EPIGRAPHY,the study of inscriptions, particularly their collection, decipherment, interpretation, dating, and classification.

i. OLD PERSIAN AND MIDDLE IRANIAN EPIGRAPHY

Definitions, classification, and method.

Inscriptions are texts carved, incised, or engraved on durable materials like stone and metal. Often they are public messages intended to be permanent, but graffiti scratched, drawn, or painted on walls are also included. Manuscripts, on the other hand, are texts written on less durable material like leather, papyrus, bast, or paper; they are primarily of administrative, juridical, economic, private, religious, or literary character. Sometimes administrative and economic documents are written on unbaked clay, pottery (ostraca), bone, or wood and thus occupy a position between inscriptions and manuscripts. As this third category of texts fills many gaps in the knowledge of epigraphy, it seems useful to include it in epigraphic studies. In a narrow sense epigraphy also includes paleography, which cannot, however, be treated separately from the paleography of manuscripts.

Inscriptions can be classified on various continua, for example, between monumental and small examples. The inscription of Darius I (522-486 B.C.E.) at B^sotu@n (qq.v) is the largest known example of the former category from the pre-Islamic period, whereas the smallest inscriptions are found on coins, seals, gems, utensils, and other portable objects. Inscriptions can also be classified as monolingual, bilingual, trilingual, and so on or as royal, priestly, official, and private or as "lapidary" and cursive. In Iranian epigraphy the lapidary paleographic style was sometimes archaizing, whereas cursive usually reflected the contemporaneous style of writing.

Iranian epigraphy of the pre-Islamic period covers mainly inscriptions in the Old and Middle Iranian languages: Old Persian, Middle Persian, Parthian, Chorasmian (see CHORASMIA iii), Sogdian, and Bactrian (see BACTRIAN LANGUAGE). Old and Middle Persian inscriptions span by far the longest period of time, from the B^sotu@n inscription until the early Islamic period, yet limiting study to inscriptions in Iranian languages cannot fully account for historical reality. Many non-Iranian traditions were incorporated into Iranian culture, which complicated the linguistic aspects of epigraphic remains from the multinational empires of the Achaemenids, the Parthians, the Sasanians, and the Kushans. As a consequence Iranian epigraphy has from the beginning required attention to materials in such non-Iranian languages as Babylonian, Elamite (see ELAM v), Aramaic (q.v. i), Greek (Huyse), Sanskrit, and Middle Indian.

Iranian epigraphy is underdeveloped, in that the publications of inscriptions are widely dispersed and the particular requirements for epigraphic work not always recognized by authors. The only attempt at collecting and studying Iranian inscriptions systematically from a consistent point of view was undertaken by the founder of the Corpus Inscriptionum Iranicarum (q.v.), W. B. Henning, who "in his later years . . . stressed the importance of making known to the scholarly world all inscriptions by photographic or other reproductions" (Frye, 1970, p. 152). It can be inferred from the volumes of the Corpus published by Henning himself that his attention was focused on epigraphic fieldwork, with all its hardships, rather than on publishing materials kept in European museums. Yet, despite his efforts and those of his successors, the majority of important epigraphic publications have appeared outside the Corpus. For example, most of the inscriptions from Persepolis and its surroundings must still be consulted in E. F. Schmidt's archeological reports, the illustrations of which are, however, usually too small for epigraphic work.

The famed archeologist and epigrapher Ernst Herzfeld documented about 100 blocks of the Paikuli inscription (1924), developing a completely new method of presenting the material. As he demonstrated, in attempting to establish a reliable text of such weathered material, it is insufficient merely to publish available squeezes and photographs; often it is necessary to retouch duplicate copies in order to make visible relevant evidence and to facilitate restoration of sequences of half-illegible characters as entire words (see, e.g., Davary and Humbach, 1976, pp. 14-15). Descriptions of graphic details, which are usually both complicated and ineffective, are thus rendered superfluous, and scholarly discussion becomes possible. Following a somewhat different approach, Rüdiger Schmitt (1991), in his masterly edition of the Old Persian text at B^sotu@n, which is visible only from a great distance, combined modern photographs with old drawings like those by Henry C. Rawlinson, which were made when the inscription was in much better condition than it is now.

The origins of Iranian epigraphy.

A.-H. Anquetil-Duperron first introduced in the West some information on the Avestan and Pahlavi languages, from Parsi sources, including strange Aramaic spellings in Pahlavi, for example, malca for "king" and boman for "son" (today transliterated MLK÷/MLKA and BRH/BRE respectively). He thus opened the way to decipherment of Middle Persian and eventually Old Persian inscriptions. Small and inexact drawings of details of such inscriptions had first reached Europe in the work of Jean Chardin, who was in Persia in the mid-17th century. The first exact documentation was provided by Carsten Niebuhr, who visited Persepolis, Naqæ-e Rostam, and Naqæ-e Rajab in March 1765 and published complete copies of several cuneiform inscriptions: the Old Persian-Elamite-Babylonian trilingual inscriptions DPa and XPe; Old Persian DPd, DPe, XPb; Elamite DPf; and Babylonian DPg (for identification and location of Old Persian inscriptions, see Kent, Old Persian). He also reproduced the three short Middle Persian-Parthian-Greek trilingual inscriptions ANRm and ANRm£ of Ardaæ^r (q.v.; 224-40 C.E.) and ˆNRb of ˆa@pu@r (240-70), as well as the Middle Persian KNRm of the high priest Karde@r (for identification and location of Middle Persian inscriptions, see Back). The last provides an informative example of the way in which poor lighting conditions can distort visual acuity, but all Niebuhr's other copies were of such accuracy that A. I. Silvestre de Sacy was able to make scholarly use of them (Figures 1, 2).

Beginning with the Greek versions of ANRm and ˆNRb, which he restored by mutual comparison, and including the short ANRm£ as well, Silvestre de Sacy succeeded for the most part in deciphering the respective Middle Persian versions. For example, he read and restored the crucial sections of the Greek version of ANRm as Masdasnou theou Artaxarou basileo@s basileo@n Ariano@n . . . huiou theou Papakou basileo@s (1793, p. 30), corresponding to Middle Persian mzdysn beh ÷rthætr MLK÷n MLK÷ ÷yr÷n . . . BRH p÷pky MLK÷ (in modern transliteration) "of the Mazdayasnian lord Ardaæahr/Ardaæ^r, king of kings of EÚra@n, son of the king lord Pa@bag." His only error here was in reading beh "better," rather than bgy "lord."

Silvestre de Sacy clearly recognized that what are now known as the Middle Persian versions are in the language of the Sasanians (1793, p. 121). He was less successful in classifying the language of the Parthian versions as Deylamite (1793, p. 123), but their true character was soon recognized, as is clear from Edward Thomas' comment (p. 267) that the alphabet "was once designated 'Parthian' . . . but has latterly been known as Chaldeo-Pehlvi." This early identification was later credited to F. C. Andreas (Meillet, p. 242), though he himself preferred the terms arsacidisch and Nordwest-Dialekt (cf. Nyberg, 1923, p. 186 n. 1). Herzfeld used the terms Parsik and Pahlavik to designate epigraphic Middle Persian and Parthian respectively. It seems to have been Henning who reintroduced the term Parthian (1934; cf. 1951).

Today little attention is paid to Silvestre de Sacy's fundamental epigraphic studies, including work on the Middle Persian inscriptions at T®a@q-e Bosta@n and the coin inscriptions of the Sasanian kings. The names of G. F. Grotefend, who, in 1802, took the first important step toward decipherment of Old Persian cuneiform script, and Rawlinson, who completed it except for a few details (1847), are better known.

At Persepolis Niebuhr had recognized three types of cuneiform, all written from left to right, with increasing inventories of characters. As it later turned out, the languages of the three are Old Persian, Elamite, and Babylonian respectively; Old Persian was the first of the three to be deciphered. O. G. Tychsen (pp. 24 ff.) identified the Old Persian word divider. Friedrich Münter (pp. 124 ff.) recognized the Achaemenid origin of the monuments at Persepolis and, drawing on Silvestre de Sacy's work, deciphered the sequence of characters for the Old Persian title "king of kings," though without their phonetic values. Grotefend, combining Silvestre de Sacy's evidence with Herodotus' report that Darius I's father had not been royal (3.70), correctly contrasted the graphic structure of DPa da@rayavauæ xsa@yaƒiya . . . viæta@spahya@ puça "Darius the king, the son of Viæta@spa" (not called "king") with XPe xæay@a@ræa@ xæa@yaƒiya . . . da@rayavahauæ xsa@yaƒiyahya@ puça "Xerxes the king, the son of Darius the king." He was thus able to derive first readings of the proper names in the two inscriptions.

During the next few decades only a few details were added to Grotefend's fundamental achievement (Weissbach, pp. 65-73), but Rawlinson's careful documentation of the B^sotu@n inscription and his admirable decipherment of its Old Persian version vastly increased knowledge of Old Persian and laid the foundation for decipherment of the Babylonian and Elamite versions.

Old Persian.

Apart from B^sotu@n, Old Persian royal inscriptions have been discovered at Pasargadae/Morg@a@b (M), Persepolis (P), Naqæ-e Rostam (N), Susa (S), and Suez (Z). Most are official proclamations carved on stone, but inscriptions on seals, weights, and utensils have also been preserved. Most of these inscriptions are of Darius I and Xerxes (486-465 B.C.E.). Although there are inscriptions of their successors Artaxerxes I, Darius II, and Artaxerxes II and III (qq.v.), most are stereotyped repetitions of well-known patterns and exhibit increasing dissolution of the grammatical rules of Old Persian. A set of gold tablets from Hamada@n, bearing the names of Ariaramnes (ArH) and Arsames (AsH) respectively, are of a later time (see ARIYAÚRAMNA; ARˆAÚMA).

Most Achaemenid inscriptions are trilingual, in Old Persian, Elamite, and Babylonian in that order. Walther Hinz and most other scholars concluded from the Old Persian version of the inscription at B^sot@u@n (DB 4.88-92, par. 70) that the Old Persian script was created on Darius' orders and first used in that inscription. Doubts on this point arose from the trilingual inscription of Cyrus (q.v. iii; 559-530 B.C.E.) the Great at Morg@a@b (CMa), the oldest Achaemenid inscription found in Persia, but the arrangement of the three versions suggests that only the Elamite and Babylonian versions were contemporary and that the Old Persian version was added later (Stronach).

Remains of a hieroglyphic text were found with the trilingual inscription DZc. Much more precious is the trilingual DSab, discovered at Susa, to which five hieroglyphic texts were also added, one of them containing a list of provinces of the Achaemenid empire (Yoyotte). Two fragments of a copy of the Babylonian version were unearthed in Babylon. (On the Aramaic version of the B^sotu@n inscription, see below.)

Babylonian.

Babylonian clay tablets earlier than the Old Persian inscriptions provide important information on the early history of the Achaemenids. On a prism dated in the year 30 of the Assyrian ruler Aææurbanipal (q.v.; 639 B.C.E.) Cyrus, king of Persia (see CYRUS ii), is mentioned as a tributary. Cyrus the Great's revolt against the Median king Astyages (q.v.) and his conquest of Ecbatana (qq.v.) are described in the so-called "Nabonidus Chronicle" (Pritchard, pp. 305-6); the defeat of Nabonidus and the transition from his rule in Babylon to that of Cyrus the Great is alluded to in the abusive "Verse Account of Nabonidus" (Pritchard, pp. 112-15). The earliest Achaemenid imperial document, preserved on the famous Cyrus cylinder and datable after the conquest of Babylon in 539 B.C.E. (Pritchard, pp. 315-16; see CYRUS iv), includes the proclamation of Cyrus the Great as "king of Babylon, king of the countries." Closely related is the Babylonian inscription of Antiochus Soter (q.v.; Pritchard, p. 317) of 268 B.C.E.

Babylonian cuneiform texts include numerous transcriptions of Old Persian proper names and titles, thus contributing to knowledge of Median and Persian onomastics and the structure of the Achaemenid administration (Eilers; Hinz, 1975; Moore). Dates by regnal year attest an uninterrupted sequence of rulers at Babylon from Cyrus I through Cambyses, Nebuchadnezzar, Smerdis, Darius I, Xerxes I, and Artaxerxes I down to the Seleucids and Arsacids.

Elamite.

Achaemenid Elamite played an important part in the court administration at Persepolis. Two assemblages of clay tablets inscribed in Elamite were discovered there by George Cameron: eighty-four "treasury tablets" dated 492-458 B.C.E., recording "disbursements of silver . . . chiefly in lieu of rations in kind" (Hallock, p. 1, after Cameron pp. 83-199), and 2,087 "fortification tablets" dealing with "administrative transfer of food commodities" in 509-458 (Hallock, p. 1). Cameron stressed the significance of both groups for Achaemenid economic and religious history, as well as their archeological and linguistic significance, yet it is mainly the onomastics that have attracted scholarly attention, as a number of Old Persian proper names appear in Elamite, stimulating etymological research (Gershevitch, 1969; Hinz, 1975; Mayrhofer, 1973).

Aramaic.

Fragments of an Aramaic version of the B^sotu@n inscription were found in the collection of papyri from Elephantine (q.v.) in Egypt (Greenfield and Porten), providing evidence that Imperial Aramaic was the language of the central Achaemenid chancery in correspondence with provincial chanceries. Among numerous lapis lazuli mortars and pestles discovered by Schmidt at Persepolis 163 items were inscribed in ink in Aramaic, the inscriptions being of economic, rather than ritual, character, as R. A. Bowman had originally thought. The objects themselves apparently originated in what is now Afghanistan, in Arachosia, which is mentioned on a number of them, but few scholars have doubted that they were inscribed in or near Persepolis. Nevertheless, the use of Aramaic, rather than Elamite, suggests that the inscriptions were added in the province where the objects were produced and collected before delivery to Persepolis. Still unpublished are "a number of brief Aramaic texts incised or written in ink or both, on small clay tablets discovered along with the fortification tablets" (Bowman, p. 19).

Owing to the scarcity of other materials, several imperial Aramaic inscriptions discovered in remote parts of the Achaemenid empire are of interest for Old Persian studies. For example, in the Aramaic version of the Lycian-Greek-Aramaic trilingual inscription from Xanthos in Anatolia the Iranian equivalent of "Apollo and the nymphs" is háætrpty w÷háwrnæ, that is, "Xæaƒrapati and the Ahura@n^s," and in the inscription of Arebsun in Cappadocia the Mazdean religion is described as "sister and wife" of King Bel.

The development of scripts of the Pahlavi type.

Scribes in the provincial chanceries probably developed their own spellings and writing customs as early as the late Achaemenid period. After the fall of the Achaemenid empire different Iranian scripts developed from these provincial styles: Middle Persian, Parthian, Sogdian, and Chorasmian, all of the "Pahlavi type," that is, with varying numbers of Aramaic elements preserved in their spelling. Some of them came over time to be read with their Iranian lexical equivalents, whereas others were replaced by purely Iranian spellings.

The development of the four scripts of Pahlavi type, with their varying orthographic and calligraphic rules, was governed not only by inner dynamics but also probably by a strong tendency among chanceries and scribal schools to distinguish themselves from one another. This tendency resulted, for example, in divergent spellings of the Iranian words for "year-month-day" in dating formulas: Middle Persian ˆNT-BYRH®-YWM, Parthian ˆNT-YRH®÷-YWM, Chorasmian BˆNT-YRH®÷-YWM, and Sogdian ˆNT-YRH®÷-YWM÷ (Tolstov and Livshits, p. 240; in Panj^kant replaced by ˆNT-m÷©-my’ and even by sr’-m÷©-my’); they are to be compared with the date B ÷LWL m÷há ˆNT 10 in the Aramao-Iranian inscription Lag@ma@n II (Davary and Humbach, 1974).

The poorly preserved inscription to the right of the entrance to the tomb of Darius at Naqæ-e Rostam (Frye, 1982) is exceptional in that no non-Iranian word is discernible. The month m÷háy sndrm(t) is all that remains of the date. In line 20 it is possible to discern the Iranian verbal form (. . .)t hyndy; the name ÷rtháæs Artaxerxes shows the typical Persian development of s from Iranian ƒr.

The Aramao-Iranian of As‚oka. Most of the stone edicts of the Mauryan emperor As‚oka (q.v.; ca. 272-231 B.C.E.), intended to propagate Buddhist morals, were composed in slightly varying Middle Indian dialects and written in Indian scripts (in the northwest usually KharosátÂh^, elsewhere Bra@hm^, q.v.). Several As‚oka inscriptions discovered in Arachosia and neighboring areas are written in Greek or Aramao-Iranian, a regional form intermediary between Imperial Aramaic and Pahlavi script, or in Aramao-Indian, that is, Middle Indian written in unvocalized Aramaic script. These inscriptions, consisting of both literal and partly free renderings of passages from the edicts, were evidently addressed to people who followed the tradition of Alexander's chancery in the eastern parts of his former empire, which had been ceded by the Seleucids to Candragupta and his grandson As‚oka. The Aramao-Iranian inscriptions from Taxila (in the Panja@b; Humbach, 1976) and the Lag@ma@n valley (I and II; Dupont-Sommer; Davary and Humbach, 1974) and the Greek inscription from Qandaha@r are monolingual. There are bilingual inscriptions as well: two fragments in Aramao-Indian and Aramao-Iranian, one from Pol-e Daru@nta (Henning 1949-50) and the other from Qandaha@r (II; Benveniste and Dupont-Sommer), and a completely preserved Greek and Aramao-Iranian inscription from Qandaha@r (I; Scerrato et al.; Schlumberger et al.; Pugliese Carratelli and Garbini).

The coexistence of Greek and Aramao-Iranian is apparent also in Parthian leather documents from Avroman (q.v.; 1st century B.C.E.) in Kordesta@n and in the inscriptions from Armazi (q.v.) near Mtskheta in Georgia, one of which is bilingual in Greek and Aramao-Iranian (middle or end of the 2nd century C.E.; Tsereteli, 1942), another monolingual in Aramao-Iranian (end of the 1st century C.E.; Henning, Mitteliranisch, p. 39 fig.; Altheim and Stiehl, pp. 243-61), and a third monolingual in Greek (period of Vespasian, 69-79 C.E.; Tsereteli, 1960).

Middle Persian.

In the Persian-speaking area the transition from Imperial Aramaic to the Pahlavi script of the Sasanian period can be followed on the legends of the early coins of Persis in the 3rd and 2nd centuries B.C.E. (Alram, pp. 162-86). The Frataraka (q.v.) governors like Artaxerxes I (who spelled his name in the Medo-Parthian way as ÷rtháætr, rather than ÷rthæs) still used Aramaic ductus and syntax, for example, in bgdt prtrk÷ zy ÷lhy÷ br bgwrt ("Baya@d, frataraka of the gods, son of Bayward"). Typical Pahlavi spellings did not occur before the rise of the governors to petty kings in the mid-2nd century B.C.E. Aramaic br "son" was then replaced by Pahlavi BRH in d÷ryw MLK÷ BRH wtprdt MLK÷ "King Dare@w [II], son of King Wa@tfrada@t" (Henning, Mitteliranisch, p. 25).

On the first coin issues of the Sasanian kings, in the 3rd century C.E., there is a sharp difference in quality between the neatly engraved portraits and the confusing paleography of the legends. The die engravers apparently had difficulty reproducing the contemporary lapidary script. Late Sasanian legends increasingly exhibit cursive features, particularly from the time of Khosrow I (531-79) and after. Leaving aside certain graphic archaisms like logograms for mint names, there seems to have been a straight line of development toward the ductus on the coin legends of the early Arab governors of T®abaresta@n (Unvala).

For unclear, perhaps technical, reasons the trilingual inscription of Ardaæ^r ANRm at Naqæ-e Rostam (Plate I) follows the sequence Parthian, Greek, Middle Persian, whereas the inscription ANRm£ on the relief of Ahura Mazda@ on the same monument provides the canonical sequence Middle Persian, Parthian, Greek, which recurs on ˆa@pu@r I's inscription ˆNRb. In ˆa@pu@r's great trilingual inscription ˆKZ, cut into the outer walls of the base of the Ka¿ba-ye Zardoæt, the Middle Persian version is on the left side, the Parthian on the right, and the Greek on the back of the building. This inscription, which provides much information on the history and inner structure of the Sasanian empire, has also contributed materially to knowledge of inscriptional Middle Persian and Parthian. The Greek version includes various renderings of such Middle Iranian proper names as krtyr/kltyr. When referring to the famous high priest, it is rendered as Greek Karteir, revealing a deliberately archaizing pronunciation, but, when referring to another person, the Greek rendering is Kirdeir, adapted to contemporary Middle Persian phonetics. Owing to religious principles, the high priest Karde@r's inscriptions at Sar-Maæhad (KSM), Naqæ-e Rostam (KNRm), Naqæ-e Rajab (KNRb), and the Ka¿ba-ye Zardoæt (KKZ, immediately below the Middle Persian version of ˆKZ) are in Middle Persian only. In this connection it is worth mentioning that the monolingual Middle Persian inscription of Abnu@n, found at Barm-e Delak (q.v.) and dated in the year 3 of ˆa@pu@r's reign, shows a composition very similar to that of Karde@r's NRb (Skjœrvø).

Two of six rectangular recesses hewn into the rock at H®a@j^a@ba@d in Fa@rs are covered respectively by Middle Persian and Parthian versions of an inscription (ˆH), in which a shooting by ˆa@pu@r is described (MacKenzie). It is not clear whether the remaining four recesses were destined for further versions of the same text, for documentation of similar feats, or both. At any rate, the parallel "shooting inscription" at Tang-e Bora@q (ˆTBq; Gropp, in Hinz, 1969, pp. 229-37) is also only bilingual, as is ˆa@pu@r's inscription at B^æa@pu@r (q.v.; ˆVˆ).

The monumental Middle Persian and Parthian inscription at Paikuli (NPi; Herzfeld, 1924; Humbach and Skjœrvø) is a description of events in connection with Narseh's accession to the throne in 293 and is thus an extraordinary document on Sasanian political practice. There is no Greek version, and Parthian appeared for the last time. In his inscription at B^æa@pu@r (NVˆ) Narseh dispensed with Parthian, as did ˆa@pu@r II and Sa@pu@r III at T®a@q-e Bosta@n (ˆTBn 1, ˆTBn 2) and the author of the inscription at Meæk^næahr in Azerbaijan (ˆMS).

A cursive ductus closer to that of Book Pahlavi appears on private stone inscriptions of the later Sasanian and early Islamic periods. Among them horticultural work is recorded at Tang-e K¨oæk (TX), but most are dakòma (places for exposure of the dead), funerary, or commemorative inscriptions like those found at Eql^d, Takòt-e Ta@u@s/EsátÂakòr, Maqsáu@da@ba@d, B^æa@pu@r, Tang-e Jelow, Shah Esma@¿^l, Ka@zeru@n, Ba@g@-e Larda, Darband (q.v.), Istanbul, and Sian in Shensi province (Gropp, in Hinz, 1969, pp. 257-60; idem, 1970; idem 1975; Kasumova, 1979; idem, 1988; Tafazµzµol^; Humbach, 1988). The tomb inscription from Sian (874 C.E.) is in Chinese and Pahlavi but is not really bilingual: The Chinese text follows the Chinese pattern of commemorating an official (perhaps of Iranian descent), whereas the Pahlavi text dedicated to his wife is typically Mazdean in character (Humbach, 1988). A bilingual construction text in Arabic and Pahlavi from Qal¿a Bahman in Fa@rs was published by Ali Hassuri, but his reading of the dates should be reconsidered. An Arabic and Pahlavi construction text from Ra@dka@n in Ma@zandara@n (ca. 411/1020-21; Blair, no. 31) is a late offshoot of the Pahlavi tradition of T®abaresta@n, as is a similar text from Rasget in the same region, in which the Pahlavi is written in the same decorative paleography as the Arabic (Bivar and Yarshater, pls. 35-40).

The 199 Pahlavi ostraca reproduced by Jean de Menasce (1957) and deciphered by Dieter Weber, most of them collected by Herzfeld near Vara@m^n and datable to the 6th century, are written in a cursive shorthand closely related to that found on the papyri from Egypt. The excavations at Dura Europos (q.v. ii) also yielded Pahlavi dipinti, one ostracon, and several graffiti.

A large number of Sasanian seals has survived, many inscribed with uncial or cursive script, though the characters are often decomposed into their single elements. Richard N. Frye (1960) distinguished three main types. The common type of private seal includes conventional figures or monograms, to which proper names, religious slogans, or both have been added (e.g., ÷pst÷n ¿L yzd÷n "(I take) refuge in the gods," YWM ˆPYR "(May my?) day (be) good," l÷st or l÷sty "truthful, righteous," l÷styhy "truth, rectitude"). Of greater historical value is the second type, personal seals belonging to officials and dignitaries, whose inscriptions expressed political power (e.g., whwdynæhpwhry ZY ÷nblkpty "Wehde@næa@buhr, master of the storehouse," w÷l÷n ZY mgw ¿L ÷twlplnbg÷n "Wa@ra@n, the magus (assigned?) to AÚdurfarnbaya@n"). The third type consists of seals of office with no figures but only writing, usually with a place name plus the office but no personal name (e.g., wh÷lthætl ætldst÷n mgwh "the 'magushood' of the city of Wehardaxæahr/Wehardaæe@r"). A special group consists of magic seals, which sometimes contain relatively long inscriptions (Gyselen, 1995). Extensive collections of available material have been published by Philippe Gignoux and Rika Gyselen (Gignoux, 1978; Gignoux and Gyselen; Gyselen, 1993; for bullae, usually found in multiples on single items, see Frye, 1973; Göbl, 1976).

Parthian.

Unlike the coin legends of the Frataraka governors and petty kings of Persis, those of the Arsacids are of limited epigraphic value. Until the time of Vologases I (51-77 C.E.) they were exclusively in Greek; then one or two Parthian letters were added, but full Parthian legends did not appear before Mithridates IV (ca. 130), and even then they were graphically unsatisfactory. With rare exceptions they included no ruler's personal name, and years are rare (Sellwood; Alram, pp. 121-37). The Gotarse@s Geopothros mentioned in a Greek inscription on a relief at Sar-e Pol in Kordesta@n (Gropp, 1986) recalls the Gotarze@s mentioned on the relief of Mithridates II (123-87 B.C.E.) hewn into the rock at B^sotu@n (Silvestre de Sacy, 1815, p. 191, pl. I) but now destroyed.

Outstanding among epigraphic sources are the ostraca discovered at Nisa (in Turkmenistan), the ancient royal capital of the Parthians. There are a total of 1,448 items inscribed with black ink (D'yakonov and Livshits, 1976-79). Many of them bear dates in the Arsacid era corresponding to years of the 1st century B.C.E. Similar materials were later excavated at other sites in Turkmenistan (Livshits, 1977, pp. 157-58; idem, 1984b). In Persia only one Parthian ostracon has come to light, at ˆahr-e Qu@mes west of Da@mg@a@n (Bivar, 1970), but more have been discovered at Dura Europos (q.v. ii).

Other Parthian materials include the inscription at Kal-e Jangal, the rock inscription of one Vologases at B^sotu@n, and a small relief, perhaps of the same ruler (Gropp, 1970, pp. 200-201, pl. 101/1). The only pre-Sasanian Parthian stone inscriptions of some relevance are the bilingual (Parthian and Greek) inscription of Arsaces Vologeses of Mesene, son of Mithridates (Morano), and that of Artabanus V, discovered at Susa, both dated in the year 462 of the Seleucid era (151 C.E.). The latter was discussed by Henning (1952) in connection with the Elymaic inscriptions at Tang-e Sarva@k in K¨u@zesta@n (2nd century C.E.) and Elymaic coins.

Chorasmian.

Whereas books in Late Chorasmian are written in Arabic script, Middle Chorasmian inscriptions (less precisely called "Old Chorasmian") are written in a script of the Pahlavi type (see CHORASMIA iii). Two small early inscriptions on clay vessels from the site of Koy-Krylgan Kala are tentatively attributed to the 2nd or 1st century B.C.E. (Livshits, 1968, p. 435). Middle Chorasmian is, however, attested mainly through coin legends (Va¥nberg), a large number of dated inscriptions in ink on ossuaries excavated at Tok Kala (Tolstov and Livshits; Gudkova; Henning, 1965), and inscriptions on silver vessels, some of them dated (Henning, 1965, p. 167). The dates are apparently in the same unknown era as those on the documents of wood and leather from Toprak Kala (Livshits, 1984a).

Sogdian.

Paleographically the Sogdian coins fall into two groups, the smaller with legends in Bukharan script, the larger with legends in Samarkand uncial script (Frye, 1949; Smirnova, 1963). Although the Bukharan script is attested only fragmentarily on these coins and a few utensils (Livshits, Kaufman, and D'yakonov), the Samarkand variant is well known, particularly from Sogdian Buddhist manuscripts written in the calligraphic "sutra style" and juridical and economic manuscripts from Mount Mugh (Panj^kant). An earlier stage of development can be recognized from the Sogdian letters found at Dunhuang (in Xinjiang) and is epigraphically attested in most of the more than 600 short rock inscriptions discovered by Karl Jettmar on the Upper Indus (Sims-Williams).

Notable inscriptions in cursive ductus include one found in the Ladakh region of the Himalayas (Müller; Klyashtorny¥ and Livshits, 1972, p. 83 no. 3); the Old Turkish-Sogdian-Chinese inscription of Karabalgasun in the Orhon region of Mongolia (Hansen; Yoshida; Hamilton); the Sogdian-Turkish inscriptions from Sevre¥ Somon on the southern border of the Gobi desert and Bugut in Mongolia (Klyashtorny¥ and Livshits, 1971; idem, 1972); and twenty-five small inscriptions (ostraca and so on) from the temple at Panj^kant (Livshits and Shkoda 1982).

Bactrian.

The only unequivocal trace of the Aramaic tradition found on the former territory of Bactria is a short inscription from AÚy K¨a@nom (q.v.) south of the Oxus (Livshits, 1977, pp. 166 n. 15, 167; Livshits and D'yakonov, in Rapin, 1992, p. 105), whereas the Greek tradition is well attested there (Huyse, pp. 113-15; Rapin, pp. 95-121, 387-92). The coins of the Greco-Bactrian rulers have legends in Greek, and even from north of the Oxus there is a Greek dedication to the deified river, dating from the 2nd century B.C.E. (Litvinski¥, Vinogradov, and Pichikiyan). Inscriptions in the Bactrian language were written in a variant of the Greek alphabet.

In the Indo-Iranian borderlands numerous coins with inscriptions in more or less correct Greek were issued at the beginning of the common era by such rulers as the Indo-Parthians Azes, Azilises (qq.v.), and Hermaios. From the epigraphic point of view, the coins of the Kushan dynasty of Bactria are the most noteworthy. The copper issues of Kujula Kadphises include bilingual inscriptions in corrupt Greek and Middle Indian, but those on the gold coins of his successor, Vima Kadphises, are quite correct. Bilingualism was abandoned a few decades later by Kanishka, whose first issues are inscribed only in Greek but who later changed to exclusive use of Bactrian language and script.

Thanks to the efforts of French and Soviet archeologists in particular, a substantial number of Bactrian inscriptions have been discovered. An outstanding epigraphic task is the decipherment of the inscription of a Kushan king named Vima (Ooe@mo Taktio) dated in the year 279 of the Seleucid era, month Gorpiaios, found at Daæt-e Na@wor (q.v.), of which only poor photographs survive (Davary and Humbach, 1976). The Bactrian text is accompanied by some lines in KharosátÂh^ and some in the same unknown script attested from Issyk and AÚy K¨a@nom (Vertogradova).

A similar date (279 or 275) occurs in an unfinished inscription from Sorkò Kotal, consisting of the beginning of the first line on a large monolith. The sanctuary was perhaps beginning to fall into disrepair in that year. Later reconstruction work, begun in the year 31 of an unknown era, is recorded in a monumental inscription preserved in three parallel versions, which, however, differ noticeably in their final lines. The earliest is version B (thirty-one blocks inscribed rather carelessly); the next is A (twenty-one blocks; for correct rearrangement of the blocks of the two versions, see Humbach, 1966-67, II, pls. 17-19). Both were reused in construction of a well (Gershevitch, 1979; idem, 1980). Version M, preserved on a monolith, includes additions in the last lines that are not present in B and A. M is thus the final and most authoritative version. Although it is carefully inscribed, decipherment is complicated by unsystematic writing of final vowels and perhaps by omission of certain simple words because of constraints on space, which has resulted in the often contradictory evidence about which Henning complained (1960, p. 47). The so-called "Palamedes inscription" from Sorkò Kotal provides a combination of a Bactrian and a short Greek text. A new long monumental inscription, giving the genealogy of Kanishka, was made known by Nicholas Sims-Williams at the conference of the Societas Iranica Europaea in Cambridge in September 1995 (P. O. Skjœrvø, personal communication).

In the area between the Hindu Kush and the Oxus Soviet archeologists have unearthed a monumental inscription at Delbarj^n (q.v.; Livshits and Kruglikova). Notable findings made north of the Oxus include the graffiti from Kara Tepe in Tajikistan, written in an early cursive ductus, one dated year 35 of an unknown era, another year 97 (Livshits, 1969; Harmatta, 1969; Humbach, 1970), as well as the monumental inscription from A¥rtam dated in the year 4 of Huvishka (Turgunov, Livshits, and Rtveladze 1981)

The lapidary script of the Bactrian stone inscriptions and coins seems to have been influenced by the cursive ductus attested at Kara Tepe. After the Sasanian conquest the cursive script itself was revived in Bactrian inscriptions on coins produced by the Sasanian Kushanshahs (Humbach 1966-67, I, pp. 50-52), followed by the Huns (pp. 53-59) and the Turkish rulers of Kabul and Gandhara (pp. 59-66). Among the latter, whom Robert Göbl took to be Huns as well (1967, I, p. 1), there are several issues with bilingual and trilingual legends in Bactrian, Pahlavi, and Sanskrit.

The latest known Bactrian inscriptions have been discovered in the Tochi valley (Waz^resta@n) in a group comprising an Arabic-Sanskrit inscription (ITAS) dated in the Islamic era 243/857, Sanskrit year 32 (= Laukika 3932 = 857 C.E.); a Sanskrit-Bactrian inscription (ITSB), with Sanskrit date 38 (= Laukika 3938 = 863 C.E.) and Bactrian 632, a synchronism suggesting that the Bactrian era began in 231 C.E.; an Arabic-Bactrian inscription (ITAB) dated in Bactrian 635 (866 C.E.); and a Bactrian inscription with traces of an unidentifiable script (ITBX) dated in Bactrian 636 (867 C.E.). Whereas the three Bactrian texts, written in cursive script, are the latest evidence for the epigraphic use of the Bactrian language, the two Arabic versions seem be the earliest evidence of the language of the conquerors in the Indo-Iranian borderlands (Humbach, 1994). See also A›OKA.

Bibliography: M. Alram, Nomina Propria in Nummis, Iranisches Personennamenbuch 4, 1986. F. Altheim and R. Stiehl, Die aramäische Sprache unter den Achämeniden I, Frankfurt, 1963. A.-H. Anquetil du Perron, tr., Zend-Avesta, 2 vols. in 3, Paris, 1771. É. Benveniste, Titres et noms propres en iranien ancien, Paris, 1966. Idem and A. Dupont-Sommer, "Une inscription indo-arame‚enne d'As‚oka provenant de Kandahar (Afghanistan)," JA 254, 1966, pp. 437-65. A. D. H. Bivar, ed., Kushan and Kushano-Sasanian Seals and Kushano-Sasanian Coins: Sasanian Seals of the British Museum, Corpus Inscr. Iran., London, 1968. Idem, Catalogue of the Western Asiatic Seals in the British Museum. Stamp Seals II. The Sassanian Dynasty, London, 1969. Idem, "The First Parthian Ostracon from Iran," JRAS, 1970, pp. 63-66. Idem and E. Yarshater, eds., Eastern Ma@zandara@n I, Corpus Inscr. Iran., London, 1978. S. Blair, The Monumental Inscriptions from early Islamic Iran and Transoxiana, Leiden, 1991. R. Borger und W. Hinz, "Die Behistun-Inschrift Darius' des Grossen," in O. Kaiser, ed., Texte aus der Umwelt des Alten Testaments I: Rechts- und Wirtschaftsurkunde, historisch-chronologische Texte, Gütersloh, Germany, 1984, pp. 419-50. R. A. Bowman, Aramaic Ritual Texts from Persepolis, Chicago, 1970. G. C. Cameron, Persepolis Treasury Tablets, Chicago, 1948. G. D. Davary, Baktrisch: Ein Wörterbuch, Heidelberg, 1982; review H. Humbach, Kratylos 28, 1983, pp. 89-95. Idem and H. Humbach, Eine weitere aramäoiranische Inschrift des As‚oka aus Afghanistan, Mainz and Wiesbaden, 1974. Idem, Die baktrische Inschrift IDN 1 von Dasht-e Na@wu@r (Afghanistan), Mainz and Wiesbaden 1976. J. A. Delaunay, "L'arame‚en d'empire et les de‚buts de l'e‚criture en Asie centrale," Acta Iranica 2, 1974, pp. 225-36. A. Dupont-Sommer, "Une nouvelle inscription arame‚enne d'Asoka trouve‚e dans la valle‚e du Laghma@n (Afghanistan)," Comptes Rendus de l'Acade‚mie des Inscriptions et Belles-Lettres, 1970, pp. 158-73. I. M. D'yakonov and V. A. Livshits, Dokumenty iz Nisy I v. do n.e. (Predvaritel'nye itogi raboty) (Documents from Nisa of the 1st century B.C.E. [Preliminary summary of the work]), Moscow, 1960. Idem, Parthian Economic Documents from Nisa, ed. D. N. MacKenzie, 4 vols., London 1976-79. W. Eilers, Iranische Beamtennamen in der keilschriftlichen Überlieferung I, Leipzig, 1940.

R. N. Frye, Notes on the Early Coinage of Transoxiana, New York, 1949. Idem, "Additional Notes 1," The American Numismatic Society Museum Notes 4, 1950, pp. 105-14. Idem, "Die Legenden auf sasanidischen Siegelabdrücken," WZKM 56, 1960, pp. 32-35. Idem, "Additional Notes 2," The American Numismatic Society Museum Notes 7, 1962, pp. 231-38. Idem, ed., The Parthian and Middle Persian Inscriptions of Dura-Europos. Corpus Inscr. Iran., London, 1968a. Idem, "Sasanian Clay Sealings in the Collection of Mohsen Foroughi," Iranica Antiqua 8, 1968b, pp. 118-32. Idem, "Funerary Inscriptions in Pahlavi from Fars," in M. Boyce and I. Gershevitch, eds., W. B. Henning Memorial Volume, London, 1970, pp. 152-56. Idem, ed., Sasanian Seals in the Collection of Mohsen Foroughi, Corpus Inscr. Iran., London, 1971. Idem, ed., Sasanian Remains from Qasr-i Abu Nasr. Seals, Sealings, and Coins, Cambridge, Ma., 1973. Idem, "The 'Aramaic' Inscription on the Tomb of Darius," Iranica Antiqua 17, 1982, pp. 85-89. I. Gershevitch, "Amber at Persepolis," in Studia Classica et Orientalia Antonino Pagliaro Oblata II, Rome, 1969, pp. 167-251. Idem, "Nokonzoko's Well," Afghan Studies 2, 1979, pp. 55-67. Idem, "The Colophon of the NOKONZOK Inscription," AAASH 28, 1980, pp. 179-84.

P. Gignoux, Glossaire des inscriptions pehlevies et parthes, Corpus Inscr. Iran., London, 1972. Idem, Catalogue des sceaux, came‚es et bulles sassanides de la Bibliotheàque Nationale et du Muse‚e du Louvre II. Les sceaux et bulles inscrits, Paris, 1978. Idem, "D'Abnu@n aà Ma@ha@n: Étude de deux inscriptions sassanides," Stud. Ir. 20, 1991, pp. 9-22. Idem and R. Gyselen, Sceaux sassanides de diverses collections prive‚es, Louvain, 1982. R. Göbl, Dokumente zur Geschichte der iranischen Hunnen in Baktrien und Indien, 4 vols., Wiesbaden, 1967. Idem, Die Tonbullen vom Tacht-e Suleiman, Berlin, 1976. J. C. Greenfield and B. Porten, The Bisitun Inscription of Darius the Great. Aramaic Version, Corpus Inscr. Iran., London, 1982. G. Gropp, "Bericht über eine Reise in West- und Südiran," AMI, N.F. 3, 1970, pp. 173-230. Idem, "Die Derbent-Inschriften und das Adur Gusnasp," in Monumentum H. S. Nyberg I, Acta Iranica 4, Leiden and Tehran, 1975, pp. 317-31. Idem, "Die parthische Inschrift von Sar-Pol-e Z¨oha@b," ZDMG 118, 1986, 315-19. G. F. Grotefend, "Praevia de Cuneatis quae Vocant Inscriptionibus Persepolitanis Legendis aut Explicandis Relatio," ed. W. Meyer, Nachrichten der königlichen Gesellschaft der Wissenschaften in Göttingen, 1893, pp. 571 ff. A. V. Gudkova, Tok-Kala, Tashkent, 1964. R. Gyselen, Catalogue des sceaux, came‚es et bulles sassanides de la Bibliotheàque Nationale et du Muse‚e du Louvre I: Collection ge‚ne‚rale, Paris, 1993. Idem, Sceaux magiques en Iran sassanide, Stud. Ir. cahier 17, Paris, 1995.

R. T. Hallock, Persepolis Fortification Tablets, Chicago, 1969. J. Hamilton, "L'inscription trilingue de Qara Balgassun d'apreàs les estampages de Bouillane de Lacoste," in A. Haneda, ed., Documents et archives provenant de l'Asie centrale, Kyoto, 1990, pp. 125-33. O. Hansen, "Zur sogdischen Inschrift auf dem dreisprachigen Denkmal von Karabalgassun," Journal de la Socie‚te‚ Finno-Ougrienne 44/3, 1930, pp. 3-39. J. Harmatta, "The Bactrian Wall-Inscriptions from Kara-Tepe," in B. Ya. Staviski¥, ed., Kara-Tepe II, Moscow, 1969, pp. 82-125. A. Hassuri, "Two Unpublished Pahlavi Inscriptions: I. A Pahlavi Inscription from Kazirun; II. A Pahlavi Inscription from Qal¿a Bahman," ZDMG 134, 1984, pp. 92-97. W. B. Henning, Mitteliranische Manichaica III, Berlin, 1934. Idem, "The Aramaic Inscription of As‚oka Found in Lampa@ka," BSOAS 13, 1949-50, pp. 80-88. Idem, Kat^baha@-ye pahlav^, Yag@ma@ 3/6, 1329 ˆ./1950. Idem, "Decipherment of Pahlavi Inscriptions: A Feat of Genius," Bulletin bimestriel de la Commission Nationale Iranienne pour l'UNESCO 3/3-4, 1951, pp. 52-43 (sic). Idem, "The Monuments and Inscriptions of Tang-i-Sarvak," Asia Major, N.S. 2, 1952, pp. 151-78. Idem, "The Bactrian Inscription," BSOAS 23, 1960, pp. 47-55. Idem, "The Choresmian Documents," Asia Major 11/2, 1965, pp. 75-87. E. Herzfeld, Paikuli, 2 vols., Berlin, 1924. Idem, Kushano-Sasanian Coins, Calcutta, 1930. G. F. Hill, Catalogue of the Greek Coins of Arabia, Mesopotamia and Persia, London, 1922; repr. Bologna, 1965. W. Hinz, Altiranische Funde und Forschungen, Berlin, 1969. Idem, Altiranisches Sprachgut der Nebenüberlieferung, Wiesbaden, 1975. Idem and H. Koch, Elamisches Wörterbuch, 2 vols., Berlin, 1987. H. Humbach, Baktrische Sprachdenkmäler, 2 vols., Wiesbaden, 1966-67. Idem, Die aramäische Inschrift von Taxila, Mainz and Wiesbaden, 1969. Idem, "Kara Tepe—Tochi—Surkh Kotal," MSS 28, 1970, pp. 43-50. Idem, "The Aramaic As‚oka Inscription from Taxila," in German Scholars on India. Contributions to Indian Studies II, Bombay, 1976, pp. 118-30. Idem, "Die pahlavi-chinesische Bilingue von Xi'an," in J. Duchesne-Guillemin and D. Marcotte, eds., A Green Leaf/Barg-e sabz: Papers in Honour of Professor Jes P. Asmussen, Acta Iranica 29, Leiden, 1988, pp. 73-82. Idem, "Bactrian kidabo, odabo, kaldabo, malabo/malbo," in Études irano-aryennes offertes aà Gilbert Lazard, Paris, 1989, p. 209-15. Idem, "The Tochi Inscriptions," Stud. Ir. 19, 1994, pp. 137-56. Idem and P. O. Skjœrvø, The Sassanian Inscription of Paikuli, 3 pts. in 4 vols., Wiesbaden, 1978-83. P. Huyse, "Die Begegnung zwischen Hellenen und Iranern. Griechische epigraphische Zeugnisse von Griechenland bis Pakistan," in C. Reck and P. Zieme, eds., Iran and Turfan: Beiträge Berliner Wissenschaftler, Werner Sundermann zum 60. Geburtstag gewidment, Wiesbaden, 1995, pp. 99-126.

O. Kaiser, ed., Texte aus der Umwelt des Alten Testaments I: Rechts- und Wirtschaftsurkunde, historisch-chronologische Texte, 4 vols., Gütersloh, Germany, 1982-85. S. Yu. Kasumova, "K tolkovaniyu srednepersidskikh nadpise¥ na Derbenta" (On the interpretation of the Middle Persian inscriptions at Darband), VDI, 1979/1, pp. 113-26. Idem, "Novye nakhodki v Derbente" (New finds in Darband), VDI, 1988/1, pp. 88-95. S. Klyashtorny¥ and V. A. Livshits, "Sevre¥ski¥ kamen'" (The Sevre¥ stone), Sovetskaya Turkologiya 1971/3, pp. 106-12. Idem, "The Sogdian Inscription of Bugut Revised," AAASH 26, 1972, pp. 69-102. B. A. Litvinski¥, Y. G. Vinogradov, and J. R. Pichikiyan, "Votiv Atrasoka iz khrama Oksa v severno¥ Baktrii" (The votive inscription of Atrasoke@s from an Oxus sanctuary in northern Bactria), VDI, 1985/4, pp. 84-110. V. A. Livshits, "Koy-Krylgan Kala," AAASH 16, 1968, pp. 435-46. Idem, "K otkrytiyu baktri¥skikh nadpise¥ na Kara-Tepe" (On the discovery of the Bactrian inscriptions at Kara Tepe), in B. Ya. Staviski¥, ed., Kara-Tepe II, Moscow, 1969, pp. 47-81. Idem, "New Parthian Documents from South Turkmenistan," AAASH 25, 1977, pp. 157-85. Idem, "Dokumenty" (The documents), in Yu. A. Rapoport and E. E. Nerazik, eds., Toprak-Kala II, Trudy khorezmsko¥ arkheologo-etnografichesko¥ ekspeditsii 14, Moscow, 1984a, pp. 251-86. Idem, "Novye parfyanskie nadpisi iz Turkmenii i Iraka" (New Parthian inscriptions from Turkmenia and Iraq), Epigrafika Vostoka 21, 1984b, pp. 18-40. Idem and I. M. D'yakonov, "Inscription arame‚enne," in C. Rapin, Fouilles d'Aï Khanoum VIII, MDAFI 23, Paris, 1992, p. 105. Idem, K. V. Kaufman, and I. M. D'yakonov, "O drevne¥ sogdi¥sko¥ pis'mennosti Bukhary" (On the ancient Sogdian literature of Bukhara), VDI 1954/1, pp. 150-63. Idem and I. T. Kruglikova, "Fragmenty baktri¥sko¥ monumental'no¥ nadpisi iz Dil'berdzhina" (Fragments of a Bactrian monumental inscription from Delbarj^n), in I. T. Kruglikova, ed., Drevnyaya Baktriya (Ancient Bactria) II, Moscow, 1979, pp. 98-112. Idem and V. G. Lukonin, "Srednepersidskie i sogdi¥skie nadpisi na serebryanykh sosudakh" (Middle Persian and Sogdian inscriptions on silver vessels), VDI, 1964/3, pp. 155-76. Idem and M. M. Mambetullaev, "Ostrak iz Khumbuz-Tepe" (Ostraca from K¨omboz Tappa), in F. Girs, E. A. Davidovich, and M. Osmanov, eds., Pamyatniki istorii i literatury vostoka (Historical and literary documents of the east), Moscow, 1986, pp. 34-45. Idem and I. Sh. Shifman, "K tolkovaniyu arame¥skikh nadpise¥ Ashoki" (On the interpretation of the Aramaic inscriptions of As‚hoka), VDI, 1977/2, pp. 7-24. Idem and V. G. Shkoda, "Sogdi¥skie nadpisi iz khrama I veka v Pendzhikente" (Sogdian inscriptions from the first-century temple in Panj^kant), Narody Azii i Afriki, 1982/5, pp. 131-41.

D. N. MacKenzie, "Shapur's Shooting," BSOAS 41, 1978, pp. 498-511. A. Maricq, "Res Gestae Divi Saporis," in Classica et Orientalia, Paris, 1965, pp. 37-75. M. Mayrhofer, Onomastica Persepolitana: Das altiranische Namengut der Persepolis-Täfelchen, Vienna, 1973. Idem, Supplement zur Sammlung der altpersischen Inschriften, Vienna, 1978. A. Meillet, "Sur les mots iraniens emprunte‚s par l'arme‚nien," MSL 17, 1911/12, pp. 242-50. J. de Menasce, "Inscriptions pehlevies en e‚criture cursive," JA 244, 1956, pp. 423-31. Idem, ed., Ostraca and Papyri, Corpus Inscr. Iran., London, 1957. M. Mitchener, The Early Coinage of Central Asia, London 1973. E. W. Moore, Neo-Babylonian Business and Administrative Documents, Ann Arbor, Mich., 1935. E. Morano, "Contributi all'inter pretazione della bilingue greco-partica dell'Eracle di Seleucia," in G. Gnoli and A. Panaino eds., Proceedings of the First European Conference of Iranian Studies . . . I. Old and Middle Iranian Studies, Serie Orientale Roma 67/1, Rome, 1990, pp. 229-38. F. W. K. Müller, "Eine sogdische Inschrift in Ladakh," SPAW, Phil.-hist. Kl., 31, 1925, pp. 371-72. F. C. C. H. Münter, Versuch über die keilförmigen Inschriften zu Persepolis, Copenhagen, 1802. C. Niebuhr, Carsten Niebuhrs Reisebeschreibung nach Arabien und andern umliegenden Ländern II, Copenhagen, 1778. H. S. Nyberg, "The Pahlavi Documents from Avroman," MO 17, 1923, pp. 182-230. Idem, "The New Corpus Inscriptionum Iranicarum," BSOAS 23, 1960, pp. 40-46. Idem, The "Pahlavi Inscription at Mishkin," BSOAS 33, 1970, pp. 144-53.

I. A. Pichikiyan, Kul'tura Baktrii (The culture of Bactria), Moscow, 1991. J. B. Pritchard, Ancient Near Eastern Texts Relating to the Old Testament, Princeton, N.J., 1950. G. Pugliese Carratelli and G. Garbini, A Bilingual Graeco-Aramaic Edict by As‚oka: The First Greek Inscription Discovered in Afghanistan, Rome, 1964. C. Rapin, Fouilles d'Aï Khanoum VIII, MDAFI 23, Paris, 1992. H. C. Rawlinson, "The Persian Cuneiform Inscriptions at Behistun," JRAS 10, 1847, pp. 187-349. U. Scerrato et al., Un editto bilingue greco-aramaico di As‚oka, Rome, 1958. D. Schlumberger, "Une nouvelle inscription grecque d'Açoka," Comptes Rendus de l'Acade‚mie des Inscriptions et Belles-Lettres, 1964, pp. 126-40. Idem et al., "Une bilingue gre‚co-arame‚enne d'As‚oka," JA 246, 1958, pp. 1-48. E. F. Schmidt, Persepolis, 3 vols., Chicago, 1953-70. R. Schmitt, Altpersische Siegel-Inschriften, Vienna, 1981. Idem, The Bisitun Inscriptions of Darius the Great: Old Persian Text, Corpus Inscr. Iran., London, 1991. D. Sellwood, An Introduction to the Coinage of Parthia, 2nd ed., London, 1980. A. I. Silvestre de Sacy, Me‚moires sur les diverses antiquite‚s de la Perse, et sur les me‚dailles des rois de la dynastie des Sassanides, Paris, 1793. Idem, "Me‚moire sur les monumens et les inscriptions de Kirmanschah et de Bi-sutoun, et sur divers autres monumens Sassanides," Histoire et me‚moires de l'Institut Royal de France, Classe de l'histoire et de litte‚rature ancienne 2, Paris, 1815, pp. 162-242. N. Sims-Williams, Upper Indus, 2 vols., Corpus Inscr. Iran., London 1989-92. P. O. Skjœrvø, "L'inscription d'Abnu@n et l'imparfait en moyen-perse," Stud. Ir. 21/2, 1992, pp. 154-60. O. I. Smirnova, Katalog monet gorodishcha Pendzhikent (Catalogue of the coins of the ancient settlement at Panj^kant), Moscow, 1963. Idem, "Numizmaticheskie zametki" (Numismatic notes), Epigrafika Vostoka 18, 1967, pp. 34-40. B. Ya. Staviski¥, ed., Kara-Tepe, 5 vols., Moscow, 1969. D. Stronach, "On the Genesis of the Old Persian Cuneiform Script," in F. Vallat, ed., Contribution aà l'histoire de l'Iran: Me‚langes offerts aà Jean Perrot, Paris, 1990, pp. 195-203.

A. Tafazzoli, "L'inscription fune‚raire de Ka@zerun II (Par^sa@n)," Stud. Ir. 20/2, 1991, pp. 197-202 (with an appendix on the inscription from Takòt-e Ta@u@s IV). E. Thomas, "Early Sasanian Inscriptions, Seals and Coins," JRAS, N.S. 3, 1868, pp. 241-358. S. P. Tolstov and V. A. Livshitz, "Decipherment and Interpretation of the Khwarezmian Inscriptions from Tok Kala," AAASH 12, 1964, pp. 231-51. G. V. Tsereteli, "Armazis bilingva: A Bilingual Inscription from Armazi near Mcªeta in Georgia," Bulletin de l'Institut Marr de langues, d'histoire et de la culture mate‚rielle 13, 1942, pp. 49-83. Idem, "Epigraficheskaya nakhodka v Mcheta—Drevne¥ stolitse Gruzii" (An epigraphic find in Mtskheta—Ancient capital of Georgia), VDI 1948/2, pp. 49-57. Idem, "Grecheskaya nadpis' epokha Vespasyana iz Mcªeta" (A Greek inscription of the period of Vespasian from Mtskheta), VDI 1960/2, pp. 125-33. B. A. Turgunov, V. A. Livshits, and E. V. Rtveladze, "Otkrytie baktrisko¥ monumental'no¥ nadpisi v Airtame" (The discovery of Bactrian monumental inscriptions at A¥rtam), Obshchestvennye nauki v Uzbekistane, 1981/3, pp. 29-48. O. G. Tychsen, De Cuneatis Inscriptionibus Persepolitanis Lucubratio, Rostock, 1798. J. M. Unvala, Coins of Tabarista@n and Some Sassanian Coins from Susa, Paris, 1938. B. J. Va¥nberg, Monety drevnego Khorezma (Coins of ancient Chorasmia), Moscow, 1977. V. V. Vertogradova, "Nakhodka nadpisi neizvestnym pis'mom na Kara-Tepe" (Discovery of an inscription in unknown characters at Kara Tepe), in B. Ya. Staviski¥, ed., Kara-Tepe V, Moscow, 1982, pp. 160-67. E. N. von Voigtlander, The Bisitun Inscription of Darius the Great: Babylonian Version, Corpus Inscr. Iran., London, 1978. D. Weber, Ostraca, Papyri und Pergamente, Corpus Inscr. Iran., London, 1992. F. H. Weissbach, "Die altpersischen Inschriften," in Grundriss II, pp. 54-74. Y. Yoshida, "Some New Readings of the Sogdian Version of the Kara Balgasun inscription," in A. Haneda, ed., Documents et archives provenant de l'Asie centrale, Kyoto, 1990, pp. 117-23. J. Yoyotte, "Une statue de Darius de‚couverte aà Susa 3: Les inscriptions hie‚roglyphiques," JA 260, 1972, pp. 263-66.

(HELMUT HUMBACH)

ii. GREEK INSCRIPTIONS FROM ANCIENT IRAN

Introduction. In April 1815 the Prussian Akademie der Wissenschaften in Berlin enthusiastically accepted the proposal by August Boeckh to produce a comprehensive thesaurus of inscriptions that would include all Greek inscriptional material published to date. When in 1853 Johannes Franz published the third of four folio volumes of this Corpus Inscriptionum Graecarum (CIG), its twenty-eighth chapter contained five inscriptions of Media and Persia then known (III 4673-6). Of these, the two short early Sasanian inscriptions in Naqæ-e Rostam (nos. 4675a+b; see below) are the ones mentioned earliest in modern western literature, having been registered by several travelers in the second half of the 17th century.

Since those early days of modern Greek epigraphy, the number of known Greek inscriptions from ancient Iran (that is, all countries which once had a population that spoke some Iranian idiom) has enormously increased. The geographic boundaries of Greek epigraphy of the "Extrême-Orient grec," a term coined by Bernard Haussoullier in 1903, have been transferred much farther to the east in Bactria and Arachosia (qq.v.). As these inscriptions represent an invaluable and varied source of information about the encounter of the Greek and Iranian civilizations, it is regrettable that no corpus yet exists. Such a collection has been planned within the framework of the Corpus Inscriptionum Iranicarum, however, for more than thirty-five years. Louis Robert had started work on it (Robert, 1960, p. 86, n. 2), but unfortunately, he did not live to complete it. His preliminary work was continued by Paul Bernard and Jean Pouilloux (Bernard, 1987, p. 111) and is now in the hands of Bernard and Georges Rougemont (personal communication from N. Sims-Williams).

In view of this situation, a chronologically arranged survey of Greek inscriptions from Iran will have to suffice. The following inscriptions have been recorded so far (for a more detailed survey on these and other inscriptions, see Huyse, 1995):

The Achaemenid period.

Persepolis. a) Five graffiti of the 6th-5th centuries B.C.E. from the limestone quarries on the Ku@h-e Rahámat near Persepolis, of which the owner inscription by one Pytharchos seems to be the most important (Carratelli, 1966, pp. 31-34). b) A clay tablet from Persepolis of about 500 B.C.E., mentioning the wine measure maris (Fort. 1771 in Hallock, p. 2).

Susa. a) An early Milesian dedication to Apollo from the 6th century B.C.E. on a large bronze astra‚galos. It was carried off in 494 from Didyma to Susa by Darius's soldiers and now is in the Louvre (Rehm, ed., p. 7; Bravo, pp. 833-35). b) An inscription of "Nikokle@s, son of Nikokle@s, Sinopian" which may belong to the 4th century B.C.E. or earlier (Cumont, 1928, pp. 79-80, no. 1).

The Hellenistic period (Seleucids, Greco-Bactrians, etc.).

Persis. a) Five inscriptions on altar stone slabs dedicated to the Greek gods Zeus Megistos, Athena Basileia, Apollo, Artemis, and Helios (first mentioned by Ernst Herzfeld in several publications; for details, see Robert, 1946-47, no. 225). b) A milestone from Pasargadae (Lewis apud Stronach, pp. 160-61).

Media. a+b) An important inscription of June 193 B.C.E., now in the Mu@za-ye Èra@n-e ba@sta@n, Tehran, recording an edict by the Seleucid king Antiochos III the Great (223-187 B.C.E., q.v.) for the polis of Laodikea (Neha@vand) on the appointment of a high priestess for the cult of Queen Laodike@ (Robert, 1949, pp. 5-29); two copies of this inscription to other addressees are extant, one in Phrygian Dodurga of May 193 B.C.E. (ibid.), the other in Kerma@næa@h of February/March 193 B.C.E. (Robert, 1967, pp. 283-96; idem, 1989, pp. 471-84). c) A short honorary inscription from Laodikea, now also in the Mu@za-ye Èra@n-e ba@sta@n for Menede@mos, governor of the Upper Satrapies, and named in (a) above (Robert, 1949, p. 23; idem, 1950, pp. 73-75). d) An epitaph for Eumene@s, son of Demetrios in Kerma@næa@h, which has now disappeared (first noticed by Herzfeld; see Robert, 1967, pp. 295-96; idem, 1989, pp. 483-84). e) An apotropaic inscription of about 300-250 B.C.E. engraved above the door of a chamber in the cave complex at Karaftu@ (Bernard, 1987). f) A dedication of a relief of Herakles Kallinikos at B^sotu@n (q.v.), dated 148 B.C.E., by Hyakinthos, son of Pantauchos, for the well-being of Kleomene@s, governor of the Upper Satrapies (Robert, 1963, p. 7; idem, 1989, p. 615).

Susiana/K¨u@zesta@n. The French excavations at Susa have brought to light numerous interesting Greek inscriptions from the 3rd to 1st centuries B.C.E., some containing decrees or honorary inscriptions (SEG VII, 2-8). Others include a dedication to the goddess Ma@h from the end of the 3rd century B.C.E. by one Apollodo@ros, son of Krateros (SEG VII, 10); a short 3rd-century dedicatory poem in elegiac distichs on the base of a statue of Apollo, now in the Louvre (SEG VII, 11); a series of manumissions records, for the most part from the 2nd century B.C.E. (SEG VII, 15, 17-26; see also Robert, 1936; Koshelenko and Novikov); a few fragments of inscriptions on terracotta pottery and a couple of Rhodian sealings (SEG VII, 28-34); and a short inscription with the title archiereu‚s (Cumont, 1938).

Hyrcania. A manumission record of Hermaios, son of Euandros, from the time of the reign of the Seleucid king Antiochos I (281-261 B.C.E.; q.v.) and dedicated to the Greco-Egyptian god Sarapis. The exact place where the inscription was found remains unknown (Robert, 1960).

Drangiana/S^sta@n. A selection of five ostraca, including a long but unfortunately very fragmentary ostracon of twelve to thirteen lines, from Qal¿a-ye Sa@m (Carratelli, 1966, pp. 34-35). Other ostraca remain unpublished (cf. Sherwin-White and Kuhrt, p. 80).

Bactria. a) A short Greek inscription in a cave at Qara Kamar in northern Bactria, on the frontier of modern Turkmenistan and Uzbekistan (Ustinova). b) The majority of Greek inscriptions in Bactria have been found during excavations by the De‚le‚gation Arche‚ologique Française at AÚy K¨a@nom (q.v., i.e., Alexandreia OÚxeiane@) from 1965 onward. Apart from a large collection of graffiti on storage vases both in and outside the treasury house (Rapin, 1983 [= SEG XXXIII, 1220-46]; idem, 1987, pp. 52-59), two finds are particularly noteworthy: the dedication of two brothers to the gods Hermes and Herakles found in a gymnasium (Robert, 1968, pp. 417-21; idem, 1989, pp. 511-15]), and a base of a statue near the he@ro@on with two inscriptions, one in two elegiac distichs stating that a certain Klearchos had made a careful copy of the Delphic maxims of the Seven Sages, the other containing the last five of these maxims (Robert, 1968, pp. 421-57; idem, 1989, pp. 515-51; Schmitt, pp. 55-56). c) A six-line fragment in elegiac distichs, on a terracotta plaque, probably belonging to a funerary epigram, found at J^ga@ Tepe near Delbarj^n (q.v.; Kruglikova, pp. 425-26). d) A dedication to Oxos of the mid-2nd century B.C.E. from Takòt-e Sang^n (Litvinski¥, Vinogradov, and Pichikyan). e) The so-called Palame@de@s inscription from Sorkò Kotal near Bag@la@n (Curiel, pp. 194-97). f) Some fragmentary graffiti on jars from Tepe Neml^k/Namal^k (west of Balkò), Emæ^ Tepe (near Balkò), and Garaw Qal¿a/Java@n (Rapin, 1983, p. 316, no. 5; Schmitt, p. 53, no. 44).

Arachosia. a) Two edicts from Qandaha@r (probably Alexandreia en Arachosia) by the Indian emperor As‚oka (r. ca. 268-232 B.C.E., q.v.) of the Maurya dynasty, admonishing his subjects to piety and abstinence. One is a bilingual Greco-Aramaic inscription (ed. Pugliese Carratelli, 1958 [improved version 1964] and, independently, Schlumberger, Robert, Dupont-Sommer, and Benveniste); the other contains the remaining parts of a Greek translation of the Prakrit Rock Edicts XII and XIII, discovered in 1958 and 1964, respectively (ed. Schlumberger, 1964; on both texts, see Schmitt, pp. 43-51). b) A fragmentary 3rd-century B.C.E. metric dedication on an alabaster base of a statue group (for further analysis and literature, see Schmitt, p. 51, nos. 37-38).

The Parthian period.

B^sotu@n. a) An inscription on a relief showing the Parthian king Mithridates II (r. ca. 124/3-87 B.C.E.) honored by four dignitaries (Dittenberger, ed., 431; Kawami, pp. 155-57). b) An inscription on another relief immediately to the right of (a), showing the Parthian king Gotarzes (II; r. ca. 38-51 C.E.) defeating his adversary in a cavalry battle (Franz, ed., III, 4674; Kawami, pp. 157-59).

Susa. a) Two poems in elegiac distichs in honor of Zamaspe@s, stratíarchos of Susa during the reign of the Parthian king Phraates IV (38/7-2 B.C.E.), expressing gratitude for his irrigation works on the river Gondeisos (SEG VII, 12-13). b) A letter of December 21 C.E. from the Parthian king Artabanus II (r. 12-ca. 38 C.E., q.v.) to the magistrates of Susa, concerning the validation of a contested city election. It is now in the Louvre (Welles, pp. 299-306, no. 75). c) A hymnus for Apollo of the 1st century C.E., now in the Louvre (SEG VII, 14). d) A portrait of a queen with an inscription of the artist Antiochos, son of Drua@s, on the crown (Cumont, 1939). e) An inscription on a mosaic, bearing the female name Mousa (Ghirshman).

Nisa. An inscription on a rhyton (Bernard, 1985, esp. p. 90).

The Sasanian period.

Naqæ-e Rostam. a) A short trilingual (Middle Persian, Parthian, and Greek) inscription of the Sasanian king Ardaæ^r I (r. 224-241 C.E., q.v.) under an investiture scene (Back, p. 281). b) A very short trilingual (Middle Persian, Parthian, and Greek) inscription under a relief of Ahura Mazda@ (q.v.; reign of Ardaæ^r I; Back, p. 282). c) A long and extremely important inscription of Ardaæ^r's son ˆa@pu@r I (r. 241/2-272 C.E.) on the Ka¿ba-ye Zardoæt near Naqæ-e Rostam, also called "Res Gestae Saporis." The king introduces himself and his empire, boasts of his victories against the Romans, and enumerates his foundations of sacred fires (Maricq; Back, pp. 284-371; Huyse, forthcoming).

Naqæ-e Rajab. A short trilingual (Middle Persian, Parthian, and Greek) inscription on the belly of a statue of the horse of ˆa@pu@r I (Back, pp. 282-83).

Bibliography (for cited works not given in detail, see "Short References"): M. Back, Die sassanidischen Staatsinschriften, Acta Iranica 18, Leiden, Tehran, and Lieàge, 1978. P. Bernard, "Les rhytons de Nisa: I. Poe‚tesses grecques," Journal des Savants, 1985, pp. 25-118. Idem, "Le Marsyas d'Apame‚e, l'Oxus et la colonisation se‚leucide en Bactriane," Stud. Ir. 16/3, 1987, pp. 103-15. B. Bravo, "Sulân: Repre‚sailles et justice prive‚e contre des e‚trangers dans les cite‚s grecques (Etude du vocabulaire et des institutions)," Annali della Scuola Normale Superiore di Pisa 10, 1980, pp. 675-987. G. P. Carratelli, "L'iscrizione greca" in G. P. Carratelli and G. L. Della Vida, ed. and tr. with comm., Un editto bilingue greco-aramaico di As‚oka: La prima iscrizione greca scoperta in Afghanistan (= SOR 21), Rome, 1958, pp. 11-14. Idem, "The Greek Section of the Kandahar Inscription" in G. P. Carratelli and G. Garbini, ed. and tr. with comm., A Bilingual Graeco-Aramaic Edict by As‚oka: The First Greek Inscription Discovered in Afghanistan (= SOR 29), Rome, 1964, pp. 29-39. Idem, "Greek Inscriptions of the Middle East," East and West 16/1-2, 1966, pp. 31-36. F. Cumont, "Inscriptions grecques de Suse," Me‚moires de la Mission Arche‚ologique de Perse 20, 1928, pp. 77-98. Idem, "Une breàve inscription grecque de Suse," Comptes rendus de l'Acade‚mie des Inscriptions et Belles-Lettres, Paris, 1938, pp. 305-7. Idem, "Portrait d'une reine parthe trouve‚ aà Suse," ibid., Paris, 1939, pp. 330-41. R. Curiel, "Inscriptions de Surkh Kotal," JA 242, 1954, pp. 189-205. W. Dittenberger, Orientis Graeci inscriptiones selectae, 2 vols., Leipzig, 1903-5. J. Franz, ed., Corpus Inscriptionum Graecarum III, Berlin, 1853. R. Ghirshman, "Cinq campagnes de fouilles aà Suse," RA 46, 1952, pp. 1-18. R. T. Hallock, Persepolis Fortification Tablets, Chicago, 1969. B. Haussoullier, "Inscriptions grecques de l'Extrême-Orient grec," Me‚langes Perrot: Recueil de me‚moires concernant l'arche‚ologie classique, la litte‚rature et l'histoire anciennes de‚die‚ aà Georges Perrot, Paris, 1903, pp. 155-59. P. Huyse, "Die Begegnung zwischen Hellenen und Iraniern: Griechische epigraphische Zeugnisse von Griechenland bis Pakistan," in Ch. Reck and P. Zieme, eds., Iran und Turan: Beiträge Berliner Wissenschaftler, Werner Sundermann zum 60. Geburstag gewident, Wiesbaden, 1995, pp. 99-113. T. Kawami, Monumental Art of the Parthian Period in Iran, Acta Iranica 26, Leiden, Tehran, and Lieàge, 1987. G. A. Koshelenko and S. V. Novikov, "Manumissii Selevkii-na-Evlee" (Manumissions from Seleucia on the Eulaeus), VDI, 1979, 2, pp. 41-54. I. Kruglikova, "Les fouilles de la mission arche‚ologique sovie‚to-afghane sur le site gre‚co-kushan de Dilberdjin en Bactriane," Comptes rendus de l'Acade‚mie des Inscriptions et Belles-Lettres, Paris, 1977, pp. 407-27. B. A. Litvinski¥, Y. G. Vinogradov, and I. R. Pichikyan, "Votiv Atrosoka iz xrama Oksa v Severno¥ Baktrii" (The votive offering of Atrosokes from the Temple of Oxus in northern Bactria), VDI 1985/4, pp. 84-110. A. Maricq, "Res Gestae Divi Saporis," Syria 35, 1958, pp. 297-360. C. Rapin, "Les inscriptions e‚conomiques de la tre‚sorerie helle‚nistique d'Aï Khanoum (Afghanistan)," Bulletin de Correspondance Helle‚nique 107, 1983, pp. 315-72. Idem, "La tre‚sorerie helle‚nistique d'Aï Khanoum," Revue arche‚ologique, 1987, pp. 41-70. A. Rehm, Didyma II. Die Inschriften, Berlin, 1958. L. Robert, "Sur les affranchissements de Suse," Revue philologique, 1936, pp. 137-52 (= Opera Minora II, pp. 1216-31]. Idem, Bulletin e‚pigraphique, 1946-47, no. 225. Idem, "Inscriptions se‚leucides de Phrygie et d'Iran," Hellenica 7, 1949, pp. 5-29. Idem, "Addenda au Tome VII," Hellenica 8, 1950, pp. 73-75. Idem, "Inscription helle‚nistique d'Iran," Hellenica 11-12, 1960, pp. 85-91. Idem, Review of "P. M. Fraser, Samothrace: II/1, The Inscriptions on Stone, New York, 1960," Gnomon 35, 1963, pp. 50-79 (= Opera minora VI, pp. 589-618). Idem, "Encore une inscription grecque de l'Iran," Comptes rendus de l'Acade‚mie des Inscriptions et Belles-Lettres, Paris, 1967, pp. 281-96 [= Opera minora V, pp. 469-84]. Idem, "De Delphes aà l'Oxus. Inscriptions grecques nouvelles de la Bactriane," ibid., Paris, 1968, pp. 416-57 (= Opera minora V, pp. 510-51). Idem, Opera minora selecta, 7 vols. Amsterdam, 1969-90. D. Schlumberger, "Une nouvelle inscription grecque d'Açoka," Comptes rendus de l'Acade‚mie des Inscriptions et Belles-Lettres, Paris, 1964, pp. 126-40. Idem, L. Robert, A. Dupont-Sommer, and É. Benveniste, "Une bilingue gre‚co-arame‚enne d'Asoka," JA 246, 1958, pp. 1-48. R. Schmitt, "Ex occidente lux: Griechen und griechische Sprache im hellenistischen Fernen Osten," in P. Steinmetz, ed., Beiträge zur hellenistischen Literatur und ihrer Rezeption in Rom, Stuttgart, 1990, pp. 41-58. SEG = Supplementum Epigraphicum Graecum VII, Leiden, 1934. S. Sherwin-White and A. Kuhrt, From Samarkhand to Sardis: A New Approach to the Seleucid Empire, London, 1993. D. Stronach, Pasargadae: A Report on the Excavations Conducted by the British Institute of Persian Studies from 1961 to 1963, Oxford, 1978. Y. B. Ustinova, "Naskal'nye latinskie i grecheskaya nadpisi iz Kara-Kamara" (Greek and Latin Rock Inscriptions from Qara Kamar), VDI, 1990, 854, pp. 145-47 (cf. supra sub Litvinski¥). B. C. Welles, Royal Correspondance in the Hellenistic Period, New Haven, 1934.

(PHILIP HUYSE)

iii. ARABIC INSCRIPTIONS IN PERSIA

In Persia, as in the rest of the Islamic lands, Arabic was the basic language for foundation and religious texts on buildings and objects. In the early Islamic period these texts were usually written in some variant of the angular script known as Kufic (see CALLIGRAPHY). From the 12th century inscriptions in Persian became more common, especially for poetic texts, and cursive scripts tended to replace angular ones. In general, inscriptions on portable objects made in Persia followed the forms and styles of monumental inscriptions (Blair, Bloom, and Wardwell).

Monumental inscriptions. Although Pahlavi continued to be used until the 6th/12th century in inscriptions in such isolated regions as the Caspian coast (e.g., the tomb tower at La@j^m in Ma@zandara@n; Combe, Sauvaget, and Wiet, no. 2331; Blair, 1992, no. 32), foundation texts on buildings in the Islamic period were usually written in Arabic. Persian first appeared in commemorative inscriptions, which were less formal than foundation texts. Beginning in the mid-4th/10th century the Buyids (q.v.) included Persian names in commemorative texts carved at Persepolis, e.g., an inscription carved in the name of ¿Azµod-al-Dawla in 344/955-56 (Plate I; Donohue, pp. 74-75; Blair, 1992, no. 7). Dates in Persian first appeared in the mid-5th/11th century (e.g., in a carved inscription of Abu@ Kal^jar dated 2 AÚba@n 438/24 October 1046; Mosátáafaw^, p. 340; Blair, 1992, no. 43). The first foundation text in Persian is from the Ilek–Khanid mausoleum at Saf^d Boland in Farg@a@na, datable to 447-51/1055-60 (Nashch and Kochneu; Blair, 1992, no. 47). In the late 5th/11th and early 6th/12th centuries Persian verses were used on such palaces as Reba@tÂ-e Malek (471/1078-79) northeast of Bukhara (q.v.) and the Ghaznavid Mas¿u@d III's palace at GÚazna (505/1111) in Afghanistan. Nevertheless, the use of Arabic for such texts remained standard in Persia even in the Qajar period (1193-1342/1779-1924).

A fully developed tradition of monumental Arabic inscriptions on Persian architecture can be documented only from the mid-4th/10th century (Blair, 1992, introd.). Earlier examples rarely survive in Persia, probably because of the impermanent materials used. According to contemporary chroniclers, however, the early ¿Abbasid caliphs' foundation inscriptions on Persian buildings resembled those used elsewhere. For example, an inscription of al-Mansáu@r (136-58/754-75) on a treasury in Azerbaijan, recorded by the early 4th/10th-century scholar Abu@ ¿Abd-Alla@h Moháammad Jahæ^a@r^ (Combe, Sauvaget, and Wiet, no. 43), is comparable to contemporary inscriptions commemorating repairs and enlargements to the sanctuary (háaram) at Mecca by the same caliph and his successor, al-Mahd^ (158-69/775-85; Combe, Sauvaget, and Wiet, nos. 40, 48-52). Inscriptions of the 3rd-4th/9-10th centuries recorded in the Caucasus by the 19th-century traveler M. N. Khanikoff were similar. Grave markers from early Islamic Persia were also inscribed in Arabic. The earliest known example is a tombstone discovered in the Ema@mzada Ja¿far at Damg@a@n (q.v.), recording the death of a Zayd^ descendant of the prophet ca. 267/900 (Adle, pp. 292-97). A handful of 4th/10th-century crested grave covers excavated at S^ra@f on the Persian Gulf and a series of steles dating from the 5th/11th century onward from the Yazd region are also inscribed in Arabic (Lowick, pp. 79-115; Afæa@r, II, pp. 909-16).

From the next 150 years, approximately seventy-five inscriptions survive. They decorated mosques, tombs, palaces, minarets, and such civil structures as walls, bridges, and cisterns. The most complete formula for construction texts consisted of the besmela (see BESMELLAÚH), the verb (generally amara be-bena@÷ "ordered the construction [of],") some mention of the structure (often indicated only by a pronoun), the name and titles of the patron, the date, and the name of the artisan. Adjectives were rarely used to qualify the structure built, though the foundation inscription in rhymed prose on the Gonbad-e Qa@bu@s ordered by the Ziyarid prince Qa@bu@s b. Voæmg^r in 397/1006-7 is one notable exception (Combe, Sauvaget, and Wiet, no. 2118; Blair, 1992, no. 19). As in the rest of the Islamic world, during the course of time the list of the patron's titles grew longer. The patrons themselves, some of whom were women, represented the myriad small dynasties that flourished in Persia and Transoxania in the 10th-11th centuries: the Hasanuyids, Kakuyids, Bawandids (see AÚL-E BAÚVAND), Ziyarids, Kòúa@razmæa@hs, Firuzanids, Shaddadids, Ilek-khanids (Qarakhanids), and the like. The artisans' signatures were also in Arabic. In addition to builders, such as the man who signed the tomb tower known as the P^r-e ¿Alamda@r in Damg@a@n in 417/1026-27, other specialists included the woodcarver who signed the columns in the mosque at Kò^va (ca. 400/1010) or the ironworkers who signed the gates at Yazd and Ganja (see EBRAÚHÈM B. ¿OT¨MAÚN) in 430/1040-41 and 455/1063 (Blair, 1992, nos. 34, 26, 41, 48, respectively). The usual formula for the signature was ¿amal ("work [of]") or ¿amela ("[so-and-so] made").

There is no standard corpus of inscriptions dating after the 12th century, but many are included in the Re‚pertoire chronologique d'e‚pigraphie arabe (Combe, Sauvaget, and Wiet), of which seventeen volumes covering the period up to 783/1381, plus a geographical index, have appeared. Other inscriptions can be found in works on individual Persian provinces and cities, for example, those by LotÂf-Alla@h Honarfar on Isfahan and Èraj Afæa@r on Yazd, as well as the volumes published by Anjoman-e a@tòa@r-e mell^ (q.v.). The general trends are thus clear. The basic formula remained the same as in earlier periods, but the texts became longer and more flowery. The patrons boasted ever more elaborate titles and epithets, and adjectives and metaphors abounded. In the foundation inscription on the mosque of Shaikh Lotáf-Alla@h in Isfahan, for example, Shah ¿Abba@s I (996-1038/1588-1629) is lauded as "the greatest sultan, the most generous king, the reviver of the virtues of his pure fathers, the propagator of the religion of the infallible Imams, Abu'l-Mozáaffar ¿Abba@s al-H®osayn^ al-Mu@saw^ al-Sáafaw^, Baha@dor Khan, may God the Exalted make his kingdom eternal and may his ships sail in the seas of eternity" (al-soltÂa@n al-a¿záam wa'l-kòa@qa@n al-akram . . .; Honarfar, Esáfaha@n, p. 402).

In addition to historical inscriptions, there were also religious inscriptions in Arabic, including koranic citations, Hadiths, poems, prayers, and pious phrases. Some koranic texts, such as the famous Throne verse (AÚyat al-kors^, Koran 2:255), are general evocations of the glory of Islam and were popular on many types of buildings in all periods. Others, such as 9:18 (ennama@ ya¿moro masa@jed Alla@h man amana be'lla@h wa'l-yawm al-a@kòer wa aqa@ma al-sáala@t wa ata@ al-zaka@t wa lam yakòæa "The mosques of God shall be visited and maintained by such as believe in God and the Last Day and perform the prayer and pay the alms and fear none but God alone; it is they who are expected to be among the guided") and 55:26-27 (koll man ¿alayha@ fa@n; wa yabqa@ wajh rabbek dòu'l-jala@l wa'l-ekra@m "All that dwells upon the earth is perishing, yet still abides the face of thy Lord, majestic splendid"), are more specific and allude to a building's function as mosque or tomb. Specific passages sometimes permitted plays on words with the name of the founder; for example, verse 17:79, (¿asa@ an yab¿at¯ak rabbok maqa@man mahámu@dan "Soon will thy Lord raise thee to a laudable station") was popular with patrons named Mahámu@d, including Shah Mahámu@d, the Muzaffarid ruler of the region who restored the Madrasa-ye Ema@m^ya/Ba@ba@ Qa@sem (755/1354) in Isfahan (Honarfar, Esáfaha@n, p. 309).

Hadiths were not as common as koranic quotations in Islamic epigraphy. The earliest example from the Persian world decorates a wooden mehára@b (niche) removed from the congregational mosque at Iskodar in the Zarafæa@n valley in Tajikistan and now in the United Republic Museum in Dushanbe (Denike‚, fig. 4; Blair, 1992, no 27). The style of floriated Kufic in one of the rectangular framing bands suggests a date ca. 400/1010. The text, which refers to the inestimable rewards for those who guard the front line, is appropriate to a site on the frontiers of Islam; it is not attested in the major concordance of canonical Hadiths (Wensinck et. al.) and may well have been invented for the occasion. Hadiths soon became more popular, perhaps because they were more adaptable than koranic citations. Some were simply familiar sayings, but most were chosen because they were appropriate to the buildings or places on which they were inscribed. In the tomb of Tu@ma@n AÚqa@, one of Timur's wives, in the ˆa@h-e Zenda cemetery outside Samarqand (808/1404-5), the saying "The tomb is a door and everyone enters it" (al-qabr ba@b wa koll al-na@s da@kòalahu) is inscribed above the doorway to the mausoleum, and "Hasten with prayer before you miss it" (¿ajjelu@ be'l-sáala@t qabl al-fawt) and "Hasten with repentence before death" (¿ajjelu@ be'l-tawba qabl al-mawt) above the portal to the mosque (Golombek and Wilber, I, p. 249; Shishkin). Often Hadiths were chosen for sectarian purposes. A mehára@b added to the congregational mosque in Isfahan in 710/1310 after Sultan Öljeitü (Ulja@ytu@, 703-17/1304-17) had converted to Shi¿ism (Plate II), is inscribed with a Hadith attributed to Imam ¿Al^ b. Ab^ T®a@leb (q.v.) to the effect that whoever frequents a mosque will receive one of eight benedictions (Honarfar, Esáfaha@n, p. 120). The Hadith was probably directed at the traditionally Sunni population in this troublesome sectarian city. Under the Safavids Hadiths became especially popular. For example, in the mosque of Shaikh LotÂf-Alla@h in Isfahan several are included in the large band of t¯olt¯ script (see CALLIGRAPHY) at the base of the dome signed by the celebrated calligrapher ¿Al^-Rezµa@ ¿Abba@s^ (q.v.) in 1025/1616-17. The text begins with a prophetic Hadith about the rewards for pious Muslims. This is followed by two additional passages about the actions suitable for someone going to a mosque. The two authorities cited are the Ahl al-Bayt (q.v.) and Imam Ja¿far al-Sáa@deq, respectively (Honarfar, Esáfaha@n, pp. 407-10).

Arabic religious poems were also inscribed on buildings in Persia. Some of the blind niches on the interior of the mausoleum of Shah Ahámed Qa@sem known as the Dar-e Beheæt outside the Ka@æa@n gate in Qom (780/1378) are inscribed with verses extolling the lineage of the deceased, ¿Al^ b. Ja¿far (Modarres^ T®aba@tÂaba@÷^, II, 46). Persian poetry was common on 9th/15th-century buildings, but under the Safavids Arabic poetry again became popular. The mosque of Shaikh Lotáf-Alla@h provides a good example. On the large band framing the arches on the north and south sides of the interior is a poem signed by Ba@qer, the builder, asking the ±aha@rdah ma¿sáu@m (q.v. "fourteen infallible ones") to intercede for Shaikh LotÂf-Alla@h in the hereafter. A similar band on the east and west arches contains a poem invoking the ±aha@rdah ma¿sáu@m. This poem, as suggested by Honarfar (Esáfaha@n, pp. 412-15), may be by Shaikh Baha@÷-al-D^n ¿AÚmel^ (q.v.), the famous Imami scholar of the period of Shah ¿Abba@s I. Pious phrases in Arabic were common in inscriptions in all periods. They included such general expressions as "God is the greatest" (Alla@ho akbar), the ninety-nine "beautiful names" of God (al-asma@÷ al-háosna@, see Gardet), and the names of revered people like the four rightly guided caliphs or the ±aha@rdah ma¿sáu@m. Many phrases reflect sectarian loyalties. A reference to "Moháammad's pure and pleasing family" in the bay over the mehára@b at the 4th/10th-century congregational mosque at Na@÷^n may well indicate Shi¿ite patronage for the structure (Blair, 1992, no. 9).

In the early Islamic period architectural inscriptions in Persia were written in plain forms of angular script used elsewhere in the Islamic world. Beginning in the 4th/10th century this simple script was elaborated in several ways. One of the simplest variants was "foliated Kufic," in which the tips of the letters swell into leaf forms. An early example occurs on the lintel at the tomb of the Samanids in Bukhara (320s/930s; Bulatov, pl. 9; Blair, 1992, no. 4). This script in turn developed into "floriated Kufic," in which independent floral motifs sprout from the letters and in the spaces between them, as in the foundation inscription framing the portal to the mausoleum at T^m in the Zarafæ@a@n valley in Uzbekistan (367/977; Pugachenkova, fig. 43; Blair, 1992, no. 11) and the superb cut-plaster inscriptions on the interior of the Ema@mza@da Yaháya@ b. Zayd at Sar-e Pol in the province of Tokòa@resta@n (Ju@zja@n) in northern Afghanistan (ca. 500/1100; Bivar; Blair, 1992, no. 75).

Both foliated and floriated Kufic were popular in the western Islamic world, where they were particularly associated with Fatimid (q.v.; 358-567/969-1171) monuments in Cairo, for example, al-Azhar and the mosque of al-H®a@kem. One variety of script that seems to have originated in the eastern Islamic world, however, is "interlaced Kufic," in which the stems or bodies of the letters are plaited. A crested grave cover from S^ra@f dated 364/975 (Lowick, pl. X) has a single interlaced letter (the ka@f in rahámateka), but another grave cover, dated less than two decades later (383/993), has extensive interlacing: Single letters are plaited, two or three consecutive letters are intertwined, and the ligature between the two la@ms of Alla@h has been transformed into a lattice of three superimposed trilobes (Lowick, pls. XII and XIII). Interlacing also occurred in architectural inscriptions. One of the first surviving examples is in the mosque at Ardesta@n (q.v.): Although the mosque seems to have been extensively rebuilt in 553-55/1158-60 (Combe, Sauvaget, and Wiet, nos. 3224, 3238), the koranic band in the arcade running south from the corner of the court can be attributed on stylistic grounds to about 400/1010; it contains an interlaced ka@f (Blair, 1992, no. 23).

Interlaced script is difficult to read and was thus usually reserved for well-known texts, like the koranic verses on the crested grave covers, in the mosque at Ardesta@n, or on the minaret at Termedò dated 423/1031-32 (Blair, 1992, no. 38 and figs. 61-62). In order to increase legibility, ornament was gradually removed to the upper zone of the inscription, so that only the ascenders and tails, rather than the bodies, of the letters were plaited. This kind of script occurs in the two foundation inscriptions (407-11/1016-21) on a tomb tower in the Ra@dka@n valley of the Alborz mountains south of Sa@r^: on a stucco plaque above the doorway and in the band encircling the tower below the roof. In both inscriptions the patron, the Bawandid prince Abu@ Ja¿far Moháammad, is named (Flury, 1920; idem, 1921).

Despite the restriction of interlacing to the upper zone, inscriptions were still somewhat cluttered. To reduce the clutter and increase the legibility, the ornament in the upper zone was the separated and enlarged so that it was equal in weight to the bodies of the letters in the lower zone. Elaborate plaiting was replaced by floral decoration. A superb example of such "bordered Kufic" is the foundation inscription, with the name of the patron, the Saljuq vizier Nezáa@m-al-Molk, formerly in the "madrasa" at K¨argerd in Khorasan (ca. 465-70/1072-77; Combe, Sauvaget, and Wiet, no. 2789; Blair, 1992, no. 57) and now in the Mu@za-ye Èra@n-e ba@sta@n in Tehran. Herzfeld (1943, p. 16, figs. 35-36) called it "perhaps the masterpiece of Kufic epigraphy in Iran" (Plate III). The stucco inscription is organized in three zones. The lower contains the bodies of the letters, outlined with beading. It is balanced by an upper zone of approximately equal width containing two tiers of palmettes, which appear to grow from the ascenders. The medial zone is allotted to the ascenders themselves, which are generally unornamented. The inscription projects about 7.5 cm from a background of floral arabesques, and the two levels of carving highlight the contrast between the organic elements in the background and the rigid ascenders that march in stately rhythm across the band.

The shift in monumental epigraphy from angular to cursive scripts can also be documented in the inscriptions dating before the 12th century. Cursive script first appeared in commemorative texts such as those carved at Persepolis; by the second half of the 11th century it was in use for foundation inscriptions on buildings. Some of the first examples survive on the stucco mehára@bs added to the Masjed-e Pa@mana@r at Zava@ra in 461/1068-69 (Honarfar, Esáfaha@n, pp. 171-75; Blair, 1992, no. 51); these cursive inscriptions, which include the name of the patron, are played off against larger framing inscriptions with koranic texts in Kufic. Persian artists quickly realized the decorative possibilities of juxtaposing angular and cursive scripts. Stone fragments from GÚazna in the name of Ebra@h^m b. Mas¿u@d (Flury, 1925, nos. 4-7; Blair, 1992, no. 69) show the sophisticated way in which the Ghaznavids exploited style to underscore content: on one panel from a mehára@b three types of script—cursive, foliated Kufic, and simple Kufic—contain three types of text, namely koranic, pious, and historical. Such a juxtaposition of scripts for rhetorical and artistic effect appeared at the same time in Saljuq lands. In the sanctuary dome added to the great mosque in Isfahan in 479-80/1086-87, for example, the monumental koranic inscription carved on the impost blocks is in cursive script, whereas the foundation inscription in brick relief encircling the base of the dome is in Kufic (Galdieri, p. 38; Honarfar, Esáfaha@n, p. 76; Blair, 1992, no. 61).

After the 6th/12th century Kufic was relegated to a secondary role and appeared primarily in repeating panels or stylized motifs. "Square Kufic" became popular under the Il-khanids and may have imitated Chinese seal script (Krachkovskaya, pp. 23-24). Often such panels simply repeat such pious phrases al-molk le'lla@h (dominion belongs to God) or the names of highly regarded people. The stucco panels in "square Kufic" in the shrine of P^r-e Bakra@n at Lenja@n outside Isfahan (703-12/1303-12) are particularly elaborate and spell out the names of God and the ±aha@rdah ma¿sáu@m (Honarfar, Esáfaha@n, pp. 256-60). Glazed bricks could also be set into surfaces in common bond so that they spelled out words in Kufic script. This technique, sometimes called banna@÷^ (builder's), was an easy way to enliven vast wall surfaces and "drench" the architecture in writing. First used in the 6th/12th century, it became characteristic of monuments built by the Timurids in Central Asia. The exterior of the shrine of the Sufi shaikh K¨úa@ja Ahámad Yasaw^ at Yas^/ Turkesta@n (799-801/1397-99) is covered with great expanses of continuous geometric patterns enclosing square Kufic designs. Lisa Golombek and Donald Wilbur (p. 210) have suggested that such inscriptions were not meant to be read as individual statements, but rather, that the repetition of sacred names was comparable to verbal repetitions in the Sufi dòekr (q.v.).

The most popular cursive script in later centuries was an elongated t¯olt¯. Beginning in the 14th century, a second inscription was often interwoven in the ascenders of the first. The earliest surviving examples in Persia are the inscriptions painted on the plaster revetment in the interior of the tomb of Öljeitü (Ulja@ytu@) at SoltÂa@n^ya (Blair, 1987, pp. 43-96). Color was also exploited to underscore the message. In foundation inscriptions framing portals, the patron's name was often highlighted in yellow in the center of a white inscription on a dark-blue glazed ground. In an early example dated 851/1447, on the portal leading to the winter prayer hall (æabesta@n) in the congregational mosque at Isfahan the name of the ruler, Soltáa@n Moháammad, grandson of ˆa@hrokò, is set off in ocher (Honarfar, Esáfaha@n, pp. 122-23).

Inscriptions on portable objects. Arabic was the standard language on coins and was also used for dedicatory inscriptions on metalwork, textiles, and ceramics, though other languages continued to be used on objects made for several centuries in such isolated areas as the mountains near the Caspian Sea or Central Asia. For example, a hoard found in Ma@zandara@n and now in the Mu@za-ye Èra@n-e ba@sta@n, included three silver bowls, two inscribed in Pahlavi with the name Wanda@d Hormazd of the Ka@rens, a local ruler in the late 2nd/8th century (Ghirshman; Henning). On the basis of a Sogdian note written in ink on a silk textile from the treasury of the collegial church of Notre Dame at Huy, Belgium, Dorothy Shepherd was able to identify a whole group with a type that medieval authors called "Zandan^j^," referring to the town of Zandana near Bukhara (Shepherd and Henning, esp. pp. 38-40; see also ABRȈAM iii., esp. pl. XII/2, where the legend has been transposed with that of pl. XII/1).

By the 4th/10th century dedicatory inscriptions in Arabic had become standard. A good example is a silver wine service found at Hama@da@n, now also in the Mu@za-ye Èra@n-e ba@sta@n. It comprises three conical bowls, two saucers, a tray, a ewer, two small jugs, a bottle, and a cup. Seven of the pieces are inscribed with the name of the amir Abu'l-¿Abba@s Valg^n b. Ha@ru@n, identified as "client of the Commander of the Faithful," a title that suggests a date ca. 400/1010 (Combe, Sauvaget, and Wiet, nos. 2154-60). Instead of the standard foundation text that appears on architecture and begins with amara be-bena@÷ (ordered the construction [of]), objects were usually inscribed with dedicatory texts invoking blessings and good wishes on the owner. The inscription on the wine service, for example, asks for God's blessing (baraka men Alla@h). This type of inscription on a silk fragment found in the church treasury at Saint-Josse-sur-Mer near Calais (Louvre, Paris, no. 7502; Combe, Sauvaget, and Wiet, no. 1507; Survey of Persian Art, pl. 981) invokes glory and good fortune (¿ezz wa eqba@l) for the commander Abu@ Mansáu@r Bakòtek^n (d. 349/960-61), who served the Samanid amir of Transoxania and Khorasan ¿Abd-al-Malek b. Nu@há (r. 343-50/954-41).

Like the foundation inscriptions, these dedicatory inscriptions became more elaborate over time. A bronze pencase in the Freer Gallery of Art, Washington, D.C., (no. 36.7; Plate IV) has a benedictory inscription invoking glory, good fortune, and twenty other benefits for its owner for eternity. In the dedication encircling the lid, he is identified as "the most grand vizier, the great, the wise, the just, the one assisted by God, the victorious, the triumphant, Majd-al-Molk, the honor of the government and religion, the flame of Islam and the Muslims, the chosen among the kings and sultans, the light of the nation, the benefactor of the people, the example of the great ones and his equals, the pillar of dignity, the lord of viziers, the king of lieutenants, the governor of Iran, the grand vizier and ruler of Khorasan, al-Mozáaffar, son of the deceased vizier, Majd-al-Molk" (Combe, Sauvaget, and Wiet, no. 3671). Herzfeld (1936-37) identified this man as one of the last viziers of the K¨úa@razmæa@hs. Another inscription between the hinges on the rim of the lid at the back includes the signature of the artist, ¿amal ˆa@dò^ al-naqqa@æ (the work of ˆa@dò^ the engraver), and the date 607/1210.

Already by the 12th century such dedicatory inscriptions were often written in Persian. The classic example is a superb bronze known as the "Bobrinski bucket," after a former owner, and now in the Hermitage, St. Petersburg (no. CA-12687), which was made for a merchant in 559/1163. It is decorated with friezes of merrymakers, animals, and animated inscriptions inlaid in copper and silver. Around the rim is a long text in Persian, in which it is recorded that the piece was ordered by ¿Abd-al-Raháma@n b. ¿Abd-Alla@h Raæ^d^; "struck" (zµarb-e; inlaid?) by Moháammad b. ¿Abd-al-Wa@háed; and "made" by (¿amal-e) H®a@jeb Mas¿u@d b. Ahámad, the decorator in Herat for its owner, the exalted K¨úa@ja Rokn-al-D^n, glory of the merchants, trustee of the Muslims, ornament of the pilgrims and the two sanctuaries, Raæ^d-al-D^n ¿Az^z^ b. Abu'l-H®osayn [sic] Zanja@n^ (Ettinghausen, 1943; Combe, Sauvaget, and Wiet, no. 3280; corrections in Melikian-Chirvani, pp. 82-83 nn. 60-61). A bronze aquamanile in the shape of a zebu feeding her calf while a leopard bites the hump on her back, also in the Hermitage (no. AZ-225), and dated 603/1206 has a dedicatory inscription in Persian (Giuzalian, pp. 102-5). On metal wares from the 9th/15th century onward, dedicatory inscriptions were replaced by Persian verses (e.g., Melikian-Chirvani, 1982, chap. IV).

The most common religious inscriptions on portable objects made in Persia are koranic verses. They occur frequently on coins and luster-painted ceramics. Koran 2:137, which contains the longest word in the Koran, fa-sa-yakf^kahom-Alla@h (God will be sufficient for you against them), seems to have served as a talisman and was inscribed in many media, from a cobalt bowl excavated at Samarra in Mesopotamia to luster-painted tiles, coins, and tombstones (Miles, pp. 155-56). Luster-painted tiles were decorated with a limited range of koranic inscriptions, especially well-known passages such as the Throne Verse and the short suras from the end of the book. These texts were chosen for their appropriateness to the site where tiles were mounted. In tombs, for example, they often contained references to the Day of Judgment and the horrors of hell (Watson, pp. 150-51).

During the Safavid period Arabic poems and phrases invoking Shi¿ite figures became common on coins and metalwork (Melikian-Chirvani, Chap. V). Pious phrases were ubiquitous in all media in all periods, and phrases like al-molk le'lla@h became such cliche‚s that they were often misspelled or stylized beyond recognition. Such stylized repeat patterns can be found around the rims of m^na@÷^ (lit. "enamel"; see CERAMICS) ceramics and on innumerable carpet borders.

There is one type of Arabic inscription that seems to have been specific to one particular type of portable object: the moralizing aphorisms found on slip-painted ceramics made in eastern Persia and Transoxania in the 4th/10th century and often associated with the patronage of the Samanid dynasty. The inscriptions praise the virtues of patience, work, intelligence, knowledge, generosity, and the like. A typical example is al-tadb^r qabl al-¿amal yo÷manaka men al-nadam; al-sáabr mefta@há al-faraj (Planning before work protects you from regret; patience is the key to comfort; Bol'shakov; Volov, p. 117). These inscriptions were sometimes written in an elaborate plaited script that makes them extremely difficult to decipher.

Like architectural inscriptions the inscriptions on objects made in Persia in the early centuries are generally written in some form of angular Kufic script and follow a similar progression from simple to more elaborate foliated, floriated, interlaced, and bordered types. Interlaced Kufic is often associated with the Samanid slip-painted wares. Kufic script became more and more stylized and was eventually relegated to repetitive phrases, whereas the more easily legible cursive script was used for dedicatory inscriptions. In many elaborate luster-painted tile assemblages, such as the cenotaph cover made by Moháammed b. Ab^ T®a@her for the shrine at Qom in 602/1206 or the mehára@b made by Abu@ Zayd for the shrine at Maæhad in 612/1215 (Watson, figs. 103-4), bands of stylized Kufic are played off against various styles of cursive. Despite this secondary role, Kufic was still used for sophisticated patterns. A tile said to have come from the Masjed-e ˆa@h in Isfahan (Art Institute, Chicago, no. 1926.1186) has a repeating Kufic inscription with the koranic phrase "God, there is no god but He" (Alla@h la@ ela@h ella@ howa). The first four words are repeated in white on a dark blue ground on the four sides of the tile, and the tall stems of the letters form an interlocking pattern in the center. The fifth word, howa, is repeated four times in turquoise above the stems of the letters and forms an internal square.

By the 12th century cursive scripts had replaced angular ones. The celebrated luster-painted plate made by Sayyed ˆams-al-D^n H®asan^ in 607/1210 (Freer Gallery of Art, no. 41.11) has a long dedicatory inscription in naskò around its rim, invoking everlasting glory on an amir and esfahsa@la@r (military commander) whose name is unfortunately lost; the poetic text inscribed on the interior and exterior of the gadrooned vessel walls are also in naskò and include an ode with alternating hemistichs in Arabic and Persian, a quatrain, and good wishes (Guest and Ettinghausen). In later centuries naskò was replaced by t¯olt¯. A group of low brass bowls inlaid with gold and silver are typically decorated with four figural roundels alternating with four epigraphic cartouches inscribed in elongated t¯olt¯ with hooked ascenders; on the basis of their titulary, they have been attributed to the province of Fa@rs in the early 8th/14th century (Melikian-Chirvani, chap. 3). As in architectural inscriptions, the ascenders of the letters are sometimes elongated, and a second inscription in Kufic is inscribed across them (e.g., Freer Gallery, bowl, 80.25; Victoria and Albert Museum, London, no. 6-1886; Melikian-Chirvani, pp. 213-14). On metalwork produced under the Timurids and Safavids, nasta¿l^q (see CALLIGRAPHY) script was common, especially in Persian verses.

One script seems to have been unique to metalwork: anthropomorphic or zoomorphic script, in which the letters or parts of them assume human or animal form. Ornithomorphic script, in which the letters are transformed into birds or the tails of the letters end in birds' heads, was used on Samanid slip–painted ceramics and later on bronzes produced in the same region. By the 12th century anthropomorphic, zoomorphic, and human-headed scripts were fully developed on eastern Persian metalwork. The best example is the Bobrinski bucket (see above). In the upper, cursive inscription the top halves of the letters have been transformed into figures of revelers, dancers, and musicians, whereas the lower halves are in bird or animal shapes or end in bird and animal heads. The middle inscription is in Kufic with interlaced ascenders. In the lower inscription, also in cursive, the letters terminate in human heads, and animals chase one another through the ascenders. The only known example of such a script in a medium other than metal work is the human-headed cursive in a Persian inscription from the harbor fortress at Baku, dated 632/1234-35 (Bretanitski¥, pp. 82-87); it is of such mediocre quality, however, that it must have been derivative from metalwork.

Bibliography: C. Adle, "Recherches arche‚ologiques en Iran sur le Kumeæ me‚die‚val. Rapport pre‚liminaire pour 1982-83," Comptes-Rendus des Se‚ances de Acade‚mie des Inscriptions et Belles-Lettres, April-June 1984, pp. 271-99. È. Afæa@r, Ya@dga@rha@-ye Yazd, 3 vols., Tehran, 1348-54 ˆ./1969-75. M. van Berchem, "Une inscription du sultan mongol Uldjaitu," Me‚langes Hartwig De‚renbourg, Paris, 1909, pp. 367-78. A. D. H. Bivar, "Selju@qid Ziya@rats of Sar-i Pul (Afghanistan)," BSO(A)S 29, 1966, pp. 57-63. Idem and E. Yarshater, eds., Persian Inscriptions down to the Early Safavid Period VI. Ma@zandara@n Province: Eastern Mazandaran I, Corpus Inscr. Iran., Part IV, London, 1978. S. S. Blair, "The Epigraphic Program of the Tomb of Uljaytu at Sultaniyya: Meaning in Mongol Architecture," Islamic Art 2, 1987, pp. 43-96. Idem, "Legibility vs. Decoration in Islamic Epigraphy: The Case of Interlacing" in I. Lavin, ed., World Art: Themes of Unity in Diversity, Acts of the XXVIth International Congress of the History of Art, University Park and London, 1989, pp. 329-31. Idem, The Monumental Inscriptions from Early Islamic Iran and Transoxiana, Leiden, 1992. S. S. Blair, J. M. Bloom, and A. E. Wardwell, "Reevaluating the Date of the 'Buyid' Silks by Epigraphic and Radiocarbon Analysis," Ars Orientalis 22, 1992, pp. 1-41. O. G. Bol'shakov, "Arabskie nadpisi na polivno¥ keramike Sredne¥ Azi¥ IX-XII vv" (Arabic inscriptions on glazed ceramics from Central Asia IX-XII centuries), Epigrafika Vostoka 12, 1958, pp. 23-38; 15, 1963, pp. 73-87; 16, 1963, pp. 35-55; 17, 1966, pp. 54-62. A. Bombaci, The Ku@fic Inscription in Persian Verses in the Court of the Royal Palace of Mas¿u@d III at Ghazni, Rome, 1966. A. S. Bretanitski¥, Zodchestvo Azerbaidzhana XII-XIV vv (The architecture of Azerbaijan XII-XIV centuries), Moscow, 1966. M. Bulatov, Mavzole¥ Samanidov (The tombs of the Samanids), Tashkent, 1976. E. Combe, J. Sauvaget, and G. Wiet, eds., Re‚pertoire chronologique d'e‚pigraphie arabe, 17 vols., Cairo, 1931—. B. Denike‚, "Quelques monuments de bois sculpte‚ at Turkesta@n occidental," Ars Islamica 2, 1935, pp. 69-83. J. Donohue, "Three Buwayhid Inscriptions," Arabica 20, 1973, pp. 74-80. R. Ettinghausen, "The Bobrinsky 'Kettle': Patron and Style of an Islamic Bronze," Gazette des Beaux-Arts 24, 1943, pp. 193-208. S. Flury, "Bandeaux ornemente‚s aà inscriptions arabes: Amida-Diarbekr," Syria 1, 1920, pp. 235-49, 318-28; 2, 1921, pp. 54-62. Idem, "Le de‚cor e‚pigraphique des monuments de Ghazna," Syria 6, 1925, pp. 61-90. Idem, "La mosque‚e de Nayin," Syria 11, 1930, pp. 43-58. E. Galdieri, Esáfaha@n Masgid-i G¦um¿a 3: Research and Restoration Activities, 1973-1978, New Observations, 1979-1982, Rome, 1984. L. Gardet, "al-Asma@÷ al-H®osána@" in EI2 I, pp. 714-17. R. Ghirshman, "Argenterie d'un seigneur sassanide," Ars Orientalis 2, 1957, pp. 77-82. L. T. Giuzalian, "The Bronze Qalamdan (Pen-Case) 542/1148 from the Hermitage Collection (1936-1965)," Ars Orientalis 7, 1968, pp. 95-119. L. Golombek and D. Wilber, The Timurid Architecture of Iran and Turan, 2 vols., Princeton, 1988. G. D. Guest and R. Ettinghausen, "The Iconography of a Ka@sha@n Luster Plate," Ars Orientalis 4, 1961, pp. 25-64. W. B. Henning, "New Pahlavi Inscriptions on Silver Vessels," BSOAS 22, 1959, pp. 132-34; repr. in idem, Selected Papers, Acta Iranica 15, Leiden, 1977, pp. 533-35. E. Herzfeld, "Postsasanidische Inschriften," AMI 4, 1931-32, pp. 14-56. Idem, "A Bronze Pen-Case," Ars Islamica 3, 1936, pp. 35-44). Idem, "Arabische Inschriften aus Iran und Syrien," AMI 8, 1936-37, pp. 78-102. Idem, "Damascus. Studies in Architecture—II," Ars Islamica 10, 1943, pp. 13-70. Abu@ ¿Abd-Alla@h Moháammad Jahæ^a@r^, Keta@b al-wozara@÷ wa'l-kutta@b, ed. M. Saqqa@ et al., Cairo, 1357/1938. N. Khanikoff, "Les inscriptions musulmanes du Caucase," JA, August 1862, pp. 57-155. L. Komaroff, The Golden Disk of Heaven: Metalwork of Timurid Iran, Costa Mesa, 1992. V. A. Krachkovskaya, "Evolutsiya kufichesckogo pis'ma v sredne¥ Azi¥" (The evolution of Kufic script in Central Asia), Epigrafika Vostoka 3, 1949, pp. 3-27. N. Lowick, S^ra@f XV: The Coins and Monumental Inscriptions, London, 1985. A. S. Melikian-Chirvani, Islamic Metalwork from the Iranian World 8th-18th Centuries, London, 1982. G. C. Miles, "Epitaphs from an Isfahan Graveyard," Ars Islamica 6, 1939, pp. 151-57. H®. Modarres^ T®aba@tÂaba@÷^, Torbat-e pa@ka@n, 2 vols., Qom, 2535 (=1355) ˆ./1976. M. T. Mosátáafaw^, Eql^m-e Pa@rs, Tehran, 1343 ˆ./1964. V. N. Nastich and B. D. Kochnev, "K attributsii mavzoleya Shaikh-Fazil" (On the attribution of the tomb of Shaikh Fa@zµel), Epigrafika Vostoka 24, 1988, pp. 69-76. G. A. Pugachenkova, Mavzole¥ Arab-Ata (The Arab-ata mausoleum), Tashkent, 1963. L. Richter-Bernburg, "Am^r-Malik-Sha@ha@nsha@h. ¿Adáud ad-daula's Titulature Re-examined," Iran 18, 1980, pp. 83-102. D. Shepherd and W. Henning, "Zandan^j^ Identified?" in R. Ettinghausen, ed., Aus der Welt der islamischen Kunst: Festschrift Ernst Kühnel, Berlin, 1959, pp. 15-40. V. A. Shishkin, "Nadpisi v ansamble Shakhi-Zinda" (The inscriptions in the ˆa@h-e zenda complex) in Zodchestva Uzbekistana (The architecture of Uzbikistan), Tashkent, 1970, pp. 7-71. M. Soroodeh, "Drei alte kufische Inschriften aus Iran," ZDMG 125, 1975, pp. 315-16. J. Sourdel-Thomine, "Inscriptions seljoukides et salles aà coupoles de Qazw^n en Iran," REI 42, 1974, pp. 3-43. L. Volov [Golombek], "Plaited Kufic on Samanid Epigraphic Pottery," Ars Orientalis 6, 1966, pp. 107-33. O. Watson, Persian Lustre Ware, London, 1985. A. J. Wensinck et al., Concordance et indices de la tradition musulmane, 6 vols., Leiden, 1936-88.

(SHEILA S. BLAIR)

iv. SAFAVID AND LATER INSCRIPTIONS

The principal characteristic of epigraphy in Persia after the advent of the Safavids (907/1501) is the emphasis on Persian poetry and pious Shi¿ite texts with an iconographic potency and deliberate frequency hitherto unknown. Arabic remained the language of koranic and Hadith quotations while Persian became increasingly prominent through its use for historical and poetic inscriptions in both architecture and decorative arts. Cursive scripts, especially t¯olt¯ and nasta¿l^q (see CALLIGRAPHY) supplanted the earlier Kufic and its variants in all but the essentially decorative repeat patterns in secondary positions.

Monumental inscriptions. There are no systematically recorded or analyzed collections of inscriptions for this period, nor are there any comprehensive architectural surveys as there are for the Il-Khanid and Timurid periods. The essential source for Safavid inscriptions is Honarfar's work on Isfahan, the city whose buildings provide the substance of Safavid monumental inscriptions in general. Apart from Isfahan, major Safavid, Zand and Qajar inscriptions survive in buildings in Ardab^l, Kerma@n, Qom, Maæhad, Shiraz, Yazd, and Tehran as well as in the Ma@zandara@n region. Persian surveys on these cities and the provinces in general, published in monographs by Anjoman-e a@t¯a@r-e mell^ (q.v.), document the relevant inscriptions. Foundation and commemorative inscriptions, royal decrees (farma@n), and endowment (waqf) texts constitute the largest official epigraphic evidence; tombstones, once documented systematically, will enhance our understanding of the popular trends in the epigraphy of this period.

Aside from recording the historical cicumstances of the construction—patron's name, builder, calligrapher, date—foundation inscriptions tend to include proclamations that delineate the main theological and political orientation of the Safavid state, above all Twelver Shi¿ism, declared by Shah Esma@¿^l I (907-30/1501-24, q.v.) as the state religion.

Several trends in the epigraphy of this period are already present in the earliest example of a full epigraphic program on a major Safavid building, namely the mausoleum of Ha@ru@n-e wela@yat in Isfahan (Plate V), built during the reign of Esma@¿^l I in 918/1513 (Honarfar, Esáfaha@n, pp. 360-69). The Arabic foundation inscription, written in t¯olt¯ and placed over the tomb entrance, includes appropriately enough a Hadith mentioning Aaron (Ha@ru@n) and states Esma@¿^l's claim of descent from ¿Al^. Esma@¿^l's reign as a caliphate and his role as "the friend of God" (wa@l^ lewa@÷ al-wela@ya) and as a warrior in the cause of God (al-g@a@z^ al-moja@hed f^ sab^l Alla@h) have been interpreted as deliberate affronts to the Sunni Ottoman usurpation of such titles (Hillenbrand, 1986, p. 762; Honarfar, Esáfaha@n, p. 361). Moreover, the concepts of wela@ya and kòela@fa were essential for the legitimization of the rule of the Safavid shahs especially in constructing genealogical links to the Imams in early Safavid historiography (Quinn, pp. 76-90). Above this inscription are tile panels with blessings on the ±aha@rdah ma¿sáu@m (q.v.) reinforcing the Shi¿ite theme of the epigraphic ensemble, and a common epigraphic feature of the majority of religious buildings of this later period.

Less esoteric than the epigraphic program at Ha@ru@n-e wela@yat is the political propaganda routinely included in later inscriptions. For a brief period the phrase "in the caliphate of . . ." (dar zama@n-e kòela@fat-e; f^ ayya@m kòela@fa) continued to be included, albeit infrequently, in some inscriptions added in T®ahma@sb's time (930-84/1524-76) to buildings as prominent as the Masjed-e Jom¿a of Isfahan or as small as the D¨u'l-faqa@r mosque in the ba@za@r of Isfahan (Honarfar, Esáfaha@n, pp. 92, 384).

Equally pointed in message is the expression of Shi¿ite sentiment in earlier Safavid inscriptions. During the reign of Shah T®ahma@sb, an elaborately worded Arabic inscription on the northern ayva@n of the old courtyard at the shrine of Fa@tÂema Ma¿sáu@ma in Qom declares T®ahma@sb, among other honorifics, "the successor to the pure and infallible Imams" (Modarres^ T®aba@tÂaba@÷^, I, p. 76). Shah ¿Abba@s I's oft-repeated personal devotion to ¿Al^ is expressed in the clearest terms on the portal inscription at the small mosque known as Masjed-e sofra±^ in Isfahan, where he is called the "loyal slave (@gola@m be-ekòla@sá) of the Commander of the Faithful ¿Al^ b. Ab^ T®a@leb" (Honarfar, Esáfaha@n, p. 475).

Some of the tenor of the message was preserved in the 17th century. Thus the phrase "the propagator of the faith of the infallible Imams" (morawwej madòhab al-a÷emma al-ma¿sáu@m^n), or "the twelve Imams" (morawwej madòhab al-a÷emma al-at¯na@ ¿aæar), became standard in foundation inscriptions of many religious buildings. The first variant of the phrase is embedded in an elaborate Arabic inscription dated 1012/1603 at the entrance to the mosque of Shaikh LotÂf-Alla@h in Isfahan (Honarfar, Esáfaha@n, p. 402) while the second version is found on an inscription also from the reign of Shah ¿Abba@s I at the shrine of Shaikh S®af^-al-D^n in Ardab^l (D^ba@j, p. 64). The formulaic expression of Shi¿ite devotion was preserved even as Persian increasingly replaced Arabic in foundation inscriptions. At the madrasa of AÚqa@ Ka@fu@r in Isfahan Shah ¿Abba@s II (1052-77/1642-66) is described in Persian as the "propagator of the rightful faith of their holiness the infallible Imams" (morawwej-e madòhab-e beháaqq-e háazµara@t-e a÷emma-ye ma¿sáu@m^n; Honarfar, Esáfaha@n, p. 606).

Shi¿ite iconography also provided fertile grounds for a thematic and visual unity in the epigraphic system of the later period. The most common passage to be found is the Prophet's saying: "I am the city of knowledge and ¿Al^ is its gate" (ana mad^nat al-¿elm wa ¿Al^ ba@boha@). Isolated in a cartouche or fitted within a larger text, the iconographic subtlety of the saying is made evident by its placement above the entrance portals and on the doors of religious buildings.

In the earliest prominent example of its use in the Ha@ru@n-e wela@yat, "¿Al^ [is] the gate" may also be a eulogy of Shah Esma@¿^l who, in some of his own verses, had compared himself to the first Shi¿ite Imam (Hillenbrand, p. 763; Honarfar, Esáfaha@n, p. 368; Babayan, p. 36). Such esoteric meanings found greater currency in decorative arts, as discussed below, than in monumental epigraphy, which quickly tended towards a more conventional use of the passage in a wide range of buildings from the prominent entrance portal of Shah ¿Abba@s I's congregational Masjed-e Shah in Isfahan to a late Qajar addition to the shrine of Fa@tÂema Ma¿sáu@ma at Qom to the small neighborhood Rahá^m K¨a@n mosque in Isfahan from the reign of Na@sáer-al-D^n Shah (1264-1313/1848-96; Honarfar, Esáfaha@n, pp. 430, 797; Modarres^ T®aba@tÂaba@÷^, I, p. 108).

Persian poetry as appropriated in Safavid and later epigraphy epitomizes this taste for an intricately layered iconography in which the functional and the literary elements are thematically unified. Cisterns, ablution basins, and doors offer some of the most sophisticated examples of the Safavid predilection for such iconographic constructs. At a cistern built in 1055/1645 by the Safavid vizier Sa@ru@ Taq^ at the eastern corner of the courtyard of the madrasa of Da@r-al-æefa@ in Qom, two quatrains in Persian are placed above the entrance to the subterranean water tank (Modarres^ T®aba@tÂaba@÷^, II, pp. 140-41). The first two verses dedicate the cistern to the memory of Imam H®osayn b. ¿Al^, who was martyred in the desert of Karbala@ still thirsting for water. The Shi¿ite sentiment of the theme is further heightened by the curse on Yaz^d, the Omayyad caliph responsible for the martyrdom of H®osayn and his followers, and the curse on Yaz^d's tomb included as the chronogram in the quatrain.

Persian poems on a pair of silver doors ordered by Shah S®af^ I in 1046/1636 for Masjed-e ˆa@h compare the Isfahan mosque with Masjid-al-H®ara@m in Mecca adding eulogies of S®af^ as the great sovereign of the age, who shall for his angel-like nature effortlessly attain entry into Mecca through these doors (Honarfar, Esáfaha@n, pp. 433-34; Allen, p. 130). The metaphor of the door as the gate opening Ka¿ba onto Isfahan (æod dar-e Ka¿ba dar Sáefa@ha@n ba@z) makes up the chronogram in the last hemistich.

The great currency of Persian poetry in later epigraphy can be documented in a wide range of buildings from the beginning of the 16th century through the end of the Qajar period. The earliest manifestations of this trend are often limited to a few verses; either giving some historical information as, for instance, the names of the patron and the architect (e.g., Ha@ru@n-e wela@yat; Honarfar, Esáfaha@n, p. 361) or evoking an image appropriate to the context as in the case of the Hadith about Imam ¿Al^ being the gate to the city of knowledge on the door of Masjed-e D¨u'l-faqa@r in Isfahan (Honarfar, Esáfaha@n, p. 384), or even more ornately executed on the silver facings of the doors to the Madrasa-ye ±aha@r-ba@g@ (Allen, pp. 130-32).

Although Persian verses were increasingly included in Timurid buildings, it was in the Safavid period when inscriptions of Persian poetry became an essential component of the architectural setting. In that capacity, poetry is employed to decorate, to describe, and to communicate the meaning of the architecture. Verses attributed to H®a@fezá, written in nasta¿l^q and enclosed in cartouches, celebrate the peaceful pleasures of life in a late 16th century mansion in Na@÷^n (Gropp and Najmabadi, pp. 194-95). They are also made to convey the theme of the painted scenes on the walls above even though the depicted stories derive from the works of other poets (e.g., Nezáa@m^ Ganjav^) rather than H®a@fezá. According to Qa@zµ^ Ahámad (ed. Minorsky, p. 143), @gazals by H®a@fez and H®osa@m-al-D^n Madda@há adorned the ayva@n and the portals of the Ùehel Sotu@n in Qazv^n.

It is, in fact, rare to find classical poetry in monumental epigraphy. More common is poetry composed specifically for a building. Characteristically, these poems lavish praise on the royal patron, on the Prophet and the Imams, and they describe the circumstances of founding, the component parts of the building, and its meaning. At the palace of Ùehel Sotu@n (q.v.) in Isfahan, thirty couplets report the event of a fire in 1118/1706 and the subsequent repairs of the building on the order of Shah SoltÂa@n-H®osayn (Honarfar, Esáfaha@n, pp. 572-74). Eulogies for the king are followed by a series of metaphorical and direct references to precise parts of the building that were repaired and embellished at the time of the inscription. The contemporary inscriptions at Madrasa-ye ±aha@r ba@g@ in Isfahan contain an unusually large number of panegyrics interspersed with metaphoric references to the architecture composed in fanciful Persian verse (Honarfar, Esáfaha@n, pp. 685-722).

The significance of Persian poetry in monumental epigraphic programs, on the rise since the Timurids, reaches its zenith in the Qajar period. The Qajar additions and repairs to famous shrines such as the shrine of Fa@tÂema Ma¿sáu@ma in Qom occasioned an impressive outpouring of versified Persian inscriptions (Modarres^ T®aba@tÂaba@÷^, I, pp. 48-111). Not only is there an overwhelming quantitative presence but poetry is made visually more prominent and legible. Previously, poetry tended to be contained either in panels often, by necessity, set in flat side walls of ayva@ns, or was written in a continuous band. Qajar designers opted instead for superposed pairs of cartouches each containing a verse (Hillenbrand, 1983, p. 358). This visually propitious organization of paired, ornately framed distichs is best exemplified on the entrance portals of two Qajar mosques: the exquisite Rokn-al-Molk mosque in Isfahan (Plate VI; Honarfar, Esáfaha@n, pp. 805-21), and the mosque of AÚqa@ Bozorg in Ka@æa@n (Nara@q^, pp. 254-62; Hillenbrand, 1983, fig. 8).

Note should also be made of a rare and curious stone tablet with twenty four couplets composed in Turkish and installed at the inner face of the mountainside that formed a natural fortification at Kala@t-e Na@der^, the stronghold briefly occupied by Na@der Shah (K¨osrav^, pp. 55-58). The poem lavishes praise on the Creator, the Prophet, the Imams, and above all on Na@der Shah (r. 1148-60/1736-47) but seems to have been left unfinished like the only major building at this site.

Other important categories of inscriptions in Persian are royal decrees and endowment texts. Both are found in large numbers in religious buildings: T®ahma@sb's decree imposing bans on theologically unlawful activities in the city of Ardab^l placed prominently at the main entrance into the shrine of Shaikh S®af^-al-D^n (D^ba@j, pp. 68-69); decrees issued by ¿Abba@s I, Fathá-¿Al^æa@h Qa@ja@r and Na@sáer-al-D^n Shah Qa@ja@r concerning tax and tariff exemptions installed in the Masjed-e ˆa@h in Isfahan (Honarfar, Esáfaha@n, pp. 444-46, 458-60); and decrees in verse issued by Shah T®ahma@sb, Shah ¿Abba@s I, etc. in Masjed-e Meyda@n in Ka@æa@n (Nara@q^, pp. 211-34). Tombstones also form an important category of