cais1.gif (153930 bytes)

CAIS Persian Text.gif (34162 bytes)


The Circle of Ancient Iranian Studies

 Persian Section.PNG (9914 bytes)


About CAIS


Daily News

News Archive


CAIS Seminars

Image Library





Contact Us


Facebook-Button.jpg (107165 bytes)



The Essential Characteristics of Parthian

and Sasanian Glyptic Art


By: Neilson C. Debevoise




Second only in number to the cylinders, those enigmatic little stamp seals of the "Later Orient" have long formed a prominent part of all large collections of Oriental glyptic art. Eighty years of oriental studies have failed to improve the label greatly and they are now usually grouped under some incorrect classification or if the curator be a cautious man, some more truthful but less illuminating title as "Mesopotamian". Yet these seals may provide us with a valuable glimpse into periods of history still only half understood and an art whose widespread ramifications and influences are scarcely now appreciated.([1])


Obviously the most important element in the identification of a seal is the design engraved thereon, but if this is not conclusive, there are several other characteristics upon which we may rely. The technique of the engraving, the shape of the seal itself and the decoration upon the back if such there be, the inscription if any, all of these may be utilized. In the periods under consideration there is little to be gleaned from inquiry into the type of stone employed, for slight variation took place during the interval concerned. Chalcedony, lapis-lazuli, banded agate, garnet, agate, nicolo, carnelian, hematite,([2]) crystal,([3]) red, green, brown and grey jasper,([4]) amethyst,([5]) were common, and more rarely ruby,([6]) meteoric stone([7]) or such metals as bronze,([8] - [9]) were used as seal material. In Achaemenid times limestone,([10]) marble,([11]) and other fine grain stones([12]) were utilized, a practice which does not seem to have been continued under the succeeding empires.


In so brief a span of years we can hardly expect to find radical changes in engraving technique, and therefore this criteria is of little value as a means of identification for Parthian and Sasanian seals. The point, the drill, and the wheel were in use from Achaemenian times onward. On the ordinary Parthian or Sasanian seal the point was seldom employed except in the case of some refractory material such as quartz (pl. I, no. 1). The drill with a simple round point was also infrequent (as an exception, see pl. I, no. 2), and by far the bulk of the stones are produced on the wheel (pl. I, nos. 3, 5-8.([13]) Probably it was only by this method that engravers could keep up with the tremendous popular demand for their products.


While the Achaemenian period (c. 550 BCE - 330 BCE) does not strictly fall within the scope of this article, the close connection and continuity of development in shape and design with the succeeding periods make a few words necessary.([14]) Both cylinder and stamp seals were used in Achaemenid times. One of the most characteristic shapes was the cone with rounded top and convex base (fig. 1-2). Another common style had flat sides with beveled edges and a square or rectangular base (fig. 3). Occasionally the sides were engraved.([15]) The scenes depicted most frequently are the combat between the king and some wild beast, various fanciful animals, hunting scenes or the worshiper before the altar from which flames arise. Winged sphinxes sometimes occur and the winged sun disk is a common symbol.


As long as the clay tablet continued to be the sole writing material, the Majority of seals were made cylindrical in form,([16]) for by the very nature of its size and the depth of its cutting the archaic stamp seal had been unsuitable for tablets. The deep engraving required too heavy a pressure on the soft clay tablet in order to force the air out of the furthermost pockets. On the other hand the rolling motion of the cylinder allowed the air to escape as the clay moved up to take its place and thus a good impression could be secured with a minimum of pressure. The cylinder was not suitable for use on round or oval bullae which are increasingly common from this time onwards and the stamp seal came into prominence because of an effort to meet the Dew conditions. The rounded surface of the clay bullae allowed the air to escape as the clay worked up into the engraving, hence seals which were principally used on bullae could be cut much deeper. Achaemenid stamp seals which would have served for use, on both tablets and bullae were generally quite shallow in cutting. Those seals with Greek influence were usually more deeply engraved since they would most frequently be employed on the bullae of parchments or papyri which oriental conservatism reluctantly adopted.' As the clay tablet was gradually discarded, so the bulky and heavier cylinder seal simultaneously went out of use and the lighter stamp seal, more suitable in shape for the work it had to perform, was adopted.([17])


Fig2a.gif (121650 bytes)

In the Seleucid period (330 BCE- 248 BCE in Persia, until 143 BCE in Mesopotamia and still later in Syria) the globular form of seal with a highly elongated base became the most popular shape (fig. 5- 7).([18]) Clay tablets were now rapidly being replaced by more modern writing materials but seals still had to be designed to cope with either tablets or bullae. Consequently we find that the cylinder has nearly if not completely disappeared, and that engraving on the majority of oriental stamp seals is quite shallow.([19]) Finger rings with a long oval base were frequently utilized (fig. 4).


Faced with a scarcity of dated or excavated examples, we must grasp whatever evidence is available for the identification of the engraved designs. Our one other source of positive information at the present time lies in the impressions made upon dated clay tablets, but this unfortunately gives only a post quem date. In the Seleucid and more especially the Parthian periods old seals were highly prized and probably a considerable number were in use.([20]) Unfortunately while many texts have been published few writers have thought it worth while to include seal impressions.([21]) Of the remaining types of evidence, the numismatic is probably the best. The portrait heads of the Seleucid period when cut by Hellenistic wasters are quickly discernable and bear a close resemblance to the coins.([22]) Details of costume and headdress when compared with the coinage will sometimes settle the matter. This is especially true of Sasanian gems but the method can also be applied to those of the Seleucid and Arsacid periods. Our greatest difficulty arises when the workmanship and subject matter are oriental in technique and feeling, in which case long continuity of design and methods of engraving of which we have scant knowledge make identification difficult without mole positive evidence from excavations.


Those seals which bear the kneeling or standing figure of a worshiper before an altar or other object of adoration might often better be classed as Neo-Babylonian or Achaemenian than Seleucid in view of their general conical shape and certain details of costume.([23]) The scorpion was a favorite subject for Seleucid seals([24]) and in increasingly conventional forms it continued in popularity in Parthian and even Sasanian times.


With the advent of the Parthian period (247, in Mesopotamia c. 143 BCE, lasting until 226 CE) we have still less evidence than for Seleucid glyptic. Clay tablets with their impressions have nearly disappeared, completely so by the beginning of the Christian era, but another though not so adequate substitute appears in the allied art of the die engraver for coins, which can by then be very closely dated. While we must expect the engraver of dies to be far more conservative than the graver of seals, who would have to conform to the fashion of the hour, still there are some valuable points to be gained along this line. From this source of evidence we may be fairly sure that the greater part of the Greek influence bad died out by the beginning of the Christian era. On the other hand most of the common seals which have come down to us were made for men of meager means by native cutters who scarcely felt the Hellenization of the upper classes where Greek workmanship would be in greater demand. For this reason seals of purely oriental design and workmanship probably continued to be made even at the height of the Hellenistic influence, especially in cities which like Nippur had a long tradition of oriental forerunners.


Ring seals were not popular in the Parthian period and -a new globular form (fig. 8) came into common usage. If the cylinder was seldom used in Seleucid times, with the rise of the Arsacid dynasty, it seems to have dropped completely out of sight. The hemispherical form, especially common in hematite began to make its appearance. In general the bases of the Parthian seals tend more to roundness than do those of the Seleucid period which are' generally oval. Besides these differences in shape, we can be reasonably sure that the engraving will be deeper on the gems of the later period. The most common technique seems to be a development of that used during the previous period, a series of straight parallel lines.([25]) This is possibly due to more extensive use of the wheel in cutting necessitated by the greatly increased demand for seals. A somewhat similar technique appears upon the coinage after the beginning of the Christian era.([26])


The human figure formed one of the principal subjects for the gem engraver, either full length, with long trousers, short coat and hair bound with a fillet the ends of which hang down behind, or a bust treated in a similar fashion.([27]) Such busts usually have no base nor were they completed with the palm leaves or flying ribbons so characteristic of the Sasanian busts.([28])


A somewhat different treatment partially dictated by the refractor); nature of the quartz and the use of the point alone in engraving may be seen in number 1, plate I and figure 9, where two figures are created with simple straight lines. There are a number of animal motifs which appear upon the coins of the latter half of the Parthian Empire and 'also, upon the seals. Doubtless some of them were employed upon the seals of early Sasanian times as well, so that their presence alone does not settle the question but merely dates the objects between these periods. Among such designs are the crouching ram or ibex,([29]) a winged horse (Pegasus ?), ([30]) the antlered stag,([31]) and the griffon.([32]) This last motif has usually been considered Sasanian probably due to its presence on such monuments as Taq-i Bostan but it was certainly in use in the Parthian period as the coins demonstrate, The eagle was employed both upon the seals([33]) and as an architectural ornament.([34]) We know that the old motif of the struggle of the king or god with some animal which represented the powers of evil, used frequently during the Achaemenian period, was still known in Parthian times and doubtless utilized by the engravers.([35])


In spite of considerable Hellenistic influence at Seleucia-on-the-Tigris there was relatively little erotic material discovered there and such scenes as were familiar from Sumerian, Akkadian, and Babylonian times were not often repeated on the clay plaques of the Parthians and do not seem to have been used on the Seals.([36])


Fig4.gif (111208 bytes)

The Sasanian period brings us to more certain ground for there is much material already in collections and definite information should be forthcoming shortly from excavations.([37]) With the rise of the Sasanian and their invasion of Mesopotamia there came about one of those revivals of art which the stimulus of such an event often seems to bring about. The change which is so obvious on the coinage([38]) probably also found reflection in the work of the seal engravers. In the better class of workmanship this meant the abandonment of the parallel line technique and the production of more rounded and naturalistic examples.([39])


The most common shape of the Sasanian period was the hemisphere with a round base, but the globe with an oval base continued and this similarity with Parthian shapes is the cause of much of our difficulties in identification. Engraving was usually fairly deep since the seals were now never used on the flat surfaces of clay tablets. In shape and material the seals differ so little from the Parthian epoch that with one or two exceptions these characteristics must be discarded as a means of identification. The decoration of the back of a seal as in figs. 11-14 seems to be essentially Sasanian. Metal seals (fig. 10), are common in this period and also in early Arabic times.


Subject matter must again be our principal criteria. Busts and full length human figures often wear distinctively Sasanian headdress. similar to those found upon the reliefs or the coinage([40]) In matters of dress considerable caution must be exercised and a thorough knowledge of Parthian and Sasanian monuments is necessary for recognition of the close distinctions in draping and modeling which mark the later Parthian and early Sasanian costumes. The mysterious symbol found on a considerable number of seals([41]) was common in Sasanian times and has not yet been found in the Parthian period. Although without any definite fixed form it preserved the same general idea throughout. Quite often the crescent which is usually the upper element is accompanied by what 'may be either the sun or a star. The whole is reminiscent of the front portion of the Sasanian crown, but the wide-spread and. varied occurrence of the design makes it unlikely that it is a royal mark of any sort. An unusual variant is to be found in number- 11, plate 11. Another very common motif is that of three flowers tied together with a band of ribbon the ends of which float gracefully away to either side (p]. 11, no. 12).


If there is an inscription on the seal, this may be used to settle the question of its Parthian or Sasanian origin. If the characters are Pahlavi the seal is probably Arsacid or may possibly be early Sasanian, if they are Parsik the seal may be tentatively placed in the Sasanid period.([42])


One of the general characteristics of the, seals of the late Parthian and Sasanian periods is a desire to fill all of the space available for the design. The late Parthian design may be spread out for this purpose as in number 4, plate 11, just as it is on architectural work of the same era. Later on, an inscription (pl. 11, no. 7) or a small border (pl, II, no. 13) was often used, or a cruder solution was found in small dots which were run around wherever a blank space presented itself (pl. 11, nos. 2,3 . Often a portion of the design was utilised for the same purpose as in number 14, plate 11, where for no particular reason a wide band of ribbon floats off behind the neck of a ram and thus fills up the space behind his head.


Final solutions to these problems must eventually come through excavation and the publication of the results. Until that time we shall have to content ourselves with generalizations based on an ever-increasing body of comparative material.




[1] Sufficient evidence is not available to admit of anything approaching a final solution of Parthian and Sasanian glyptic problems and the present article is simply an attempt to carry forward the discussion so ably presented by Dr. H. H. von der Osten, "The Ancient Seals from the Near East in the Metropolitan Museum", The Art Bulletin, XIII, no. 2, 1931, pp. 1-21, to which the reader is referred for extensive references. Many of the conclusions reached by the present writer are based on excavated evidence not yet published or material in the hands of dealers. See also N. C. Debevoise, Parthian Seals in Pope (ed.), Persian Survey, forthcoming shortly.

[2] Von der Osten, Art. Bull., XIII, 1931, p. 11.

[3] L. Delaporte, Cylindres Orientaux, I, p. 80. D. 172; p. 81, D. 200; p. 83, D. 266; p, 84, D. 270.

[4] OP. cit., 1, p, 81, D. 193, D. 204, D. 212-13; p. 82, D. 218, D. 220-21, D. 224-25, D. 228, D. 231, D. 240-1, D. 250; p. 83, D. 259, and others.

[5] OP. cit., I, p. 81, 1). 202, D. 206, D. 222.

[6] OP. cit., I, p. 80, D. 183; p. 83, D. 265.

[7] Von der Osten, Art. Bull., XIII, 1931, P. 16, fig. 53; p. 17, fig. 70.

[8] OP. cit., P. 14, fig, 15-16, 20; p. 17, fig. 65, 73; p. 18, fig. 82, 88.

[9] Delaporte, Cyl. Orient., 1, p. 81, D. 211.

[10] Von der Osten, Art. Bull., XIII, 1931, P. 13, fig. 2; p. 14, see below, p. x, note 10, on this example.

[11] Delaporte, Cyl. Orient, Il, p. 174, A. 782; p. 175, A. 791.

[12] Op. cit., 1, p. 15, D. 104 "pierre brune"; Il, p. 175, A. 789 "pierre rosâtre".

[13] Cf. H. B. Walters, Catalogue of the Engraved Gems and Cameos, Greek Etruscan and Roman 2, in the British Museum, (1926), P. xviii-p. xxi.

[14] Unless otherwise noted, seals used as illustrations were purchased by the writer in the winter of 1932 in Baghdad from a local dealer. They come as usual from provenances diverse and unknown, chiefly "the other side of the Diala River." Examples from the Museum of the American University, Beirut, Syria, pl. ii, 3, 7, were made available through the courtesy of the Director, Dr. H. Ingholt, and still others are from the collection of Dr. Douglas Cruikshank of the medical staff of the same University, pl. i, 4, 5, 7, pl. ii, 9, 10.

[15] Von der Osten, Art. Bull. XIII, 1931, p. 4, fig. 14a and b; classed on p. 14, fig. 14a as Parthian, but more probably Achaemenid. See below p. x, note 10.

[16] W. H. Ward, Cylinders and Other Ancient Oriental Seals in the Library of J. Pierpont Morgan, p. 12.

[17] Ward, op. cit., p. 20.

[18] Delaporte, Cyl. Orient., 11, pl. 122 f. (impressions) ; for others see von der Osten, Art Bull., XIII, 1931, p. 10, note 12.

[19] Cp. impressions mentioned in note 2 and pl. 1, no. 3.

[20] H. V. Hilprecht, The Babylonian Expedition of the University of Pennsylvania, 1, p. 335 f.; a number of Sumerian seals were found in the Parthian levels at Seleucia-on-the-Tigris.

[21] A considerable number appear in Delaporte, Cyl. Orient. See above note 2.

[22] Delaporte, Cyl. Orient., II, pl. 122, 5b(A. 805) and 7 (A. 807) ; R. McDowell, Bullae, in Leroy Waterman, Preliminary Report upon the Excavations at Tel Umar, Iraq, pl. V, nos. 19 and 2 1.

[23] Von der Osten, Art Bull., XIII, 1931, p. 12, figs. 10-12, states that this subject was used "until Seleucid times in Mesopotamia" but he classes the seals under the heading "Seleucid or Arsacid".

[24] Pl. I, no. 3 (Seleucid) and no. 4 (Sasanid) Delaporte, Cyl. Orient., II, p. 220, cp. A. 1276, A. 1277.

[25] Pl. 1, nos. 5-8; Delaporte, Cyl. Orient., 1, p. 79, D. 152, D. 153; von der Osten, Art Bull., XIII, 1931, p. 8, fig. 113.

[26] See the plates in Wroth, Catalogue of the Coins of Parthia, (British Museum), or any other volume on Parthian coinage.

[27] Pl. I, nos. 8-11; cp. the Parthian bust from Dura-Europos, Wells, Rostovstzeff and Bellinger, eds., The Excavations at Dura-Auropos, Third Season, pl. XVII; for other examples see Delaporte, Cyl. Orient., 1, p. 83, D. 252-3, D. 258-9; von der Osten, Art Bull., XIII, 1931 p. 4, fig. 17 and 18.

[28] Pl. I, nos. 13, 14; pl. II, nos. 8, 12; see also von der Osten, Art Bull., XIII, 1931, p. 8, fig. 120.

[29] J. de Morgan, Numismatique de la Perse Antique, ler fasc., Introduction-Arsacides, (in part III of E. Babelon, Traité des Monnaies Grecques et Romaines) pl. XXIV, 19 and 20, text no. 305, (coins of Mithradates IV, 130-47 CE).

[30] Pl. 1, no. 12; cp. op. cit., Pl. XIII, 6-8 and text 123 and 123a.

[31] Pl, 1, nos. 13-15; cp. op, cit., pl. XIII, 16 and 17, text 130 (Orodes 1, 57-38/37 BCE; cp. Tarn, Cambridge Ancient History, IX, P. 613).

[32] Pl Il, nos. 1-3; cp. op. cit., pl. XXIII, 17 and 18, text 291 (Osroes 106/7-130 AD ?)

[33] Pl. II, no. 4; cp. op. cit., pl. XIX, 15 and 16, text 222 and 222a (Vardanes 1, 41/42-45 CE).

[34] S. Yeivin, Architectural Notes 1929-30, in L. Waterman, Prelim. Report, p. 24, fig. 3.

[35] Wells, Rostovstzeff and Bellinger, eds., Dura-Europos, Third Season, pl. XI, fig. 2, and pl. XII, fig. 1-2.

[36] Cp. von der Osten, Art Bull., XIII, 1931, pl. p. 4, 14b and text p. 12 fig. 14a and b, where he says, "The erotic scene appears frequently on glyptic objects of this period (see L. pl. 52; 20b, 21)", that is, of the Parthian period. The seal in question both from treatment and from shape to me seems better classed as Achaemenian, and the examples quoted from Delaporte do not appear to be Parthian.

[37] Kish; Ctesiphon; Damghan.

[38] Cp. the coinage of the later kings in Wroth, op. cit., and the first of the Sasanian in Paruck, Sassanian Coinage.

[39] Pl. 11, nos. 5, 6, 7; A. Furtwângler, Die antiken Gemmen, II, p. 245-46, no. 50; 1, pl., L, no. 50.

[40] Delaporte, Cyl. Orient., 1, p. 83, D. 262; von der Osten, Art Bull., XIII, 1931, p. 8, fig. 123; A. Mordtman, "Studien ûber geschnittene Stein mit Pahlavi Inschriften," ZDMG., XVIII, 1864, P. 5 f., pl. 1, no. 5; p. 9 f., pl. I, no. 10; p 34 f, pl. II, no. 106; p. 45, pl. II, no. 155; P. Horn, "Sasanidische Gemmen aus dem British Museum", ZDMG XLIV, 1890, p. 658, pl. Ia, no. 569; p. 671, pl. ID, no. Se.

[41] Pl. II, nos. 8-11; the type is very commonsee also Charles C. Torrey, "Pahlavi Seat Inscriptions from Yale Collections," JAOS, LII, 1932, pl. facing p. 206, no. 8a, d, c; 10a, b: 11-43; 76; pl. facing p. 207, no.4496.

[42] See E. Herzfeld, Paikuti, p. 77; Debevoise, Parthian Seats, in Pope, (ed), Persian Survey, forthcoming.






top of Page

my_Iran.jpg (13682 bytes)

"History is the Light on the Path to Future"


Persian_NOT_Farsi_by_Shapour_Suren-Pahlav_3D2.gif (177309 bytes)


Encyclopaedia Iranica

BIPS.jpg (15695 bytes)

The British Institute of Persian Studies

"Persepolis Reconstructed"

Persepolis_reconstructed2.jpg (36944 bytes)


The British Museum

The Royal

Asiatic Society

Persian_Gulf_Facebook.jpg (1935028 bytes)

The Persian Gulf

Facebook Page

Please use your "Back" button (top left) to return to the previous page

Copyright © 1998-2015 The Circle of Ancient Iranian Studies (CAIS)