Cet ouvrage fait partie de la bibliothèque YouScribe
Obtenez un accès à la bibliothèque pour le lire en ligne
En savoir plus

Which Triad ? A Critique and Development of Dumézil's Tripartite Structure - article ; n°1 ; vol.221, pg 5-21

De
18 pages
Revue de l'histoire des religions - Année 2004 - Volume 221 - Numéro 1 - Pages 5-21
The author argues that there are two distinct triads in Indo-European ideology. One is a simple set of three terms while the other is a complex set that consists of three different relationships. Since Dumézil was unaware of the second of these, he interpreted all the meaningful triads that he encountered as belonging to the simple series and this led him to misidentifications of the places of the gods in the overall structure. If the present argument is accepted, it will lead to the reinforcement of Dumèzil's fundamental premise concerning the existence and key importance of a tripartite structure while developing a new way of perceiving the gods in the pantheon.
Quelle triade ? Critique et développement de la structure tripartie de Dumézil.
L'auteur argumente qu'il existe deux triades distinctes dans l'idéologie indo-européenne. L'une est un ensemble simple composé de trois termes, tandis que l'autre est un ensemble complexe composé de trois relations différentes. Dumézil n'étant pas conscient de la deuxième triade, il a interprété toutes les triades significatives qu'il a trouvées comme appartenant à la série simple, et cela l'a mené à une identification erronée de la place des dieux dans la structure générale. Si la présente argumentation est acceptée, elle conduira à un renforcement de la prémisse fondamentale de Dumézil concernant l'existence d'une structure tripartie et son importance clé, tout en développant une nouvelle manière de percevoir les dieux dans le Panthéon.
17 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.
Voir plus Voir moins

Emily Lyle
Which Triad ? A Critique and Development of Dumézil's
Tripartite Structure
In: Revue de l'histoire des religions, tome 221 n°1, 2004. pp. 5-21.
Abstract
The author argues that there are two distinct triads in Indo-European ideology. One is a simple set of three terms while the other
is a complex set that consists of three different relationships. Since Dumézil was unaware of the second of these, he interpreted
all the meaningful triads that he encountered as belonging to the simple series and this led him to misidentifications of the places
of the gods in the overall structure. If the present argument is accepted, it will lead to the reinforcement of Dumèzil's fundamental
premise concerning the existence and key importance of a tripartite structure while developing a new way of perceiving the gods
in the pantheon.
Résumé
Quelle triade ? Critique et développement de la structure tripartie de Dumézil.
L'auteur argumente qu'il existe deux triades distinctes dans l'idéologie indo-européenne. L'une est un ensemble simple composé
de trois termes, tandis que l'autre est un ensemble complexe composé de trois relations différentes. Dumézil n'étant pas
conscient de la deuxième triade, il a interprété toutes les triades significatives qu'il a trouvées comme appartenant à la série
simple, et cela l'a mené à une identification erronée de la place des dieux dans la structure générale. Si la présente
argumentation est acceptée, elle conduira à un renforcement de la prémisse fondamentale de Dumézil concernant l'existence
d'une structure tripartie et son importance clé, tout en développant une nouvelle manière de percevoir les dieux dans le
Panthéon.
Citer ce document / Cite this document :
Lyle Emily. Which Triad ? A Critique and Development of Dumézil's Tripartite Structure. In: Revue de l'histoire des religions,
tome 221 n°1, 2004. pp. 5-21.
doi : 10.3406/rhr.2004.898
http://www.persee.fr/web/revues/home/prescript/article/rhr_0035-1423_2004_num_221_1_898EMILY LYLE
University of Edinburgh
Which Triad ?
A Critique and Development
of DuméziPs Tripartite Structure
The author argues that there are two distinct triads in Indo-
European ideology. One is a simple set of three terms while the other
is a complex set that consists of three different relationships. Since
Dumézil was unaware of the second of these, he interpreted all the
meaningful triads that he encountered as belonging to the simple series
and this led him to misidentifications of the places of the gods in the
overall structure. If the present argument is accepted, it will lead to
the reinforcement of Dumèzil's fundamental premise concerning the
existence and key importance of a tripartite structure while developing
a new way of perceiving the gods in the pantheon.
Quelle triade ? Critique et développement
de la structure tripartie de Dumézil.
L'auteur argumente qu'il existe deux triades distinctes dans
l'idéologie indo-européenne. L'une est un ensemble simple composé de
trois termes, tandis que l'autre est un complexe
de trois relations différentes. Dumézil n'étant pas conscient de la
deuxième triade, il a interprété toutes les triades significatives qu'il a
trouvées comme appartenant à la série simple, et cela l'a mené à une
identification erronée de la place des dieux dans la structure générale.
Si la présente argumentation est acceptée, elle conduira à un
renforcement de la prémisse fondamentale de Dumézil concernant
l'existence d'une structure tripartie et son importance clé, tout en
développant une nouvelle manière de percevoir les dieux dans le
Panthéon.
Revue de l'histoire des religions, 221 - 1/2004, p. 5 à 21 •
.
:
;
6 EMILY LYLE
The theory put forward here is the culmination of work done
over the last quarter of a century and, although I: have published,
many of my ideas in a book and in journal articles, the final»
result оГ bringing . together, the various strands of my thinking
about a structural model that- would fit the Indo-European ev
idence may strike the reader as a radical revision and it is rather
surprising even to me: , The • basic : design that : has emerged is a
coherent and elegant one that rests on triadic organisation; as
Dumézil postulated, but it includes two triads and not just one.
When T was searching through the Indo-European materials
for indications of structure, as Dumézil had done, I! agreed that,
the evidence supported his basic * three- function finding, but» Г
also came upon a second, triad, that I felt hadi structural, signif
icance. - This was . a triad . of- polarities expressed by the ancient
Greeks as , hot/cold, dry/wet and < dispersing/attracting which; I
named the А; В and С axes and argued could be shown to be a
cosmogonie series. I did not at that time see a way in which the
two triads were related and discussed them both in isolation and
it is only now that they are being presented as tightly connected
sequences.
The two come together most clearly when we think in terms
of time. I had always been interested in the temporal dimension
and at an early stage had discussed the three seasons ofi spring,-,
summer and winter as one register of the functional set, and had:,
sketched ■■. in connections with the human life , cycle. . It was only <
later that I explored the phases of life in . terms of age :
grading and saw, as > Kim McCone had done, that a triad of male:
age grades was : a more probable base for. the functional • series
than the . social-class divisions once suggested; by Dumézil. In
going ; into - detail about . how such ■ an age-grade . system : would
have operated, . I \ found that the ■ A, . В \ and С polarities were
connected respectively with the relationships between paired age
classes, generations (consisting of two classes) and half-classes, in
that order. Since they were also connected with the three transi
tions of marriage, initiation and death-and-birth, this made
it possible to i look with fresh -, eyes ; at the focal points . of the :
ritual * year, which . I had -. already seen as • correlated with these
transitions. TRIAD? 7 WHICH
After this brief introduction on the development of my ideas,
I will now turn to fuller study of the two triads in relation to
Dumézil's work and consider the implications of my proposed
new approach for an enriched understanding of the structure of
the pantheon.
THE SIMPLE TRIAD
The simple triad can be identified by the colours white, red
and black and relates, as Dumézil argued, to his functions of the
sacred, physical force and fertility and prosperity. Dumézil ini
tially found these concepts embodied in society and connected
with priests, warriors and cultivators, as in the Indian varna and
the Iranian pištra that were the subject of fruitful studies by Ben-
veniste and himself in the 1930s.1 Later, Dumézil was more incl
ined to speak in terms of the ideology rather than its embodiment
in social groups.2 His position when he published Mythe et
épopée I in 1968 was stated in this way in the course of an
account of his work after 1938.
Je ne puis ici résumer le travail des trente ans qui ont suivi. Je dirai
seulement qu'un progrès décisif fut accompli le jour où je reconnus,
vers 1950, que Г "idéologie tripartie" ne s'accompagne pas forcément,
dans la vie d'une société, de la division tripartie réelle de cette société,
selon le modèle indien ; qu'elle peut au contraire, là où on la constate,
n'être (ne plus être, peut-être n'avoir jamais été) qu'un idéal et, en même
temps, un moyen d'analyser, d'interpréter les forces qui assurent le cours
du monde et la vie des hommes.3
Dumézil contemplates the possibility that there never was a
real social division into three parts and it has been recognised
that the tripartite social division which he did consider — into the
1. G. Dumézil, "La préhistoire indo-iranienne des castes", Journal asia
tique, 216, 1930, 109-130. É. Benveniste, "Les classes sociales dans la tradi
tion avestique". Journal asiatique, 221, 1932, 117-134.
2. D. Dubuisson, "Contributions à une epistemologie dumézilienne :
Tidéologie", Revue de l'histoire des religions, 208, 1991, 123-140.
3. G. Dumézil, Mythe et épopée, I, L'idéologie des trois fonctions dans
les épopées des peuples indo-européens, Gallimard, 1968, p. 15. •
.
8' EMILY LYLE
classes of priests, warriors and cultivators — could not have exis
ted •'■ in a primal form ; of society. Accordingly, whether the triad
was simply an • ideological mental construct or it was
based ona real-life . set of v social classes, the schema does \ not
claim. very ancient roots in terms of social organisation. Neither,
of Dumézil's formulations leads to the positing of an actual par
titioned society in remote prehistory operating in ■' terms of the
functional triad.
The suggestion fairly recently made by McCone and myself
that the: triad; in terms of divisions of society, relates primarily
to: male age grades (old; men, young; men, mature men) rather
than social classes leads to a different understanding of the situa
tion. Although McCone was able to point to a number of;traces
of age grading throughout the Indo-European; area,4 we do not
have a widespread fully; operative age-grade system within ,it in
historical, times, and consideration of a triad of age grades inevi
tably directs us back, from the historical record into prehistory.
Accordingly, we : would ; have to consider at least i two stages.
1 / One is a prehistoric stage for, which, by definition,, no
documentary evidence is available. This is the posited stage when,
age grading was really operative in society and was an important
component in the creation of the tripartite conceptual structure,
although there is no need, to give it an exclusive role.
2 / The second • is the historical stage when ■ there are only
remnants of an age-grade system remaining but in which the tr
ipartite conceptual, structure that, was partially founded on it lin
gers on.
A question that arises about stage 1 is how we can arrive at
a valid- view of the posited society when it is a prehistoric one..
So far. as I am aware, age-grading does not leave evidence of a
kind that can be found -in the archaeological record and so we
have to turn for support to the evidence of age-grading in -
societies in recent times. Drawing on studies of the East African
4. K. McCone,- "Hund, Wolf >шкЬ Kriegerbei; den Indogermanen",
in Studim zum indogermanischen . Wortschatz, ed. W. Meid, p: 101-154,;
Innsbruck. 1987. See also M: Pettersson, Cults of Apollo. at Sparta: The
Hyakinthia.the Gymnopaidiai and the Karneia, Stockholm, 1992, p. 78-90." ■
WHICH TRIAD? 9
pastoralists, , I have shown how connections cam be made bet
ween the. Indo-European, indications, we have and the age-grade
systems operated in s these : societies.5 A\ major difference is . that '
the groups I. mention — the Maasai, . the Jie and- the Karimo-
jong — do not have- kings whereas sovereignty seems to be
deeply embedded; in the Indo-European , conceptual : structure. , I
had- thought < that . known: age-grade societies - were all acephal
ous, as Bernardi stated,6 but a recent book called ; Kings of-
Disaster treats a Sudanese society which has both kings and age
grading7 and' this instance removes what might have been regar
ded as a- difficulty in- seeing: the society, where the Indo-
European conceptual structures originated as having both three
separate : elements • consisting of age-grade divisions and also an .
overarching sovereignty.
THE COMPLEX TRIAD *
The suggestion that t the simple triad -. was one ■ of age grades
opens up a whole new range of possibilities. The important point,
for, the approach taken in this article, is that' it gives a sequence
in ï terms of time and enables us ; to relate • the - functional series
with greater immediacy to both human. life and the year.. The
simple triad • is > that ; of < the three : adult « life , periods of the young
men (= 2 physical force),,the mature men (= 3 fertility anďpros-
perity) and-.the old men (= L the: sacred). Similarly, in < the year,
where the ; sequence • follows the hierarchical order, the simple
triad is that of «the three seasons of spring (= 1 the sacred), sum
mer (= 2 physical force) and winter (= 3 fertility and prosperity).
5. E. Lyle, "Age grades, age classes and alternate succession: A restat
ement of the basis at: the societal level of Indo-European symbolic parti
tion", Emania, 16, 1997, p. 63-71; translated by A. Le Borgne, "Grades
d'âge, classes d'âge et succession alternée : nouvelles vues sur, l'origine des
partitions symboliques indo-européennes au niveau societal", Ollodagos, 16,
2001, 111-145.
6. E. Lyle, Archaic Cosmos: Polarity, Space and -Time. Edinburgh,
1990, p. 131-133, and; B. Bernardi, Age Class Systems. Cambridge,. 1985,
p.-. xiv,. 153, 157.
7. S. Simonse, Kings of Disaster: Dualism,'- Centralism and the Scape
goat King in Southeastern Sudan, Leiden, 1992. ,
,
:
10 EMILY LYLE
Both < these series are punctuated • by ritually ; marked transitions
and it is argued here that the complex triad concerns these transi
tion points, which are death and birth, initiation, and marriage,
and their representations in the year. Although it seems likely that
each ! transition has a. special connection with the member of the
simple triad that follows, it is essential to distinguish the transi
tions as a separate series since they mark relationships and cannot
be defined as single entities.
The year is organised < by means of- special * ritual > points; the -,
seasons of the year belong to ordinary time and it is the transi
tions between them that are sacred. Bernard Sergent has recently
argued that, the evidence of a threefold division of > the year that
survived in Iceland and India can be taken to indicate the proto-
Indo-European structuring, and in both his illustrations it is the
three ritual points that are mentioned.8 In the case of Iceland,
there were three sacrifices, one about the time of midwinter "for
growth",", one in spring "for victory" and one in autumn for. "a
good ' harvest/year".9 It is possible • to explore these three ■ annual
sacrifices in connection • with • the three life-cycle transitions and
the . match us a particularly good i one in : the : case : of the death ■•
period which i occurred at Yule.10 Ins the case: of. Vedic India,
where the festivals took- place at • intervals of four months, one
festival , includes a . rite "which ; follows the rules for offerings to
the * dead", the second is not : immediately informative for^ the
present purpose, but the third involves the sacrificer and his wife
in creating representations of a ram andewe "with.the marks of
sex made prominent".11 The first festival can readily be related to
death; and; the; thirds to marriage. Although the: total \ picture:
concerning annual festivals ; within . the Indo-European - area is
rather confused, one point that; comes out clearly, both* in
8. B. Sergent, . Les Indo- Européens : histoire, langues, mythes, Payot,
1995, p. 360-361.
9. Snorre Sturluson, tr. Erling Monsen and A. H. Smith, Heimskrin-
gla, Cambridge, 1932, p. 6.
10. E. Lyle, "Interpreting Yule and other transitions in the year", Cos
mos, 16 (forthcoming)..
11. A. B. Keith, The Religion and Philosophy of the Veda and Upanis-
hads, Cambridge, MA, 1925, p.- 321-323; WHICH TRIAD? 11
Sergenťs discussion and in one of. Dumézil's early books,1.2 is the*
striking ■ importance ' of a carnival ■ period at the ; changement du •■
temps which; is found at either midwinter (Christmas) or spring
(Easter). This is the period connected with the rites of the dead;
and the strong evidence for the existence of this dominant festi
val can be taken as a good indication that we are on the right
track in; relating- the transitions * in the year to corresponding
points. in. the life cycle.
As 1^ have noted, V do. not! find that the transitions, are
simple entities. . Each transition • appears to connect with one . of
the relationships between two terms that I have called the. three
axes of polarity.. I have; identified them by the: alphabetic
sequence А, В and С since, as I -understand it, they are built up*
in a specific order in, the cosmogony.13 The С axis is that of the
living versus the dead and the interface between them is marked
at the death-and-birth point in the cycle. The В axis, marks the:
opposition t between • the . qualities that distinguish the . first • and
second 1 functions — magical ! power versus . physical power or, . as
McCone ' puts it, the mental \ versus: the. martial14 — and- 1 argue -
that contest between them is connected with the initiation point
in the cycle. The A axis distinguishes. the above from the below
and ' the . male from the female and is * appropriate to the marr
iage point.
THE SIMPLE AND; COMPLEX TRIADS- IN COMBINATION
Figure 1 shows the members of the simple (white, red, black)
triad interspersed by the of the complex (A, B,ř C axes)
triad. The table includes boys in- square brackets.. Uninitiated •;
boys are nonentities so far as the age-grade system is concerned'
12. G. Dumézil, Le problème des Centaures : étude de mythologie com
parée indo-européenne, Paris, .1929.
13. E. Lyle, "Modeling Feature and Mark in Old- World Cosmology",
Semiotica, 106, 1995, 171-185. For ease of cross-reference to the hierarchi
cally ordered functional series, in the present article I list the B-axis compon
ents in -/+ order. while the A- and C-axis components are listed in the:
usual +/- order.
14. K. McCone, Pagan Past and Christian Present in Early Irish Literat
ure, Maynooth, .1990, 125-128. 12 EMILY LYLE
summer- winter spring
old men [boys] men mature men young
3' 2 physical force' 1 the sacred fertility
white red! black
\ \ \
transition death living/dead С and axis festival birth • transition mental/martial initiation. В axis festival transition male/female marriage A axis festival
Fig. 1
but the overlap of the old men who live beyond the ritually reco
gnised point of. death with the boys accounts for the position of
the old men in the sequence/
THE MITANNI AND UPPSALA TRIADS
When we turn now to expressions of totality that were parti
cularly influential in Dumézil's formulation of ideas about the
pantheon > in •. relation ; to ; the ■ three functions, we can see that it
should not be taken for granted 'that an expression of a totality
in three parts necessarily refers to the simple triad. It- is at least a
possibility that it is the. complex triad of: the relationships be
tween the living and the dead, the mental and the martial, and
the male and the female, that is being expressed! I' shall argue
that two of the keys to Dumézil's thinking about the structure of
the Indo-European • pantheon — the : Mitanni i set • of' gods and the
Uppsala triad— make better sense when seen in relation- to the
complex triad than to the simple one and that, accordingly, any
of Dumézil's ideas about the - pantheon that were based .ona ,
"simple" interpretation of them are .without foundation; .

.
TRIAD?" 13 WHICH
Even as late as the posthumous volume entitled Le roman des
jumeaux et autres essais published in -1994, the importance of the.
Mitanni series is still being stressed: Dumézil says::
La célèbre liste des "dieux indiens" de Mitanni est dans nos études
d'une importance qu'il ne faut pas se lasser de souligner. Elle se pré
sente comme un tout "arya" isolé au milieu de mentions anarya dans ■
un \ document non •■ pas i liturgique mais diplomatique, ce . qui r donne - à
penser. qu'elle s'imposait, en tant que tableau. tout. fait, à qui: voulait;
mettre un traité sous la garantie, à côté d'autres dieux allogènes, des
"principaux dieux" arya.15
Elsewhere in the. volume he speaks of- the Indian set consis
ting of: "Mitra- Varuna, Indra,, les Nâsatya,. complétés- à
l'occasion : par SarasvatîV as the , most ! elaborated \ form ? of the
series and gives the Scandinavian. Uppsala . set, consisting : of
Odin, . Thor and Freyr," as : an> instance of* the * most compact
form.16 Of course, Dumézil does make use of» liturgical land narr
ative material in addition !■ to the • Mitanni ■> treaty - and Adam -, of
Bremen's account . of the . Uppsala s temple but he places conside
rable reliance on ; these brief texts; .
Dumézil: first- gave a trifunctional* interpretation of the
Mitanni treaty in 1941, 17 and deals. with it at length in a chapter
on. "Les dieux Arya de Mitani" in> Naissance, d'archanges (1945)
where: he ; .acknowledges that' Arthur < Christensen had, already-
understood in i the 1920s that the gods formed a set; although, he
did not see it as a triadic one.18 It is useful to return to this earl
ier, interpretation which has something to offer, us : today. Chris
tensen draws attention; to the division of. the gods, into the Asu-
ras and the; Devas and s notes that, "le: pouvoir: asurique est
attribué spécialement à, Mitra et à^ Varuna". He continues:.
Or, si le document de Boghazkiôi [the Mitanni treaty] nous donne
les quatre noms divisés en deux groupes; dont le premier comprend le.
couple -Mitra- Varuna i appartenant au cercle des Asuras, on pourrait
15. G. Dumézil, Le? roman des jumeaux et autres essais; Gallimard,.
1994; p. 81.
16. Ibid:, p. 232.
17. G. Dumézil, Jupiter Mars Quirinus. Essai sur la conception indo
européenne de la société et sur les origines de Rome, . Paris, 1941, p. 61.
18. A. Christensen,. "Quelques notices sur les plus anciennes périodes
du zoroastrisme", Acta Orientalia, 4, 1925-1926, 94-95.

Un pour Un
Permettre à tous d'accéder à la lecture
Pour chaque accès à la bibliothèque, YouScribe donne un accès à une personne dans le besoin