Ērān ud Anērān
Webfestschrift Marshak 2003


Bullae from the Pahlavi Archive at the University of California, Berkeley

Guitty Azarpay*

It gives me great pleasure to contribute to a volume that honors my colleague and friend Boris Marshak whose valuable research, field-work and publications, particularly in Sogdian art and archaeology, have lent a new and enriched dimension to Iranian studies. I wish to thank Dr. Prudence Harper, whose study of the physical characteristics of Sasanian bullae from Qasr-i Abu Nasr, in the collection of the Metropolitan Museum of Art, in New York, appears in the present volume, and whose comments and observation have prompted this paper.


The Pahlavi Archive, recently acquired by the University of California, Berkeley, is a collection of 390 bullae and 260 manuscripts in Middle Persian, datable to the early post-Sasanian age in the seventh and eighth centuries. This collection is currently being classified by Professor Philippe Gignoux and Dr. Rika Gyselen in preparation for its publication and its digitization for a Bancroft Library website.

As a compliment to the paleographic and iconographic studies of the Pahlavi Archive by Philippe Gignoux and Rika Gyselen, the present study focuses on the use of bullae and on the method of their attachment to manuscripts in Sasanian Iran, This extraordinarily large corpus of Pahlavi manuscripts and bullae now casts a new light on questions pondered in earlier studies and it attests to, 1) the use of bullae of differing shapes for sealing manuscripts rather than goods and merchandize as once believed, 2) the placement and method of attachment of bullae to documents, and 3) the creation of folders of manuscripts. The folder is made by means of the binding together with a bulla of several sealed documents.

Because many bullae from the Pahlavi Archive are found intact on manuscripts, they provide important clues for the study of unattached bullae, such as those from the archaeological site of Qasr-i Abu Nasr, in the collection of the Metropolitan Museum of Art, New York, discussed in a contribution to this volume by Dr. Prudence Harper.

The Pahlavi Archive at Berkeley: History and Content

The Pahlavi Archive at Berkeley comprises 260 silk and leather manuscripts, 82 of which still have one or more clay bulla attached (N°53 and N°154, each have two bullae attached). This collection is currently being classified by Professor Philippe Gignoux and Dr. Rika Gyselen in preparation for their eventual publication and digitization of the documents and bullae for a Bancroft Library website. This substantial collection of Pahlavi manuscripts was probably even more extensive prior to its acquisition by the Bancroft Library, as evidenced in the strong similarities between manuscripts and bullae at Berkeley and those from the following smaller collections. A manuscript reportedly from the 1960's 1 in the ‘Abbas Mazda Collection, six Pahlavi manuscripts and a related group of 71 complete specimens of bullae in the California Museum of Ancient Art, in Los Angeles 2, and over thirty specimens of manuscripts and related bullae in Berlin that are soon to be published by Dr. Dieter Weber3. As in other cases that involve the passage of antiquities through the hands of dealers and collectors, there is no record of the temporal context and original provenance of this collection, though it was reportedly found in northwestern Iran. The documents so far examined by Philippe Gignoux appear to be economic texts, lists of remittance and receipt of goods, dated to years from an as yet undetermined era 4. It is hoped that the most recent scientific test by Dr. Timothy Jull, cited below, will help toward the determination of this era, and that the archive's provenance may be identified by examination of place names and other evidence found in the documents, and by information obtained from the collections' physical characteristics and properties.

The corpus of Pahlavi manuscripts and bullae at Berkeley was first brought to my attention in 1988 when a dealer offered them for sale to the University of California at Berkeley. In the spring of 1988, random samples of manuscripts from this archive were submitted for C14 testing to Professor D. J. Donahue, Department of Physics, at the University of Arizona. The test result, sent to us in May of that year, indicated a date of approximately AD 700-1100 for the archive. In the same month we consulted with Dr. Prudence Harper, most familiar with the physical characteristics of the Sasanian bullae excavated at Qasr-i Abu Nasr, and upon the suggestion of my colleague Martin Schwartz, also with Philippe Gignoux, an expert in Pahllavi. These scholars then visited Berkeley in order to examine the Pahlavi Archive at an informal meeting, an occasion that led to the subsequent publication of a number of the manuscripts from the archive by Philippe Gignoux 5.

After years of negotiation, and several visits to examine the collection by Philippe Gignoux and Rika Gyselen, the acquisition and donation of the Pahlavi Archive to the University of California finally occurred in 2001 and 2002, when the collection was anonymously given to the Bancroft Library in two parcels. The bulk of the documents and bullae was donated in May, 2001, exactly 14 years after our first viewing of the collection at Berkeley. After its receipt by the Bancroft Library, a sample of a leather manuscript from the collection was sent for a new C14 testing to Dr. Timothy Jull, Senior Research Scientist, at the NSF-Arizona AMS Laboratory, the University of Arizona. Dr. Jull's report on the result of the test, dated November 7, 2001, sent to Anthony Bliss, Curator of Rare Books and Literary Manuscripts, at the Bancroft Library, gave the radiocarbon age of the manuscript as 1,323+ or -77 before present (year BP), with a calibrated age range of AD 651-776 (1 sigma, 68% confidence), and AD 600-888 (2 sigma, 95% confidence). A smaller portion of the documents and bulllae (mss N°217 through N°260 and 63 unattached bullae) was placed on permanent loan to the Library in January 2002. At present, of the total of 308 detached bullae in the Pahlavi Archive, 245 are in the donated group and 63 in the permanent loan group.

Some Physical Characteristics

The Pahlavi Archive in its entirety includes 390 bullae, with 82 bullae still attached to manuscripts, and 308 bullae that are detached. 6 Of the 82 bullae found on manuscripts, 27 are on silk and 55 on leather documents. The 308 detached bullae had presumably fallen away, or were separated from documents to which they were attached prior to our examination of the archive in 1988. Many of the detached bullae in the Berkeley collection are preserved with fragments of silk or leather from the manuscripts to which they were originally attached (figs. 1-4). The latter bullae, and bullae still attached to the 82 manuscripts in this collection are clearly associated with the sealing of documents rather than with the closure of goods.

Based on evidence from a closed Pahlavi manuscript in the collection of the Free University, Berlin, it may be assumed that each manuscript was originally rolled and tied with a string that was threaded through the center of the page, below the bottom line. 7 The string was then tied and knotted around the middle of the scroll and the manuscript scroll was closed by means of a moist clay bulla that was pressed around the knot and stamped with one or more seals. The impressions on the back of some bullae that show the grid-pattern of the weave of silk manuscripts (fig. 1), and occasionally the creases of leather manuscripts, suggest that the manuscript was pushed up from the back against the bulla with one hand, as it was sealed with the other. Whereas the smaller manuscripts were presumably rolled and tied and sealed with a string around the middle of the scroll, larger manuscripts were probably rolled, then folded before they were tied and sealed. Since all the manuscripts at Berkeley had been unrolled prior to their acquisition, the exact method of closure of these manuscripts remains conjectural. The string impressions and concavity on the bottom of some bullae would have occurred, as explained by Dietrich Huff, when the knot around the folded scroll was held up at a certain distance above the object as the moist bulla was pressed around the knot. 8 Dr. Huff‘s explanation is clearly substantiated by the particulars of a closed and sealed scroll bundle in the collection of Pahlavi manuscripts at the Free University Berlin.

The majority of the bullae at Berkeley are made of buff to light grey colored tempered clay, resembling potter's clay, but a few specimens are made from reddish clay, and a fraction from a coarse, friable clay with organic particles. The small bullae were seemingly rolled between the thumb, the index and middle fingers into a roughly cone-shaped lump, about 2 to 3 cm long and 1 to 2 cm wide. A few disc-shaped bullae are larger, about 4 to 5 cm in diameter. Thumb and fingerprints and cracks are clearly detectable on the sides of most bulla, especially at pressure points around the face of the bulla which bears the seal impression (figs. 1-3). The bottoms of the bullae are often slightly concave with either a smooth surface, or with the impression either of creased leather or of the grid pattern of the silk manuscript to which they were attached.

Archival and Administrative Practices

Notable in the Pahlavi manuscripts in the Berkeley Archive is the attachment of the bulla below the bottom line, often in the center of the document. Manuscripts that lack the bulla at the bottom of the page generally have slits in that area (fig. 10). These bullae were evidently placed over the knot that sealed the scroll, and therefore served as a closure seal, as evidenced in the unopened Berlin scroll. 9

Occasionally several documents, each bearing a bulla below the bottom line, are bound together at the top center of the page with an additional bulla with one or more seal impressions (figs. 5-7). Carefully cut slits found at the top center of many unbound manuscripts from the Berkeley collection suggest that a bulla originally bound together other documents in a similar fashion (fig. 8-9). Since such bound manuscripts had evidently been opened before they were bound in antiquity, leaving the original closure seal still in place below the bottom line, one can only speculate about the reason for sealing of bound manuscripts. Was the seal that bound sets of documents perhaps that of the archivist who thus kept a record of the original texts (fig. 5)? It is of interest that among recently uncovered Bactrian documents are texts preserved in duplicate form, one intended to remain open to be read, and the other rolled, tied with string and sealed with closure bullae.10 It is assumed that the sealed copy would have been opened for validation in case of a dispute.

One manuscript, Berkeley N°37, shows, in addition to the carefully cut slits at the top and bottom of the page, a round perforation at the top corner through which is passed leather straps connected to a large, disc-shaped bulla (figs. 9-10). Close examination of this manuscript and its bulla reveals a careless puncturing of the leather document and seemingly haphazard knotting of the bulla straps on the verso, the result either of an ancient or later manipulation of the manuscript and its bulla. One of the distinguishing features of this exceptionally well preserved manuscript is the presence of traces of a short inscription close to the bulla on the reverse of the document (fig. 9). This may be compared with the use of short inscriptions on the reverse of some Bactrian documents, cited by Sims-Williams, giving the names of the vendors and witnesses written besides the holes for the seal-string. 11

In offering a summary of the history and content of the Pahlavi Archive at Berkeley, the present writer takes this occasion to thank Philippe Gignoux, for his untiring effort toward the decipherment of the Pahlavi documents, and Rika Gyselen for her excellent work on the classification of the Archive's sealings. The foregoing observations on the physical characteristics of the bulllae in the collection, which deserve a more extensive treatment, allude to yet another line of inquiry that, among others, is necessary for a full understanding of the Pahlavi Archive and its larger context.12


Figure 1

[Missing Image]
Bulla N°B129, 2X1.6 cm, showing back and side views, with remains of leather straps and silk manuscript. Pahlavi Archive, Berkeley.

Figure 2

[Missing Image]
Bulla N°B129, see fig. 1, showing side view.

Figure 3

[Missing Image]
Bulla N°B129, see fig. 1, showing face with sealing.

Figure 4

[Missing Image]
Bulla N°B64, 1.5X1.2 cm, showing back view with fragment of leather straps and manuscript. Pahlavi Archive, Berkeley.

Figure 5

[Missing Image]
Leather document, N°ms 43, with five pages, each originally with a bulla at the bottom of the page, bound together with a bulla at the top center, maximum measurements 19X10 cm. Pahlavi Archive, Berkeley.

Figure 6

[Missing Image]
Leather document, N°ms 154, with one bulla at bottom of page, and a second bulla at the top center, placed over fragments of other leather manuscripts, 18X8 cm. Pahlavi Archive, Berkeley.

Figure 7

[Missing Image]
Detail of bulla at the top center of document N°ms 154, see figure 6.

Figure 8

[Missing Image]
Slit shown on the top center of the page of a silk manuscript. Pahlavi Archive, Berkeley.

Figure 9

[Missing Image]
Horizontal slit cut at the top center of the page adjacent to a bulla suspended from leather straps and knotted at the back of the document, N°ms 137, see figure 10.

Figure 10

[Missing Image]
Leather document, N°ms 137, with suspended bulla and horizontal slits at center top and bottom of page. Pahlavi Archive, Berkeley.


* University of California, Berkeley.

1 Richard N. Frye, ``Sasanian Seals and Sealings'', Mémorial Jean de Menasce, Fondation Culturelle Iranienne 185, Louvain 1974, P. 156, FIG. 3; idem, ``Unfinished Projects in ancient Iranian Studies'', Iranica Antiqua XXXVII, 2002, p. 107, fig. 3.

2 Philippe Gignoux, ``Six documents Pehlevis sur cuir du California Museum of Ancient Art'', Bulletin of the Asia Institute 10, 1996, pp. 63-72.

3 To appear in, Corpus Inscriptionum Iranicarum.

4 The era referred to in these documents, earlier thought by Gignoux to be based on the reign of Xusro II (AD 590-628), is now shown by Gignoux , in his reading of the silk documents from the collection, to have exceeded the regnal years of Xusro II, see Gignoux, ``Nouveaux documents pehlevis sur soie'', op0. cit., pp. 9-10.

5 Philippe Gignoux, ``Une nouvelle collection de documents en pehlevi cursif du début du septième siècle de notre ère'', Comtes Rendus de l'Académie des Inscriptions et Belles-Lettres 1991, Paris pp. 683-700; idem, ``Six documents Pehlevis sur cuir du California Museum of Ancient Art'', Bulletin of the Asia Institute 10, pp. 63-72; idem, ``Nouveaux documents pehlevis sur soie'', Philologica et Linguistica, Historia, Pluralitas, Universitas, Festschrift für Helmut Humbach zum 80. Geburtstag am 4. Dezember 2001, hrsgg. von Maria Gabriela Schmidt und Walter Bisang, Trier 2001, pp. 281-301; idem, ``Une esquisse de comparaison entre des textes économiques en bactrien et pehlevi'', Actes du Colloque de Budapest, October 2002 (in press); idem, ``Une liste pehlevie des noms de mois et de jours (document Berkeley no. 38)'', Festschrift Bo Utas, Uppsala (in press); idem, ``Sept documents économiques en pehlevi'', Mélanges W. Skalmowski, Leuven (in press) [re-edition of the article that appeared under Festschrift Humbach].

6 I wish too thank Dr. Rika Gyselen for her organization and numbering of these bullae which will be presented in her future publication of the bullae and of their sealings from the Pahlavi Archive at Berkeley.

7 D. Huff, ``Technological Observations on Clay Bullae from Takht-i Suleiman'', Mesopotamia XXII, Firenze 1987, p. 390. I wish to thank Dr. Maria Macuch for making an image of this manuscript available to me.

8 Ibid, p. 385.

9 Ibid., p. 390.

10 Nicholas Sims-Williams, "New Light on Ancient Afghanistan, the Decipherment of Bactrian", An Inaugural Lecture delivered on 1 February 1996, School of Oriental and African Studies, University of London, London 1997, p. 14, and "Four Bactrian Economic Documents", Bulletin of the Asia Institute 11, 1997,p. 3.

11 Ibid.

12 I wish to express my deep gratitude the Bancroft Library staff, and particularly to its Director, Charles B. Faulhaber, for their interest in housing the Pahlavi Archive, and to Anthony Bliss, Curator of Rare Books and Literary Manuscripts, for his help in negotiating the transfer of the Pahlavi documents to the Bancroft, arranging the digitization of the collection, and planning its future conservation. I am also grateful to my friends and colleagues, Martin Schwartz and Denise Schmandt-Besserat for their invaluable advice and inspiration on the preservation and study of the Archive from the very beginning of its history at Berkeley.

© The Author/s

Actualizado el 24/07/2004

Valid HTML 4.01!
Any Browser!
Creative Commons License
Como lo hacemos?