Greek Maenadism Reconsidered

Author(s): Jan N. Bremmer

Source: Zeitschrift fr Papyrologie und Epigraphik, Bd. 55 (1984), pp. 267-286
Published by: Dr. Rudolf Habelt GmbH
Stable URL: http://www.jstor.org/stable/20184041
Accessed: 12-08-2016 22:24 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Dr. Rudolf Habelt GmbH is collaborating with JSTOR to digitize, preserve and extend access to
Zeitschrift fr Papyrologie und Epigraphik

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

267

GREEK MAENADISM RECONSIDERED

One of the most fascinating aspects of Euripides' Bacchae is the description

of the maenads, the female worshippers of Dionysos. These women run over
mountains, attack men, move like birds, are invulnerable to iron, fire and
snakes, tear apart animals, children, and even the Theban king. Recently, this
behavior has been the subject of two detailed studies which adopt widely di
vergent approaches. E.R.Dodds compared certain features of maenadism, such as
the shaking of the head and the carrying of fire, with the behavior of known
religious hysterics, and concluded that maenadism had developed from a kind
of collective hysteria, which was channelled into organized rites in later
1
)
2)
times.
Followin
sidered
to
be
th
the
worshippers
of
Euripides'
d
Dodds'
approac
3)
main
arrows
of
the
Bacchae
elem
projected
on
to
cult
had
becom
4)
tangle.
His
seco
understood
as
a
historical
maen
religious
identi
exhaustion than with an abnormal state of mind."

5)

It is the intention of my paper to advance our knowledge and understanding

of precisely these two points (1) by focussing on criteria which allow us to

distinguish the mythical elements in the Bacchae from those that were really
1) E.R.Dodds, The Greeks and the Irrational (Berkeley/Los Angeles 1951)
270-282. This is a partial reprint of Dodds, "Maenadism in the Bacchae,"

Harvard Theol.Rev. 33 (1940) 155-176.

2) J.E.Harrison, Themis, 19131 (Cambridge 19272) 45-48. Harrison had been
greatly influenced by the sacrificial theories of W.Robertson Smith, see A.
Henrichs, "Loss of Self, Suffering, Violence: The Modern View of Dionysus
from Nietzsche to Girard," HSCP 88 (1984) esp. section 4.

3) A.Henrichs, "Changing Dionysiac Identities," in B.F.Meyer and E.P.

Sanders (eds.), Jewish and Christian Self-Definition III (London 1982) 137
160, 213-236, esp. 143-147, 218-224. The epigraphical evidence for maenadism

has been analysed by Henrichs in his "Greek Maenadism from Olympias to Mes

salina," HSCP 82 (1978) 121-160. These studies have virtually renewed the

study of maenadism.
4) The distinction between mythical and cultic maenads was made, in an

article still worth consulting, by A.Rapp, "Die M?nade im griechischen Cultus,

in der Kunst und Poesie," RhM 27 (1872) 1-22, 562-611.
5) Henrichs, "Changing Dionysiac Identities," 146f.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

268

J.N.Bremmer

part of historical maenadic ritual, and (2) by studying the maenadic ritual
in a systematic way. In addition we will also consider the development of the

maenadic ritual and its function in Greek female life.

I The problem of myth and ritual

We have at least two criteria at our disposal which can help us to de

termine whether a motif belongs to ritual or only to myth. First, we can
compare a motif with similar reports from ancient Greece or other places.
For example, according to Euripides the maenads carried snakes which even
licked their cheeks (102-04, 698, 768). Dodds (The Greeks, 275) has inter

preted the snake as Dionysos himself, but for this identification almost no
evidence exists. Dodds also compared the Dionysiac snakehandling with the
handling of rattlesnakes practised by sects in backward parts of America.
Even though in the latter instance the snakehandling is founded on Christ's
words according to St. Mark (16.17f): "they shall take up serpents; and if
they drink any deadly thing, it shall not hurt them," the resemblance is
rather striking: in both cases the true worshipper is safe in the hands of
71
8)
his
god.
we
do
n
snakes
i

6)
See
m
Psychiat

Serpents

West

Vi

The
Snak
255-262;
(Nashvill
(Cranbur

see
W.M
en d?lire, extases collectives (Paris 1947) 163f mentions a s?ance of the
A?ss?o?a of Algeria in which the ecxstatic members slashed themselves, and
devoured scorpions and vipers. In both cases the god is apparently supposed
to protect the worshipper against the lethal power of the snakes.
7) Power over snakes was long to remain a topos even in the lives of
Christian Saints, cf. A.A.Barb, "Der Heilige und die Schlangen," Mitt.Anthrop.
Ges. Wien 82 (1953) 1-21; R.Wildhaber, "Beda Venerabilis and the Snakes," in
Folklore Today: Festschrift R.M.Dorson (Bloomington 1976) 497-506. For snake
handling in modern European cults, see D.S.Loukatos, Religion populaire ?
C?phalonie (Athens 1950) 151-59, and especially A.M. di Nola, Gli aspetti
magico-religiosi di una cultura subalterna italiana (Turin 1976) 31-178 (on
the cult of S.Domenico from Cocullo in the Abruzzi Mountains). For the Marsi
and other professional snake handlers in antiquity, see Louis Robert, in
Charisterion A.K.Orlandos I (Athens 1965) 343-347; G.Piccaluga, "I Marsi e

gli Hirpi," in P.Xella (ed.), Magia (Rome 1976) 207-231, esp. 207-10; B. de
Gaiffier, Receuil d'hagiographie (Br?ssel 1977) IX,167-172.
8) See the iconographical studies of maenadism: L.Lawler, "The Maenads,"
Mem.Am.Rome 6 (1927); H.Philippart, "Iconographie des Bacchantes d'Euripide,"
Rev.Belg.Phil.Hist. 9 (1930) 5-72; E.Coche de la Fert?, "Les M?nades etc.,"
RA 38 (1951) 12-23; M.Edwards, "Representation of Maenads on Archaic Red
figure Vases," JHS 80 (1960) 78-87; S.McNally, "The Maenad in Early Greek
Art," Arethusa 11 (1978) 101-135.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

269

according to whom Olympias exaggerated the customary rites by introducing

9)

into the thiasos a number of large and tame snakes. It has therefore been

inferred that in picturing the women of Thebes as handling snakes Euripides

has transferred an act which in his time existed only at the mythical level

on to the ritual level. 1 0)

There are, however, two more texts which need to be taken into consid
11 )

tion. Aeschines handled innocent snakes in the cult of Sabazios. The pre
sence of books and an Athenian wedding formula in the Sabazian ritual clearly
demonstrates that the original Phrygian ritual (whatever that may have been)
1 2)
had been considerably changed when transferred to Athens. Moreover, the
mention of ivy, the liknon, and the cry euhoi points to a considerably Dio
nysiac influence on the ritual. This could well mean that the snakes also
derived from the Dionysiac (maenadic?) ritual. Snakes are also mentioned by
the late Hellenistic poet Andromachos (apud Galen XIV.45 K?hn), according to
whom oi xcp Aiovuccp ?axxeOovxec searched for snakes in the spring and rent
them. The masculinity of the worshippers and the mention of spring exclude

the possibility of a maenadic ritual, but the connection of Dionysos with

snakes is clear.

The absence of any mention of snakehandling in the maenadic ritual af

the Bacchae and the observation by Plutarch make it likely that Eurip
indeed took his snakehandling maenads from the mythical (iconographica
tradition, but the iconographical tradition and the passages from Aesc
and Andromachos strongly suggest that snakehandling was practised by m
in more archaic times.

We come to our second example. When the maenads raided the Boeotian vil

lages they "carried fire upon their locks, and it did not burn them". '

1 3)

9) Plut.Alex.2 although the wording of this passage is not especially

clear; for Olympias' attachment to Dionysiac religion, see also Duris FGrH
76 F 52; Athen. 14.659f. D.J.A.Ross, "Olympias and the Serpent, the Inter
pretation of a Baalbek Mosaic and the Date of the Illustrated Pseudo-Cal
listhenes," J.Warburg-Courtauld Instit. 26 (1963, 1-21) 21 wrongly sees in
these snakes a rationalisation of the supposed snake at Alexander's birth.
10) This is the conclusion of Rapp (above, note 4) 13; Dodds, The Greeks,
275.

11) Dem. Cor.26 0. For the identity of the snakes, see now L.Bodson, "L
grecs et leur serpents," Ant.Class. 50 (1981) 57-78.
12) For Sabazios, see more recently Ch.Picard, "Sabazios, dieu Thraco
Phrygien: expansion et aspects nouveaux de son culte," RA 48 (1951) 129-17
S.E.Johnson, "A Sabazius Inscription from Sardis," in J.Neusner (ed.),

Religions in Antiquity: Essays in Memory of Erwin Ramsdell Goodenought (Leid

1968) 542-550; F.Fellmann, "Belege zum Sabazioskult im fr?hkaiserzeitliche

Legionslager von Vindonissa," in Festschrift F.K.D?rner I (Leiden 1978) 2
294; E.N.Lane, "Towards a definition of the iconography of Sabazius," Num
27 (1980) 9-33; R.Kraemer, "'Euoi, Saboi' in Demosthenes De Corona: in whi
honor were the women's rites?," Seminar Papers Soc.Bibl.Lit. 1981,229-236.
13) Eur. Bacc. 757f. Curiously, Dodds, Roux (ad loc), E.C.Koppf in his
recent Teubner edition (1982), and H.Oranje, De Bacchae van Euripides: het

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

270

J.N.Bremmer

Manipulation of fire was not unheard of in antiquity. Egyptian priests actually

lighted a kind of wine on their head which because of their thick hair burnt
1 4)
up too quickly to become dangerous. However, manipulation of alcoholic
drinks seems unknown in Classical and Hellenistic Greece even though it was
known from libations that alcohol could burn (Theophr.Ign. 67). Moreover, in
post-Homeric times Greek women were strictly forbidden to drink wine, and the
transportation of wine into the mountains would certainly have aroused male
1 5)

curiosity. Finally, unlike the priests, maenads did not have to impress
an audience in their secret rites.

In their commentaries Dodds and Roux compared the maenadic firehand

with the reports of ritual fire walking, a rite which occurred (and st

occurs) in many parts of the world, such as Spain, modern Greece (the fam

case of the Anastenaria), Oceania, India and Ceylon, China and Japan.

1 7)

stuk en de toeschouwers (Diss. Amsterdam 1979) 151-154 = idem, The Bacchae

of Euripides. The Play and its Audience (Leiden 1983) Appendix 3 (who argues
that the text is perhaps interpolated) have not noted the imitation by Nonnos
D.43.356f.

14) Hipp.Hef. 4.31, cf. R.Ganschinietz, "Hippolytos' Capitel gegen die

Magier" = Texte u. Unters. 39,2 (Leipzig 1913) 46f.; H.Diels, Abh.Ak. Berl
1913, Phil.-hist.-Kl. no.3, 21 ff.; E.O. v.Lippmann, Beitr?ge zur Geschich
der Naturwissenschaften und der Technik (Berlin 1923) 60ff.; Th.Hopfner,

Griechisch-?gyptischer Offenbarungszauber II (Leipzig 1924) 75. Liv. 39.1

mentions manipulation of fire during the Roman Bacchanalia, for which see
most recently J.North, "Religious Toleration in Republican Rome," Proc.Camb

Phil.Soc. 1979, 85-103; J.-M.Pailler, "La spirale de l'interpr?tation: le

Bacchanales," Annales ESC 37 (1982) 929-952 (with very useful Forschungs

schichte); idem, "Les pots cass?s des Bacchanales," M?l. de L'ec. Franc,
Rome 95 (1983) 7-54; O. de Cazanove, "Lucus Stimulae. Les aiguillons des
Bacchanales," ibidem, 55-113.

15) For women and wine, see F.Graf, "Milch, Honig und Wein," in Perenni
Studi in onore di Angelo Breiich (Rome 1980) 209-221; Henrichs, "Changing
Identities" (above, note 3) 140f.; J.-L.Durand/F.Frontisi-Ducroux," Idoles
figures, images: autour de Dionysos", RA 1982, 83-108; Bremmer, The Earl
Greek Concept of the Soul (Princeton 1983) 109f.
16) R.B.Onians, The Origins of European Thought (Cambridge 19542) 166 n
suggested a connection between the Bacchants and the stories of flames ab
the heads of Servius Tullius and Salvidienus Rufus. However, these flames
were a royal portent, a meaning certainly excluded for the maenadic ritu
cf. Th.K?ves-Zulauf, Reden und Schweigen (M?nchen 1972) 249f.
17) Spain: L.Armstrong, "Fire-Walking at San Pedro Manrique, Spain,"
Folklore 81 (1970) 198-213. Modern Greece: W.Puchner, "Beitr?ge zum thra
schen Feuerlauf" (Anastenaria/Nestinari) etc., Zs.f. Balkanologie 17 (1981
47-75 (with exhaustive bibliographiy); add now W.D.Furley, Studies in the
Use of Fire in Ancient Greek Religion (New York 1981) 211-235. Oceania: A
Lang, "The Fire Walk Ceremony in Tahiti," Folklore 12 (1901) 446-455; E.
Martino, II mondo m?gico (Turin 1948) 29-35 (many references). India and
Ceylon: E.S.Thomas, "The Fire Walk," Proc. Soc. Psych. Research 42 (1934)
292-309; L.Feinberg, "Fire Walking in Ceylon," The Atlantic 203 (1959) Ma
73-76; K.Indra-Kumar, Fire Walking - The Burning Facts (Ceylon 1972); K.W.
Bolle, "Firewalking: A Note on Empirical Evidence," in Ex Orbe Religionum
Studia Geo Widengren oblata II (Leiden 1972) 3-10; H.-J.Klimkeit, "Die
"Teufelst?nze" von S?dindien," Anthropos 71 (1976) 555-578. China and Jap
A.Lang, Magic and Religion (London 1901) 270-294.
Interesting are also the notice with photo in The Times, September 18,
1935 of a test carried out by the University of London Council for Psych

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

271

Similar feats already took place in antiquity where fire walking is reported
18)

of the priests of Perasia in Asia Minor and of the Italic Hirpi Sorani. How

ever, in none of these rites has a single example been reported of people
1 9)
carrying fire on their head.
In the messenger's speech the carrying of fire is combined with the report
of the maenads' invulnerability to the villagers's arms. The combined mastery
of fire and arms is typical of ecstatic cults, as appears from Tibullus* de
scription of the followers of Bellona (1.6.45-48):
haec ubi Bellonae motu est agitata, nee acrem
flammam, non amens verbera torta timet:

ipsa bipenne suos caedit violenta lacertos

sanguineque effuso spargit inulta deam

We may also compare Iamblichus' report of some unspecified ecstatics:
"Many, through divine inspiration, are not burned when fire is introduced
to them, the inspiring influence preventing the fire from touching them. Many,
also, though burned, do not apprehend that they are so, because they do then

not live an animal life. And some, indeed, though transfixed with spits, do
not perceive it; but others that are struck on the shoulders with axes, and
others that have their arms cut with knives, are by no means conscious of
what is done to them." 20)

Outside Greece, in shamanistic seances, for example, we find the same combina
tion. Among the Tungus, the shaman reportedly plays with burning coals and
red-hot iron and stabs himself deeply without shedding any blood. And among
the Kazak-Kirgiz, the shamanoid baqsa is reported to walk over red-hot iron
barefooted, and to put a lighted wick into his mouth; he also slashes his
21 )
face with a razor sharp knife, which leaves no visible marks. We find the
same combination among the Islamic Isawiyya, whose founder, Muhammed ibn *Isa

(AD 1465-1524),"adopted ecstatic practices, whereby the dervishes became

Research with a young Indian, and J.M.Freeman, "Trial by Fire," Natural
History 83 (1974) 55-63 on a walk during which many people were wounded. Note
that in the ordeals by fire, fire is never put on the head: E.Benz,"Ordeal
by Fire," in J.M.Kitagawa and Ch.Long (eds.), Myth and Symbols. Studies in
Honor of Mircea Eliade (Chicago 1969) 241-264.
18) Perasia: Strabo 12.2.7; Iamb. Myst. 3.4 ; L. Robert, in A.Dupont-Sommer and
L.Robert, La d?esse de Hi?rapolis Castabala (Paris 1964) 53-64. Hirpi Sorani:
A.Alf?ldi, Die Struktur des voretruskischen R?merstaates (Heidelberg 1974)
77f, 98f (hardly convincing); G.Piccaluga (above, note 7) 211-231. Note also
Vita Pachomii1 c.8.
19) Oranje, De Bacchae, 252 n.118 ( = The Bacchae, n.402) also stresses
the difference between fire walking and carrying fire on the head.
20) Iamb. Myst. 3.4, tr. T.Taylor, Iamblich us on the Mysteries of the
Egyptians, Chaldeans and Assyrians (London 18952) 122.
21) Tungus: S.M.Shirokogoroff, Psychomental Complex of the Tungus (London

1935) 365. Kazak-Kirgiz: J.Castagne, "Magie et exorcisme chez les Kazak

Kirghizes et autres peuples turcs orientaux," Rev.Et Islamiques 1930,53-151,
esp. 87,90.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

J.N.Bremmer

272

immune to sword and fire."

The occurrence of the idea of immunity to the lethal powers of culture and
nature in maenadic ideology is highly probable. Maenadic rites were performed
in the rough mountains of Greece in the heart of winter. Yet the women will
not have felt any pain or discomfort because of their state of ecstasy (below).
This must also have been the conclusion of Nonnos (D. 14.384f) who noted that
the rocks did not scratch the naked foot of the Bacchant. We cannot absolutely
exclude the possibility of the existence of fire manipulation among the x?xvai
of the Dionysiac orgia - [Aristotle] (Mir. 122) mentions a fire miracle for
Dionysos - but outside the Bacchae fire handling in maenadic ritual is
nowhere reported; neither need we believe in collective invulnerability. The
inference therefore seems most plausible that Euripides does not give us a
factual description of handling fire, but he presents as a reality the idea
of the maenads' insensibility, even invulnerability, to the lethal powers of
fire and iron. Moreover, he does so in a hyperbolic way: even though the
23)
maenads' hair caught fire, it did not burn.
Euripides exaggerated the maenads' insensibility to pain by representing
it as invulnerability. We touch here on a characteristic element of mythical
discourse, as the following examples show. In the pharmak?s ritual the scape
24)
goats were only expelled, but according to the myths, they were killed. In
ritual the Locrian maidens had to spend only one year in Troy, but myth spoke
25)

of a life long stay. We know that in initiatory rites boys were dressed
as girls, but initiatory myths, such as that of Leukippos, could speak of an
26 )
actual transformation into a girl. During the annual Lemnian ritual of the

new fire, wives had to abstain from sexual intercourse with their husbands,
27)
but myth spoke of the wives murdering their husbands. In the ritual the
22) J.Spencer Trimingham, The Sufi Orders in Islam (Oxford 1971) 86.
23) Another example where comparison with non-Greek reports is useful
occurs in the messenger's speech (v. 748) where we are told that the maenads

moved 'like birds lifted by their flight'. Roux (ad loc.) interprets these

words as a description of l?vitation, yet another Dionysiac miracle. However,

in all the reports of l?vitation we find no example of crowds of flying women.
L?vitation is always the activity of the individual, who never travels more
than one yard; l?vitation is therefore impossible in this case, cf. O.Leroy,
L?vitation (London 1928); A.M.Turi, La levitazione (Rome 1977). For classical

reports of l?vitation, see E.R.Dodds, The Ancient Concept of Progress (Oxford

1973) 203-205; add the case of St. Augustine's mother Monica: Acta Sanctorum

Maii Tom.I, p.486.

24) For this detail and similar examples in scapegoat rituals, see Bremmer,
"Scapegoat Rituals in Ancient Greece," HSCP 87 (1983) 299-320.
25) Cf. F.Graf, "Die lokrischen M?dchen," Studi Storico-Religiosi 2 (1978)

61-79.

26) Cf. W.Burkert, Structure and History in Greek Mythology and Ritual

(Berkeley etc. 1979) 29; D.Z.Nikitas, "Zur Leukipposgeschichte," Hellenika 33

(1981) 14-29.
27) Cf. W.Burkert, "Jason, Hypsipyle, and New Fire at Lemnos," CQ 20 (1970)

1-16

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

273

Arrephoroi passed through a subterranean tunnel, but myth spoke of suicide.

This particular difference between myth and ritual has received almost no
29)

attention. As a preliminary contribution - further investigation is neces

sary - we may note the following aspects of the phenomenon. First, myth can

represent symbolic acts as reality: whereas ritual has to dress boys as girls,
myth can actually change them into girls (Leukippos). Thus myth can realize
the intention of ritual at a higher level than ritual itself. Second, myth
can contain the comments of society on ritual: by representing the maenads'
leaving of the home as a tearing apart of their children, myth expresses male

horror at this act (below). Third, myth represents the ideal reality and can
never fail, whereas ritual is fallible: for instance, in the scapegoat myth
people always offer themselves voluntarily, in reality slaves often had to be
30)
bought or other people lured by rewards. Finally, a general observation.
Whereas ritual can often manipulate people over a certain length of time, myth
has to be effective the moment it is communicated. It therefore often needs

exaggeration to drive home the point it wants to make. But whatever aspect

of the difference may be relevant in a particular case, one thing is certain.

This difference cannot be interpreted as a development in time: we may

safely assume that boys never changed into girls, and that wives never an
nually killed their husbands. This insight into the essential difference
between myth and ritual can help us to reach a better understanding of some
other important details of maenadic behavior in the Bacchae.
Modern scholars often describe maenadic behavior as a kind of infectious

disease, a mass hysteria, and ever since Nietzsche a comparison has been made
31 )
with the dancing epidemics of the Middle Ages. However, we know that in
historical times maenadic rites were not celebrated spontaneously by all the
women of a city but only by restricted groups at a fixed period every other
year (below ? 3); in Miletus the thiasoi were even closely regulated by the
32)
city. In Amphissa the women had to ask their husbands' permission to
28) Cf. W.Burkert, "Kekropidensage und Arrephoria," Hermes 94 (1966) 1-25
and Homo necans (Berlin/New York 1972) 169-173= idem, Homo necans. The Anthro

pology of Ancient Greek Sacrificial Ritual and Myth (Berkeley etc. 1983) 150
154; G.S.Dontas, "The True Aglaurion," Hesperia 52 (1983) 48-63; A.Henrichs,
"Die Kekropidensage im P.Here. 243: Von Kallimachos zu Ovid," Cronache Ercol.
13 (1983) 33-43? N.Robertson, HSCP 87(1983) 241-288.

29) M.Delcourt, Oedipe ou la l?gende du conqu?rant, 19441 (Paris 1981) 44

considered the difference typical of aetiological legends which had to ex

aggerate their point to make it clearer. However, the phenomenon certainly

transcends the genre of aetiological legend and is more complicated, as I

argue. Note, however, Burkert, Homo necans, 44 ( = Amer. ed. 34).
30) Bremmer, "Scapegoat Rituals," 307f. We may compare the perceptive
comments on the difference between a bear festival and a real bear hunt by
J.Z.Smith, Imagining Religion (Chicago 1982) 53-65, 143-145.
31) For references, see Henrichs, "Changing D, Identities" (above, note 3)
145f with notes; add Smith, Imagining Religion, 113.
32) Sokolowski, LSAM, 48.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

J.N.Bremmer

274

accompany the maenads who had collapsed in the agora to the edge of the city:
33)

certainly no mass hysteria was present there. Taking into account the
fundamental difference between myth and ritual, we can now see that many de

scriptions by modern scholars are strongly misleading, based as they are on

an extrapolation from the mythical accounts only. Moreover, the medieval

dancing epidemics, which only took place in situations of stress such as

famine and plague, always involved both sexes, and were closely linked with
34)
existing rituals and institutions.
When we differentiate between myth and ritual, we can also make progress
with another much discussed problem. In the parodos of the Bacchae the themes

of 'sparagmos' and omophagy are sounded at an early stage: the chorus

mentions the "blood that flows when a goat is killed, the joyful act of eating
raw meat" (138). The theme is later elaborated upon when the women tear cattle

apart (734-47), and when cannibalism is hinted at. The idea is older than
the play; vases from the sixth century show Pentheus' 'sparagmos1, and on a
vase of the second quarter of the fifth century a maenad is represented with

a dagger and a goat leg. Virtually all scholars have taken these savage
activities at face value, if not for the Classical Age, then at least for
37)

earlier times. Here, too, a comparison with the historically documented

ritual is illuminating.
In a Milesian inscription we read: "Whenever the priestess performs the
rites of sacrifice on behalf of the city, nobody must 'omophagion embalein'
33) Plut. M.249ef. The presence of Phocians in Amphissa dates the incident
to about 353 B.C. (during the Third Sacred War), as P.St?dter, Plutarch's
Historical Methods (Harvard 1965) 79f persuasively argues. Sir Lawrence Alma

Tadema painted 'The Women of Amphissa' in 1887; the painting is now in the

Sterling and Francine Clark Art Institute, Williamstown, Massachusetts. See

R.Borger, Drei Klassizisten, Alma Tadema, Ebers, Vosmaer (Leiden 1978) 14
no. 278 for a recent bibliography.
34) Cf. J.F.C.Hecker, Die grossen Volkskrankheiten des Mittelalters (Berlin
1865) 143-192 (18321); H.Liebscher, Ein kartographischer Beitrag zur Geschichte
der Tanzwut (Diss. Leipzig 1931); E.L.Backman, Religious Dances in the Christ
ian Church and in Popular Medicine (London 1952); G.Rosen, Madness in Society
(New York 1968) 195-225; M.Broekman, "La dansomanie de 1374: h?r?sie ou
maladie?," Revue du Nord 63 (1981) 339-355.
35) Eur. Bacc. 1184 and 1241-4 7, cf. C.Segal, Dionysiac Poetics and Euripi

des' Bacchae (Princeton 1982) 188 n.46. Oranje loc. cit. (above, note 13)

also compares v. 754; according to him, Nonnos D. 45.2 94ff shows that Nonnos
did not understand the kidnapping of the children whereas the Athenian

audience will have inferred that the children will be torn apart.
36) See, besides the literature cited in note 8, F.T. v. Straten, "Archeo
logische bijdrage tot de bestudering van Euripides' Bacchae," Lampas 9 (1976)
51-77; K.Schauenburg, "Das Motiv der Chimairaphonos in der Kunst Unteritali
ens," in Studies in Honour of Arthur Dale Trendall (Sydney 1979) 149-154;
J.-J.Maffre, "Quelques sc?nes mythologiques sur des fragments de coups at
tiques de la fin du style s?v?re," RA 1982,195-222, esp. 203-207 ("Dionysos

MAINOMENOS").

37) For the evidence, see the enumeration by Henrichs, Changing D.

Identities, 218 (above, note 3); add H.S.Versnel, "Pentheus en Dionysos.

Religieuze achtergronden en perspectieven," Lampas 9 (1976, 8-41) 23.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

275

before the priestess has done so on behalf of the city." Although the exact
meaning of the Greek phrase is obscure, it is clear that in Miletus omophagy
consisted of a sacrifice and that, as Henrichs observes, "the Milesian maenads
will have left the scene of their sacrifice, at the worst, with bloodstained
hands and clothes but hardly with raw meat in their teeth and blood dripping
38 )
from their mouths. The inference seems inevitable that once again myth
exaggerated ritual: the maenads who ate raw meat operated only on the level
of myth - reality was much less breathtaking or disturbing.
II The maenadic ritual

Having discussed various ways of disentangling myth from ritual, we

now proceed to a systematic analysis of the maenadic ritual by tracin

successive steps. Until now scholars have concentrated on the individu
elements of the ritual without presenting an integrated picture in wh
the known elements are seen as indispensable parts of a multi-media
formance. We will pay particular attention to the effects the ritual m
have had on the maenads, and we will not hesitate to invoke the help
physiological research for a better understanding of the various ritua

elements. Of course I am well aware of the preliminary character of man

modern findings, but an application of neurophysiological data such a

paper attempts will at least show that new insights in the study of G
39)
ritual may be expected from this direction.
The fact that a group of women left the city must have caused a gr
in the Greek polis, and it is hardly imaginable that this event was no
tized in such a way as to make this peculiar behavior legitimate. I ta
therefore that the maenadic ritual started with some public manifesta

before the women marched into the mountains. On the occasion of this event

the sacrificial act of 'omophagion embalein', which we already mentioned,

will have taken place. In Miletus this act was performed 'on behalf of the

city' ('hyper poleos') and was clearly localized near or in the city, and

certainly not in the mountains.

Why did the maenads handle this raw meat? In Greek ideology eating raw meat
38) Sokolowski, LSAM, 48, cf. Henrichs, "Greek Maenadism," (above, note 3)

151. H.S.Versnel, in Le Sacrifice dans l'antiquit? (Entretiens sur l'anti

quit? classique 27, Geneva 1981) 38f., suggests, not convincingly, that the

sacrifice had no direct ritual predecessor, but was derived directly from
the mythical material.
39) For the physical and neurophysiological effects, see S.S.Walker,
Ceremonial Spirit Possession in Africa and Afro-America (Leiden 1972) 10-25
("The Neurophysiology of Possession"); B.Lex, "The Neurobiology of Ritual
Trance," in E.G. d'Aquila e.a. (eds.), The Spectrum of Ritual (New York 1979)
117-151. Note also the plea for the use of neurophysiological data in the
study of ritual by J.Verrips, "Wie zijn hoofd niet koel houdt gebruikt zijn

hersens" = EUROMED Working Paper No.35 (Amsterdam 1983). For this section of

my paper I have also greatly profited from suggestions by the neurophysio

logists Dr. C.Deinema (Rijksuniversiteit Utrecht) and Prof. Dr. F.H.Lopes da
Silva (University of Amsterdam).

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

276

J.N.Bremmer

was typical of marginal sects, and it was frequently ascribed to neighboring

peoples, that is to say to those who were in opposition to the ordered world
40)
of the polis. Maenadic behavior thus fits in well with other important
female rites like the Thesmophoria, Skira, and Adonia, which all enacted
elements of disruption or inversion, such as sexual abstinence or promiscuity,
41 )
sterility, and women leaving home. In the maenadic myths the tearing apart
of the children dramatized the separation from home; the eating or touching
of raw meat had the same function in the maenadic ritual.

42)

After their dramatic start the maenads, perhaps in three thiasoi, marche
into the mountains. During the march the maenads will have shouted the cry

'eis oros', or 'to the mountain,' a cry which is repeatedly mentioned in the
4^)
Bacchae (116, 165, 986). ' In the Bacchae (704ff, 1051f) the mountain
as a lush place where it is very pleasant to be, and Ca?ame has compa
44)
Bacchic 'paysage' with the flowery nature of Artemisian landscape.
40) Cf. C.Segal, "The Raw and the Cooked in Greek Literature," Cla
69 (1973/74) 289-308; Henrichs, in Le Sacrifice, Entretiens Hardt 27
1981) 220.

41) For female rites as an inversion of the polis' values, see J.Gould, J
(1980) 51. Thesmophoria: K.Dahl, Thesmophoria: en graesk kvindefest
hagen 1976) with most of the texts; Burkert, Griechische Religion de
chaischen und klassischen Epoche (Stuttgart 1977) 365-370 (with prev
bibliography); M.D?tienne, "Violentes 'eugenies'," in M.Detienne/J.-P

(eds.), La cuisine de sacrifice (Paris 1979) 183-214; F.Zeitlin, !?Cult

Models of the Female: Rites of Dionysos and Demeter," Arethusa 15 (1

157; R.C.T.Parker, Miasma: Pollution and Purification in Early Greek
(Oxford 1983) 81-83; E.Simon, Festivals of Attica (Madison 1983) 18Burkert, Homo necans, 161-168 ( = Amer. ed. 143-149); Simon, Festiva

Attica, 22-24. Adonia: D?tienne, Les jardins d'Adonis (Paris 1972) w

vious bibliography; Burkert, Structure and History in Greek Mythology

Ritual (Berkeley etc. 1979) 105-111; S.Ribichini, Adonis, Aspetti "o

di un mito greco (Rome 1981) with the substantial review by O.Loret

Forschungen 13 (1981) 323-325.
42) Eur. Bacc. 680 (which will have influenced [Theoc] 26.2); Proper
17.24; Sokolowski, LSAM, 48 attests a three-fold organization in Mag

which was derived from Thebes. The three Proitids (below, note 79) and

three Minyads (below, note 82) suggest that an organization in 3 th

also occurred outside Thebes.

43) As was already seen by W.Schmid, Geschichte der griechischen L

III.1 (M?nchen 1940) 670 n.1; note also the cry e?c cxorc?Aouc in Nonn
172. Henrichs, "Changing D. Identities," (above, note 3) 156 unnecessa
suggests that in Hellenistic inscriptions the cry might derive from t

Bacchae.

44) C.Ca?ame, Les choeurs de jeunes filles I (Rome 1977) 262f. The simi
larity as regards the landscape is also noted by Ph.Borgeaud, Recherches
sur le dieu Pan (Rome 1979) 122. For Dionysos and the mountains, see Roux
on Bacc. 32-33 to which has to be added Pratinas TGF 4 F 3 and Anth.Pal. 6.
134, a description of a Hellenistic (according to D.Page, Further Greek Ep
grams, Cambridge 1981, 138; J.Labarbe, "Les trois bacchantes d'Anacr?on,"

M?langes Jacques Stiennon (1983) 403-416 maintains the authorship of Anacreon

painting of three Maenads returning ??j ?peoc; Segal, Dionysiac Poetics

(above, note 35) 111-118. Female Dionysiac ?peioyu?oec now apparently appe
in an inscription of Ephesos (I. Ephesos 106 = SEG XXVI. 1237). The Dionys

interpretation of the Erythraean Miuctvxo?axnc (I. Erythrae und Klazomenae I

64; cf. also ZPE 38 (1980) 150= SEG XXX.1327) is rightly rejected by Henric
"Changing D. Identities (above, note 3) 224 n.98.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

277

there is a great difference between the worlds of Dionysos and Artemis, for
the Dionysiac rites took place in winter, but the Artemisian ones in spring. The

Bacchic landscape therefore represents a Utopian vision, not a ritual reality.

Yet this vision in due time could influence the choice of cult locations. In

Magnesia, a plane tree had been the scene of a miraculous epiphany of Dion
It can hardly be fortuitous that in Greece plane trees usually grow in a
45)
shadowy, moist place: the epiphany took place in the god's 'own country'.
It is only in the mountains that Euripides, perhaps in imitation of the
ritual, lets the Theban women change themselves into maenads proper. The
messenger (v. 695ff) relates that in the mountain the women left their ha
down and pulled up their fawnskins; here they will also have removed thei
shoes. Loose hair and barefootedness are typical signs of liminality in Gr
46)
in this case stressing the separation from the ordered world of the polis.
This liminal state also made the women similar to each other in appearance
removing the individualising ornaments of hairbands and shoes: for the go
all worshippers were equal.
The rites were opened perhaps with the sacrifice of cakes, as described
47)
in the strange maenadic poem in the Theocritean corpus. Diodorus Siculu
has recorded that during the rites the women sang, calling for the presen
48)
of the god. Unfortunately, our sources do not tell us whether any other

ritual activities took place.

Charles Segal has perceptively connected the cry 'eis oros' with the

ecstasy-producing or self-hypnotic techniques of the chanted repition of

single word or phrase from shamanistic seances (or, we may add, modern Zen).
This shouting will have helped the women to reach the proper Dionysiac mood,
but the transformation from quiet housewife into raging maenad is a psycholo
gical transformation which needs more than mere shouting. How then was this

brought about? As the English anthropologist Lewis notes, "trance states can
be readily induced in most normal people by a wide range of stimuli, applied
either separately or in combination. Time-honored techniques include the use
of alcoholic spirits, hypnotic suggestion, rapid overbreathing, the inhalation
of smoke vapors, music and dancing, and the ingestion of such drugs as

45) I. Magnesia 215 (a) 5-7, cf. Henrichs, "Greek Maenadism" (above, note 3)
132f. For plane trees, see Frazer on Paus. 4.34.4.

46) Maenadic loose hair: Roux on Eur. Bacc. 150; Henrichs, "Greek Maenadism"

(above, note 3) 157 n.113; add Callistr. Stat.2; Nonnos D.14.345f., 15.76.
Barefootedness: Roux on Bacc. 665; add Nonnos D.14.384f, 19.330, 46.147. In
general: Graf (above, note 25) 67-6 9.
47) Theoc. 26.7 with Gow's commentary.

48) Diod. Sic. 4.3.3 xai xad?Aou x?v rcapouc?av ?uveiv xou Aiovucou. The
singing was apparently antiphonic: Eur. Bacc. 1057; Ennius fr.52 Jocelyn.

These passages were overlooked by Rohde, Psyche, 9 who noted: "Wir h?ren
nichts von Ges?ngen."

49) Segal, Dionysiac Poetics (above, note 35) 112.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

49)

J.N.Bremmer

278

mescaline or lysergic acid and other psychotropic alkaloids." Such tech

niques have been called "driving behaviors." Which ones did the maenads
employ?
In the Bacchic rites a prime 'driving behavior' will have been the music
of the tympanon and the aulos (commonly translated 'flute' but in sound much
more similar to the oboe or clarinet). These instruments also figured in the
orgiastic rites of Cybele and the Corybantes, and Dodds has adduced examples
of their combined occurrence in the late medieval dancing cult of St.Vitus
51 )

and in dancing epidemics in Italy. Why did the maenads use these in

struments and what were their effects? We will start with the tambourine.

Percussion instruments are employed in rituals all over the world to induce
52)

trance-like states; the drum in particular is very popular. Continuous

subjection to the rhythms of the drum, or, as in our case, the tambourine,
has a synchronizing effect on the activity of neuronal cells in some centres
of the brain. The external rhythm becomes the synchronizer of the activity
of the brain; as a result the sound and the action 'possess' and control the

participant. In addition the photic stimulation of torches will have in

tensified the influence of the music. We all too often forget that the dances
took place at night, so that the torches produced visual flicker effects due

to shifts in ocular focus and the movement of dancers between an individual

54)

and the light source.

Continual and violent shaking of the head by the maenads also intensified

the synchronizing effect. Dodds presented a number of parallels from other

55)

cultures for the tossing of the head in ecstasy, but he completely over
looked that this tossing is attested in antiquity not only for orgiastic cults
50) I.Lewis, Ecstatic Religion (Harmandsworth 1971) 39.

51) Dodds, The Greeks, 273.

52) Cf. R.Needham, "Percussion and Transition," Man 2 (1967) 606-614; and
the reactions by W.C.Sturtevant, "Categories, Percussion and Physiology,"
Man 3 (1968) 133f.; A.Jackson, "Sound and Ritual," Man 3 (1968) 293-299.
53) I have formulated the influence of rhythmical music in a more careful
way than is normally the case in anthropological studies which usually quote
the study by A.Neher, "A physiological explanation of unusual behavior in
ceremonies involving drums," Human Biology 4 (1962) 151-160. In fact, Neher's
study is rather debatable, see G.Rouget, La musique et la transe (Paris 1980)
249-252; V.Erlmann, "Musik und Trance, symbolische Aspekte des Bori Beses
senheitskultus der Hausa in Maradi (Niger)," Africana Marburgensia 15 (1982)
3-24 and idem, "Trance and Music in the Hausa B?orii Spirit Possession Cult
in Niger," Ethnomusicology 26 (1982) 49-58. On the other hand, Rouget goes

much too far in rejecting any connection between music and trance, see A.
Zempl?ni, L'Homme 21 (1981) 105-110.
54) Nighttime dances: Eur.Ant. 1150; Roux on Bacc. 486; Nonnos D.16.386,

401, 27. 214. Torches: Henrichs, "Greek Maenadism," (above, note 3) 144

n.75; Nonnos D. 12.391.

55) For maenadic headshaking, see Dodds, The Greeks, 273f. Its importance

was already noted by Rohde, Psyche, 11 n.3. In the Epilogue to his The Devils
of Loudon (1952), Aldous Huxley also noted the phenomenon of 'head-wagging'.
One wonders whether he had already read Dodds1 The Greeks (1951), since in

chapter 6 he compares the possessed nuns to a 'pack of Maenads'.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

279

in general but also more specifically for the cults of Cybele (whose etymology

was even explained by the tossing), Bona Dea, Bellona, and the Dea Syria;
even the ecstasy of the Pythia and other prophets was often connected with it

57)

Dodds explained the practice from religious hysteria but this is hardly con
vincing: the practice can also be observed in the case of autistic children.
The shaking activates the body's balancing organ, just as the whirling character
59 )
of the dancing does, and this helps the dancer to concentrate on the external
stimuli, such as music and light. This synchronizing of the brain activity
together with a focussing of the attention on specific stimuli must have been
accompanied by a reduction of attention to the world around the maenads.
The importance of rhythm in connection with trance has only been 'disco
vered' in recent times; the Greeks connected madness in particular with the
other instrument important in the Dionysiac rites, the aulos. The sound of the
aulos must have had a relatively high number of high-pitched notes, since its
two pipes, being rather long in comparison to their diameter, had to be blown

with force. The music of the aulos, especially when in the Phrygian mode,
had the power to bring people to ecstasy, as the ancient texts repeatedly
61)
assure us. Unfortunately, we cannot say why this would be so, since w

unable to reconstruct the Phrygian mode with any certainty. At this point

further musicological research is necessary.

Amidst the pandemonium the maenads frequently shouted euhoi, the Dion
62 )

cry par excellence. ' It is noteworthy that in the orgiastic cult of Tarantism

56) Servius Aen. 3.111 (also 10.220) arc? xou xu?icxav x?v xecpaA?v; Et.
Magn. 543.11.
57) In general: Alciphron 2.38 ; Quint. 11.3.71; Ulpianus apud Dig. XXI.
1.1.9. Cybele: Men. Theoph. fr. dub. 7 Sandbach; Call.fr. 193.35 Pf.; Anth.
Pal. 6.51 (anon. XLII Gow-Page), 6.94 (Philip XIV GP), 6.173 (Rhianus VII GP),
6.218 (Alcaeus XXI GP), 6.219 (Dioscorides XVI GP), 6.281 (Le?nidas XLIV GP);
Varro Sat.Menipp. 132 Buch. ( = 140 C?be); Ov. Fa. 4.244; Statius Theb. 10.173;
Lucianus Dial.Deor. 12; Lampr. Heliogabalus 7.2. Bona Dea: Juv. 6.316.
Bellona: Luc. 1.566. Dea Syria: Ap.Met. 8.27; Florus 2.7.4. Pythia: Sen.

Oedip. 230. Mopsus: Val.Flacc. 1.208f. Sibyl: Luc. 5.172; Statius S. 4.3.121.
58) As Christine Bremmer pointed out to me.

59) Eur.Bacc. 569 eiAiccou?vac Maiva?ac; Nonnos D. 44.275. Rohde, Psyche,

9 n.4 overlooked these passages but nevertheless inferred the whirling
character of the maenadic dances from other orgiastic dances by comparing
Verg.Aen. 7.377ff; Alex.Aphrod. Pr. p.6 Us.; Heliod. Aethiop. 4.17; Et.Magn.
276.32. L.Gernet, Anthropologie de la gr?ce antique (Paris 1968) 81 also

noted "le vertige de la choree."

60) For the aulos, see Th.Reinach, "Tibia," in D.aremberg-Saglio V, 300

332; G.Wille, M?sica Romana (Amsterdam 1967) 53-56 ("Die Musik im r?mischen

Bacchuskult"); Rouget, La musique, 297-315 (with some mistakes in inter

preting the Greek); W.Boetticher, Aulos, Kleine Pauly I (M?nchen 1975) 755
760; J.-P.Vernant, "L'autre de l'homme: la face de Gorg?," in M.?lender (ed.),
Pour L?on Poliakov (Brussels 1981, 141-156) 148f.
61) Cf. Aesch, f r. 71 Mette u?Aoc, uocviac ?rcaycoy?v ?uoxAav; Soph. A j. 610;

Eur. Her. 871, 879, 895; Plato Symp. 215c.

62) Cf. Henrichs, "Changing D. Identities," 156; add Wankel on Dem. 18.
260; Th?s. Ling. Lat. s.v. euhans and euhoe.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

J.N.Bremmer

280

too the women shouted a cry of two syllables, a-hi. From the reports it
appears that this word was shouted in various intonations, sometimes close t
barking; euhoi may well have been shouted in a similar way. Since chanting h
a liberating effect but also causes hyperventilation, the shouting of euhoi
was not only part of the generally excited atmosphere but also contributed
to it.

The music and shouting accompanied and stimulated the dances o

which consisted mainly of running and jumping about. The charact

maenadic dances also becomes clear from the figurative use of

minology which in Greek and Latin literature was widely used to d

and raving. ' The dancing took place in the mountains, and, w
overlooked, during the winter, so that the combination of phys

thinner air and low temperature helped the women to achieve thei
trance much earlier than normal circumstances would have allowed. ' The

relative lack of oxygen will also have had an influence on the activity of
the neuronal cells and the metabolism of the brain, which helped to cause a
certain intoxication and to promote feelings of lust. These effects were
reinforced by the deprivation of sleep which disturbs the circadian rhythm
of the body and therefore induces the relaxed atmosphere of the nighttime
66 )

ritual. Low blood glucose and the production of adrenaline, resulting

from overexertion, lead to a change of degree of awareness and increase

susceptibility to visions: the mention of the smell of Syrian frankincense
and the sight of milk and honey as described in the parodos of the Bacchae,
6 7)
may well be connected with real maenadic visions.

63) For the cry, see D.Carpitella, in the fundamental study of Tarantism
by E. de Martino, La terra del rimorso (Milano 1961); I quote from the French
edition: La terre du remords (Paris 1965) 365.
64) Jumping: N.J.Richardson on h.Dem. 386; Nonnos D. 20.8, 28.36. Dodds

on Bacc. 111 perceptively compares the wearing of the fawnskin by the maenads
with the leaping of the fawns; and the panther, whose skin the maenads wear

on the oldest representations, was known for his leaps (Nonnos D.22.50, 36.
314, 41.191). Maenadic swiftness: Dodds on Bacc. 1090-3. Greek terminology:
S.G.Cole, "New Evidence for the Mysteries of Dionysos," GRBS 21 (1980) 223

238, esp. 226ff. Latin: Thes.Ling.Lat. s.v. bacchari.

65) For the sometimes severe character of the winter, note especially Plut.

M. 953cd.

66) Relaxed atmosphere: L.Ziehen, RE 17.2 (1949) 629-32; Borgeaud, Pan, 246

252. Both overlooked the important text, Artemidorus 3.61.

67) For the sweet smell, see Roux on Bacc. 144; add W.Deonna, "Euco?ia.

Croyances antiques et modernes: l'odeur suave des dieux et des ?lus," Genava
17 (1939) 167-263; E.Benz, Die Vision (Stuttgart 1969) 371-77; Richardson on

h.Dem. 275ff; B.K?tting, "Wohlgeruch der Heiligkeit," in Gedenkschrift A.

Stuiber (M?nster 1982)168-175; G.Bounoure, "L'odeur du h?ros. Un th?me ancien

de la l?gende d'Alexandre," Quad, di storia 1983, 3-46. Milk and honey: Graf,
"Milch, Honig und Wein," add Euphorion Suppl. Hell. 430 ii 24 Lloyd-Jones/

Parsons.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

281

It is hard to know precisely what the maenadic state of mind was during
the ritual. The name 'maenad'(uociv?c) was essentially a poetic word which had

decidedly pejorative connotations, and which evidently reflected male dis

68)
approval of the female worshippers of Dionysos. Yet this disapproval does
not necessarily invalidate the male view, and it must be noted that the close
connection of 'madness' and Dionysiac rites also appears from the poetical
69 )
application of the language of Bacchic frenzy to Lyssa. Plutarch (M.291A)
mentions that during their rites the women tore ivy to pieces. This activity
and the ecstatic character of their dancing indicate that, even if 'madness'
is not the right word, the maenads must certainly have been in a, what we

call so inprecisely, altered state of consciousness.

70)

The dancing will have reached a climax when the dancers began to fall, as

is mentioned in the Bacchae (136): "welcome in the mountains whosoever from

the running bands falls to the ground, wearing the sacred cloak of fawnskin,

hunting blood that flows when a goat is killed, the joyful act of eating raw,
71 )
speeding to the mountains of Phrygia, of Lydia." ' It is the falling which
in this enumeration of Bacchic activities receives the greatest emphasis,
since it is the only activity which is described in the verbum finitum; all

others are described by the participle. In ecstatic cults a fall often

signifies the climax of the ritual, the sign that the god has finally taken
72)
possession of the worshipper. Given this emphasis on falling in the parodos,
a fall may well have had this significance in the maenadic ritual; when the

maenad resumes dancing, she is really entheos. The exhaustive dancing must

have led to a decrease in the blood supply to the brain so that fainting will
have occurred regularly; moreover, the flicker of the torches may well have

elicited epileptic seizures in susceptible participants in the ritual.

In collections of Greek proverbs the expression 'like a Bacchant' is
73)
explained as applicable to people who remain silent. It seems not impos
sible that this expression arose from the silence which must have occurred

at the end of the ritual when the women would have been too exhausted to

talk anymore.

68) Henrichs, "Changing D. Identities" (above, note 3) 146.

69) Soph.fr. 941 Radt; Eur. Her. 899, Bacc. 977.
70) Henrichs, "Changing D. Identities" (above, note 3) 146f. seems too

reluctant to ascribe to the maenads "an abnormal state of mind."

71) For the textual problems connected with this verse, see most recently
W.J.Verdenius, Mnem. IV 34 (1981) 308-310; Henrichs, in H.D.Evjen (ed.),
Hulley Memorial Volume, Scholars Press 1984.

72) Cf. J.Belo, Trance in Bali (New York 1960) 98; De Martino (above, n.
63) 69 and photo 9; A.Zempl?ni, "La dimension th?rapeutique du culte des rab
nd?p, tuuru et samp, rites de possession chez les Lebou et les Wolof," Psych
pathologie Africaine 2 (1966) 295-439, esp. 310, 314, 400, 414; G.Cossard,
Contribution ? l'?tude des candombl?s au Br?sil I (Paris 1970) 158f; S.
Ferchiou, "Survivances mystiques et culte de possession dans le maraboutisme
tunisien," L'Homme 12 (1972) 47-69; Rouget, La musique, 73f.
73) Suidas B 56, S 1021; Diog. 3.43; Apost. 4.71.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

282

J.N.Bremmer

And so, all through the night, the maenads danced their wild and ecstatic
dances. And at the end? The chorus sings in the parodos of the rc?vov n??v
xduaxov x' e?x?uaxov, or 'sweet toil and labor that is well-labored'
74)
(v. 66f.). ' This must refer to the feelings of the maenads after their ex
hausting dances. The pattern of sensory stimulation, along with the physical
exertion causing changes in the blood circulation, resulted in a feeling of
euphoria. It is possible that this state was accompanied by an increase in
endorphins, the body's natural opiates, in the brain, as the most recent in
75)

vestigations seem to suggest. It was in this euphoric mood that the

maenads will have returned home where the harsh realities of Greek female

life must quickly have brought them down to earth.

Ill The origin and function of maenadism

According the Henrichs, "the problem of maenadic origins, and of the
reflection in maenadic myth and ritual, is unsolved, and conceivably incapable
76 )
of any convincing solution by scholarly means of inquiry." This statement
seems unduly pessimistic, since a comparison of maernadic myths and rituals sug
gess at least one direction into which we have to look for a solution.
We are relatively well informed about the occurrence of maenadism. Besides
the maenads proper of Thebes, Delphi, and Athens, there were some other sacred
colleges of women which were clearly maenadic in origin, such as the Sixteen
in Elis and the Oleiai from Orchomenos; and, as repeatedly has been seen,
tbe myths of the Minyads and Proitos1 daughters also belong in this con
77 )
text. Within the Theban maenads Euripides distinguishes three categories
of women: the young, the old, and the maidens; the latter participated also
in the Delphic ritual. According to Diodorus Siculus (4.3.3) the maidens could
only carry the thyrsos; apparently they later passed into the class of mature
women. This points to a pre-matrimonial origin of the maenadic ritual, as does
indeed some other evidence.

78 )

74) For the physical exhaustion, note also Plut. M. 249ef; Ov. Am. 1.14.
21f; Prop. 1.3.5f.
75) J.L.Henry, "Possible Involvement of Endorphins in Altered States of
Consciousness," Ethos 10 (1982) 394-408; R.Prince, "Shamans and Endorphins,"
ibid. 409-423.

76) Henrichs, "Changing D. Identities" (above, note 3) 218 n.53.

77) Except for Proitos' daughters, this was already seen by Rapp (n.4
E. Gerhard, "?ber die Anthesterien und das Verh?ltnis des attischen Diony
zum Koradienst," Berl.Akad. 1858, 166 (like various scholars after him)
nected the college of the Athenian Gerarai too, but this seems less likel

see Bremmer, The Early Greek Concept of the Soul, 111. C.Brown, "Dionysus
the Women of Elis: PMG 871," GRBS 23 (1982, 305-314) 307 wrongly connects

Anthesteria with the Elean Thyia.

78) Three categories: Eur. Bacc. 6 94. Against Dodds, W.J.Verdenius, Mn
IV 15 (1962) 350 rightly maintains that Euripides speaks of three differe
groups. For maidens in the ritual, see Eur. Bacc. 694, Ion 551, Phoen. 65

1751-57, Antigone fr. 213 Mette (Lustrum 23/4, 1881/82 = POx. 3317), Hypsi
fr. 752 N2 = p.2 3 Bond. Maidens are also frequently mentioned by Nonnos D

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

283

The pursuit of the daughters of Proitos ended in a marriage. A wedding

was also the resolution of Bacchic disorder in Chios. In his version of the

Minyads1 myth Aelian merely mentions that the Chian women were once stung

the 'oistros bacchikos', but Seleukos informs us that at the Chian Dionysia

women once became mad and started to fight with the men - a fight which end
80 )

in a group marriage. Finally, The Elean Sixteen organised a race for adoles
cent girls. Since the race was typical in Greece of pre-matrimonial rituals,

and this one was run during the Heraea in Elis, a pre-matrimonial function
for the Sixteen seems very likely. This college also organised chorusses for
Pbyskoa and Hippodameia. Unfortunately, the age of the participants in these
choruses is nowhere mentioned, but Ca?ame has argued persuasively for the pre
matrimonial character of Physkoa's chorus. According to him, the cult of Hip
podameia was attached "? la figure de la femme mari?e", but the myths about
Hippodameia all revolve around her wedding so that a pre-matrimonial character
81 )
seems not unlikely in this case, too.
Our last instance envisages a slightly different situation. The daughters
of Minyas refused to follow the other women of Orchomenos in the Bacchic rites,
although they were already married and had children. This means, as Ca?ame
bas seen, that the Minyads refused to accept that they had become married
women, whereas the Proetids refused to become married women. The connection

of this myth with the passage from adolescence is stressed in two different
ways. First, the Minyads are characterised as adolescents by the 'hippie'
names they carry (Arsippe and Lysippos); Antoninus Liberalis (10.3) actually
calls the Minyads 'girls' (k?pac). Second, Plutarch (M. 299) mentions that
the Minyads still lived in the house of their father - in other words, they
had refused to settle in their husbands' homes as Greek women normally did
after their wedding. Even though this myth cannot be called a pre-matrimonial
one in the strict sense of the word, it is still concerned with the passage

261, 30.213, 33. 168, 36.257, 44.42. Verg.Aen. 7.389ff also points to the
presence of maidens in the maenadic rites, and it is perhaps presupposed by
the aetiological myth of Dionysos Pseudanor, cf. Polyaenus Strat. 4.1; Schol.
Persius 1.99. For the pre-matrimonial character, note also Burkert, Griechische
Religion der archaischen und klassischen Epoche (Stuttgart 1977) 436: "Alte
Formen der Pubert?tsweihe d?rften gerade in den sexuellen Ritualen weiter
wirken; nicht die Jungfrauen, nur Frauen konnten B?kchai sein"; F.Graf, in his

Nachwort to J.O.Plassmann, Orpheus, Altgriechische Mysterien (K?ln 19822) 165

speaks of "die Form des initiatorischen Kultbundes."

79) For the daughters of Proitos, see M.Massenzio, Cultura e crisi per
manente: la "xenia" dionisiaca (Rome 1970) 91-98; Burkert, Homo necans, 189
194 ( = Amer. ed. 168-73). A.Henrichs, "Die Proitiden im hesiodischen Katalog,"
ZPE 15 (1974) 297-301; Ca?ame, Les choeurs I, 214-220; P.Scarpi, "Melampus e i
"miracoli" di Dionysos," in Perennitas. Studi in onore di Angelo Brelich (Rome
1980) 431-444; H.Maehler, Die Lieder des Bakchylides 1.2 (Leiden 1982) 196-202.
80) Ael. VH.3.42; Seleukos apud Harpokration s.v. *Ounpi?ai. The group
marriage is an archaic feature, cf. Gernet, Anthropologie, 39-45.
81) Elean Sixteen: Paus. 5.16.6f; Plut. M.251ef. Ca?ame, Les choeurs I,
326 (Physkoa), 212 (Hippodameia and the married woman), 417ff (Hippodameia

and Pelops).

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

J.N.Bremmer

284

from adolescence to the state of married woman.

At first sight the presence of old women in the strenuous Bacchic rites is
ratber surprising. This was also the Greek male view, for in the case of older
women acting in a manner unsuitable for their age it was proverbially said:
"the old woman acts the Bacchant." Yet the mythical ancestors of the Elean
Sixteen were said to have belonged to the most senior women, and it seems
therefore reasonable to infer that the historical Sixteen too were advanced

in age. The Theban maenads who were sent to Magnesia will also have been ad
vanced in age, since it seems unlikely that their kyrioi would have allowed
women to go who were still capable of bearing children. In Elis the Sixteen
were the initiatory supervisors: this will also have been the original role
of the old women in Thebes, as it was for old women among many 'primitive'

peoples. '

In the course of the Archaic Age the maenadic rituals seem to have lost

their pre-matrimonial character, but they did not become obsolescent. Why

not? There is of course no a priori reason why a dysfunctional institution

should not survive for a long time in society. The functionalist approach
all too easily overlooked the many examples of institutions which survived
solely because they had a quasi-independent institutional life of their
84 )
own. ' Yet there are a number of reasons why maenadism should survive, even
though not all the reasons that have been adduced are convincing.
Jeanmaire saw in maenadism a cure for illness like that offered by the

African zar and bori cults but, as Gernet observed, maenadism is a collective
85 )

trance, not an individual treatment. Maenadism is also not an expression

of women whose socio-sexual status was in a flux or who were in need of a

more defined and stable social status. There is nothing in our evidence that
points in this direction. On the contrary, wherever we have more detailed in
formation about them, maenads appear to belong to the elite. The mythical

82) For the Minyads, see the analysis by Burkert, Homo necans, 194-197
( = Amer. ed. 173-76) with all sources; Massenzio, Cultura, 85-91; Ca?ame,

Les choeurs I, 242f, 417 n.131 (on the adolescent flavor of the names

Leukippe and Arsippe); Scarpi (above, n.79); A.Schachter, Cults of Boiotia

I (London 1981) 179-181 .

83) For ypauc ?axxeuei, cf. Pherekrates fr. 35K, Zen. 2.96; Diog. 3.74,
4.10; Et.Magn. 266d; Photios A 1485 Theod.; Suidas A 1883. A.M. Komornicka,
QUCC 9 (1981) 63 has overlooked the Bacchic coloring of the verb ?vaduav in
this context (like LSJ s.v. before him). Elis: Paus. 5.16.5. Magnesia: I.
Magn. 215 with the extensive analysis by Henrichs, "Greek Maenadism" (above,
note 3) 123ff. Initiatory supervisors: see e.g. A.Richards, Chisungu. A girl's
initiation ceremony among the Bemba of Zambia (London/New York 19822); D.

Taylor, "Daughters of the Desert," The Guardian August 8, 1980.

84) For a trenchant critique of the functionalist approach, see L.Stone,
The Past and the Present (Boston 1981) 9-11.
85) H.Jeanmaire, Dionysos (Paris 1951) 124ff; Gernet, Anthropologie, 80.
Before Jeanmaire, R.Ganszyniec, "Dionysos i Mainady," Przeglad Historyczny
31 (1933/34) 279-294, 309 (French summary) had already explained the Maenads
as neurotics.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

Greek Maenadism Reconsidered

285

ancestors of the Elean Sixteen were of the highest social circles, and so will

have been their historical descendants. The leaders of the Theban maenads in

Euripides Bacchae belonged to the royal family, as did in historical time the
Oleiai in Orchomenos. Clea, the leader of the Delphic Thyiads, was apparently
well-to-do, as we learn from Plutarch and the inscriptions. Alkmeionis, the
Milesian Bacchant, had a name which, according to Merkelbach, "die Tr?gerin
an den milesischen Uradel ankn?pft." These references are few in number but
86 )
they do not point in the direction of deprived women.
Various scholars have argued that the tenor of maenadic mythology and
87

ritual is one of female rebellion against male authority. This is ce

true, but at the same time it should be observed that maenadism is only
half-hearted rebellion and only a pseudo-liberation: the maenads were u

class women who undoubtedly had slaves to look after the children that w
left behind, and their trip into the mountains would only last for a sh

ritually prescribed period. Yet it is the nature of Greek women's life

provides the clue to maenadism.
The restricted nature of Greek women's life has recently been stress
88 )
again. ' There are at least three reasons, then, why women would have liked
to continue practicing the maenadic ritual, even if it had lost its original
meaning. First, it enabled women to leave home. The contrast between maenadic
freedom and non-maenadic confinement appears clearly in the case of the women
of Amphissa (above, n.33), who had to win their husbands' consent to accompany

86) Contra R.S.Kraemer, "Ecstasy and Possession: the Attraction of Women

to the Cult of Dionysus," Harvard Theol.Rev. 72 (1979) 55-80. Elis: Paus.

5.16.5. Oleiai: Plut. M.299ef. Clea: St?dter, Plutarch's Historical Methods,
2f; E.Kapetanopoulos, BCH 90 (1966) 128-130. Alkmeionis: R.Merkelbach, "Mile
sische Bakchfen," ZPE 9 (1972) 77-83.
87) See the enumeration of these views by Henrichs, "Changing D. Identities,"
(above, note 3) 145, 221. During the discussion of this paper in Oxford it
was suggested that the stories about infanticide were the product of the
female mind, which in this way subconsciously expressed its horror at always
being tied down by the children. It seems dangerous to infer a subconscious
horror, when we do not know anything about a conscious horror. Moreover, do

we know of any Greek myths undeniably originating from women?

88) Cf. J.Gould, "Law, Custom and Myth: Aspects of the Social Position of
Women in Classical Athens," JHS 100 (1980) 38-59, esp. 40, 46-49. It should
be noted that the severity of women's seclusion had already been described
with a wealth of evidence by W.A.Becker, Charikles III (Leipzig 18542) 265ff.
Of this seclusion I note two aspects which deserve further investigation.
First, J.Minces, The House of Obedience: Women in Arab Society (London 1982)

41 observes that the seclusion creates a feeling of solidarity among women.

This could also have been the case in Greece, cf. Antiphon 1.14f. Second,
this seclusion may have played a role in the development of a specific 'Frauen
sprache', of which the existence in Greece has been argued by G.R.Solta, Die
Stellung des Armenischen im Kreise der indogermanischen Sprachen (Wien 1960)

464; J.Knobloch, "Problemi e metodi della paleontolog?a ling?istica" in

Paleontolog?a ling?istica (Brescia 1977) 19-38, esp. 33-36.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms

J.N.Bremmer

286

the stranded maenads to the boundaries of the city. The maenadic shuttle from
Athens to Delphi must actually have enabled the Athenian women to stay away
from home in a legitimate way longer than any other period in their lives.

Second, the maenadic ritual, like other female cults, made it possible for
women to mix freely in a female company, the membership of which transcended
89 )

the immediate neighborhood. Third, in Greek society a woman's identity was

fixed in terms of her social role. Escape into a more authentic self-expres
sion could only come through the temporary, if regularized, experience of
possession trance.
After the trance the women would return home and resume their dull and

isolated existence, which the maenadic ritual, like other female cults, helped
91 )
them to endure. ' As such, the cult was an integrative factor in Greek social

life of the Classical and Hellenistic period, just as visits to the disco,

where we find the same phenomena of auditory and photic driving, headshaking,
92)

and strenuous activity, help modern youths to get through the boredom of
everyday life: maenadism as a Saturday Night Fever avant le lettre - a
93)

sobering thought.

Rijksuniversiteit Utrecht

Jan N. Bremmer

89) For female contacts being confined to the immediate neighborhood, see
W.K.Lacey, The Family in Classical Greece (London 1968) 173; add Men. Sam.
35-38.

90) Cf. E.M.Zuesse, Ritual Cosmos: The Sanctification of Life in African

Religions (Athens, Ohio 1979) 187.

91) The element of 'escape' was also noted, in a revealing passage, by
Wilamowitz, Griechische Trag?dien IV (Berlin 1923) 130f: "Rasen und toben

wird die Maenade, bis sie ersch?pft zusammenbricht. Wenn sie erwacht, wird

sie totmatt sein, ern?chtert, aber sie hat sich ausgetobt, und durch die

Entladung der in dem dumpfen Getriebe des Tageslebens niedergehaltenen

Sehnsucht nach ungebundener Lebenslust erleichtert mag sie unter das Joch

des Alltags zur?ckkehren. Der Gott hat sie erl?st. Wir sehen an unsern

Kindern, dass sie sich austollen und austoben m?ssen, und wehren ihnen nicht.
Das Bed?rfnis erstirbt auch in uns nicht, wenn wir es auch nicht mehr wie

die Maenade befriedigen k?nnen. Sie hatte es besser als wir."

92) The resemblance of behavior in the disco to Dionysiac rites is noted
by H.Cox, The Feast of Fools (Cambridge, Mass. 1969) 108-112; Henrichs,
"Changing D. Identities" (above, note 3) 222 n.80.
93) This article is a revised version of a paper read at the meetings of
the American Philological Association in Philadelphia in December, 1982,

and at Oxford in March, 1983. For information and comments I would like to

thank, besides the audiences at both occasions, W.Burkert, Fritz Graf,

Nicholas Horsfall, Charles Segal, and especially Albert Henrichs. Susan
Guettel Cole and Robert Parker corrected the English in a way which im
proved the manuscript in other respects.

This content downloaded from 87.20.220.105 on Fri, 12 Aug 2016 22:24:15 UTC
All use subject to http://about.jstor.org/terms