is forced to undergo other trials.

We have until now left the shaman alone with himself, alone with his obsession and his god. However, he lives with other men and other shamans. The conduct of the apprentice-shaman causes a certain resonance within the group, which leads the latter to adopt certain attitudes vis-à-vis the former. This immediately begs a question common to every expression of shamanism: is the shaman mentally ill? The Khevsurs have answered this question for us: nobody thinks of looking after, of caring for this young child, this young adolescent whose behaviour is "disturbed". He is abandoned to his agonies for ten or fifteen years. But many Khevsurs suffer from mental illness; they are recognized as such and are led to undergo various treatments (which we will describe later in this work). This is the first reason which prevents us from equating the kadag with madmen. Let us now consider the development of the disturbances which we have just described: everything is done to ensure a transition from abnormality to normality; from the cry to speech, from torture to communion, from ignorance to knowledge, from hallucination to divine vision. The actions of the future kadag are certainly strange, not ordinary; but this is due to the fact that he undergoes an experience which is itself hardly ordinary, direct contact with the sacred. The Khevsurs make no mistake and are perfectly able to recognize the exact nature of the possession: illness or divine calling. They can make out this beautiful progression which leads from physical pain to a kind of superior knowledge; as a kadag said, the teenager who has been called goes through three main stages, which themselves almost constitute the modes of knowledge: first, the vital organs, heart, liver, are afflicted and torment the adolescent; then he begins to lose his mind; lastly, he shamanizes. These do constitute three "orders": the acute awareness of the body, the excessive and disorderly understanding of the soul, and finally the supreme knowledge which constitutes the permitted contact with the divinity. This is why the Khevsur informant may say: 'He who will become kadag est tormented in such a way that he becomes a kadag himself and shamanizes'. The Khevsur do indeed say that the kadag "goes mad", but a madness which seems to have imediate meaning to all is not madness.

This spiritual evolution is obviously accompanied by a proper initiation—both on the mythical and social levels. As this initiation is roughly the same for thos kadag called from childhood as for those whose calling came later, we will first describe this second category. Let us mention immediately that the latter are recruited from those who are ill. These cases in particular have led some authors to diagnose the kadag as mentally ill.[*] Such a conclusion pays no heed to reality, for, firstly, these kadag are people who have recovered from their illness, or else they would not have been able to shamanize; secondly, the Khevsur do not consider them to be mad, as we will see further on.
Georgian mountain tribes classify diseases into three categories:

1. Visible physical ailments: measles, smallpox, boils, angina, gingivitis, diseases of the eyes, &c.

2. Invisible physical ailments, from simple fevers to the most serious illnesses.

3. Illnesses of the soul. These include all mental illnesses, from the most benign to incurable madness.

Wounds are not included in this classification, as the latter are not illnesses. One should note that all illnesses can be phenomena of possession. Three kinds of entity are responsible for the oncome of illness: the khat'i, demons, and lords-of-diseases, who are specialized genies. The two first categories—illnesses of the body, whether visible or invisible—are either due to possession by a khat'i or to the presence of a specialized genie. But these genies can never be responsible for illnesses of the soul, which can only be the result of the fact of being possessed by a khat'i or by a demon. If the sick person is possessed by a god, his recovery is guaranteed. If he has fallen prey to a demon, two cases are possible: either the rites of liberation succeed, and the sick person recovers; or they fail to free him, in which case the possessed is considered definitely incurable and as belonging to the category of the mentally ill. Thus is the classification explicitly created and professed by the Georgian-speaking mountain tribes, In Khevi, in Khevsureti, in Pshavi, in Mtiuleti and in Tusheti. We have here but summed it up, for we will find this classification applied quite often. It is generally rigorously applied; certain cases, however, are subject to different interpretations, like diseases of the eyes among children (see section 4, chapter 2).

Patients are sometimes brought to kadagoba simply by falling ill. We will deal with this process briefly, for this is not so much a case of being called as of being recruited; additionally, in most cases, the patient recovers by becoming a priest or a dastur (a servant of the shrine). It is very rare for the patient to recover by becoming a kadag; nevertheless, this does sometimes happen. A Khevsur suffers from a physical ailment—usually invisible, but sometimes visible. He suffers from headaches, or his eyes or kidneys hurt, or he coughs and chokes. The affliction is usually quite vague and fleeting. If, however, it keeps returning, he goes to consult the kadag. The latter questions him and helps him search his memory: did he not recently fail to obey a religious prohibition by mistake? For example: Farming sacred land, or consuming products harvested on sacred land, or leading one's animals into the khat'i's fields, &c. He need only have walked in a forbidden place i.e. the khat'i's lands, woods or meadows. It is rare for the person not to end up confessing to such acts, regardless of whether or not he really did so or whether he merely believes he did. The kadag then begins to shamanize and informs the "guilty party" that he has displeased the gods: this results in a particular khat'i "concerning" him, which is why he fell ill. The kadag prescribes certain additional offerings to the displeased divinity—to be made once or repeatedly. If the patient recovers after having made such offerings, the matter is forgotten. Otherwise, he returns to the kadag, who then advises him to become a servant of the shrine, either as a dastur or as a simple worker. When the disease appears to be serious, the patient becomes khutsi (priest) or sometimes even kadag (O., p. 16). In any case, the person concerned is not answering a "call": it is the serving kadag who decides his fate. We will busy ourselves no further with this kind of recruitment, for we will come across it again in the form of a shamanistic cure.

In most cases, beyond a revelation during childhood, the "call" appears as an ailment of the soul. When the symptoms and crises become stronger and more frequent, the patient becomes aware of his calling and the possession then turns towards shamanism, progressing along the same stages as those followed by children or teenagers. The possessed, the villagers and the kadag recognize the "hand of the khat'i", and the possessed is then considered to be (and considers himself as) a khel-katsi, literally a "man of the hand", that is to say a man "upon which the khat'i has laid his hand" (O., p. 15-16). This term is widely understood, and refers to any man whom the khat'i has chosen to serve him: it therefore includes all the shrine's "staff", including the kadag and other soothsayers, from the most basic servant to the priest himself. The possessed must therefore choose between two possible careers: priest or kadag. In reality, it is the strength of his calling which dictates this choice. If the "call" is too weak and the possessed undecided, or if on the contrary the call is too strong and the possessed too "lost" to be able to decide on hiw own, he goes (or is brought) to consult the kadag. The decision is theirs; more often than not, they declare that the possessed must become a priest. But, in general, kadagoba forces itself onto the possessed without the intervention of others. The possessed is forewarned by the specific evolution of his ailment and by premonitory dreams (O., p. 16-17). Here is the story of a Khevsur kadag:

'I was ill, I felt terrible. I lost all my hair, I constantly had visions. I was in no pain, however—that's how I was ill. I could no longer tell night from day, I lost consciousness, I went half-mad, I was unconscious all the time. The khat'i Kopala appeared before me, everywhere, even during my sleep. I was with them (= "with the gods") all the time. I was twenty years old back then. I was single (litteraly "no woman had been brought before me"). I had not had a woman before the Angel appeared' (57 years old, 1945, O., p. 18-19).
This is the most common process. It barely differs from that followed by those called from childhood other than in terms of the age of the person concerned. As we pointed out earlier on, some callings are not strong enough for the possessed to become aware of his calling himself. In such cases, another kadag will help him realize and accomplish his calling. The intervention of the kadag replaces the obvious nature of the calling. Kadagoba must force itself upon the possessed with the binding strength of destiny. If the possessed does not find kadagoba within himself, he will be compelled to follow its path by being ordered to do so by the kadag. It is inconceivable that this order be disobeyed. On the contrary, should one possessed consult a kadag and be ordered to become a priest, he may choose not to do so. He risks much in doing so, but he can refuse; here is what a remarkably stubborn Khevsur says regarding his refusal to become a priest:

'I was ill. I had smallpox. Then my brother fell ill. The kadag told me: "Become a priest."; but I absolutely refused to do so—and then my brother died. Then my grandson also fell ill: "Become a priest, or he will die." He died. Then I was told the same regarding my son; and he, too, died. I still refused to become a priest. The I also fell ill. I saw priesthood in my dreams (that is to say: "I saw myself as a priest"). My wife then went to consult the oracles. The kadag told her: "Tell him he will be priest, or else he will forever remain in this state." Nothing doing! And I was still ill: my heart ached, and my kidneys sometimes hurt, my heart suffered. I was at death's door. I had no visions; I only saw myself as a priest (by this, the Khevsur means that the khat'i did not appear before him). I said to myself: if they do not leave me alone, they will end up forcing me to become a priest' (O., p. 119).

We do not know how this deadly refusal ended, but it speaks volumes. Despite all the risks and all the tragic deaths, priesthood was refused. In the event of kadagoba, however, such refusal is unthinkable. A 17-year-old Khevsur was possessed, but the calling was weak and was limited to physical sickness. The help of a kadag was therefore necessary:

'His face and eyes were swollen, his hands trembled. He is taken to see the person who knows dreams and kadagoba. He brings him a candle. The kadag shamanized thus: (this is the god speaking) "I took you. I forbid you to enter another family (= I forbid you to marry)." The kadag tells him he will be a holy man and that he, too, will shamanize' (O., p. 19).
There is no question here of refusing, and the young man becomes a kadag, for the session of kadagoba played the same role as him being called.

The different cases which he have examined up until now are exceptions. For the kadag is an exceptional being, different from the mass of those possessed. But this distinction alone cannot account for reality. One should add: he who shamanizes is not always a shaman. A category is therefore missing from our list of modes of possession (we are not attempting to draw up a typology, but merely to enumerate the different possible ways in which possession can evolve based upon all the cases of possession which have been described in Khevsureti):

1. Possessed from childhood, the person is irrevocably destined for kadagoba (without the intervention of another kadag). Special case: Possession only reveals itself when the child becomes a teenager.

2. A person falls ill, at any age, and recovers by becoming a priest or a kadag (following the intervention of another kadag).

3. A person falls ill and recovers following a shamanistic cure. Special case: The person does not recover, despite the cure, and remains ill for life.

4. A person falls ill and—with or without the intervention of a kadag—begins to shamanize. He does so intermittently (once or twice a year) but without ever becoming a kadag.

We have just introduced a new category of possessed—the richest, in some ways. Many of the mistakes made when dealing with Pshav-Khevsur kadagoba sprang from the basic misunderstanding which considers those intermittently possessed to be kadag. Thinking he was describing the kadag, the observer was actually describing someone who was simply possessed. He does indeed "shamanize", but is not a shaman. This misunderstanding is also the origin of contradictory information concerning the shamanism of women: according to some, a woman is never—or only very rarely—a kadag, whereas others assure us that the vast majority of kadag are women. The former had in fact observed real shamans; the latter, on the other hand, were describing the possessed of our fourth category. This confusion between kadag and possessed women is also due to a lack of geographic precision: when dealing with kadagoba, some authors did not distinguish the highlands from the lowlands and took their information from materials collected in both areas, mistakenly attributing them to the whole of Georgia. In the highlands of Georgia, kadag are always men—with the exception of one shrine, which will be described later (chapter 7). Some, equally wrong, have also equated kadag with the young girls and teenagers who are called upon to communicate with the souls of the deceased.: the mesultane and the mesulete; but we will see that they in fact belong to a distinctly separate category, with no relation to kadagoba (section 3, chapter 4). Also, the situation in the lowlands is the opposite of that in the highlands, for there it is the women who, more often than not, carry out the functions of the kadag. (See below, chapter 5.) Finally, among the highlanders themselves most of those possessed who shamanize from time to time are women.

What differences are there between those who are possessed, both women and men, and the kadag? The frequency of the séances of shamanism cannot be taken into consideration, for the intervals between an authentic shaman's sessions of kadagoba can be very long. It can even happen that the latter shamanizes less often than those who are simply possessed. On the other hand, both remain at the mercy of a sudden attack of possession for their whole lives. It often happens that the possessed—when no-one expects it—collapses in violent convulsions and then begins to shamanize: the divinity which holds him captive has brutally manifested itself. The kadag can also suffer from such unexpected attacks:

'Sometimes the kadag is forewarned of the attack by signs and omens, he lies prostrate and suffers; but sometimes the khat'i all of a sudden manifests itself when no-one expects it.' (1944, O., p. 39)

Lastly, because the predictions and the orders of the possessed are followed as scrupulously as those of the kadag, it is impossible to distinguish one from the other according to the attention they are paid by their audience. The three main criteria for distinguishing a kadag from a person possessed are therefore the following: 1. The kadag has a social status and social functions. 2. He is tested i.e. made to undergo trials, whose whole constitutes an "initiation" (this word being understood in its widest possible sense). 3. He has magical-religious powers which can also be brought to bear upon him and upon his relation with the khat'i: he thus has the ability to enter into a state of trance and to make the divinity intervene: 'My father-in-law was a kadag,' a Khevsur woman says, 'the khat'i manifested itself in him. At home, the khat'i would sometimes manifest itself suddenly; but it also came to him when he [the kadag] wanted it to.' (59 years old, 1945, O., p. 39-40.)

An informant from Mtiuleti perfectly described and distinguished kadag and person possessed. In the following text the first two cases apply to the kadag; the first paragraph describes the calling with the intervention of another kadag, and the second without outside intervention. The third paragraph defines the person possessed who shamanizes without being a kadag (the informant divided the paragraphs himself):

'Firstly: a child, from its youngest age, is weak and often ill: headaches, heartbeats [battements de cœur in the French original—perhaps the informant meant to describe the symptoms of a condition such as e.g. cardiac dysrhythmia? —A.B.], sweating, insomnia, strange thoughts. When the child "matures" at 13 or 14, 20 or 25, he falls seriously ill. He is then taken to see the oracle. The oracle tells him: "this is what you have, you have this and that, it is this or the other khat'i who is taking you as kadag." He can only be taken by the khat'i whose slave he is (that is to say the khat'i of his clan).

'Secondly: it can happen that he displays the symptoms of his illness since childhood but that, in the end, instead of falling seriously ill he goes mad. Come the day of the year when people celebrate the cult of his khat'i, he runs away, he tears his clothes to shreds, he rushes around aimlessly, he prays, he cries, he eats handfuls of earth. At the same time, he shamanizes in a disorderly, frantic way, but after the second or third year he begins to shamanize more correctly. After having shamanized, he is normal, as usual. Then, even if he shamanizes, he no longer goes mad. And even if he does suffer from bouts of madness, he no longer needs to go to see the oracle.

'Thirdly: it is also possible that the child suffers from no illness—neither of the mind nor of the body—but that he goes mad. He behaves in the same way as described above, he invokes the khat'i, he pulls at his hair, he eats soil, he tears his clothes, he scratches and bloodies his face, he speaks; at first, he too speaks in riddles and cries a lot, sometimes dancing, sometimes crying; everything is mixed up. After this, he "declares" himself and begins to speak [il se déclare, la parole se manifeste]. Every year, when people celebrate the cult of his khat'i, he attends and shamanizes. This is called being possessed by the khat'i, the khat'i possesses him (G. იჭერს, "ich'ers"). When a child is ill, one often considers that he has become possessed by a khat'i. The case of those people who have become "concerned" (damizezebul) is a different matter: "to be concerned" is not at all the same thing as being possessed, as it presupposes that one has trespassed in some way or committed some offence. The possessed need not have done something wrong; the khat'i "takes" them just like that, for no reason. "To be concerned" is one thing, "to be possessed" is another.' (Mikheil Kedeladze, 1946, O., p. 28-29.)

This type of person possessed is quite different from the person whose possession is linked to illness [possédé-malade], for he has fallen under the spell of a kind of calling. One can see in him an evolution identical to that which characterizes the kadag and in which cries evolve into words. Instead of seeing the kadag as some sort of possessed person, it would be better to see a sort of kadag subdivision among the possessed. A possessed person's powers are limited and incomplete, reduced to a passive and intermittent contact with the divinity. With the kadag, on the other hand, the contact is permanent and implies a certain degree of activity on behalf of the kadag. The possessed are not subject to any trials, restrictions or obligations, whereas the kadag is held to obey certain prohibitions and restrictions all his life, the latter at the same time constituting the trials by which he accedes to the fullness of his powers. We will now see how the kadag attains his powers (his initiation), how he uses them (his behaviour) and lastly which place in society these powers grant him (his social functions).

2: The "initiation" of the kadag

[Note: the following subdivision of the text into "2.a.", "2.b.", &c. is my doing —A.B.]

          2.a. The secret "language of the Cross"

The term most common and most used to refer to the shaman is meene, "linguist" [G. მეენე, from G. ენა, "ena" = "tongue" —A.B.]. The apprenticeship of the young kadag will therefore above all consist of his acquisition of the secret language of the khat'i, the jvart ena [G. ჯვართ ენა, from the G. ჯვარი, "jvari" = "cross" —A.B.] (= "language of the Jvar", that is to say "of the gods"). As its name indicates, this language is none other than that used by the divinities. In reality, however, the jvart ena is not secret: it is its use which is strictly reserved to the priests and the shamans (at least in theory), but the language's terms are known to all. Proof is the fact that one frequently comes across them in popular poetry as rhetorical tools used to heighten style or to underline particular expressions. One also comes across them in satirical poetry, but every time a poet makes use of them he does so knowingly. This stylistic specialization shows that the vocabulary of the jvart ena is not yet completely in the "public domain"; the fact that certain terms are used in popular poetry among the arsenal of literary accessories as "internal exoticisms" is an indicator of their purity. Although occasionally encountered in everyday language, they are conceived of as remaining foreign to such use and therefore remain quite close to their esoteric origin.

          2.b. The kadag: adolescence and the introduction into the social and religious group

There is a complete lack of information on the mode of transmission of the jvart ena. We only know that the future kadag remains a long time in the company of the most experienced "oracles" and that he learns from their advice (O., p. 16). It is likely that, during this period, the "accomplished kadag", the "elders", do not stop at giving advice but also teach and initiate the neophyte in the techniques of kadagoba. It is also they who, during this same period, explain to the young kadag the sexual prohibitions to which he must submit himself from that moment on. The latter is fifteen or sixteen years old when the obligations and the "tongue" which will become his are revealed to him. At the same time, two important events take place in his existence: the divine voice declares itself in him and he is introduced into the social group. This last point has led to some confusion which T. Ochiauri alludes to (pp. 20-21) although he himself is not fooled.

Some have attempted to establish a link between the "tongue declaring itself" and the adolescent's physiological development. This assimilation is apparently based upon Sternberg's works on Uralo-Altaic shamanism (Sternberg, 1929, p. 12; 1936, pp. 141-142). It is not up to us to decide whether or not the evolution of the signs of possession is linked to the crises brought about by sexual development and puberty. We do, however, know for sure that the adolescent becomes a member of the social and religious community when he turns sixteen. We will add the following remark: there are two ways in which contact with the divinity can be made. The first, individual, is that which defines the kadag from childhood: the direct link which is established between a man and a supernatural being. The other is social and is only conceived of within the framework of worship and social community. Until he turns sixteen, the young kadag knows only the first, that of personal contact. When he reaches the age of sixteen, he becomes part of the social and religious community, and it is at that precise moment that the foremost social phenomenon of kadagoba appears—the language, communication between the possessed and the group: the language declares itself. As long as he was outside of the group, the possessed shamanized without communicating, prophesizing for himself, as in a kind of monologue. But at sixteen, he undergoes a double metamorphosis: he ceases to be a child and becomes a social individual; his lunacy transforms itself into a language and in leaving soliloquoy behind he accedes to communication. The beginning of the shaman's career cannot be separated from his entrance into the life of the group, this correlation marking the transition from a personal relationship with the divinity to a social one.
It would be pointless to attempt to recreate what the initiation of the kadag might have been like long ago according to the precise manner in which ethnography and the history of religion understand this concept. It is nevertheless possible to describe the trials by which he acquires and preserves his power and to discern clues to the themes of his initiation.

          2.c. Sexual prohibitions and the problem of ghabushi

The basic nature of the prohibitions is the negative attitude to sexual life which is demanded of the kadag. As soon as the khat'i appear before him in dream or as visions, the young kadag is declared to be tsminda [= "holy, ritually pure" —A.B.] (tsmida in the Khevsur dialect of Georgian) and must obey the rules of tsmindoba, that is to say a "holy and healthy" life. This purity essentially measn avoiding any place which has been soiled by impure women [souillé par les femmes impures]: kadag are forbidden from passing close to the samrevlo or kokhi, the huts outside of every village in which menstruating women isolate themselves during their period; from crossing through stables, for it is there that women give birth; and lastly from speaking to women during their period. But these are contacts which every Khevsur man—be he a kadag or not—does everything in his power to avoid. For kadag, however, this prohibition also encompasses the sexual life allowed to ordinary men.

Until he marries, every young Khevsur has more or less complete sexual relations with a young girl of his clan, whose "sworn brother" he is ("brother-husband" among the Pshav). All normal young men entertain such relations before marriage, when they then marry a young girl belonging to another clan than theirs. But the khat'i not only exclusively calls celibate adolescents: young kadag are also prohibited from practising tsatsloba [the name given by the Pshav to these pre-marital sexual relations, which the Khevsurs know as stsorproba —A.B.]:

'When a khat'i chooses a young man, he must remain pure; he can never pass through impure places, nor pass by a pregnant woman. The khat'i takes a young man, he takes him as long as he is without wife or child. Tatarai, for example, was without wife, Khtisai as well and all the others' (64 years old, 1945, O., p. 19).
'As soon as the khat'i appears, he (the adolescent kadag) must remain pure. He isolates himself in order to avoid seeing impure women. He is forbidden to practise stsorproba, and is also forbidden to take a wife' (80 years old, 1946, O., p. 19).

As long as his "initiation" lasts, he cannot have contact with women. But what happens after he has become a "confirmed" shaman? Information concerning this question is unclear and sometimes contradictory. According to field interviews conducted among the Pshav in 1930 by S. Makalatia, he is forbidden to marry for as long as he lives (Makalatia, Pshavi, pp. 181-184). In the lowlands of Kartli, the "slave of the khat'i" was also forbidden to marry; but if the khat'i took a married man as its slave, the latter was bound to put an end to all relations with his wife and, more often than not, to go and live elsewhere: 'If the khat'i chooses a kadag after he has married, he must separate himself from his wife. If he is unmarried, he is forever forbidden to marry' (recorded in Kartli, 1947, O., p. 62).

In Khevsureti, the problem is more complicated (or the information more complete) than in Pshavi. It would seem that the Khevsur distinguish two ways of shamanizing, according to whether or not the kadagoba comprises a phenomenon called ghabushi [G. ღაბუში]. This term refers to the excessive and violent manifestations which contact with the khat'i may give rise to: delirium, convulsions, self-mutilation, falls, &c.; in sum, all the pathological aspects of shamanism. In theory, ghabushi should disappear during initiation; but this is rarely the case and ghabushi continues to plague the shamanizing of the initiated kadag during the first few years of his career. According to the Khevsur, ghabushi and marriage are absolutely incompatible, and the kadag does not marry for as long as he shows signs of ghabushi. If ghabushi remains present for too long, the kadag goes to consult another, more famous or older kadag. After having shamanized, the latter tells him the number of years to which the khat'i which possessed him has condemned him to remain pure: usually three to five years. Once these years have past, the kadag is delivered from the effects of ghabushi, shamanizes in a calmer manner and has the right to take wife.

It often happens that the kadag stops shamanizing when he marries:

'The khat'i prefer single men. The kadag often no longer shamanizes after his marriage. Sometimes he purifies himself after marriage for three or five years. Ghabushi is impossible for a married man. He goes to see another kadag, &c. After these years of purity, he can bring his wife back [ramener sa femme]. Once he lives with her, he begins to shamanize again, but not as he used to: he is no longer subject to the effects of ghabushi as before. He is forbidden to sleep with his wife, the khat'i prevents him from doing so' (1945, O., p. 20).

          2.d. A dangerous ménage à trois: The kadag, women, and the jealous khat'i

The relationship between married kadag and his khat'i are not always easy, and sometimes give rise to serious tensions. This is a very common theme in Khevsureti, which often dons legendary or anecdotical overtones. The stories in which this theme appears also clarify the question of the relations between the kadag and his wife. A popular example tells the story of a shaman called Likokeli [i.e. Likok-eli, "from the village of Likoki" in Khevsureti —A.B.]: he was possessed by a khat'i who appeared before him and perched on his hand [khat'i were often thought to take the shape of birds, hence "perched" —A.B.]. Likokeli was not married. One day, his khat'i takes him along to pay a visit to the neighbouring tribe (the Tush). But the family of the kadag uses this opportunity to introduce a wife into his house. The khat'i learns of this and returns from Tusheti, having left its shaman behind. Its vengeance was, one says, swift and terrible (O., p. 58).

Another shaman and his khat'i leave one day to pay a visit to the neighbouring Kist. They meet a woman on their way: 'In his heart, the shaman desired her and wished to sleep with her'. But the khat'i, 'who can read [people's] hearts', became irritated and flew back to its shrine, alone. The shaman, having learnt of this, also returns, furious at the khat'i's lack of trust in him. Having made his way to the shrine, he sees the khat'i, perched in the darkness. He bitterly chides the khat'i and spits at it. The irate khat'i strikes him down dead on the spot (O., p. 58).

Gigia Chincharauli tells the story of something which happened to his ancestor, Saghira, who was a kadag: 'When the khat'i took him, Saghira was single. He married shortly afterwards. The khat'i authorized Saghira to sleep with his wife three times, telling him that he would have a son the first time, a son and a daughter the second time and a daughter the third time. And so it came to pass. But Saghira returned to his wife a fourth time, who then gave birth to a stillborn child. My ancestor stopped seeing his wife and children altogether after that: they give me the impression of snakes, he used to say' (O., p. 58-59).

Many are the kadag who leave their wife and build themselves a hut close to the shrine of their khat'i in which they remain for the rest of their lives. Totia Arabuli, for example, was called by the khat'i when he was already married and the father of several children. He left his family and retired to live like a hermit. Another kadag called Moklia was also already married when the khat'i called him; he stayed with her but ceased to have any contact with her for the rest of his life. Chincharauli remembers the kadag he knew:

'Tatarai was something of a kadag and a soothsayer. He married late in life. He undertook this with much piety: he made countless sacrifices of oxen and rich offerings to the khat'i before the wedding. He resumed these sacrifices when he brought his wife [en amenant sa femme, to "introduce" her to the shrine, presumably —A.B.]. He did not cease greatly honouring the khat'i. And only afterwards did he live with his wife. He only had a single child. He was forbidden to have any more.
'Another kadag called Kholiga began to shamanize when he was seventeen, and died aged eighty. He took a wife but was never allowed to sleep with her. He had her at home, and that was it. How does one find out whether or not one can sleep with one's wife? In dreams. The khat'i appears in your dreams and shows you that you are forbidden to touch your wife' (O., p. 59).

This information suggests that the sexual prohibitions are proportional: the stronger the relation between the kadag and the divinity, the stronger the prohibition with regard to the wife. Tatarai, who 'was something of a kadag', was allowed to have a child. But Kholiga, who was a kadag from age 17 to 80, never had any relations with his wife. In the same vein, kadag who are subject to ghabushi i.e. who are violently possessed are strictly forbidden to marry. On the other hand, married kadag shamanize more infrequently, their bond with the khat'i loosens. The proportionately inverse relationship between the degree of possession and the degree of intimacy between the shaman and his wife accounts for the divergent explanations given above. Essentially, it could be said that authentic kadag are forbidden to have any kind of sexual relationship.

V. Bardavelidze questioned the origins of this prohibition; her explanation, although different, does not contradict the one we have just suggested. The essential element of kadagoba is 'the belief according to which the shaman represents the object of the divinity's incarnation'. The author sums up the relationship between the khat'i and the kadag in the following manner: 'According to popular belief, the kadag was elected by the divinity. It was in him that the khat'i established itself, that it settled, and it was through his mouth that the khat'i communicated its commands and wishes to the community, using the kadag as a go-between in its relationship with its slaves. The belief in the incarnation of the khat'i was the very basis of the practice of kadagoba, whose sessions took place in houses or at the shrine. [...] The khat'i demanded that its kadag remain pure and reserved in his dealings with women; for the khat'i took part in the sexual life of its kadag—not directly and immediately—by via the intermediary of the kadag and of the other people it had possessed, in whom it had incarnated itself. In reality, any sexual relationship between the kadag and a woman must be interpreted as if the khat'i itself had entered into a relationship with her. In a word, and according to popular belief, we should think of the khat'i as incarnating itself in a kadag who enters into sexual relationships with a woman, and it is perhaps in this that we should search for the origin of the sexual prohibitions which the person elected by the divinity must obey: the khat'i does not want to have a sexual relationship with an "earthly" woman, with a woman capable of awakening the kadag's desire, which would inevitably be the case if it had not forbidden the latter from having such relations. I therefore believe that these prohibitions can be based upon the incarnation of the khat'i in its kadag' (Bardavelidze, ivris pshavlebi, pp. 109-110). The author adds that only the prohibition is explicit among the Pshav and the Khevsur; its origin, the incarnation and the divinity's risk of becoming "contaminated" as a result of incarnating itself having remained unformulated or having been forgotten.

It is clear that V. Bardavelidze's explanation implies taking into consideration the khat'i's psychological motives. One could say that the divinity imposes certain restrictions for personal and subjective reasons. If one accepts for the time being this homerical conception of the motives which guide the actions of the gods, one must go further and admit the possibility of jealousy being a further motive. Re-read the documents quoted above, and particularly the two stories in which the khat'i exacts a cruel and swift vengeance upon its shaman whom it suspects of having been interested in or desired a woman: are these not crimes of passion? They show the abnormal symptoms of jealous love: suspicions erupt, triggered by the slightest indication of a possible infidelity, followed by injustified reproach blind to the truth, and the drama ends in an exaction of revenge completely disproportionate with reality.

This clinical picture we have painted is not exactly confirmed but is certainly rendered plausible by a series of documents concerning the trials the divinity inflicts upon the shaman.

These are legends, myths or prophecies which develop the well-known theme of love between a human being and a supernatural woman, goddess, genie or fairy. We know that the Soviet specialists of shamanism considered such sexual relations to be one of the ways in which shamanic powers were acquired. As for the Georgian highlands, all the texts refer to a shrine and a divine being, the khat'i of Kakhmati and the "goddess" Samdzimari.

          2.e. Samdzimari

Samdzimari [G. სამძიმარი] is one of the supernatural creatures who are on a level between the khat'i and the demons and whom the Khevsur refer to as Dobil. The term means "sworn sister". The Dobil are the "sisters" of certain khat'i; they do not share the latter's dignity but are worshipped just the same. (We have dealt with this matter in more detail in chapter 7.) The most famous and most powerful of the Dobil is Samdzimari, the sister of Giorgi of Kakhmati [one of Khevsureti's most important khat'i —A.B.]. She has sexual relations with most of the priests and shamans of this shrine. But she does not stop there, and frequently seduces adolescents or married men who do not belong to the servants of any shrine.

When she comes to sleep with [partager la couche de] a man who is not a shaman, this event is a sign of future 'happiness and good fortune':

'Samdzimari sleeps with some [men]. She comes into their bed, or one suddenly sees her standing next to the bed, and if she comes to sleep with you this means that good fortune awaits you, the birth of a son and luck in every domain' (O., p. 50).

She always appears at night, in hallucinations or during a dream. More often than not she prefers to seduce the khelkatsi [G. ხელკაცი, ხელი = "hand" i.e. "the hand of the khat'i" and კაცი = "man", so "a man touched by (i.e. chosen, selected, designated by) the khat'i" —A.B.], the men linked to the shrine:

'This Samdzimari walks around at night in the guise of an earthly woman. She enjoys fooling around with the men of the shrine—and more than that. She often merely pleases them with her hand (et'ot'aveba)' (O., p. 49).

The most common texts are those which speak of the relations between Samdzimari and famous living, dead or legendary kadag. These texts are written in verse and belong to the repertoire of the kadag. This is Samdzimari herself speaking:

'So powerful was I then, I walked the earth,
I carried [or "wore"; the context is unclear —A.B.] ç'ima-lakht'ar (plants); so powerful was I then:
I seduced Abuletaur Saghira, who desired me as if I were a [real] woman.
I lay down next to him, I embraced him, I pressed my breasts against him'
(O., p. 47, K).

The shamanic texts generally speak of kadag belonging to or having belonged to the clans of Abudelauri or Megrelauri. Samdzimari either takes on the appearance of their wife or of a woman they find particularly attractive. The most frequently mentioned kadag is Abuletaur Kholiga of the village of Kistani [near Shat'ili —A.B.]. A Khevsur mentions:

'This Kholiga was pure, he was a kadag. He was married, but the Cross [the khat'i —A.B.] of Khakhmat'i [G. ხახმატი] forbade him to have sexual relations with his wife and to have children. He had to remain pure and chaste, like the holy monks. This Kholiga lived with Samdzimari for a long time, she acted like his earthly wife, they lived and slept together' (O., p. 44).

But Samdzimari was not sparing with her favours; a text in verse tells us that:

'I seduced Abuletaur Kholiga, I was like his wife.
I shared his bed, in the guise of an earthly woman.
But I also took pity on Gogoturi, I slept with him, caressed him in all sorts of ways'
(O., p. 47, i).

As Samdzimari bestows her favours equally upon initiated kadag and upon "unelected" men or men who will become shamans later, we must ask ourselves the following question: does having sexual relations with a Dobil lead to the acquisition of shamanic powers, or is it on the contrary merely the mark of such powers? Any event which directly links a human being with a supernatural being can be a sign of a shamanic calling; in this sense, having sexual relations with a divinity predispose the individual to the calling of the kadag. On the other hand, since any man might have such relations without necessarily becoming a kadag and since those who are the most likely to have them are kadag, one may come to the conclusion that such relations are not a decisive factor in the acquisition of these powers. In reality: 1. With reference to the shamanic calling, the notion of causality loses its meaning; union with Samdzimari is a sign, and a sign is both cause and effect. 2. The event cannot be considered on its own; it must be linked to the personality of the person who experiences it. A man who constantly finds himself in a relation with the sacred lives in a universe in which human history no longer has the same meaning; everything that happens to him necessarily has a supernatural aspect, and every event in his life bears a religious dimension which the same event in the life of an ordinary man would simply not have. This is why the meaning of a union with Samdzimari is different for ordinary Khevsurs and for kadag or future kadag. When Samdzimari chooses to unite herself with a mere mortal, this union is no more than a good omen, but when she unites herself with an initiated or future kadag, this is or will have been a sign. 3. The life of a kadag and every event which takes place therein have, so to speak, two faces: one sacred and the other social; these two equalize each other, are complementary. In other words, any event in the life of a kadag expresses itself in two directions, one towards the divinity, the other towards the social group. In order to replace the union with Samdzimari in this perspective, we must first emphasize the importance of the following:

Firstly, Samdzimari does not have sexual relations with her own kadag [i.e. the shaman of her own shrine —A.B.], for—uniquely in Khevsureti—the kadag linked to her is a woman. This woman is either a female kadag or a particularly gifted possessed woman. The possibility of union with Samdzimari is not the privilege of a particular kadag but extends instead to all the kadag in Pshavi and Khevsureti. Also, Samdzimari is by no means the only Dobil; there are others, less famous, who also come to spend the night with such and such a kadag. We mean by this that sexual union with a divinity is the fate of any shaman, that the possibility is part of the shamanic condition in general. (As for Samdzimari's preference for the men of particular clans, see section 7, chapters 6 and 7.)

Secondly, the relations which the kadag have with the Dobil are in no way manifestations of a supernatural sexual power, but are instead experiences of sexual impotence or ineffectiveness. These unions take place in two ways: firstly, in dreams. In such cases, as the Khevsur tell us, 'it doesn't work, and the seed is lost'; the man is united with the woman-khat'i in a dream, but wakes up at the last moment, sees that she is not there, and 'the seed is lost' (O., p. 44, 45 and 49). In the second case, the Khevsur is not asleep; he tries to fulfil his conjugal duties, but fails: it is through this manifestation of impotence that he realizes that the woman-khat'i had adopted the guise of his real wife. Union with a Dobil therefore means, in sexual terms, impotence, and in human terms the impossibility of fulfilling a normal sexual act. Finally, relationships between a kadag and a supernatural woman are both an acquisition in sacred terms and a loss in human and social terms. To witness—in dream or while awake—the supernatural experience of union with Samdzimari thus means that the relationship the kadag entertains with the divine is strengthened but also implies the adoption of a completely negative attitude towards sexual life—both on the physical and social plane.

In sum, and this applies to kadagoba as a whole, that which the kadag acquires in terms of divine relationships he loses in terms of social relationships. This is why it is no exaggeration to compare the effeminacy of the Asiatic shamans or the Scythian oracles with the renunciation of his own gender which characterizes the Khevsur kadag. That which Herodotus' Scythian [? Enarées] and the Siberian shamans—who both lead a "feminine" existence—and the Khevsur kadag share in common is their asexual nature. When a shaman acquires his supernatural powers, he becomes or commits himself to becoming a non-man [un non-homme]—both biologically and socially. This can be seen, as Mircea Eliade did, as a kind of consequence of the shaman's sacred promotion: 'The bisexual nature... is due to the fact that the priest is considered to be an intermediary between two cosmic planes... This is a ritual androgyny, an archaic formula well-known to the divine bi-unity.' 'This hermaphroditism,' Eliade continues, 'is based upon the sacred value of the intermediary, upon the need to abolish polarities.'[*] This can be put somewhat differently: without any mediation, a man brutally acquires a dual structure, accepting sole responsibility for the meeting of two universes, that of the gods and that of humans; at the same time, he becomes self-sufficient in the sense that he also shoulders the masculin/feminine relationship. In sum, the shaman stands in for social group, both in terms of the sacred/human relationship and in terms of the man/woman relationship.

But to these remarks we need add another, which seems to us more important. Dumézil demonstrated the relationship which should be drawn between supernatural power, the magical specialities of certain gods or heroes and their characteristic mutilations. For Odin, for example, 'the loss of the earthly eye was the means by which the god-magician acquired the immaterial eye, clairvoyance, and all the supernatural powers this ability entails.' And in the case of Tyr, the one-handed: 'Tyr did perhaps not lose his right hand to become the Law-maker; it was precisely because he was the Law-maker that, alone among the gods, he lost his right hand.'[*] An increase in supernatural abilities is thus accompanied by a decrease in physical and human abilities. A human deficiency is the price of supernatural power. Is this not the case, all things being equal, of the kadag?

The Khevsur put into practice, though without formulating it, a sort of law of conservation of energy, of power in the widest sense of the term. They seem to conceive human society, that is to say the world of men, as having a certain amount of power which must remain constant. To preserve this constancy, this invariability, they resort to symbolic means: when Khevsur society loses a soul i.e. following the death of an individual, it makes sure that it acquires another symbolically (cf. section 3, chapter 4). Now, when a Khevsur becomes a kadag, both he and the social group acquire an extra amount of power: it is normal that, at the same time, the social group rids itself of a certain amount of power in another domain. When a new kadag is initiated, society benefits from new relationships with the divine; but in doing so it deprives itself of particular relationships with other clans: for the kadag will be neither father nor uncle &c. The social group becomes richer in religious terms but poorer in social terms. And the same applies to the kadag: he becomes an Übermensch of sorts, but also a kind of Untermensch on the social plane. Is it not this duality which expresses itself in the union between the kadag and the woman-khat'i—which confirms his victory in the divine order but consummates his defeat in the world of men? This is the dual meaning of Samdzimari's visit: the sign of a divine calling, a symptom of physical impotence, her visit reveals to or reminds the man that he has become a shaman but has ceased to be a man.

          2.f. Purity, the attraction to water and stags with flippers for hooves [sic]

The prohibitions which are imposed upon the shaman and which are at the same time the condition and the mark of his power are collectively referred to [in the Khevsur dialect of Georgian —A.B.] as tsmidoba, "purity". The Khevsur distinguish two kinds of "purity": one active, consisting in the observation of what Mauss called negative rites; besides the sexual prohibitions, the kadag must also respect nutritional taboos: he must abstain from eating onions, garlic, eggs and poultry (O., p. 17). A sick man goes to see a kadag and asks him to shamanize: if the latter has recently eaten garlic, he asks the sick man to return some other day, for the khat'i 'would refuse to manifest itself' (O., p. 87). kadag also drink alcohol, be it beer or vodka, only very rarely. These prohibitions are more or less strict: for established kadag, they are absolute and suffer no exception; for occasional shamans or simple fortune-tellers, the prohibitions are part of the preparations before shamanizing. The other kind of purity is, so to speak, undergone: the Khevsur use the same term tsmidoba to refer to the irresistible attraction which water exercises upon uninitiated kadag. This kind of purity is undergone in the sense that the possessed plunges into the water against his will and without later being able to account for his behaviour:

'My father was a kadag. He told me: I was young when the Cross elected me. I began by being constantly drawn to the water. I ran towards the river and threw myself into it like a madman. Be the water icy cold or the weather freezing, I ran to the river and had to fall in. Then I began to shamanize. I was ill, the khat'i had clearly taken me, held me. Purity is needed. If I was drawn to the water, it was maybe because purity was needed' (54 years old, 1945, O., p. 18).

All those questioned agreed with one another on this point: the kadag is drawn to the water despite himself, and this is one of the main signs of his calling. For example:

'The kadag go mad when they are adolescents. They bathe in even the coldest water.' Another Khevsur: 'The khat'i chooses children. He takes care of them (of the possessed). When the chosen one is fifteen or sixteen, he is drawn to the waters' (O., p. 18).

Being drawn to water is precisely one of the symptoms of ghabushi, that disorderly and violent state of being incompatible with marriage: 'The khat'i compells the possessed to go towards the rivers; he forces them to throw themselves in' (O., p. 20).

'The attraction to water,' wrote Marie Delcourt, 'belongs to the very depths of human belief'; after reminding us of the legend of the Carian spring 'which was capable of dissolving all energy' and of Ovid's story of the metamorphosis of a boy into a hermaphrodite in the waters of this spring, the author goes on to demonstrate the relationship which exists between water and androgynous dreamings.[*]

What interests us here is the conception of water as a place of ontological transformation, as an environment favourable to metamorphosis. Now, it would seem that the Khevsur sensed this relationship between water and transformation. When a highlander drowns, the soul remains prisoner to the water and may be preyed upon and tormented by demons; a special ceremony is required to free the soul and to enable it to become itself again. It is important to note that there is a veritable re-creation of the soul from water. In the waters, the soul is gradually deteriorated and it is from the water that it may regenerate. In a myth, very widespread in Pshavi and Khevsureti, we see the khat'i Kopala undergo a transformation following his dive into the river: pursued by demons, Kopala plunges into the water and turns into a stag (see section 5, chapter 1).

As we shall see further on, the Khevsur clearly oppose the human and animal worlds; the symbolic representation of animality versus humanity is, however, not the role of carnivores but that of herbivores—particularly the stag. This animal turns up in many Georgian archaeological finds—and notably on objects made of bronze found in the western Georgian region of Racha, [west of] Khevsureti, which appear to date back to the VIth century B.C.: 'the study of Georgian art during the Bronze Age shows that the culture of bronze was dominated by the representation of fantastical creatures, depicted in motion,' wrote the archaeologist Shalva Amiranashvili, who added that most of these animals resemble the stag, an animal which therefore seems predisposed to metamorphosis. One of the most common ornamental designs to have been unearthed in Racha represents a running stag with flippers for hooves [sic]. One often notices, says Amiranashvili, 'the stag's relationship with the aquatic element'. In his history of Georgian art, the historian includes several photographs of brooches made of bronze upon which a leaping, "flippered" stag is depicted.

The relationship between the stag, water and metamorphosis is interesting for the following reasons: 1. the metamorphosis into a stag takes place in water; 2. this is no simple magical accident but a veritable transition from one order to another, from the human social kingdom to the animal kingdom. Water is therefore not only a medium favourable to metamorphosis, but is actually a place which facilitates the meeting of two universes and one's access to another. It is in the Carian spring that the son of Hermes and Aphrodite loses his virility and acquires femininity; it is in a river that Kopala abandons the social state for the animal kingdom. Is it not the passage through water that helps the possessed to cross over from the earthly, human universe to the supernatural world? When he becomes a shaman, the possessed—hitherto normal—undergoes a profound change, which affects the deepest levels of his state of being. Are not his insane jumps into the river so many attempts to make this ontological transformation come about, to facilitate it? May we be permitted the suggestion.

3: Kadagoba and illness

          3.a. Kadagoba described

Once the kadag has definitely mastered his power, he remains forever tied to the khat'i which called him. With one exception: the female kadag of Khakhmat'i can contact any khat'i, as will be explained further on. Apart from this unique case, the kadag may only contact the khat'i of his social group, the protector of the social and religious community he belongs to. The kadag of the village of Likoki, for example, is possessed for life by the khat'i Kopala, the clan divinity of Likoki. It is inconceivable that a kadag of Likoki could contact other khat'i such as Kviria, Iakhsar, Khirchla, &c. The limits of the religious community vary according the scale or the perspective in which one considers them: the village of Barisakho is consecrated to the khat'i Kviria and consists of only one clan and one shrine—the kadag is therefore "recruited" from among the villagers. In the village of Akhieli, on the other hand, where three clans live together but each have their own shrine, the population is socially and religiously-speaking divided into three sectors: Mgelika, for instance—a famous kadag from Akhieli—was in contact with the khat'i of his social and religious sector, Mtis-Veshag [G. მთის–ვეშაგ, "mountain-dragon", from მთა = "mountain" and ვეშაგი = "dragon" —A.B.]; the two other shrines also had their own kadag, chosen from among their respective clans.[*]

Things are more simple in Pshavi: twelve clans each have their own shrine with its own kadag. In Khevsureti, on the other hand, socio-religious groups—despite being founded on clan membership—are extremely numerous and their organization is very complex.

The tribal shrine of the Khevsur, called Gudanis Jvari [G. გუდანის ჯვარი = "the Cross of Gudani" —A.B.] is shared in common among the three great clans of Arabuli, Chincharauli and Gogochuri [G. არაბული, ჭინჭარაული, გოგოჭური]; the members of these clans inhabit numerous villages across the region. The kadag of Gudani may therefore come from one of these villages and belong to one of the three clans.[*] In any case, the choice of the kadag and his relationship with the khat'i are determined by his social membership. It is as the member of a social group that he is called by the khat'i.

The Khevsur use a special term borrowed from the language of the gods to refer to the relationship which unites the kadag and the divinity: madena. More specifically, madena denotes the treatment which the khat'i forces the kadag to undergo and its manifestation in him. The khat'i manifests itself in three ways: 1. suddenly, when the kadag is asleep; 2. suddenly, when the kadag is awake; and 3. when the kadag calls upon it to do so. The third of these is evidently the most common, for this is precisely the power the kadag holds (O., p. 39-40).

Unlike Chikovani states in his Handbook to Georgian Folklore [Manuel du folklore géorgien], kadagoba in the highlands has neither musical accompaniment nor material preparation. [*] This is precisely why shamanism in the mountains is so original compared to that practiced in the lowlands. In Khevsureti, language is the defining feature; the preparations of the kadag to "receive" the divinity are not accompanied by any rituals apart from the lighting of three candles. The session [or, more properly in English, séance —A.B.] calls for the presence—real or virtual—of three partners: the kadag, the khat'i and the social group, the latter sometimes reduces to its most simple expression: the person who has fallen ill. One must distinguish several categories of shamanic séance, according to time and place. As regards the place: kadagoba may take place either in the shrine or in the kadag or someone else's house. The time: some rituals of the religious calendar are necessarily accompanied by a shamanic séance. In such cases, kadagoba is called for on a certain day; these séances may only take place in the shrine (O., p. 64-65). Kadagoba in the shrine may also be called for by particular events and therefore be unplanned: the Khevsur may, for instance, wish to consult their divinity and find out its opinion on an important matter such as war, pillage, revenge or any other problem the community may feel the need to consider, or people who have fallen ill may go to the shrine during particular celebrations or rituals and call upon the kadag to cure them. Kadagoba in the home is always linked to an event: a Khevsur requires the services of the shaman, for a member of his family or for his cattle, generally in cases of illness or for other reasons which we will describe later (O., p. 68, 70, 72-73). The nature of kadagoba at the shrines of Karati or Khakhmat'i is different, and will be described later during the description of these two shrines.

The shamanic séance, considered without reference to either time or place, takes place in an identical manner. Initially, nothing distinguishes the kadag from an ordinary man. He lights three candles and prays to the khat'i, asking it to manifest itself:

'The kadag sits down; those in attendance remain motionless. Everybody waits. He beseeches the khat'i, then begins to tremble, to foam at the mouth, to pant. He behaves in this manner for a while. He beats his heart with his closed fists. The strength and length of this agitation varies from kadag to kadag. Dzia Batsaligoeli, for instance, whipped his head backwards with such violence that one could have thought he would break his neck. Then, [the kadag] shamanizes' (1945, O., p. 66).

Before he enters into a trance, he is standing or sitting; during the fit, he sometimes rolls around on the ground, but as soon as the khat'i manifests itself and speaks, [the kadag] regains his composure and once again stands or sits. He never vaticinates (prophesizes) lying down:

'I saw Khichla shamanize at [the shrine of] Gudani. He stood before the altar. The mezare (G. მეზარე, from ზარი, "bell" i.e. the man responsible for ringing the shrine's bell) rang the bell; if the khat'i manifested itself immediately, Khichla would shamanize as he stood. Otherwise, if the khat'i did not appear immediately, he would sit down and would then shamanize as he sat. In any case, making the khat'i come is difficult; he does not come "for free" (G. იაფად, iapad, from იაფი, iapi, "cheap")' (62 years old, 1946, O., p. 66).

According to most informants, the kadag sits down to shamanize; even if he beseeches the khat'i to appear while standing, he then sits down after the fit. Here is a prayer used to call the khat'i:

'Great Archangel Michael [the name of this particular khat'i; although pagan, the divinity was subsequently named after the Christian archangel —A.B.], listen to the cries of he who calls you! I am ignorant, I am made of flesh! Do not turn away from your servant, come to him!' (O., p. 67).

The role the candles play appears to be similar to that of the prayer i.e. to attract the khat'i. Baliauri, in an unpublished manuscript, describes séances of kadagoba and speaks of the 'angels and crosses circling the candles'[*] As we will see later on, the khat'i, when they come closer to the world of humans, adopt the appearance of birds and, despite being invisible, soar through the air above holy places. It is probably this characteristic which Baliauri was referring to. Also, lit candles are the best mediator and occupy an important place in the rituals and symbolic representations destined to establish an ontological relationship between an ordinary human being and a divinity.

          3.b. The khat'i speaks

As soon as the khat'i "enters" his body, the kadag freezes and remains completely motionless; he cries out the divinity's words:

[N.B. The Georgian text is in a mountain dialect of that language (probably the Arkhotian dialect of the Khevsur dialect of Georgian) and even native speakers of Georgian would be unable to fully comprehend it. It is also important to remember that this is the khat'i himself talking through the kadag whose body he temporarily controls —A.B.]

'Hai! Hai! I am the Archangel, I have strength and power! | My slaves no longer respect me! | They have levelled the arm to the arm, the neck to the neck [i.e. they consider themselves equal to the divinity, they have lowered me to their level]! | They are dragging me into soiled waters! | I am no longer different from others, but I am nonetheless the Archangel, I nonetheless have strength and power! | I advise my slaves to be wise! | May they not treat me without respect, otherwise I will punish them, I will not renounce my rule, for I hold it from the heavens!' (G. 'ჰაი! ჰაი! მე ორ მთავარ–ანგელოზი, მე მაქვს ძალიდ’ შაძლებაი! | ჩემთ ყმათ აღარაად გავაჩნიორ! | მხარი მხარს გამისწორესად კისერ კისერს! | მათრევენ სიმურშიად’ რიოშანში! | აღარაადვის გავაჩნიორ, მაგრამ მე აისივ მთავარ–ანგელოზი ორ, მე აისივ ძალიდ’ შაძლებაი მაქვ! | ჩემ ყმათ ურჩევ, რო ჭკვით იყვნან! | მე ნუ გადამაგდებენ თორე ვანან, ჩემს წესს მე არ დავიყრი, რაიც ხთისგან მაქვ მალოცვილი!') (The prophecy continues: the khat'i goes on to deal with various matters concerning the community; see below) (O., p. 69).

Almost all shamanic texts begin like this: the khat'i is angry and complains about his followers—his "slaves"—who are neglecting his worship. The passage from 'my slaves no longer respect me...' to 'I am no longer different from others' is enormously impressive to those in attendance. From the Khevsur point of view, it could be said that the khat'i is polishing his style and knows how to address the crowd; 'my slaves no longer distinguish me from others', that is to say: they treat me as an equal, as if I were one of them. The very mention of this accusation, obviously exaggerated, evokes sacrilege, but the image which follows reinforces this impression: 'they have levelled the arm to the arm, the neck to the neck', i.e. the khat'i's arm and neck have been lowered to the level of those of simple men. The terrible accusation 'they are dragging me into soiled waters' (simurshiad' rioshianshi) is stylistically expressed through an unthinkable combination of words: water = simur, a sacred term belonging to the divine "language of the Cross" + filth = rioshi; the words themselves are being profaned. The following sentence sums up the accusation: 'I am no longer different from others'; the transition between accusation and threats is very cleverly achieved: it uses the opposition between the shameful manner in which the khat'i has been treated and the affirmation of his powers: 'but I am the Archangel...'. All this is very well structured and one may compare this, the beginning of a vaticination, with the exhortations of a speech destined to impress an audience. (One need notice the symmetry and the metathesis [O., p.E.D.: 'The transposition of sounds or letters in a word, or (occas.) of whole words or syllables; the result of such a transposition. Formerly also: †substitution of one sound or letter for another (obs. rare). Now chiefly Linguistics.' —A.B.] between the start of the accusation and the transition from accusation to threats: 1. 'I am the Archangel...', contrasts with: 'my slaves no longer respect me'. 2. 'I am no longer different from others...', contrasts with: 'but I am the Archangel...'; the same contrast, but inverted).

When the khat'i has ceased to harangue his flock he instantly leaves the kadag, who stops, exhausted, sometimes unconscious: 'When he finished shamanizing, Khichla felt his strength fade: when the khat'i left him, he began to feel weak' (62 years old, 1946, O., p. 66).

During the vaticination, the kadag is unconscious; his eyes are turned up or closed, 'his awareness is gone', say the Khevsur (O., p. 66). When he finishes, those present or the person who has fallen ill almost always ask him to explain the meaning of the khat'i's words, but the kadag refuses to answer. For him to be able to interpret the prophesies, he must be told what he cried out during his fit; to an ill person who asks for more details, for instance, the kadag might answer: 'How do you expect me to know what the khat'i made me cry out? You heard it yourself, didn't you? All I can tell you is to do what the khat'i ordered you to do, and he will cure you.'

          3.c. Curing the sick: the scapulimancy of the mk'itkhavi

In most cases, the goal of the shamanic séance is to cure someone who has fallen ill. We have seen that illness is always thought of as a phenomenon of possession. But the kadag is not the only physician, even though as a last resort he may be asked for his views on any condition. In general, the task of healing the sick is divided up among the following people: the shaman or kadag, the soothsayer or mk'itkhavi (G. მკითხავი), the doctor or diast'kari [G. დიასტქარი according to Charachidzé, but the word seems to be extremely obscure in Georgian —A.B.], and the flag-bearer or mk'adre (G. მკადრე). The mk'adre plays an important but very specialized role, which will be studied separately (section 3, chapter 1). Ailments or illnesses "of the soul" are the exclusive responsibility of the kadag. Invisible illnesses "of the body" are sometimes treated by the mk'itkhavi, sometimes by the kadag. Visible physical illnesses are treated by the native doctor [médecin indigène in Charachidzé, presumably the diast'kari —A.B.] and sometimes, if complications arise, by the mk'itkhavi or even by the kadag. Wounds are the exclusive responsibility of the Khevsur doctor. This division of responsibilities is naturally variable at times, e.g. following an erroneous initial diagnosis or if the illness evolves. Conjunctivitis, for example, at first treated by the doctor, can reveal itself to be the sign of an illness "of the soul" and therefore become the responsibility of the kadag.

In 1925, a new element intervened in Pshav-Khevsur medicine and introduced changes to this division of responsibilities. For the first time in the region's history, a medical centre was set up in Pshavi, complete with doctor and assistant. Dr Tedoradze later published an interesting work called Five Years in Pshav-Khevsureti, in which one may find much useful information on all the aspects of life in the mountains. One thus learns that between 1925 and 1930 the mountaineers maintained all their folklore and pagan beliefs. This book and the experience it relates are of particular interest to the ethnographer and the historian: how did Pshav culture react to the appearance of this new person? The doctor quickly and quite naturally found himself integrated into the existing system; society conferred upon him a role which was in harmony with the pre-existing structure and which did not affect it. As soon as the local population became convinced of the therapeutic effectiveness of Dr Tedoradze's treatments, the latter occupied an intermediary position between the Khevsur doctor and the mk'itkhavi. Up to a certain degree, visible illnesses "of the body" were treated by the native doctors. When the latter, however, seemed to be unable to treat them, people called upon Dr Tedoradze—this consultation often being suggested by the mk'itkhavi himself. As for invisible physical ailments, for whose treatment Khevsur doctors were at a particular disadvantage, people either consulted the mk'itkhavi or the government doctor. The latter thus encroached upon the domain of the native doctor as well as upon that of the mk'itkhavi. The domain of the kadag, however, remained intact and was never encroached upon by Dr Tedoradze's work. Lastly, concerning wound and fractures more specifically, the doctor shared the work with his local counterparts. According to him, Khevsur doctors were astonishingly good surgeons who were particularly skilled in performing cranial trepanations. The Khevsur doctor Alek'o Ochiauri, for example, performed over 350 serious operations during his career, 150 of which called for a cranial trepanation. Of these 150 cases, only three resulted in the death of the patient.[*] Essentially, the presence of a foreign doctor did not interfere with the functioning of the local system; when, for example, Dr Tedoradze said that the patient was in a desperate condition, the patient's family would politely thank him and would then immediately call for the kadag, who would in turn begin his own form of treatment.[*]

All things considered, the role of the native doctor and the nature of his calling are no different from those of the kadag. One must also not lose sight of the fact that the mountaineers did not think of illness as an abnormal state of being but instead as a supernatural state. "Ill" is not the opposite of "healthy", but instead of "profane". To be ill is, one way or another, to find oneself in contact with a supernatural being. That is why any cure presupposes the intervention of a person who entertains a privileged relationship with the supernatural. This conception of illness is perfectly illustrated by the belief according to which doctors—like the kadag but to a lesser extent—are in contact with the sacred. Native doctors hold their powers by the khat'i; their knowledge of medicinal plants is thought of as the result of a revelation; lastly, they entertain direct contact with the khat'i, who appear to them in dreams and instruct them to prepare such and such a potion (O., pp. 89-90). Besides—as Ochiauri and Dr Tedoradze explain—the taking of a medicine is always accompanied by prayers addressed to the khat'i by the doctor or the mk'itkhavi, and the effectiveness of the remedy is linked to their doing.[*] There is a continuity between doctor and kadag, which rests upon their relationship with the divinity; this relationship, however, is very powerful in the case of the kadag, and quite weak or "relaxed" in the case of the doctor.

The mk'itkhavi occupies an intermediary position—one closer to the kadag, however, than to the doctor. His role is important and deserves closer inspection. The term mk'itkhavi is often used to denote the kadag; the expression kadag-mk'itkhavi is quite common. In the "Language of the Cross", one of the synonyms of mk'itkhavi is metsar (G. მეცარ). Essentially, metsar means "one who knows, one who has acquired the knowledge", but in the XVIIIth century Saba Orbeliani defines the word as "one who pretends to be able to read the scapula [the shoulder-blade]" [le faux connaisseur de l'omoplate], or, according to the commentary by Marr and Brière, "a sorcerer, an astrologist, a false prophet".[*] The analogy between metsar and mk'itkhavi can also be found in Georgian folklore: All the variants of a tkmuleba (G. ტქმულება) or folk story called "How a demon entrusted a cow to a man from Kiziq'i" allude to the existence and the personality of the mk'itkhavi. This story is part of the rich répertoire of folk singers, the mest'vire (G. მესტვირე), who accompany their stories on the bagpipes (for more on the mest'vire and the st'viri [G. სტვირი] or "bagpipe", see ch.5); the story can also be found among Georgia's body of folk tales. The demon, or devi, come to claim his cow from a peasant, talks with him and praises the qualities of its master, but the peasant mocks everything the demon says: "The Peasant:—Your master must be deaf.—The Devi:—Deaf? He is nothing of the kind. When ants fight he goes to separate them.—The Peasant:—So he is a gul metsar (= he who knows in his heart).—The Devi:—A gul metsar? He is nothing of the kind. Two of his oxen were stolen and he has never been able to find the thief."[*] The singer Z. Eradze defines a gul metsar as being a mk'itkhavi. In the conversation between the peasant and the devi, one sees how a mk'itkhavi is defined: he who "knows in his heart" what happens among even the smallest of creatures (e.g. among ants) and who would know, if he were really, who stole his cattle. He is therefore an oracle. [N.B. Charachidzé has some trouble explaining his reasoning in this last paragraph; the original French is just as obscure —A.B.]

Unlike the kadag, however, the mk'itkhavi's power of prophecy is linked to a technique: "reading" a shoulder or shoulder blade. A "shoulder-blade-oracle", as Marr and Brière rightly say.[*] This practice has survived among the Georgian highlanders, particularly among the Khevsur, who continue to resort to it to this day. "Reading" the shoulder blade is done on the shoulder of a sheep or an ox; it enables one to predict the future and to diagnose certain illnesses. The vision of the future thus obtained restricts itself to the destiny of one family, and never encompasses the collective affairs of the wider social group. Every time a Khevsur family slaughters a sheep or an ox, they call the mk'itkhavi; he takes part in the consumption of the meat and uses his oracular gift. In exceptional cases—e.g. repeated misfortune or if a family member or domestic animal falls ill—the Khevsurs sacrifice an animal solely to enable the mk'itkhavi to prophesy. Here follows a description of how the mk'itkhavi "reads" a shoulder blade.

Essentially, every one of the bone's anatomic singularities, every detail of its shape, is a sign. The inevitable variations in the shoulder blade's colour are also "read". If the top end features an osseous growth: a horse will be stolen. If an indentation is ossified instead of being "open": a family member will be injured. One of the shoulder blade's edges is darker in colour: the harvest will be good. The shoulder blade is thin and translucid: this is a good omen in general. If the shoulder blade is somehow particularly thin and translucid: a horse whose owner thought was lost was actually stolen by someone. If the lower part of the shoulder blade features a small hole [for the passage of e.g. the scapular circumflex vessels —A.B.]: a member of the clan will die. A bony growth at the lower end: the birth of a son. If, on the contrary, the end is absolutely smooth: the birth of a daughter. The lower end is pierced: a family member is or will fall ill. Spots against the white background of the shoulder blade: a cow will fall into a gully.[*]
This method of foretelling the future from the shoulder blade of cattle is also practiced in Khevi (albeit a lot less than in Khevsureti[*]) and in Pshavi.
Resorting to this practice is obviously quite costly, since it requires the sacrifice of a sheep or an ox. In case of illness, the Khevsur therefore try to combine this sacrifice with the one they offer to the shrine to beg for the patient to be cured. Instead of sacrificing an animal without knowing to which particular divinity it should be offered, the Khevsur prefer to first find out the name of the khat'i capable of healing the person who has fallen ill. The mk'itkhavi is often called upon to visit people who have fallen ill in their homes, but in such cases he resorts to a less costly method of divination whose specific function is to identify the name of the khat'i responsible for the illness. The method is the following: in the patient's house, he asks the family to bring him a wooden bucket, a wooden spoon and a thread of wool or cotton. He winds the thread around the spoon, balances the spoon on the edge of the wooden bucket full of water, and quickly unreels the thread into the water. Every time the end of the thread is reached, the mk'itkhavi utters the name of a khat'i; if the thread unreels itself completely without stopping, the khat'i responsible for the illness is the one whose name is being uttered. The operation is repeated until a result has been obtained. In any case, even if the illness was not the result of divine wrath, the khat'i whose name was discovered during this ritual is the one to whom a sacrifice or an offering must be made in order to cure the patient.[*]

          3.d. Parallels with other regions of the Caucasus

This practice must have been quite widespread in the Caucasus mountains (and not only among the Georgian highlanders) for it can also be found, under various guises, among the Ossetians. Their sorcerers [sorciers] 'are in great demand, writes M. Dumézil... Mostly, one consults them to discover the cause of someone's illness and the nature of the cure: one must identify the divinity [Charachidzé uses the word génie] responsible for having "sent" the illness, which expiatory victim this divinity requires, &c.' To do this, the sorcerer holds thin wooden rods laid out in a particular way and recites a list of divinities: when the rods begin to rise, the name has been found. Herodotus described an identical technique for divination used by the Scythians: 'The Scythians have many oracles, who make use of thin wooden rods made of willow; they bring bundles of such rods, set them down and untie them; when they have finished laying out all the rods, they fall into a trance. While they utter their prophecy, they tie every rod, one by one, back into a bundle' (IV, 67). Another method used by the Scythians to foretell the future should perhaps be compared to the practices of the Ossetian sorcerers. Herodotus: 'The Enariots, who are men-women [hommes-femmes in the original French], claim to hold their oracular powers from Aphrodite. They use the bark of lime-trees; they split it into three, twist the strips around their fingers, unwind them, and begin to prophesy.' Ossetia: the witch [la sorcière] ties a knot into a piece of cloth, whose length she then measures from the knot. She recites a list of divinities and re-measures the cloth at every new name: when the result differs from the first measure, the name has been found (G. Dumézil, 1930, pp. 155-156). Vs. Miller, who noted this similarity, points out that the use of a piece of cloth for prophecy is restricted to women, just as divination through lime-tree bark among the Scythians was the privilege of the men-women.[*]
We can see that the principle of this method is the same in every case: reciting a list of divinities and the manipulation of objects to identify the divinity responsible for a particular case of illness. The Georgian practice of unravelling a thread seems even closer to the Scythians' use of lime-tree bark in divination than the equivalent Ossetian practice (Note: lime-tree bark is in great demand among the Pshav and the Khevsur, as it can be used to roll cigarettes).

          3.e. Mk'itkhavi and kadag—a comparison and practical examples

These are the two divinatory methods employed by the mk'itkhavi in Pshavi, Khevsureti and Khevi. The kadag, as we know, resorts to no such manipulations. And this is not the only difference between the two: unlike the kadag, the mk'itkhavi can be man or woman. He [or she] fulfils no social function—his [or her] competence remains within the bounds of the family. Lastly, he does not heal the sick, but can only, in simple cases, discover the cause of an illness. His powers stop here. It is true that this is sometimes enough to bring about the patient's recovery. As in every civilization, it can happen that someone who has fallen ill feels better as soon as he is told the name of his illness. This is precisely the role of the mk'itkhavi, who reveals himself by the very methods he employs: naming the illness, identifying the divinity responsible. V.V. Bardavelidze wrote of a case which illustrates this phenomenon rather well: a young Khevsur's eyes became swollen and began to suppurate. A few days later, his condition had already clearly improved, and his eyes were almost back to normal. To Bardavelidze's astonishment at his amazing recovery, the Khevsur answered: 'my mother consulted her kadag and he told her that I had been possessed by the khat'i Kviria'. Whatever the cause of his recovery, the young Khevsur believed it to have been brought about by his being told the reason for his illness (on cases of possession such as this, see sect. 4, ch. 2).
The mk'itkhavi is often unable to diagnose patients, even if the symptoms of their illness are clearly visible. Or the mk'itkhavi identifies the khat'i responsible and the patient offers a sacrifice at the corresponding shrine—but his illness does not go away. In cases such as these, the Khevsur begin to believe that a patient's torment at the hands of a khat'i is not a sign of divine anger, but of a divine calling: he will have to serve [i.e. work for —A.B.] the shrine. In any case, the decision rests with the kadag:
'If the fever does not abate, writes Baliauri, we turn to religion. An example: 'The disease remains, and nothing can be done. The fever rises and rises. One must go to see the kadag. The head of the family goes to see the kadag, first of all to the kadag of the shrine, taking three candles with him. He explains the problem: a member of my family has fallen ill, and all our efforts have been in vain. He has fever. The remedies we have given him have been without effect. Come on, maybe you can understand. Light your candles. If you can find out what's wrong with him, good, and if not then at least we will have tried. He'll die and that will be that. The kadag lights the candles and implores the khat'i: 'Your power is great. You are the great ones, God's dignitaries. May Morige the God preserve you, angel whirling around these candles. If your slave has angered you in some way, help him, throw the remedy of joy upon the roof of he who is suffering. Do not confine him to his bed, do not close his bed (=do not let him die), here are the candles, take them, they bear your name.

'With these words, the khat'i comes to the kadag [i.e. completely possesses him, occupies his body; note that the following paragraphs contain some rare transcriptions of a khat'i speaking —A.B.], sending him into a trance and speaking through him:
'Three shadows and three morning stars, perhaps [si on le veut in the original French], then you will understand my purpose (mizezi). Three morning stars I declare the period to be. If this can happen [s'il peut en être ainsi], may peace descend upon thy home; add a sheep to my flock within the bounds of my shrine' (quoted in O., p. 87).

After three shadows and three morning stars i.e. three days and three nights, the patient might recover, as long as he or she sacrifices a sheep at the divinity's shrine. In the example we have just mentioned, the kadag was consulted in his own home. Here is another, where this time the kadag goes to visit the patient:
'I used to be in great pain because of my kidneys (წელი, ts'eli, in Georgian); you know, like when a shotgun is "broken" open and is folded in two, well I was walking around bent in two like that. It was terrible! They brought the woman to see me[*]. She was so pale you would have been terrified had you seen her. She was possessed by Khakhmat'i. So they bring this woman. We all light candles, there were three (literally: "the candle had three feet"). She began to wail, to shout:—I have come to earth, I, Samdzimari, I followed the rays of the sun. You have failed to honour me, to please me. Do not... (here, two incomprehensible words that Maia Ochiauri, the patient, reads as "I had to be pure, to stay away from the menstruation and birth-hut [la maison d'isolement in the original text]"). Why should I help you unless you adorn my collar with buttons (=make a donation to my shrine)? Give, you, and I will give in turn. I will give you my belt if you give me buttons for my collar.

'We made a sacrifice, and on the second morning I took the cattle out to pasture, my kidneys were no longer causing me any pain.

'Another time, my eyes hurt. They were swollen, as big as this. My face was covered in spots, which often burst. So someone went to fetch a woman-kadag, one of those women with oracular powers. The khat'i of Khakhmat'i came to her and told her the cause of my illness:—Sacrifice an animal to the paternal shrine. Don't bother to offer a horned animal to your parents' shrine; there you need only bring a young, weak-hearted animal. If you can, it would be a good idea to sacrifice a victim at the shrine your father worships, where you grew up.
'I sacrificed a ram and a kid-goat. The ram at my father's shrine, the kid-goat at Khakhmat'i. My eyes got better and my spots disappeared' (O., p. 87-88).

          3.f. The women-kadag of the shrine of Khakhmat'i

To understand this text, one must know the following: the patient is married to a man from the community of Khakhmat'i and therefore belongs to the shrine and clan of Khakhmat'i. But when she was young and until she was married, the woman belonged to another clan, was the slave of another shrine. When a married woman falls ill, the kadag must find out whether the woman is being "concerned" by her husband's khat'i or by that of her father. In the case mentioned above, the woman-kadag found a compromise: a kid-goat for the husband's khat'i and a ram for the father's.

The women possessed by the khat'i of Khakhmat'i are often called upon to intervene throughout Pshavi and Khevsureti, but exclusively when a woman or cattle have fallen ill. When a woman or an animal fall ill, it is very rare to call upon a male kadag. One either goes to see the mk'itkhavi, or a woman-shaman from the shrine of Khakhmat'i. Among her other responsibilities, the female divinity Samdzimari is the protector of women and cattle. It would be unthinkable for a male khat'i to incarnate himself in a woman's body, and unworthy of a male kadag to fall into a trance on her account or that of a domestic animal. She may sometimes risk the wrath of the khat'i, who exacts revenge and manifests himself by causing illness, but if she is really possessed she could only be so by a demon or a female khat'i (unless of course the illness is one of the soul, in which case everything is possible).

When cattle falls ill or dies for no clear reason, or if the milk turns sour, the Khevsur call upon a mk'itkhavi or a female kadag from the shrine of Khakhmat'i:
'If someone falls ill, we go to see the mk'itkhavi. We also go to see him if the milk goes bad, when the cows stop giving milk.'
'We go to find the mk'itkhavi when something is wrong with the cattle, or if a woman falls ill in the menstruation hut [la maison d'isolement in the original French text]. For the cattle, we also call upon the women of Khakhmat'i. We sacrifice a kid-goat to Khakhmat'i' (1944, O., p. 90).

          3.g. Exceptional cases: epidemics and the dying

We will see later on that the male kadag also distinguishes between men and women, and adopts different cures accordingly when the séance takes place within the shrine. Treatments which take place not in the home but within the sacred enclosure generally accompany a normal ritual, except in cases of epidemics. In such cases, the priests and the oracles summon the faithful and carry out a special ceremony led by the kadag. He proceeds as usual: invoking the khat'i, falling into a trance, after which the divinity manifests itself, possessing the kadag and speaking through him. The aim of this ritual is to find out why the khat'i is angry and how the faithful can appease him i.e. which sacrifices they need to make (O., p. 85). Epidemics are usually caused by specialized demons, or rather are these demons; but the khat'i is nonetheless responsible for their outbreak, for out of anger with his human flock he may have given free rein to the Lords of Disease [les Seigneurs-Maladies in the original French] to spread disease among the community or may even have invited them to do so. Every patient is cared for individually within his or her family, without the intervention of the khat'i, but at the same time the social group fights the presence of the epidemic collectively at the shrine. The epidemic's disappearance is secured by a shamanic séance within the sacred enclosure, whilst the sick are cured without calling upon the kadag and his khat'i.

In very exceptional cases, the priests and the oracles [devins] decide to hold a special ceremony to save a dying person. The priest sacrifices a victim in such cases; the faithful pray to the khat'i, pleading with him one last time to spare the person's life; lastly, the kadag goes into a trance and begins to prophesy. He then goes to visit the dying person's bedside; if the latter shows no signs of recovering and if the kadag sees that death is inevitable, he leaves and the mesultane is called for. Her speciality is communicating with the world of the death (O., p. 85; for more on the mesultane, see sect. 3, ch. 4).

          3.h. Description of the different rituals by which instances of illness and possession are cured

In more normal circumstances [i.e. when the sick are neither the victims of an epidemic or dying —A.B.], the ritual by which the sick are cured follows or coincides with one of the ceremonies foreseen by the community's religious calendar. The healing always begins in the same manner: the kadag prays, goes into a trance and prophecizes. Several different cases are possible, however; these can be divided into two categories according to the nature of the intervention required from the khat'i. If the cause of the illness is still unknown, the khat'i is asked to reveal it. If the patient is possessed in the strict sense of the word, the nature of his possession is known, having been expressed by the symptoms of his illness: the goal of the shamanic séance is then to "free" the person. The shamans recommend or practice three different kinds of cure: 1) expiatory offerings and sacrifices; 2) the patient entering into the service of the shrine [i.e. becoming a lay servant of the shrine —A.B.]; 3) the patient being released from his possession through a shamanic ritual.

When the priest has led the faithful in worship, he stands aside and the kadag seats himself in front of the altar. The first shamanic séance is devoted to collective matters, as we shall see later in the next chapter. The goal of following séances is to cure the sick. The kadag must repeat the entire process for each patient: the prayer which calls the khat'i, the trances, the manifestation of the divnity and the prophecizing; every contact with the divine leaves the kadag in a state of exhaustion and forces him to rest for a few minutes before he starts again with the next patient. He is often incapable of uttering the initial prayer, which his assistants then recite for him:

'This man (the patient) and his parents and all of us, we beg you to reveal your command. Hear our prayer and tell us what we must do, &c.

'After this invocation, explains the kadag Sumbata Arabuli, the khat'i comes to him. He trembles, foams at the mouth and cries out. Then he proclaims the khat'i's will and stops, exhausted' (O., p. 86).

"Diagnosing" patients is not always easy, for unlike the kadag they are not necessarily solely possessed by the khat'i of the social group and clan to which they belong. The patient [as an ordinary member of the community —A.B.] can be "concerned" by three different divinities: that of his clan, that of the clan of his maternal uncles, and that of the clan of his brothers-in-law. The Pshav poet Vaja Pshavela wrote in 1903: 'When a man consults the oracle, his ailment can just as well be caused by his father's shrine as by the khat'i of his mother's brothers or by the shrine of his wife's brothers'[*]. The clan's khat'i is therefore asked to reveal to his shaman whether or not the illness is the result of his own intervention or instead that of one of his sworn brothers, one of the other khat'i. This initial uncertainty is expressed in the following prayer spoken by the faithful:
'We beg of you, tell us the reason for your anger or for that of your sworn brother or sister. Secure the forgiveness and release of the one who is suffering. If other khat'i have chosen to torment him, find out the cause of their anger and tell us how to appease them' (O., p. 85).

When the khat'i has delivered his oracle, the subject knows the names of the sanctuaries to whose divinities he must sacrifice expiatory victims. This is the simplest remedy, suitable for benign illnesses of the body.
Some invisible illnesses of the body, e.g. heart conditions, are immediately synonymous with early stages of possession; people suffering from such ailments can only be cured if they join the other servants of a shrine. Baliauri declares: 'If a man is having trouble with his heart and if the sacrifices he makes to the khat'i do not cure him, people say: perhaps the khat'i wishes him to serve in his shrine as a man of god (khelosani) [G. ხელოსანი, a word now used to refer to a workman —A.B.] and perhaps this is why the khat'i is "concerning" him and harming his heart. The kadag then declares that he must become a priest or simply treasurer or even the shrine's flag-bearer [= the m'kadre —A.B.]. Only the Crosses [= the khat'i —A.B.] can cure heart conditions, the informer concludes' (O., p. 88-89).

When someone is suffering from an illness of the soul, two causes are conceivable:

1. The patient is possessed by a khat'i. One need then return to the previous stage: the kadag establishes the identity of the khat'i and the possessed is cured by becoming a man of god in the shrine.

2. The kadag reveals that the patient has been struck down by bneda [G. ბნედა], that is to say the divinity has abandoned his soul, which has fallen prey to a demon (O., p. 89). Three cures are possible:

a) If the kadag considers the demon's hold over the man to be weak, he recommends that the Khevsur become a member of the shrine; his links with the divinity will be re-established by the very fact that he will be a man of god, and the demon will be forced to relinquish his soul:

'A certain T'ot'ia from the village of Amgha [in Arkhot'i, north-west of central Khevsureti, beyond the main mountain ridge and watershed i.e. technically in the northern Caucasus —A.B.] went mad. He went to the shrine, but the demons would not leave. The oracles then said that he should serve at the shrine as a flag-bearer, in order for the demon to leave him and for the Cross to become his master; thus would he be cured. His family therefore made him enter the shrine as flag-bearer, and his madness disappeared' (Baliauri, manuscript, in O., p. 89).

b) Any woman who becomes possessed has inevitably fallen prey to a demon. When a woman showing signs of madness is brought to the shrine, she is dealt with by the mk'adre [the flag-bearer, G. მკადრე —A.B.] instead of the kadag. The mk'adre purifies and delivers the possessed woman by wielding the sacred banner, according to a ritual which we will describe later called dak'otch'va [G. დაკოჭვა].

c) If it is a man who has gone mad, then it is the kadag who delivers him through a ritual called ganatvla [G. განათვლა] (a word formed around the verb natvla, "to illuminate, to baptize"). Despite the fact that this is one of the most interesting therapeutic rituals and that it could tell us much about the fundamental problem of the dynamic relationship between the possession of the kadag and that of the patient, we unfortunately lack a detailed description of ganatvla. We know its main elements but ignore the way in which they were expressed. The ritual is initiated by the kadag and ended by the priest. The former, by the usual method, reveals that the patient is possessed by a demon, whose identity is irrelevant. The possessed also falls into a trance-like state and prophesies in the chaotic and violent manner which the highlanders call ghabushi (see above, ch. 2). Lastly, the priest sacrifices five or seven victims the patient's family has brought as offerings; he drenches the possessed in the animals' blood and, in doing so, purifies him and restores him to the religious community, from which his madness had excluded him. We understand the link between the two sequences and their significance: during a double scéance of possession, the person who had become alienated is freed from the clutches of the demon which had possessed him, and the priest need only reaffirm and consolidate the person's link to the shrine. What needs to be established, however, is how the first part is carried out: is the possession of the kadag somehow being transferred to the patient? or are the two instances simply parallel, without interference? We will only know the answer to this question when the Ethnographic Institute in Tbilisi will have finished publishing its Materials.

Lastly, we must mention a kind of shamanic cure which also involves twin instances of possession. A Khevsur goes to the shrine, knowing that he is possessed by a particular khat'i. As soon as he crosses into the sacred enclosure, he falls into a trance, prey to a violent nervous attack, and shamanizes in a wild and disordered manner. He finally collapses, exhausted. People often fall into a trance before the ceremony has even begun, and are dragged to the shrine by two brawny Khevsurs. His chaotic utterings are nonetheless prophetic, and all those present listen to them carefully. The kadag then prophesies himself, and through him the khat'i reveals the fate which he has reserved for the madman: either he declares that the séance has cured him, or he announces that he has chosen him to serve the shrine along with its other servants (O., p. 86-87).[*]

Throughout this chapter, one may have noted that in many cases the best way of being cured is to become a "man of god" i.e. to serve as a priest, treasurer, flag-bearer or even kadag. Such recruitment is, however, the prerogative of the kadag, for it is he who reveals to the sick what cure the divinity recommends. By merely exercising his therapeutic powers, the kadag plays an important social role. But the servants of a shrine are not necessarily all former or recovering patients: we will now see that they are all nonetheless promoted to their sacerdotal functions by the kadag, the divinity's spokesman. The kadag therefore not only possesses certain powers, but is himself the source of all power.

4: The kadag—his social role

The kadag's social functions are exclusively exercised during the religious ceremonies held at the shrine, in the presence of all the worshippers. Apparently, this shamanic séance is no different from those which we have already described in the previous chapter, but the personal preparation which the kadag must undergo for this ceremony is much greater. A month before the date, he stops all contact with his wife; beyond mere sexual abstinence (which is year-round in any case), this also entails absolute isolation: he ceases to see her, to speak to her. He sleeps in a special bed which no woman has ever touched. He no longer eats dishes prepared by women and only uses plates and cutlery which he has specially recovered from their hiding-place in preparation for the ceremony. Several days before the ceremony is held (the duration of this period varies from place to place), he carefully washes himself [= purification —A.B.], dons a clean change of clothes and goes to live in the shrine, where he now submits himself to a rigorous period of fasting (O., p. 65).

          4.a. Choosing the dast'ur

Every year, the kadag chooses new dast'ur for the khat'i [économes du Xat'i = the shrine's i.e. the khat'i's servants, stewards e.g. the priest, the oracle, the treasurer, the flag-bearer, &c. —A.B.]. We have already seen that to all intents and purposes he designates the priests (khutsi) and the flag-bearers (mk'adre)—former possessed whose calling is channeled and regulated by the kadag's shamanic intervention. We have also seen that the kadag tells some patients to serve the shrine, but dast'ur recruited in this manner (accidentally, so to speak) are considered supernumeraries. This is why some shrines "employ" dozens of dast'ur despite the fact that the number of real i.e. actual, serving dast'ur rarely exceeds thirteen; these titular stewards are nominated for a year, as is the khat'i's treasurer (megandzure) [G. მეგანძურე, from განძი, განძეული—gandzi, gandzeuli = "treasure" —A.B.]. They are generally chosen on New Year's Day or, more rarely, on the day upon which the clan's cult is celebrated. In all parts of Khevsureti except two, the dast'ur are designated by the kadag during a shamanic ritual.

Al. Ochiauri tells us more:

'On New Year's Day, matters of justice and other important questions were discussed at the Cross (= the shrine), notably the choice of dast'ur. At some shrines, they were elected to their position or lots were drawn at random, whereas at other shrines a shamanic ritual was held. The kadag was asked to prophesy at Likok'i, Gudani, Roshka and Khakhmat'i, for example, and this used to also be the case at the shrines of Arkhot'i and Shat'ili. Later, however, the people of Arkhot'i and Shat'ili abandoned this custom, but it survived everywhere else' (Manuscript, quoted in O., p. 64).

In 1946, the inhabitants of Shat'ili told T. Ochiauri [Tinatini Ochiauri, Aleksandre Ochiauri's daughter, like her father a noted ethnographer; she published most of his writings: see OCHIAURI, Tinatini, ქართველთა უძველესი სარწმუნოების ისტორიიდან ("From the history of the Georgian's most ancient religious beliefs"), Tbilisi: 1954 —A.B.]:

'We designate the dast'ur every year, without calling upon the kadag. A long time ago, apparently, it was the kadag who designated them, but we no longer practice kadagoba around here. We only go to consult the kadag to learn the khat'i's intentions, to find out the reasons for illnesses and to discover how they can be cured. But as we no longer have a kadag, we go elsewhere to other villages such as Kist'ani or Barisakho, for example' (informant in the village of Mutso, O., p. 64).

So in these two areas—Arkhot'i and Shat'ili—dast'ur are chosen by drawing lots, almost certainly because kadagoba is no longer practised as frequently as it used to be. One should also consider the fact that Shat'ili and Arkhot'i are located to the north of the great chain of the Caucasus Mountains, whereas all the other parts of Khevsureti are to the south of this chain, which acts as a barrier. It takes several days' journey to cross it. This isolation relative to the centre of Khevsureti—which the highlanders refer to as Bude-Khevsureti, "the nest" [from G. ბუდე, bude —A.B.]—may explain the slow disappearance of a custom so alive elsewhere. One would also need to find out whether these two districts have indeed followed their own independent cultural evolution, different from that of the rest of Khevsureti. Several facts seem to indicate this: the slow disappearance of kadagoba; the fact that funerary rites are not quite the same as to the south of the Caucasus watershed; an essential difference: it is in the Shat'ili region that standing (i.e. above ground) burial structures are to be found, evidence of a notion of death and the afterlife quite different from those which prevail among all the other Georgian highland tribes. These "houses of the dead" stand to this day, even though they are no longer used. They were described by Radde in the nineteenth century [Gustav Radde, 1831-1903, a German naturalist and ethnographer who travelled widely throughout the Caucasus on behalf of the Imperial Russian Academy of Sciences; see Wikipedia entry —A.B.], who in 1878 declared the structures to be around forty years old.[*] Kovalevskii gives a somewhat erroneous description of them and claims to have seen identical structures in Tusheti.[*] ["Houses of the dead"—G. აკლდამა, ak'ldama—smaller than the famous and uniquely(?) large ones in Anatori a few miles north of Shat'ili but almost identical to the ones in Muts'o further to the north-east still stand in the ruined village of Ts'aro above the confluence of the Ts'ovatis-tsq'ali and Gomets'ris Alazani rivers in western Tusheti; until they migrated to Kakheti in the early nineteenth century, Ts'aro and the 7 or so other ruined villages further up the valley of the Ts'ovatis-tsq'ali River were inhabited by the Bats (or Tsova-Tush) people—another Georgian tribe of highlanders who most certainly had close links with the peoples of the North Caucasus viz. with the neighbouring (and probably related) Chechens and Ingush to the north as well as with the Andi of Daghestan to the east —A.B.] Mummified human remains have been found in these stone mortuaries, indicating that the dead were placed therein for all eternity.[*] We know that similar structures were widespread among the Ossetians, the Chechens, the Ingush, &c.[*] The Ingush to the north, the Chechens to the north-east and the Tush to the east are the immediate neighbours of the Khevsur regions of Arkhot'i and Shat'ili, indicating that the inhabitants of the latter two areas may have been influenced by north Caucasian practices. In any case, the custom of interring the deceased in these "houses of the dead" seems to have been completely forgotten; it is not even mentioned in the very complete description of funerary rites in Arkhot'i published by the Ethnographic Institute in Tbilisi.

There are two categories of dast'ur in Khevsureti: the "great", who assist the priest during important religious ceremonies, and the "small", who only take part in the observance of secondary cults. 'The kadag first names the great dast'ur, then the small dast'ur' (O., p. 68). This ritual is carried out in the same way as we have described above; only the formulas expressed are different. To the prayer addressed to the khat'i, one adds: 'Change for us the inheritance of this happy year [i.e. the previous year's choice of dast'ur —A.B.], by your strength and your will, declare, o blessed one, the [new] dast'ur according to your wishes; we are ready to obey'. When the khat'i manifests himself, he tells his followers: 'You must see to your duties, and I to mine. My dast'ur are old, they are finished. I no longer want them, I no longer recognize them, I tell you this through the mouth of my linguist [i.e. the kadag, from the Georgian term meene, "speaker"; see above —A.B.]. I have chosen two dast'ur to serve me for a year (two, in some cases), Mgelik'a and Torghva (for example)'. More often than not, the khat'i also names both a man responsible for the beer and the harvest and a treasurer: 'I entrust my jars and barley to Khtiso, I entrust my silver coins and my treasures to T'ariel' (Al. Ochiauri, manuscript in O., p. 69).
We now add some particular details concerning the shrines which we will study later on:

The shrine of K'op'ala in Likok'i (sect. 5):

'In Likok'i, the shrine of K'arat'is Jvari [G. კარატის ჯვარი, "The Cross" i.e. shrine "of the khat'i K'arat'i" —A.B.] has four dast'ur. The kadag designates them for two years during the ceremony to mark the arrival of the New Year. We all gather, men and women. The kadag places himself upon a special stone (dast'urt saq'eno kva) used for the election of the dast'ur... (Here, description of the séance of possession)... The kadag, when the khat'i comes to him, first designates the "great" dast'ur, the one who will bear the sacred flag (i.e. the mk'adre). Then, one by one, the other dast'ur' (1945, O., pp. 67-68).

The shrine in Khakhmat'i (sect. 7):

'The kadag designates the dast'ur during the Feast of Giorgi, the day of the ceremony. There is a dast'ur for the village of Biso, another for that of Khakhmat'i (the two villages which belong to the shrine). The kadag prophesys on the Saturday at noon. Saturday morning, we gather the shrine's herd of livestock. Last year's dast'ur assist the priest, but they are no longer allowed to tend to the oxen [Charachidzé uses bouefs in French i.e. oxen as opposed to taureaux, bulls —A.B.]. The great dast'ur, those who take part in the great feast of Kakhmat'i, are named that day. The small dast'ur, those who assist during the secondary ceremonies, we designate in winter, on New Year's Day, by drawing lots' (62 years old, 1945, O., pp. 66-67).

The tribal shrine of Gudani (sect. 6, ch. 2):

As we know, the shrine of Gudani brings together the three main Khevsur clans—Gogotch'uri, Tch'intch'arauli and Arabuli [G. გოგოჭური, ჭინჭარაული, არაბული]—which are themselves subdivided into several smaller clans, who live scattered all over Khevsureti in many different villages. So two kinds of dast'ur were designated at Gudani: the ones attached to the shrine itself, the others being chosen there for different khat'i but belonging to one of the three clans. All the nominations take place during the festivities which mark the arrival of the New Year, but the location of these festivities varies according to the origin of the dast'ur and their destination: those belonging to the shrine are designated by the kadag of Gudani within the sacred house [i.e. the stone structure which stands at the heart of every shrine, the main "chapel" —A.B.], whereas those belonging to other shrines are designated outside the sacred house [of Gudani] by kadag come from elsewhere. The latter, being possessed by divinities foreign to Gudani, are not allowed to penetrate the holy of holies:

'When the kadag are from here (from our Cross [i.e. shrine]), they enter the room (begheli [G. ბეღელი, "grain store"]). Those who come from other villages cannot shamanize in the begheli. They sit on the bench reserved for elders, outside the begheli, along with their dast'ur. Only the priest, the kadag and the mk'adre enter the begheli' (52 years old, 1946, O., pp. 65-66).

A man from the village of Ghuli explains that the communities without a kadag go to Gudani to have their dast'ur chosen:

'There is no kadag in our village. We can only go to Gudani. We go there specially to have our dast'ur designated. We take the candidates with us, along with the villagers from Uk'ankhadu. We then pray to the khat'i: "Our blessed altar, our Cross, do not be less generous that the other khat'i which reside at the door to God's court [the khat'i are thought of as intermediaries between their worshippers and the supreme God, Morige Ghmerti—G. მორიგე ღმერტი, "God the Ruler" —A.B.], you are no less honoured than they are, you are no less well provided for: come to the aid of your saq'mo (= the social and religious community)"' (75 years old, 1946, O., p. 67).

          4.b. Dealing with the problems of the Cross

Once the dast'ur and the treasurer have been named, the kadag goes into another prophetic trance during which he "deals with the problems of the Cross", i.e. foretells the community's future for the coming year. This is his most popular activity, the one which everybody has been waiting for, for the kadag is more appreciated as a prophet than as a healer. The Khevsur base their agricultural work on these predictions e.g. when deciding the time will be right for sowing, &c.; the predictions generally concern the quality of the coming year's harvests, the weather, the health of livestock, enemy attacks, the health of the community, illness and death. The following is an example given by a Khevsur informant of what the khat'i says when he "comes" to the kadag:

'Where are my mighty powers? Where have they gone? Everything used to be thrown before me—golden barley, glorious cups [of sacred drink —A.B.], oxen and young bulls! Back then I had power and strength—back in the days when I hunted down the old mothers of demons and smote them, scattering their pieces to the nine corners of the world! And now that I busy myself at the door to God's court on behalf of my slaves, do my slaves show me their gratitude? I have seen no golden barley grow! The sun-woman [la femme-soleil in Charachdizé's text] has taken her place up on the mountains above and my crops are ruined, along with the golden barley of my slaves! Morige-God has cleaved the ram's horns in three! [Charachidzé writes mouflon, i.e. the wild mountain sheep Ovis orientalis, but the khat'i might be referring to either the Eastern or Western Caucasian "Tur" or wild mountain goat—Capra caucasica and Capra cylindricornis respectively; all are endangered and are on the International Union for Conservation of Nature and Natural Resources' red list of threatened species —A.B.] Beware! I may be able to save you, and will intercede on your behalf at the door to God's court! Yes, I hope to spare you through my victory and my intercession! I will accumulate sacrificed cattle, thus will I mollify K'viria, I will do everything in my power during the shadows and the dawn, I will stack up victims and offerings in heaps! I will entreat Giorgi, the sister of Giorgi who wears a necklace [i.e. Samdzimari; see above —A.B.]—may they grant me their help! Go, bedeck the Cross of Khakhmat'i! Fill the sky, be generous with your sacrifices! Live in hope—not fear!' (76 years old, 1946, O., pp. 67-68).

The message contained in this prophecy is clear: misfortune threatens to befall you, but if you offer up many victims to the shrine you will be spared. The "language of the Cross" [the secret jvart ena, see above —A.B.] uses various expressions to refer to different evils: the sun-woman has taken her place up on the heights: drought; the ram's horns are split: violent deaths in battle. The khat'i promises to intercede with the supreme god Morige and with K'vriria, the agent of divine power. Giorgi and his sister are the divinities of the shrine of Khakhmat'i, one of which specializes in protecting the rural economy (cf. sect. 7).

The prophesies uttered at the shrine of Gudani concerned not only the local community but instead Khevsureti as a whole. Most of the region's villages sent representatives to the shrine in order to keep abreast with the results of the great shamanic séance, and villagers would continue to exchange and discuss this information long after the news had been brought back to their village. The inhabitants of Khevsureti's more distant district where kadagoba was rarely practiced, such as Arkhot'i, or had completely disappeared, such as in Shat'ili, impatiently waited to hear what had been prophesied in Gudani. The people of Arkhot'i, for instance, among whom kadag continued to exist but whose sole competence was healing the sick, could not themselves discover their khat'i's intentions for the year to come; instead, it was the kadag assembled in Gudani who asked their respective khati what the intentions of the Cross of Arkhot'i were:

'Shamanizing was done in Gudani. People even mocked those who came from Arkhot'i, who asked the different kadag of Gudani—"You must shamanize for us: what does our khati (that of Arkhot'i) demand from his slaves? What are his intentions?" Throughout Khevsureti one heard, for example, that the Cross of Arkhot'i had drawn its sword dripping with blood. If the Cross of Arkhot'i drew its bloody sword from its scabbard, then the people of Arkhot'i would defeat the infidels (i.e. any enemy foreign to Khevsureti). If the Cross did not, then men would be killed or their livestock would be taken' (O., p. 71).

The Khevsur distinguish between prophesies uttered by the kadag during ceremonies foreseen by the calendar and between those uttered by the kadag during impromptu séances held in response to particular events:
'During the annual ceremony, the kadag proclaims what will happen to the village, or what the many Angel-khati require be done to honour them. He speaks of illness, of misfortune, and of what needs to be done to avoid them [Charachidzé uses pour les détourner = to divert them, to send them someplace else —A.B.], the victims which will need to be sacrificed. He also predicts hailstorms: "I fear hail, God-Morige is displeased (...)" But he does not prophesy injuries, nor dangers arising from enemy actions. It is not his custom to deal with matters concerning war, but instead only with matters concerning the village' (O., p.70).
The kadag's annual prophecy is therefore essentially devoted to work and daily life; special problems such as war or armed expeditions call for an extraordinary i.e. impromptu séance.

          4.c. Consulting the khat'i before the hunt or war

Before setting off to hunt or to avenge some wrongdoing, the Khevsur consult their various khat'i; equally, when a group of hunters or warriors are far away from home, those who have remained in the village go to the shrine to discover the fate of their expedition. If the news is bad, the village offers up sacrificial victims to the shrine, imploring the khat'i to fly to the hunters' or warriors' aid (Al. Ochiauri, Ms. in O., p. 71). In such circumstances one consulted the local khat'i; if men were instead planning to raid or fight an enemy, then all the Khevsur turned to the most illustrious of the khat'i and to Gudani especially.

Just as the various khat'i devote most of their ethereal existence to fighting demons, the Khevsur never missed an opportunity to plunder neighbouring tribes: Tush, Ingush, and particularly Chechens. Such raids, which were still carried out as late as in 1918, were often nothing more than tit-for-tat acts of revenge whose main goal was to recover livestock or avenge the dead. The highlanders, as we shall see later on, believe that the clannish khat'i fights by their side in person and accompanies them throughout all their raids and battles; this is why they absolutely must consult him before going to war, for if he is not willing to fight alongside them then his slaves will certainly be defeated.
In Arkhot'i, the khat'i Mika (or the Archangel Saint Michael) reveals his will during a vision: the mk'adre "sees" Mika astride a white horse preparing for battle, then sees him 'charging down the snow-covered slopes—his pack of savage dogs, their fangs red with gore, running alongside him'.

At most of the other shrines, the divinity communicates its wishes through the mouth of the kadag. In Pshavi, Vaja Pshavela wrote in 1888, 'it is via the intermediary of the oracle that the clan consults its khat'i before every armed expedition'.[*] A. Ochiauri describes the going to war of the entire Khevsur tribe:
'When the people made ready for war, everybody gathered in Gudani. At least one victim was offered up in sacrifice, and the kadag was then asked to shamanize. All the warriors knelt on bended knee and uttered the prayer: "Blessèd Cross of Gudani, you are great, and great is the power of thy hallowed name. Come set in motion the tongue of this mortal: shall we or shall we not be victorious if we go to war? Is it or is it not the time for your power to bring us victory over our enemy? Will your standard march before us for the greater glory of your slaves? Even if defeat once again awaits us, tell us, o blessèd one, through the mouth of your mortal being [= the kadag —A.B.]. We will not leave until you reveal to us your wishes."

'While the men prayed, the kadag sat a short distance away. In turn, he beseeched the Cross: "Do not mock me, have me speak your heart's desire, so that your saq'mo [G. საყმო] (social and religious community) may leave to do thy bidding and be victorious. Without your power, without your victory, we mortals are powerless."

'The kadag, his eyes popping, began to shake and to wring his clenched hands, and then began to prophesy: he proclaimed the wishes of the khati, announcing whether the Cross was willing to immediately go into battle—or, if not, when the right time would come. Those present listened to his prophecy with their hearts. Sometimes the khati refused to go to war, sometimes he accepted, saying: "You will not see me during the battle, but I will be fighting alongside you"' (Ms. quoted in O., pp. 75-76).

If the Khevsur went on an expedition, they had to stop on their way and consult the khat'i anew to find out if he had not changed his mind since they left. For this reason the kadag accompanied the warriors, along with the mk'adre holding aloft the sacred standard which guaranteed the symbolic presence of the divinity. Stories of such expeditions are very widespread in the Pshav-Khevsur folklore. The most famous of them tells of how the Khevsur, gone to plunder the region of Mitkho [probably the community of Mutso, north-east of Shat'ili —A.B.], which was inhabited by Chechens, stopped half-way; the khat'i told them to turn back, but they disobeyed him and pressed on until they were severely defeated:
'One fine day the Khevsur decided to go and ravage Mitkho. They gathered in Gudani, and the khat'i declared: "Go, brandish my standard, I will lead you to Mitkho. I will fight alongside you and will crush your enemies."
'They set off: the mk'adre marched at their head with the standard, behind him three kadag and the priest, followed by the warriors. On their way they met up with the men from Shat'il and, having reached the mountain gorges which dominate Kist'eti [G. ქისტეთი = "the land of the Kists", from ქისტი, kist'i = a Chechen], they set up camp and had the three kadag shamanize. The kneeling warriors prayed: "O great khat'i of Gudani, repeat thine orders and intentions for our victory, have thy wishes not changed?" The three kadag went into a trance, but the khat'i appeared to only one of them: "Since the moment I ordered you to go to war, I have held counsel with the other khat'i at the gate to God's court. My sword was taken from me for a year. Do not go to Mitkho now, for you would be beaten: I have been forbidden from helping you. My sword will be returned to me in a year's time—then will we go to destroy Mitkho!"' (Al. Ochiauri, Ms. in O., p. 76).

Al. Ochiauri's story ends there, but many other examples exist in verse which can tell us what happened then. We will quote one of these almost in its entirety, for it reveals particularly interesting information concerning the relationship between the Khevsur, spoiling for a fight, and the kadag inspired by the khat'i:
     Seven days and seven nights they prepared for war, awaiting the purity of the heavens.
     In Datvisi they gathered, bedecking mountains and valleys.
     Through Datvisi marched the black-cloaked host,
     The host with white rifles took up its positions in the gorges of Guro.
     Once we are past Shat'il, down there will be Mitkho.
     In Shat'il, among three kadag the khat'i came to Punchia:
     — "Go back, my slaves. Victory is not yours. My bloody sword has been taken from me for three years.
     — Punchia is afraid. His prophecy comes from his fear, not from the Cross.
     — If I am afraid, then let me be the first to advance; and if you return in peace, then let my head be cut off.
     — We will cut off the hem of your tunic if we return victorious.
     The Khevsur did not falter, our clan never retreats.
They attack the Chechens, and have the upper hand at first, destroying all in their way. 'Accompanied by spiralling vultures, they cut paths where none had been.' But in the end Chechens pour into the battle from all around and massacre half of the Khevsur host.[*]
The event is historic, apparently, and is said to have taken place exactly as in the story at the end of the nineteenth century. As recently as 1920, another tit-for-tat raid was organized against the people of Arkhot'i: in the autumn, taking advantage of the fact that all the men were away selling livestock in the lowlands (across the entire Caucasus, the narrator says), the Chechens attacked the shepherds of Arkhot'i, killing four of them and stole some animals. (To surprise the shepherds, their vanguard had pretended to be a government delegation! This raid was a punitive expedition: the previous summer, the shepherds of Arkhot'i had killed a Chechen shepherd and had stolen a sheep.) The Khevsur made ready to attack the Chechens the minute they returned to their villages from the lowlands; but as they were preparing, gathering and consulting their different kadag, the Georgian government was informed of the matter and had to intervene quite energetically to settle the quarrel.[*]
When the Pshav and the Khevsur united their forces against a common enemy, the warriors, priests and oracles of both tribes gathered at the Pshav shrine of Lashari:
'In times of war, the Pshav and the Khevsur gathered at [the shrine of the khati] Karat'i, in Gudani—or rather Lashari. They consult each other, they discuss various matters, and they seek to find out the opinion of the khat'i. When the Russians tried to enter Georgia, for example, the Pshav and the Khevsur rose up in arms; they gathered at Lashari. Then we set off to attack the Russian army. On the way to battle, the Pshav kadag (of Lashari) prophesied: "Do not go," he proclaimed, "the Russians' victory is assured." The Pshav obeyed and returned to their villages, but the Khevsur angrily thought: "Those Pshav must be afraid of our enemies!" We were beaten by the Russian army, and left many dead' (O., p. 77).

Popular poetry contains many references to warlike expeditions between various Khevsur clans, whose purpose was generally to "raise tribute". The inhabitants of the three shrines of Gudani, Batsaligo and Khakhmat'i, for example, decided to levy a tax upon the shrine of Arkhot'i:
     Gaidauri prophesies: the Cross of Gudani commands:
     'Assemble at my door, slaves of three shrines.
     May all three brandish the standard, may all three ring the bells.
     The Khevsur are blessed, their victory is assured.
     May they tax the people of Arkhot'i, may they take one sheep for every home.'
     The Cross of Arkhot'i heard [the Cross of Gudani's] words, and reached for his sword:
     'It is I who collects tribute [around here]. Where do you think you are going, you rotten swine?'[*]

The insult which the Cross of Arkhot'i profers to its enemies—"rotten"—seems to apply particularly to the shrine of Batsaligo, which the Pshav and Khevsur often refer to as the "rotten" khat'i. Such wars between shrines were commonplace before 1914 and were generally triggered by overly-long vendettas, which turned clans into mortal enemies even when the initial "debt" of blood was settled. We will see later on that the kadag also plays a role in the peaceful resolution of vendettas.

          4.d. The kadag as the highest legal authority

But the higlanders do not only consult their khat'i in matters of war: any entreprise or reform concerning the shrine and its rituals is only admissible following a special shamanic séance:

'The clan would only consult the kadag if it wished to discuss a matter concerning the entire community, if the people wanted to do something together. Old laws, for example, sometimes need to be reviewed and brought up to date: the work of the shrine, high days and holy days; how ceremonies used to be held, how and when one performed certain acts or rituals (in the morning, at mid-day or in the evening?); what order the rituals should follow, things like that. If one forsakes the old order, what is there left? People no longer know how things were done, and it is to settle matters such as these that one consults the kadag' (O., p. 72).

This is one of the most important functions of the kadag: he is the guardian of liturgical science, of the correct way of holding rituals, a doctor of law. He does not take part in the normal carrying out of rituals—he essentially has no role to play in the religious cult itself—but he is tasked with maintaining or re-establishing the ancient ways and ritual traditions. In this he distinguishes himself from the priests; just as he is the source of the priest's powers without personally exercising them, he is also the source of the correct manner in which the religious cult should be observed without taking part in the latter. This central but passive role of his is illustrated by the myths which we will study later on. The shrine of Karat'i, for example, was founded by a couple of mythical shepherds. The woman (who by now is alone in the story) established the rules, rituals and religious solemnities; she was a kadag. But when her founding role was complete, the inhabitants—who had until then held her in high honour—chased her and stoned her to death. Unable to celebrate the rituals which she had created, responsibility for this task was given to the ancestor of the clan of Likok'i, the shrine's first priest. We see that the role of the kadag is understood in the same way in both the myth and in the day-to-day observance of religious life: he only intervenes in the former to restore the rules to their origins, and he creates the latter before standing aside. This regulatory nature is characteristic of most of the kadag's functions (for the myth, see sect. 5, ch. 6).

His role as a restorer also expresses itself in the physical world: when the question of building new structures for the shrine is raised, the final decision rests with the kadag. Moreover, when he feels that a sacred building is too run-down or badly sited, he demolishes it himself and has it rebuilt somewhere else. A case such as this took place in Shat'ili in 1930: the kadag declared that the shrine's kiosk was not standing where it should; he demolished it and supervised its reconstruction in a different place (O., p. 73).

Essentially, any religious or physical modification of the shrine is permitted or initiated by the kadag. Among the Khevsur, as we shall see in a later section of this book, there are several ways in which land can be appropriated for the shrine; tradition, however, has it that all the sacred lands were initially acquired in the following way: a kadag fell into a trance and, inspired by the khat'i, seized the sacred banner and ran towards the mountain, marking out the boundaries of a space which from that moment onwards would become the shrine's domain (O., p. 73). At least that is what the legend says. Whether it springs from a distant memory of bygone rituals or whether it is acted out on the spiritual plane, the behaviour this legend describes still exists today. The district of Roshk'a, for instance, comprises five villages united in a clan-based saq'mo (a social and religious community), all the inhabitants belonging to the same clan. The shrine of Didgori in the village of Roshk'a itself is the religious expression of this social unity.[*] One day, the kadag fell into a trance and—prey to divine madness—ran without stopping from the shrine to one of the five villages, Paparena; having arrived, and still in a state of ecstasy, he drew the boundaries of the plots of land belonging to the khat'i (O., p. 73). By acting in this way, he brought no innovation, the limits of the sacred plots of land being well-known, but renewed a founding event and strengthened the relationship between the khat'i and the space which belonged to him. This mythical form of behaviour corresponds perfectly to the kadag's essential role, which is to go back to the source, to bring rules closer to their original state, to remind and reinforce the relationship between religious observance and its creation.

If one transposes this tendency to the economic plane, one notices that the kadag's influence favoured the preservation or the development of the shrine's possessions to the detriment of private property. V. Bardavelidze follows this direction when she says that:

'When the shrine's economic interests and territorial claims clashed with or were in opposition to those of profane landowners, clans or families, — the kadag, whose prophesy was a powerful weapon in the fight against private landowners, was one of those who defended the shrine's interests the most effectively. According to the oracle, the offerings with which the divine master was most pleased were individual plots of land and draught animals for ploughing; entire estates belonging to a family line or to an individual couple thus became the property of the shrine for various reasons, which the priests and particularly the oracles gave on behalf of the clannish divinity.'[*]

Basing himself upon the kadag's economic role, the same author compares the two forms of kadagoba, at home and at the shrine, and concludes that the latter is older than the former:

'The economic foundation of kadagoba in the home is provided by the farm, an important part of whose produce was presented to the shrine thanks to the intervention of the kadag and benefited those who made up the shrine's administration. Kadagoba at the shrine, however, whose main goal was to regulate the religious economy (naming the dast'ur, increasing sacred land or its confirmation as such, &c.), must have developed along with the archaic, clannish form of society based upon collective agricultural production.'[*]
We will later return to these economic problems; for the time being, let us remember that the kadag's activities are linked in practice to all forms of social organization.

The kadag also has an important role to play in terms of the law and justice. Alongside war and religion, customary rights are a major part of the daily life of the Georgian highlanders. The Khevsur are notorious for their partiality to long, drawn-out legal wranglings and their almost immoderate inclination towards chicanery. It could even be said that every important problem the group comes to face expresses itself along three main lines: religious, martial and legal. When two clans are in disagreement over something, be it a banal incident or a blood-feud, the Khevsur envisage three possible solutions (given here in order of preference): to settle the matter by fighting, to resolve it through the offering up of religious sacrifices, or to resort to "the law". Logically, it seems unexpected and even contradictory that the kadag may intervene in matters of justice, for both he and "the law" defer to an apparently infallible basis for their authority—"the law" to a body of laws and rules, and the kadag to divine will. Why bother relying upon the law to solve a difficult question when it would be easier to simply consult the oracle? The Khevsur are aware of this theoretical incompatibility, which is why when one asks them whether or not the kadag involves himself in matters of justice, they answer that such matters are none of his business:

'Why should the kadag become involved in matters of justice? The council [of elders] is authority enough—I have seen this happen more than a dozen times. Why involve the Cross in legal matters? We never do. If someone told you the contrary, they are lying' (60 years old, 1945, O., p. 81).

The Khevsur also feel that a profound antinomy exists between the law and kadagoba, between the discursive, measured and predictable nature of the former and the affective, violent and uncontrollable aspect of the latter. This antimony is, in a sense, one of the kinds of opposition between legislator and magician which Dumézil underlined [*]. Opposing the kadag to the "men of the council", a Khevsur uses the term mts'ereltashvil (Khevsur dialect), which literally translates as 'son of a learnèd man, meaning: the lost man [l'égaré], the violent man, the immoderate man [le démesuré], the psychopath' (O. 81, no. 3):

'We do not consult the kadag on legal matters. If the men of the khat'i advise us, we do not listen to them. The kadag are mts'ereltashvil, and one needs wise men [instead] to advise upon legal matters' (70 years old, 1945, O. 81).

          4.e. Duelling: the kadag and vendetta

Yet the kadag does intervene in the legal domain; first of all in specific circumstances such as vendetta and "duels", called parik'aoba [G. ფარიკაობა] or tch'ra-tch'riloba [G. ჭრა-ჭრილობა] [*]. In a way, tch'ra-tch'riloba represents the legal form of vendetta if the protagonists belong to different clans. Tch'ra-tch'riloba means 'a cut, a wound': the two opponents kneel facing each other, the sword in the right hand and the shield in the other [see Image ref. 1789-23/3710434 on this page—A.B.]; they cannot move away from each other. They may only strike their opponent's face using the point of their sword [l'épée étant maniée d'estoc]; wounds must be light and not down to the bone. The fight takes place within the limits of the shrine; it is ordered to be held by the men of the council, the judges, either to settle a debt of blood between two clans, or as an ordeal to decide between two plaintiffs belonging to a same clan. If one of the fighters is seriously wounded, the other fighter or his clan must "buy back" the blood shed. The wound is measured in grains of cereal, each one corresponding to a cow due in compensation. This has led to the practice of some Khevsur doctors, at their patient's request, to deepen a wound down to the bone [*]. If the seriousness of a wound is disputed or the good faith of the "wound surveyors" ["arpenteurs de la plaie"] is cast into doubt, tch'ra-tch'riloba sometimes revives a vendetta instead of ending it. In such cases, the "judges" are obviously incompetent: when they decided that the fight should be held, they had in some sense already relinquished their jurisdiction over the case and entrusted the matter to divine judgement. If even that judgement is invalidated, there is no need to return before the court: one applies instead to the divinity, i.e. one takes the matter to the kadag. This must have happened quite frequently, if one is to believe the persistence, the popularity and the violence of tch'ra-tch'riloba. Vaja Pshavela tells of how he once counted over 50 scars on the face of a single man [*]. G. Eladze, describing this custom in 1949, concludes by recommending government intervention in order to transform these bloody fights into a simple sport [*]. Furthermore, resorting to a trial by ordeal in order to extinguish a vendetta also means that all the peaceful solutions such as "buying back the blood" with cattle have failed or have turned out to be impossible. Consequently, the kadag intervenes in the blood feud (1) if compensation is impossible [si le rachat par prestations est impraticable] or (2) if the result of the ordeal is disputed.

Kovalevskij noted another case of shamanic intervention which continues to exist to this day:
'In only one case was an agreement other than "buying back the blood" impossible: when the kadag publicly proclaimed that an ill person belonging to the victim's clan would inevitably die if the murderer were not forgiven. The plaintiffs, if they chose to obey these words [s'ils se conformaient à ces paroles], asked for the murderer to be brought before them and told him—May the people and the khat'i know that we forgive you. We will no longer attack you on your land, we will no longer stop you if we meet you on our way; may your house and your roof be yours only from now on' [*].

In 1946, T. Ochiauri's main Khevsur informant said:
'Bats'q'via, a man from Datvisi, had killed someone from Barisakho. The victim's cousin, Arabuli, met the enemy in a gorge, and shot at him, but missed. The cousin [Note: Charachidzé says 'the enemy', but the story then seems to make no sense—A.B.] returned among his clan, but fell gravely ill. In order to cure him, the kadag was asked to shamanize, and declared: "death stalks him. If you do not bring this man (Bats'q'via, the murderer), nothing will save the patient. Go and fetch Bats'q'via. Make a sacrifice, and let the murderer be forgiven." It is not easy to kill a man whom the khat'i protects [un homme qui est sous la main du khat'i]; they therefore had someone fetch Bats'q'via, swore their oath, gave up victims in sacrifice...' (O. 82).

Now we can better understand how the intervention of the kadag in a vendetta not only undermines the incompatibility underlined above. The kadag only intervenes when the course of events escapes human justice and becomes part of the divine: 'It is not easy to kill a man whom the khat'i protects'. Evidently, the reconciliation of enemies takes on a new significance: instead of marking the end of the blood feud, it becomes a condition of healing. The reconciliation loses its normal links with history and with the course of the quarrel and becomes part of the illness-possession-healing complex. The case is now no longer a matter of relationships between society and its laws, but instead of the relationship between the divinity and the people it protects: that is to say, it is no longer the responsibility of human judges but instead that of the divine intermediary, the kadag. His role in the duel can be explained in a similar way, as we have seen: there too, therefore, there is no overlap of responsibilities.

Until now, the circumstances were specific ones, but we will now see that the role of the kadag in legal matters goes further. Al. Ochiauri notes:

'When certain legal questions were not correctly resolved, the shrine's kadag (more often than not, the kadag of the shrine of Gudani) was asked to shamanize in order to settle the matter. What the kadag said was final, and all respected and obeyed this law.

'If a problem could not be solved according to Khevsur law and if legislation needed to be changed, one then asked the khat'i to do so through his kadag. In any case, when the kadag shamanized at Gudani on New Year's Day, the judges were in attendance and, if necessary and based upon the oracle's declarations, worked to modify and enrich legislation.' (Al. Ochiauri, manuscript in T. Ochiauri, op. cit., p. 83.)

One of the functions of the kadag is therefore to carry out legislative reform. One could say that he is the custodian of laws—be they the Khevsur code or religious rules. His actions in the legal domain are similar in form to those he adopts in the religious domain: on the one hand, to create customary laws, and, on the other, to make sure that they are "correctly" applied, that their origin is not distorted. Considering kadagoba as a whole, we see that this is the essential course of his influence: the source of power, he does not wield it, but instead contributes to maintaining it, to renovating it and to enforcing it. His role, as both founder and regulator, extends to all the domains of social and religious life, preserving the same characteristics in each one of them: he declares war, but does not lead it; he acquires land, but does not work it; he orders the shrine's buildings to be moved, but does not officiate therein; lastly he guarantees justice, but does not dispense it. Moreover, his calling as a creator is constantly reaffirmed through an essential aspect of his conduct: repetition. One could say without exaggeration that, in all the spheres of his activity, his behaviour consists in repeating an initial event, as it was accomplished "back then, a long time ago", at the beginning, at the very origin of mankind's social and religious life. His task is to relive the creation and the foundation in all authenticity. Therein lies his power, and he owes this ability to the most remarkable aspect of shamanism: the kadag relives that which was the sign of a divine call, the initial possession; and in doing so he relives the experience of the original revelation which the possibility of a relationship between mankind and the gods implies.

         From Le système religieux de la Géorgie païenne—Analyse structurale d'une civilisation
         Paris: Francois Maspero, 1968 (reprinted by La Découverte, 2001), p. 113 onwards, my translation.