Mycenaean Plumes 9781463221065

Holland uses sculpture to suggest that these ornaments were meant to be worn in womens' headdresses as a developmen

227 131 4MB

English Pages 33 [37] Year 2009

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Mycenaean Plumes
 9781463221065

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Mycenaean Plumes

A n a l e c t a Gorgiana

231 Series Editor George Kiraz

Analecta Gorgiana is a collection of long essays and

short

monographs which are consistently cited by modern scholars but previously difficult to find because of their original appearance in obscure publications. Carefully selected by a team of scholars based on their relevance to modern scholarship, these essays can now be fully utili2ed by scholars and proudly owned by libraries.

Mycenaean Plumes

Leicester Holland

1 gorgia* press 2009

Gorgias Press LLC, 180 Centennial Ave., Piscataway, NJ, 08854, USA www.gorgiaspress.com Copyright © 2009 by Gorgias Press LLC Originally published in All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. 2009

1

ISBN 978-1-60724-460-8

ISSN 1935-6854

Extract from The A^merican Journal of Archaeology, vol. 33 (1929).

Printed in the LTnited States of America

HOLLAND:

MYCENAEAN

PLUMES

205

from the other did not occur in Greece, but that the two were imported either contemporaneously or successively from the eastern Mediterranean, where the development may have taken place not long before. LEICESTER B . H O L L A N D PHILADELPHIA

archaeological SnHtitute o! America MYCENAEAN PLUMES IN a recent article entitled "The Ring of Nestor" Sir Arthur Evans 1 mentions that among other objects found in a tomb at Thisbe, from which the ring in question came, there was " a whole set of perforated glass paste objects, with holes below for the attachment of glass disks, which Mr. Wace has now conclusively shown to belong to necklaces." Formerly Evans had stated 2 that "the exact application of these objects remains uncertain, but it seems clear that they served as brackets or supports;" and still earlier 3 had described them as consoles "forming part of a cornice." And Bell,4 evidently following Evans's lead, says of them, " I t has been suggested that they were used side by side as a continuous ledge or string course, but their small size—about three inches in l e n g t h seems inconsistent with their use in architecture. A brass matrix in the Ashmolean museum implies that they were moulded for use as applied ornament such as ears for vases. They are chiefly interesting as giving a primitive form of modillion which is characteristic of the Corinthian order of architecture and which occurs exceptionally on the north door of the Erechtheion." There is something arresting in the extraordinary versatility of a bit of abstract ornament which can be considered chiefly interesting as a prototype of the Corinthian modillion, and then by simple inversion in the hands of Mr. Wace, be cordially accepted as part of a necklace. The contrast between these two interpretations is so great that the question inevitably arises whether either is certainly correct. Wace's accepted conclusion is reached in a careful study of the object—the only one so far made, I believe—appended to the report of his excavations at Mycenae. 6 Instead of calling it a "modillion" he christens the object in question "the curled leaf ornament"; both names are really suitable to the form, though they apply best to different stages of its development. The earliest examples found in Greece are made of gold (Fig. 1) and consist of three elements: (a) long oblanceolate leaves, with central ribs and ribbed edges, which curve outward from the base to the broadest part, and from 1 2

J.H.S., 1926, p. 1; also published separately, Macmillan, London. Palace of Minos, I, p. 488. 3 B.S.A., IX, pp. 65-66. 4 Hellenic Architecture, p. 29. B.S.A., X X V (1921-23), pp. 398-402. The theory was almost simultaneously advanced by Maiuri (Ann. della Scuola di Aiene, VI (1923), p. 101, fig. 19) that the objects were strung together in the position suggested by Wace and worn above the forehead in such a way that the series of square plaques formed a continuous diadem from which the "leaves," imitating a series of curled locks, hung like a bang. The principal objection to this theory is that it is the later forms only which resembled frizzed hair. See below. 5

173

174

AMERICAN

JOURNAL

OF

ARCHAEOLOGY

there to the tip are rolled inward in a complete circular loop; alternating with these are (b) lanceolate leaves, broadest at the base, ribbed in the center and at the edges, which curve simply outward to a point in line with the greatest projection and greatest length of the adjacent roll. These lanceolate leaves usually have chains of small gold disks and rings fastened to their tips. Both sets of leaves spring from (c) rectangular plaques, approximately square, with beaded ribs at top and bottom which form continuous bands if two or more elements be set side by side. In the center of the plaque there is almost always a circular area, sunk as if to hold an inlay of some sort. Usually two leaves, one of each type, spring from the ribbed edge of a plaque, and from behind their tips spring

f ì

Ì

F I G U R E 1. "CURLED-LEAF ORNAM E N T ' ' IN G O L D F R O M MYCENAE (a-c) AND A R G O S (d) size)

similar secondary leaves, and from behind these again come a third pair, with even a fourth pair occasionally found. The terminal curled leaf is usually longer than the others, sometimes it is slightly broader also, in which case the accompanying lanceolate leaf becomes narrower or shorter in relation to the rolled leaf. Apparently we have to do with a conventionalized natural form of some sort, which is eminently suited to execution in gold. In the later examples of the ornament (Fig. 2), glass paste, cast in a mold and then gilded, is substituted for true metal work. The result is a progressive series beginning with a labored imitation of the goldsmith's technique, where the pairs of leaves are cast separately and then joined to make the compound unit (Fig. 2, c + d = e) and ending in a form which gives little indication of its metallic ancestry, where the plaque and three or four pairs of leaves are all cast in a single piece (Fig. 2, a, b). At the same time the conventionalization increases until the whole perhaps suggests some structural architectural member rather more than any natural object. In one

HOLLAND:

MYCENAEAN

PLUMES

175

example 1 which probably comes at the very end of the series, the form has degenerated still further into a mere series of rolls ending in a crudely modelled female head and bust (Fig. 3). Wace says of this: "the original form of the ornament was by now forgotten and obscured," 2 and he is undoubtedly correct. Both gold and paste examples have holes bored in them in such a way that they could be strung together side by side, or fastened in that position to some background, presumably of fabric. The length varies from .02 rn. to over .08 m.; the largest examples, it is needless to say, are of paste, but the majority, both of gold and paste, measure from .035 m. to .045 rn. The ornament occurs so frequently in Mycenaean tombs of the better class that it would

FIGURE 3. GLASS PASTE ORNAMENT FROM MYCENAE

(natural size)

FIGURE 2. " C U R L E D - L E A P O R N A M E N T ' ' IN G L A S S P A S T E F R O M M Y C E N A E (a, c-i) AND SPATA (6) (}4 size)

seem to have formed a conventional object of personal adornment. The evidence of the excavations bears out the chronological arrangement which the development of form suggests. Wace 3 believes the earliest known example in Greece to be that from Argos (Fig. 1, d). This he dates in L. H. II and considers that all others in gold are of the same general period, while all the paste examples he places in L. H. III. The fact that one gold example from Mycenae (Fig- 1, a) has its plaque decorated with basket work instead of with the usual sunken circle, would seem to place it at one end or the other of the series; since it can be dated L. H. II we may safely say that at that period the form was in its youth, not far removed from some natural prototype, and had not yet crystallized into a set convention. 1 2 3

Praktika, 1886, p. 78; Arch. Eph., 1887, pi. X I I I , 23. Op. cit., p. 401. Ibid., p. 39S.

176

AMERICAN

JOURNAL

OF

ARCHAEOLOGY

In Crete specimens are as rare as they are common in Greece. Wace 1 cites one L. H. I l l example in glass paste from Kalyvia near Phaestos, 2 a steatite mold from Knossos 3 and a curled leaf of ivory which resembles part of the gold form in question, but much more closely the handles of an alabaster vase from Mycenae.4 Evans 5 says sweepingly, "faience specimens of such objects have been found both at Knossos and in mainland Greece," but he mentions specifically only an example from Messenian Pylos,6 which as Wace 7 points out, is of glass paste. Wace mentions no examples in faience, and so far as I am aware, none has been found on the mainland or in Crete.8 The fact that the paste examples were often gilded in imitation of the original gold ornaments, and that faience was unsuited for gilding, readily explains why this material was not commonly used. The steatite mold from Knossos, which Evans 9 dates as "probably belonging to L. M. I," was made, he says, for casting faience. Of course nothing about the mold itself indicates whether it was for faience or for paste; and since the dating is qualified with a "probably"—the details of the provenance, " a dependency to the North West of the palace," are not given10—while the form of the object corresponds exactly, even to the circular sinkage in the base plaque, with the L. H. I l l examples in paste from the mainland, it seems to me highly improbable that it antedates the gold prototypes of the mainland castings. The authoritative examples from Crete then, reduce to two; one in paste from Kalyvia, of L. M. I l l , which as Wace 11 suggests may 1

Op. tit., p. 398. Mm. Ant., XIV, pp. 632-633, fig. 106. Evans, Palace of Minos, I, p. 488, figs. 349-350; B. S. A., IX, pp. 65-66, figs. 42,43. 4 Fourth shaft grave (Sohliemann, Mycenae, p. 246, No. 356). The alabaster handles were made separately and attached with pins. The ivory curl is much smaller than the alabaster handles, but the form is strikingly similar, even in section. I should suppose that the Cretan curl came from a small vase made of gold and ivory. The shape of the alabaster vase suggests a metal prototype; it also recalls shapes shown on Egyptian monuments of the eighteenth dynasty. Wainwright (Liverpool Ann., VI (1913-14), p. 48 and pi. XIV, No. 6) considers it 5 related to a Syrian type (ibid., pi. IX, 7). Palace of Minos, I, p. 488. 7 °Arch. Eph., 1914, p. 103, fig. 5. B.S.A., X X V (1921-23), p. 398, note 2. 8 The only faience examples of which I am aware, are a series of twelve, found in tomb X X X I , Ialysos, Rhodes, (Ann. della Scuola di Atene, VI (1923) p. 165, fig. 92). The fact that rosettes of gold leaf were set in the sinkages in the plaques, while similar examples, of blue glass paste from tomb IV (op. cit. p. 101, figs. 17-19) had inset rosettes of greenish paste, suggests that the faience examples were ungilded save for the rosette, while the other series may have been gilded in part, or not at all. 9 Loc. cit.; Wace (op. cit., p. 398), probably because of obscurity in this passage in the Palace of Minos, gives the still earlier date, M. M. I I I . 10 The " N o r t h West Building" of Vol. I is probably the same as the " N o r t h West Treasury" of Vol. II, but in the description of the latter I can find no mention of the steatite matrix. It is perhaps significant that among other objects found at that point is the painted clay base of a circular stand which unquestionably dates from the end of L. M. I l l (Evans, op. cit., II, fig. 390, pp. 619-620). 11 Op. cit., p. 399. 2

3

HOLLAND:

MYCENAEAN

PLUMES

177

be an importation from the mainland; the other, the mold for a similar object from Knossos, of uncertain date, but possibly also L. M. III. To all intents and purposes the ornament is purely a mainland form, appearing in L. H. II and developing throughout L. H. I l l , at which period occasional examples probably made their way, along with other expressions of mainland culture, to Crete. 1 The holes for fastening the ornament and the asymétrie arrangement of one curled-leaf and one lanceolate-leaf element to every unit of the base plaque, leaves no doubt that the units were to be set close together to form a continuous band. But there may be doubt that this band was a necklace. Wace's conclusion that it was such, seems to be based on the assumption 2 that the plaque was at the top of the ornament with the leaves pendant from it, and on the observation 3 that "the top (i.e., the plaque) is always narrower than the bottom (the curl) so that when threaded the ornaments, as the units of a necklace, make a wide curve." Among the examples which he illustrates, however (Figs. 1 and 2), there is only one (Fig. 1, b) in which the curl is noticeably wider than the plaque, and in this one the lanceolate leaves are lacking. If the latter were made separately, as in other gold examples, and inserted on plaques of their own between the plaques of the curled leaves, the wedge shape of the latter would lose all significance. In some of the examples shown (e.g., Fig. 1, a; Fig. 2, f, h, compare Fig. 3) the sides do not appear to me to diverge at all, while in those which are unquestionably somewhat wedge-shaped the divergence is so slight that if joined side by side on a flat surface, the resultant circle would be far too great for a necklace or even for a girdle.4 In order to form a circle small enough to be worn upon the neck the elements would have to lie almost at right angles to the plane of the circle, thus producing a collar with nearly vertical sides. But if they were joined in this way there is no certainty as to the size of circle they formed: it might have been small enough for a bracelet—with some of the smaller examples at least—or large enough for the hem of a coat. As for the plaque being at the top, I can see no evidence in support of this except that it would make a better necklace so, and even this would be of little weight with a necklace of the collar type. On the other hand, in the late example with the human head (Fig. 3) it is very certain that the head was right side up, and the plaque con1 Examples, of L. H. I l l date, have been found as far afield as Corfu (Cavvadias, 458 p l V T n o t e 8)PX' TV$ KKXà5"'¡• p - 3 6 7 ' > a n d Mysos, Rhodes (see above,

Op. cit., p. 398. s ibid As an example, take fig 2 b, from Spata. The width across the center of the plaque is 0.0195 m., and at the widest part of the curl 0.0225 m. The distance between these points of measurement is 0.06 m. If the pieces were laid flat, side by side, the circle through the center of the plaques would have a diameter of 0.78 m 2

1

178

AMERICAN

JOURNAL

OF

ARCHAEOLOGY

sequently at the bottom. Wace 1 says this "was apparently inverted," but it is hardly credible that an ornament of long and fixed tradition as to form would ever be worn upside down. To invert a well known object, notoriously produces burlesque. While I think Wace is surely right in considering the ornament an article of personal adornment rather than an architectural motif, the evidence seems to me against rather than in favor of its being a necklace. Furthermore in all the considerable repertoire of Mycenaean costume shown in fresco fragments there is no representa-

A FIGURE

4.

A,

FRESCO

B F R A G M E N T FROM M Y C E N A E ; T R I A D A SARCOPHAGUS

B,

FIGURE

FROM

HAGIA

tion of any necklace which at all suggests this ornament. On the other hand, a fragment of fresco illustrated by Rodenwaldt 2 (Fig. 4, A) shows a series of triple incurled forms which for linear indication are as accurate representations of the "curled-leaf ornament" as one could wish; and here certainly a necklace is not shown. This fragment is, unfortunately, so small that it shows nothing but three repeats of the ornament and an adjacent detached stripe which formed part of the enclosing frame of the picture. By assuming this stripe to be at the top, the curls fall into the proper place to form the front edge of a crown. This interpretation of the fragment is accepted by Miss Lamb in the publication of Mycenae, 3 and therefore, I presume by Wace himself. Interpreted in terms of the "curledleaf ornament," the crown shown in the fresco would be formed 1 B.S.A., X X V (1921-23), p. 401. Fries des Megarans von Mykenae, p. 50. 3 B.S.A., X X V (1921-23), p. 166.

2

HOLLAND:

MYCENAEAN

PLUMES

179

of a series of units joined side by side in a slightly flaring cylinder, with the flat band formed by the juxtaposed plaques at the bottom. Confirmation of this restoration is furnished by a crown depicted on the famous L. M. I l l sarcophagus from Hagia Triada 1 (Fig. 4, B). The indication here has been somewhat simplified by representing only a single row of curls. These are drawn in red line on a yellow ground, evidently representing gold, and spring from a flat band which encircles the head. But the head-dress here is more than a simple crown, for from the center rises a stiff lily-like object from which a long soft tassel or plume floats out and down the wearer's back.2 Presumably, then, the crown on the fragment from Mycenae should be restored with a similar central tassel. Evidently the flexible gold circlet was fastened around a close fitting cap of leather or fabric, from which the tassel sprang. The delicacy of some of the actual gold specimens of the "curled-leaf ornament" is certainly such that they could hardly have been used without some backing. The cap, of course, has left no traces in the tombs where the ornament has been found, and the tassel has naturally vanished completely too, but the central lily-like socket of the tassel might have been made of something that would have lasted, yet so far as I know, nothing of the sort has been identified. Now if the "curled-leaf ornament" is the material remains of tasseled crowns such as are shown on the Hagia Triada sarcophagus, a very special interest attaches to it; for this particular type of headdress is otherwise well known, and has a'ready received considerable attention. 3 It is unmistakably represented on fresco, ivory, gold, glass paste and on pottery, and these representations come from Crete, from mainland Greece and from Rhodes and Cyprus.4 Its 1

Mon. Ant., X I X (1908), pl. II. "Rien ne permet d'assurer que cette longue queue que le vent semble entraîner soit formée de plumes. Il s'agit peut-être plutôt d'un faisceau de cheveux ou de fils (dorés dans le Xoos and 0â\os. Some Homeric helmets

fledged apparently had two ia\os, IL, V, 743; XI, 41); some had four (rerpà^aXos, II., X I I , 384; X X I I , 315). The i4>a\oi and X60os. Cf. above, p. 200, note 1. 3 Perrot, Histoire de l'Art, VI, pp. 936-937; "c'est une calotte qui colle au crâne; les points dont elle est hérisée indique peut-être les poils d'une peau de bête."

HOLLAND:

MYCENAEAN

PLUMES

203

sharp upper line; a skin cap could hardly be sufficiently hirsute to warrant such a pictorial expression. Altogether I feel no hesitancy in seeing on the "Warrior Vase" feather crowns very similar to those of the east.1 The wearers may, of course, be allies or enemies of the warriors in the horned and crested helmets, in either case—• since there is no reason to think the vase imported ware—both apparels were evidently familiar to the Mycenean painter, and probably therefore both were to be found in the Argolid.2 If one only be local, it should be the feather crown, for a ninth century geometric vase 3 from Mycenae shows (Fig. 11, e), in even more schematized drawing, head-gear which is undoubtedly meant to be the same as that on the "Warrior Vase." 4 I even consider it quite possible that a still more degenerate representation of the feather crown is shown on another geometric vase 6 from Mycenae (Fig. 11, f). In geometric design the manes of horses are regularly indicated by short parallel lines, and a similar treatment is given to the long plumes—presumably horse tails—that float from Attic warriors' heads, but where representation is limited to the most essential elements, it seems unlikely that an artist would trouble to indicate the hairs of a man's head. And if it be not hair depicted in Fig. 11, f, then it can hardly be a different head-dress from that shown in Figure 11, e.6 1 A. D. Frazer, in an unpublished dissertation at Harvard University , reached the conclusion some years ago that feather crowns might be represented on the "Warrior Vase." , Perhaps introduced by the Lycians, who, according to legend, built Tiryns for Proetus and Mycenae for Perseus. One of the carved stele erected over the shaft graves at Mycenae was, some time later, covered with stucco and painted (Arch. Eph. 1896, pis. 7, 2). Two zones are fairly well preserved. The upper one shows a file of five warriors which in costume, pose and drawing are strikingly like the second troup on the "Warrior Vase." Unfortunately the heads are almost wholly obliterated; but behind the head of the first man is the stiff neckpiece of a helmet like that of the horned and plumed helmets on the vase, while five short radial lines above the head of the second man suggest that he in turn wore the feather war-bonnet. The colors of shield and kilt alternate from man to man throughout the troup, whence it may be conjectured that an alternation of the two types shown on the "Warrior Vase" was represented here, evidently in friendly association. In the lower zone are shown four deer and a smaller spotted animal with curved and bristling back, very suggestive of the warrior's head-gear. This animal looks more like a hedge-hog than anything else, but there seems little reason for representing a hedge-hog in a Greek hunting scene. 3 Wide, Jb. Arch., X I V (1899), p. 85, fig. 44. 4 Pernier (II disco di Phaestos, in Ausonia, I I I (1908-9), p. 282) agrees with Hall (B.S.A., VIII, p. 185) in considering this geometric vase to show feather crowns; on the other hand, Reinach (Rev. Arch., XV (1910), p. 62, note 2) considers that both here and on the Warrior Vase we may be dealing with skin caps. 6 Wide, op. cit., fig. 42. G A silver rhyton from the fourth shaft-grave at Mycenae (Arch. Eph., 1891, pi. II, 2; Stais, Ath. Mitt., X L (1915), pp. 45 ff., p. 112; pis. VII, VIII), which represents an attack upon an Aegean city by the sea, shows the naked defenders

204

AMERICAN

JOURNAL

OF

ARCHAEOLOGY

In résumé the points I have endeavored to bring out are: I. The " curled-leaf ornament " appears first in Greece in L. H . II, approximately contemporaneously with the change from the " Shaftgrave D y n a s t y " to the "Tholos-tomb D y n a s t y " ; it does not reach Crete till the later part of the following period. II. The "curled-leaf o r n a m e n t " formed the coronet of the tasselled crowns which are represented in L. H. I l l , and possibly in L. H . II, as worn by women and sphinxes. These representations are found in Greece, Crete, Rhodes, and Cyprus, but mostly in Greece. The crown was probably a religious vestment, and the women who wore it priestesses of the tree goddess. I I I . The sphinx is an oriental creature, perhaps originally of Egypt, but common in Asia before its introduction to Greece. The cults to which it belonged were probably brought to Greece from the eastern Mediterranean at the time its head-dress—the tasselledcrown—appears there, under the aegis of the Tholos t o m b dynasty. IV. The "curled-leaf o r n a m e n t " seems to be a conventionalization derived from a feather crown. Feather war bonnets are found in the eastern Mediterranean from the fourteenth to the fifth centuries before Christ; it is probable t h a t the Luku-Lycians wore them in the latter part of the thirteenth century. They were known in Crete and Aegina by the end of L. M. I. They were established in Mycenae from a t least the twelfth to the ninth century. 1 V. The tasselled crown appears to be a religious head-dress peculiar to priestesses and sphinxes, while the stiff feathered war bonnet—with no central tassel—is the head-gear of fighting men. The two m a y be quite independent creations, but the crown of relatively short incurling feathers on the Pisko-Kephalo head, and t h a t of stiff feathers worn by the great goddess of Beth-Shan suggest a common origin. I t seems certain t h a t the development of one with crests on their heads which may be upstanding hair or may be feather crowns (so Hall, J.H.S., X X X I (1911), pp. 119 ff.; cf. illustration in Evans, Palace of Minos, I, p. 668), while the corseleted warriors on the sea wear conical helmets marked into horizontal bands, like the Mycenaean boar's tusk helmets, from which plumes float out. The value of this object as a historical illustration is undoubtedly great, but for our purpose seriously lessened by uncertainty as to where it was made, where the scene is laid, whether the men in boats are allies or enemies of the landsmen, and whether the latter wear artificial head-dresses or not. 1 To those who realize the important part which misrepresentation of ancient practices plays in the formation of legend, it may be interesting to speculate on the possibility that the winged cap of Hermes is a reminiscence of an obsolete feathered war-bonnet. Compare the early fifth century terra-cotta from Locris Epizephyrii (Ausonia, I I I (1909), p. 187, fig. 40), where Hermes is shown with a head-dress which resembles a feather crown quite as much as it does a winged cap, and is certainly not a petasos.