Canepa_Achaemenid Sites Sasanian Identity.pdf

8/14/2019 Canepa_Achaemenid Sites Sasanian Identity.pdf

1/34

American Journal of Archaeology 114 (2010) 56396563

Technologies of Memory in Early Sasanian Iran:Achaemenid Sites and Sasanian Identity

MATTHEW P. CANEPA

AbstractThis article analyzes the techniques by which the kings

of the early Sasanian dynasty engaged the past and shapedthe experience of future generations. I concentrate onthe innovations and legacy of the first two kings of kingsof the dynasty, Ardashir I (r. 224239/40 C.E.) and hisson Shapur I (239/40270/2 C.E.). These sovereignsfashioned a new and politically useful vision of the past toestablish their dynastys primacy in Persia and the widerIranian world, eclipsing their Seleucid, Fratarakid, andArsacid predecessors. I identify and examine the artistic,architectural, and ritual means by which the early Sasani-ans conformed the built and natural environment of theirhomeland to their grand new vision of the past. I arguethat the Achaemenid patrimony of the province of Parsplayed an important role in these efforts, serving as inspi-rations and anchors for the Sasanians new creations.*

introduction

The Sasanian empire (224642 C.E.) was the lastgreat Iranian empire to rule over Mesopotamia, Iran,

and portions of south and Central Asia before thecoming of Islam.1Although the Sasanians were a newregime that disturbed the status quo of nearly five cen-turies of Arsacid rule, they were quick to assert thatthey were rightful heirs of an ancient line of Iraniankings and heroes. Up to this point, scholarly debateson Sasanian memory have largely centered on textualsources, with archaeological and visual evidence play-ing a subordinate role. These debates have tendedto focus on how well the Sasanians knew the Achae-menid dynasty and whether they wanted to rebuild the

Achaemenid empire, one small subset of what I argueis a more complex problem.2Scholarship is largely inagreement that although the early Sasanians countedthe Achaemenids as ancestors, they had imperfectknowledge of them and did not set out to recreate theAchaemenid empire.3While they often praised theirancestors, no Sasanian ever mentions the Achaemeniddynasty in a primary source.4They seem to have under-

*This article was finished with support from a Charles A.Ryskamp Research Fellowship from the American Council ofLearned Societies (ACLS). It was researched while represent-

ing the Archaeological Institute of America as a North Amer-ican Fellow of the German Archaeological Institute, Berlin.Thanks to Dietrich Huff for reading an earlier version of thismanuscript and for sharing his experience with many of thesesites; any mistakes or misunderstandings are my own. Trans-lations from all languages and all images and maps are myown unless otherwise indicated. Transliterations for all Irani-an languages follow those ofEncyclopaedia Iranica, as does theconvention of using the New Persian names of Sasanian kings.I use diacriticals only where a specific Middle Persian word isused or in quoted Middle Iranian source material. Refer toEncyclopaedia Iranicafor the proper transliterations.

1 The Sasanian empire began in 224 C.E. when ArdashirI overthrew Ardawan IV, his Arsacid overlord (Schippmann1987, 650). The Arab victory over the Persian army at Niha-vand in 642 and the death of Yazdgerd III in 651 mark the endof the empire and the dynasty, respectively, although Yazdg-erd IIIs son and grandson continued to militate for the resto-ration of the Sasanian empire from their exile in Tang China(Pulleyblank 1992, 42526; Compareti 2003; Canepa 2010a).

2Foundational works: Nldeke 1879, 3 n. 1; Yarshater 1971,1983. These offered an important corrective to the assump-tion that had prevailed previously in both centuries (reflect-ed, among other places, in Rawlinson 1876; Ghirshman 1954)that the Sasanians understood the Achaemenids as modernscholarship did and that they sought to resurrect the empire

whole-scale (Shahbazi 2001).3 On the character of the Iranian national history, see Yar-

shater 1983, 36970 (for its differing post-Sasanid literary

manifestations). For studies that stress discontinuity and lossof memory, see Yarshater 1971; Kettenhofen 1984, 1994; Roaf1998; Shayegan 1999, 77157, 2012; Rubin 2000; Shahbazi2001; Huyse 2002, 2008. Shayegan (1999) provides a thor-ough account of the earlier debate. For studies that considerthe evidence of an engagement with the past, see Wiesehfer1982, 1986; Frye 1984, 293; Winter 1988, 2644; Gnoli 1991;Wolski 1993; Wiesehfer 1994b; Daryaee 1995, 20012002,2006, 2009, 2010; Winter and Dignas 2001, 7584; Harper2006; Huff 2008; Canepa 2009.

4Although Shapur I states, in his Kaba-ye Zardosht in-scription (KZ) that Pars and E

-ra- nahr was the land of his

ancestors (Middle Persian ahe-naga-n; Greekprogonoi), wherethey held property, he did not find it necessary to clarify ex-actly whom he understood these ancestors to be: And the[captured soldiers] that came from the kingdom of Rome,from Ane-ra- n, we deported to E- ra- nahr: to Pars, Pahlaw, Xuz-esta-n and Asu- resta-n and other lands, where we, [our] father,[our] grandfathers and [our] ancestors held property, [andthere] they [were] settled (ZK 30; Gnoli 1991, 589). Thelack of specificity in ZK 30, Narsehs Paikuli inscription (NPiB1.03-B3-4.04), and the Persepolis inscriptions of Shapur theKing of the Sakas do not prove they lacked knowledge of theAchaemenids (Gnoli 1989, 13637, 178; Daryaee 1995, 13233, 140; 20022003; 2010; Shayegan 1999, 8392; Boyce 2001,12728).


2/34

MATTHEW P. CANEPA564 [AJA 114

stood those whom scholarship calls the Achaemenidsto be the last scions of the legendary Kayanid dynastybefore the invasion of Alexander, although, in thisregard, no primary source exists that can attest to anearly Sasanian claim of Kayanid ancestry either.5Thisdebate, largely based on textual sources, offered a valu-

able corrective to earlier notions that the Sasaniansconsciously sought to resurrect the Achaemenid em-pire as the Achaemenid empire. It does not, however,offer a useful framework for analyzing the early Sasani-ans own understanding of their place in Persian andwider Iranian history nor the close physical proximityand parallels between Sasanian and Achaemenid visualculture, architecture, epigraphy, and ritual.

The line of inquiry I wish to introduce shifts theemphasis to archaeological evidence and the archi-tectural, visual, and ritual techniques by which theearly Sasanian dynasty shaped the past.6The last twodecades have nurtured a fluorescence of scholarship

on cultural memory in the work of historians, arthistorians, archaeologists, anthropologists, and soci-ologists. Although none speaks directly to the LateAntique experience, this debate offers some broadlyuseful insights. The term site of memory, or lieu demmoire, has become a common critical term to speakabout issues of the past, place, and collective memoryin such a context.7My use of the term only applies toLate Antique Iran and contrasts with its sense in con-temporary discourse where such sites function as sym-bols of modern societies alienation from their past.In the Late Antique Iranian world, a site of memorymore often than not was the portal to the past andthe means by which the kings of kings actively partici-pated in cultural memory (table 1). I argue that the

Sasanian kings of kings approached the past, couldgain control of it, or introduce dramatic changes toit through the natural and built environment of theirempire. I refer to these joint practical, artistic, and ar-chitectural efforts as technologies of memory wherebycertain images, structures, and activities facilitated avital and compelling experience of the past.8To fullyunderstand the early Sasanians efforts to come toterms with the past, as archaeologists and art histori-ans, we must widen our conceptual categories to viewthe interrelation of these elements. I concentrate onthe efforts of the first two kings of kings of the Sasanian

5 The Avesta, the most ancient collection of texts of the Zo-roastrian religion, contains the earliest evidence of the Ira-nian epic tradition, which celebrated the mythical Kayaniddynasty (Avestan Kauui-). This epic tradition coalesced in awritten form only in the late Sasanian empire as the Xwada-y-na-magand reached its final form in Ferdowsis a-hna-ma(Bookof Kings). Although they shared the same home province asthe Achaemenids, spoke a descendent of their language, andlived among the ruins of their monuments, the legendary

Kayanid kings and heroes eventually became the dominanttradition for the Sasanian dynasty and wider Iranian world.(Yarshater 1971, 1983; Daryaee 1997, 2002, 2010; Huyse 2006,18289). The Achaemenids were already influenced by theAvesta and used its concepts to bolster their royal ideology(Lincoln 1996; Skjaerv 2005; Soudavar 2010 [my thanks tothe author for providing me a copy before the volume was re-leased]). In this regard, Sasanian interest in the Kayanids canbe seen as another mark of continuity rather than rupture.

6For the theory of the reciprocal nature of art, architecture,and ritual in late antiquity, see Canepa 2009, 721; 2010a.

7 Halbwachs 1925, 1941, 1992; Foucault 1986; Connerton1989; Lincoln 1989; Le Goff 1992; Nora 1992; Wood 1994; Ol-

ick 1999; Alcock 2002; Nelson and Olin 2003; Van Dyke andAlcock 2003; Young 2003. For a critique of anthropologicaldiscourse on history and memory, see Radstone 2000; Her-zfeld 2001, 5589. Although directed more at contemporaryissues, Kansteiner (2002) presents a useful critique as well.For modern misappropriations of the pre-Islamic heritagefor nationalist agendas (both Iranian and Euro-American),see Abdi 2001; Majd 2003.

8 By using the terms techniques or technologies, I en-

gage both sociological theory that analyzes the means bywhich memory is recorded and replicated (Olick 1999) and alarger debate that has considered how power shapes culture.These technologies of power focus primarily on achievingthe subjection of bodies and control of populations, creat-ing the perception that such things as truth, justice, history,and the divine are independently (a priori) existing absolutes(Foucault 1965, 1973, 1977a, 1977b, 1978). In this regard, Ifocus on the manipulation of the past on the part of thosein power rather than a popular collective memory, anoth-er concept that has received a great deal of attention (e.g.,Connerton 1989). While Achaemenid-inspired images ap-pear in wider expressions of Sasanian visual culture, such as

Table 1. Achaemenid Kings of Kings.

Name Regnal Dates (B.C.E.)

Cyrus the Great ca. 558530

Cambyses 530522

Bardiya (Gauma-ta) 522

Darius I 521486

Xerxes I 486465

Artaxerxes I 465424

Xerxes II 424423

Darius II 423405/4

Artaxerxes II 405/4359/8

Artaxerxes III 359/8338

Arses (Artaxerxes IV) 338336

Darius III 336330

Bessos (Artaxerxes V) 330329


3/34

ACHAEMENID SITES AND SASANIAN IDENTITY2010] 565

dynasty, Ardashir I (r. 224239/40 C.E.) and his sonShapur I (239/40270/2 C.E.). These sovereigns inno-vated a repertoire of early Sasanian memory practicesthat at once shaped the early Sasanians experience ofthe past and inspired the activities of their successors.Reacting in part to the activities of their predecessors

and competitors, they modified august Achaemenidruins such as Persepolis and Naqsh-e Rostam while cre-ating grand, new Persian monumental zones. In bothcases, they cloaked their often radically new visual andritual innovations with forms and practices carefullydrawn from ancient Persian tradition. Their ambitiousbuilding campaigns and ritual activities meaningfullyconnected these ancient sites and their new creationsinto the living experience of the empire. They wereintended to yield a convincing perception to theirsubjects and vassals of what was an ideologically co-herent and useful, though, perhaps to the historian,less than accurate, past.

Iranian history between Alexander and Islam (ca.331 B.C.E.642 C.E.) is an extremely challenging pe-riod to study because of its fragmentary textual sourcesand often unprovenanced visual material. I adhere toa hierarchy of sources that I have adapted to includevisual and archaeological material (see tables 13).9This hierarchy privileges unquestionably contempo-rary and authentic materialsuch as well-attributedrock reliefs, inscriptions and seals, and archaeologi-cally excavated structures and artifacts, which formthe primary sourcesover unprovenanced or post-Sasanian materialsuch as post-conquest Pahlavitexts or unprovenanced silver platewhich forms thesecondary sources. Non-Iranian material, includingGreek, Latin, Arabic, and Chinese textual sources andnon-Iranian visual material are tertiary.

the persian roots of sasanian memorypractices

The Sasanian dynasty had its roots in the province ofPars in southwestern Iran, the homeland of the Achae-

menid empire (fig. 1).10Although their empire hadbeen defunct for centuries, the ruined palaces, sacredsites, and tombs of the Achaemenid kings of kings stillloomed large on the physical and ideological horizonsof Pars long after their fall. The vestiges of this great,yet half-understood, Persian heritage confronted all

who held power in the province and eventually stimu-lated the Sasanians own memorial and monumentalpractices. The most impressive concentration of vis-ible Achaemenid remains in Pars lay at the westernend of the Marv Dasht plain. Here, the plain meetsthe mountains, and the Polvar River divides the moun-tains into two spurs, the Hosayn Kuh to the north andthe Kuh-e Rahmat to the south. Persepolis massiveplatform rose below Kuh-e Rahmat, while, about 6.25km to the north, the Achaemenids royal necropolis,called today by its New Persian nickname of Naqsh-eRostam, marked the final spur of the Hosayn Kuh. Be-tween these two ancient sites grew Staxr, post-Achae-

menid Pars principle city and religious center.11Fromat least the early Sasanian period, the inhabitants ofPars conceived of Staxr, Persepolis, and Naqsh-e Ros-tam as a whole.12With their colossal architecture andfine relief sculpture, Persepolis and Naqsh-e Rostamemerged as objects of special pride and fascination forthe post-Achaemenid rulers of Pars and, eventually,became the raw material out of which the Sasanianscrafted their early expressions.13

While they presented themselves as the true stew-ards of the ancient Persian sites, the Sasanians memo-rial activities owed a great deal to their more proximatepredecessors in the region. Indeed, the Sasaniansinitially drew from, and reacted to, the accumulatedHellenistic, Arsacid, and local post-Achaemenid Per-sian reinterpretations of the sites. After entering Parsain 331 B.C.E., Alexander held victory games and abanquet at Persepolis, a celebration that culminatedin the destruction of the palace.14While, in this in-stance, Persepolis served as a monument to Hellenicvengeance, other Achaemenid structures retained

in seals and silver (Roaf 1998; Harper 2006), here my focusis discretely the monumental and ritual productions of the

Sasanian court.9 Gignoux 1979, 1984; Gyselen 2001; Canepa 2009,xviixviii.

10 The historical province extended beyond the confinesof the modern province of Fars (Wiesehfer 1999). I use theMiddle Persian version of the name to refer to the provincein late antiquity.

11 New Persian Estakhr; Middle Persian Staxr (stxl), mean-ing fortress, reconstructed as Old Persian *pa-rsa.staxra(Stronghold of Parsa); cf. Avestan staxra- (strong, hard) (Bar-tholomae 1904, 1591; Herzfeld 1935, 45; Bivar 1997, 643).Staxr lies ca. 2 km southeast of Naqsh-e Rostam and 5 kmsouth-southwest of Persepolis.

12Shahbazi 1977, 2001; Bivar 1997, 643.13 Sovereigns in many eras, including the Pahlavis, cultivat-

ed Persepolis as a site of memory that could imply a connec-tion with the past, no matter how tenuous or absurd (Shahbazi1977; Mousavi 2002). For the Friday mosque constructed outof Achaemenid remains, ca. 660 C.E., see Whitcomb 1979,36366. For foundational work on Qajar use of ancient Irani-an visual and archaeological heritage, see Grigor 2007, 2009.

14 Despite the propaganda of revenge, his most impor-tant goal was to prevent Persepolis symbolic and monetaryresources for kingship from falling in the hands of a usurp-er (Curt. 5.7.110; Diod. Sic. 17.72; Plut. Vit. Alex. 38; cf. Arr.Anab. 3.18.1112; Hammond 1992; Sancisi-Weerdenburg1993; Wiesehfer 1996, 1046).


4/34


their original significance. Alexander made a show ofcaring for the Tomb of Cyrus to associate himself withthe founder of the Achaemenid dynasty.15Persepolis

significance as an aggressive, Macedonian victory mon-ument did not endure long after Alexanders death.Despite its damaged state, the multivalent symbolicpotential of the site attracted Alexanders successorsquickly thereafter for different goals (see table 2). In316, Peukestas, Alexanders companion whom he hadappointed governor of Parsa, staged an elaborate ban-quet for his army at Persepolis before the showdownbetween Eumenes and Antigonos Monophthalmos,where he conducted lavish sacrifices to Alexanderand Philip.16The banquet hosted both Macedonianand Iranian contingents, and its seating arrangementsand sacrifices evoke Persian protocol.17This suggests

that Peukestas, popular and trusted among the Per-sian nobility, intended to capitalize on Persepolis asan open-ended symbol that could speak to the events

different constituencies.18Herzfeld discovered five inscribed stone slabs in the

vicinity of the platform of Persepolis, likely originallyconnected with cult furniture.19These slabs bear Greekdedications to Zeus Megistos, Athena Basileia, Apollo,Artemis, and Helios as discrete deities, with no overtsuggestion of assimilation to Iranian deities.20Whilethe Macedonians could have created them in a cal-culated act of triumphal imperialism after the initialsack of Persepolis, Peukestas banquet provides theirlikeliest original context.21Whatever their origin orpurpose, the installation of these slabs provides archae-ological evidence of an important shift in Persepolis

15Arr. Anab.6.29.414; cf. Strabo 15.37.16 Diod. Sic. 19.223; Plut. Vit. Eum.14.3; Billows 1995, 35;

1997, 93.17 Calmeyer 1982, 185; Wiesehfer 1991, 130; 1994a, 534;

Henkelman 2007.18Wiesehfer 1996, 107; 2007, 389.19 The slabs were not associated archaeologically with the

later Fratarakid Temple, though Herzfeld (1935, 446)guessed that that was their origin; see also Wiesehfer 2007,

39.20 Herzfelds hypothesis that they arose from Iranian pa-

tronage and honored Iranian gods assimilated to the Greekpantheon should not be discounted, nor should a later rein-terpretation in an Iranian manner be rejected (Boyce 1991,107). Both practices were important religious and politicaltools in Hellenistic and early Arsacid Iran.

21Wiesehfer 1996, 108.

Fig. 1. Map of principal cities and sites of the Sasanian empire.


5/34


significance away from its original function, as palaceand ceremonial center of the Achaemenid empire, toone that capitalized on a more generalized symbol-ism deriving from its new, more malleable status asa ruin. This also marks the beginning of centuries ofMiddle Iranian ritual and artistic activity focused on

recovering, reanimating, and eventually reinventingthe sites power.

By the beginning of the second century B.C.E., lo-cal rulers, rather than Macedonian governors, admin-istered Pars (table 4).22They remained loyal vassalsof the Seleucids while that dynasty held power; how-ever, within their province, which they ruled largelyautonomously, they pursued a vigorous program ofengagement with the Achaemenid past.23They calledthemselvesfrataraka-, a title that derived from the Old

22

Wiesehfer 2001.23Wiesehfer 1996, 10910; 2007, 413. The majority ofour evidence is numismatic, for which Alram (1986) servesas the standard reference. The coinage of Fratarakid Pars canbe classified into four groups: Group A (Alram 1986, 51143),Group B (Alram 1986, 54463), Group C (Alram 1986, 56486), and Group D (Alram 1986, 587655). With some varia-tion in details and minor elements, Group A (ca. early secondcentury B.C.E.) and B (ca. second century B.C.E.) portraythe rulers head facing right on the obverse, with a rectangu-lar structure topped with three crenellations, venerated by astanding figure. As well as becoming progressively more styl-

ized, Group B is distinguished by a change in the crenella-tions, showing two flanking step structures instead of three.Group C (ca. first century B.C.E.) marks a break with the ear-lier tradition and parallels the provinces new incorporationinto the Arsacid empire: the rulers display Arsacid hairstylesand regalia and now face left. The reverses show a simplifiedscene of a standing figure holding a barsomin front of a simplefire altar. Group D (which encompasses the early Sasanians)shows greater variation in reverse types, portraying diadems,investitures, and heavenly bodies, among other symbols. For auseful overview, see Haerinck and Overlaet 2008, 2089.

Table 2. Macedonian Kings and Satraps of Iran.

Name Dates of Control over Iran (B.C.E.) Dynasty

Alexander III the Great 330323 Argead

Peukestas 324316 satrap of Parsa

Antigonos Monophthalamos 316312 Antigonid

Seleukos I Nikator 312281 Seleucid

Antiochos I Soter 291281 (coruler);281261 (sole ruler)

Seleucid

Antiochos II Theos 261246 Seleucid

Seleukos II Kallinikos 246225 Seleucid

Seleukos III Keraunos 225223 Seleucid

Antiochos III the Great 223187 Seleucid

Seleukos IV Philopator 187175 Seleucid

Antiochos IV Epiphanes 175164 Seleucid

Antiochos V Eupator 164162 Seleucid

Demetrios I Soter 162150 SeleucidAlexander Balas 150145 Seleucid

Demetrios II Nikator 145139 Seleucid

Table 3. Early Sasanian Kings of Kings.

Name Regnal Dates (C.E.)

Ardashir I 224239/40

Shapur I 239/40270/2

Hormozd I 270/2273Bahram I 273276

Bahram II 276293

Bahram III 293

Narseh 293302

Hormozd II 302309

Shapur II 309379


6/34


Persian title of a subsatrapal Achaemenid governor,and, while they did not hold imperial pretensions, theycontinued to use many of the old Achaemenid names,including Darew (from Old Persian Da- rayavau) andArdashir (from Old Persian Artaxaa).24The ar-chaeological and visual evidence firmly indicates that

the Fratarakids engaged Achaemenid visual cultureas a prestigious starting point for their own official

representations.25The early obverse coin portraits ofthe Fratarakid rulers show the influences of Hellenis-tic royal portraits and incorporate some symbols ofHellenistic kingship, such as the diadem. The domi-nant symbol of their rule, however, was the *kurpa-sa,headgear originally worn by satraps who served the

Achaemenid empire.26The reverses of tetradrachmsof the Fratarakid Baydad portray the ruler enthroned

24 Frye 1984, 272; Boyce 1991, 11016; Wiesehfer 1996,10910; 2001. An inscription on a Hellenistic-style bowl inthe Getty refers to these rulers as da-ra-ya-naga-n(of the lineageof Darius [I]) and presents a rare non-numismatic piece ofevidence of the persistent memory of the Achaemenids (Skja-erv 1997; Callieri 2007, 131).

25 de Jong 2003; Panaino 2003; Potts 2007.26 Rendered by Greek historians as kyrbasia, reconstructed

as Old Iranian *kurpa-sa(Shahbazi 1992). Some Greek sourcesrefer to this satrapal headgear as the tiara apage-s(Wiesehfer2007, 43).

Table 4. Pre-Sasanian Rulers of Pars (Reconstructed from Coinage).

Name Approximate Date Title (Suzerain)

Baydad end of third century/beginning of second century B.C.E.

Frataraka-(sub-Seleucid)

Ardaxshir I first half of second century B.C.E. Frataraka-(sub-Seleucid)

Wahbarz first half of second century B.C.E. Frataraka-(sub-Seleucid)Wadfradad I mid second century B.C.E. Frataraka-(sub-Seleucid)

Wadfradad II ca. 140 B.C.E. a-h (sub-Arsacid)

Unknown King I second half of second century B.C.E. a-h (sub-Arsacid)

Darew I third quarter of second century B.C.E. a-h (sub-Arsacid)

Wadfradad III first half of first century B.C.E. a-h (sub-Arsacid)

Darew II beginning of first century B.C.E. a-h (sub-Arsacid)

Ardaxshir II second half of first century B.C.E. a-h (sub-Arsacid)

Wahshir second half of first century B.C.E. a-h (sub-Arsacid)

Pakor I first half of first century C.E. a-h (sub-Arsacid)

Pakor II first half of first century C.E. a-h (sub-Arsacid)Nambed mid first century C.E. a-h (sub-Arsacid)

Napad second half of first century C.E. a-h (sub-Arsacid)

Unknown King II end of first century C.E. a-h (sub-Arsacid)

Wadfradad IV first half of second century C.E. a-h (sub-Arsacid)

Manchir I first half of second century C.E. a-h (sub-Arsacid)

Ardaxshir III first half of second century C.E. a-h (sub-Arsacid)

Manchir II mid second century C.E. a-h (sub-Arsacid)

Unknown King III second half of second century C.E. a-h (sub-Arsacid)

Manchir III second half of second century C.E. a-h (sub-Arsacid)

Ardaxshir IV end of second century C.E. a-h (sub-Arsacid)

Shabuhr (New Persian: Shapur;brother of Ardashir I, founder ofthe Sasanian empire)

beginning of third century C.E. a-h


7/34


in an Achaemenid-style throne holding a flower in hisleft hand. His costume is clearly satrapal rather thanimitative of Achaemenid royal regalia. Although thefigure holds a scepter as in the Achaemenid reliefs,Seleucid rather than Persian models inspired the formof the scepter.27

In addition to their regalia, the Fratarakids incorpo-rated into their coins aspects of the most prominentfeatures of Achaemenid royal architecture and archi-tectural ornament that still existed around them.28Most of the Fratarakid coins depict a winged disk witha male bust emerging from it, recognizable on everyAchaemenid royal tomb, many prominent reliefs atPersepolis, and many seals (fig. 2).29On most issues,this divine figure hovers over a stepped rectangularstructure with coffering or coffered doors that recallsAchaemenid architectural forms and post-Achae-menid crenellations at Persepolis. Speculations onthe identity of the structure on the reverses of the

coins have proliferated; however, the most cogent in-terpretation of the iconography, not to mention theonly one grounded in primary source material (i.e.,archaeological evidence), argues that it was inspiredby Achaemenid architecture, possibly the Achaemenidtowers such as were built at Naqsh-e Rostam and Pasar-gadae.30Although we will likely never know the exactfunction or identity of the structure, on most issues,a male figure stands next to it in a posture directly in-spired by the composition and posture of the Achae-menid kings of kings on their tombs: the figures faceright, raising their right hand to the winged figureabove. The figures hold their bow with the bowstringfacing away from the object of veneration. This is thesame posture of respect shown by the Achaemenidsovereigns on the tomb reliefs. The bow, however, isof a contemporary, recurve style, rather than a directcopy of those on the Achaemenid reliefs.31

A discrete break with early Fratarakid coin types oc-

curs only after Pars submitted to the Arsacids.32Thecoinage of Wadfradad II, the first ruler of Pars thoughtto acknowledge Arsacid suzerainty, marks a transition,and after Darew II, the obverse portraits clearly followArsacid royal iconography.33The reverse types changeas well but do not follow Arsacid models. Achaemenidiconography appears also to have inspired these newtypes. The reverses of most of these portray a malefigure in profile, facing a fire altar broadly similar tothe fire altars on all Achaemenid tombs and many ofthe seals.34

The Fratarakids built both on and near the plat-form of Persepolis.A group of structures located 300m north of the platform show characteristics of apalace and a shrine where the inhabitants honoredthe gods with a statue, a fire, or some combinationthereof.35 These post-Achaemenid mudbrick struc-tures employed some carefully chosen and reworked

27Wiesehfer 2007, 43.28 Boyce 1991, 11618; Trmpelmann 1992, 525; Wiese-

hfer 2001.29Alram 1986, pls. 17.533, 18.53359, 19.56063.30 This is by far the most common reverse type in the Fra-

tarakid coinage (Alram 1986, pls. 17.51834, 17.54547,18.53559, 19.56067; de Morgan 1979, 27781, figs. 34051). Palace H in Persepolis preserves remains of post-Achae-menid crenellation that is very close to that on the roof of thetower on the reverses of the Fratarakid coinage (Gnoli 1989,12425; Trmpelmann 1992, 534). Potts (2007) presents auseful historiographical overview of this debate as well as co-gent evaluation of the scholarship and evidence. Haerinckand Overlaet (2008) argued that the images on the coins rep-resent Graeco-Roman-style open-air altars, such as the Ara Pa-cis. Haerinck and Overlaet present a very detailed argument;however, without archaeological evidence of the presence ofsuch altars in Pars, I am not prepared to discard the possibility

that the structures relate to the Achaemenid architecture ofPars and the provinces Achaemenid and post-Achaemenidtowers.

31Wiesehfer 2007, 43.32 They exchanged the ancient title offrataraka-for that of

king (a

-

h), matching Arsacid conventions (Potts 2007, 276).The upheavals following Antiochos IVs invasions of Egyptprovide the backdrop to the Seleucids eventual loss of Pars.See Mittag (2006) for in-depth analysis of these events.

33Potts 2007, 276; Wiesehfer 2007, 45.34Alram 1986, nos. 56486 (Group C); Garrison and Root

2001; Root 2008.35 Herzfeld 1935, 46. This room yielded column bases and

a large stone block that could have served as a statue base(Schmidt 1953, 556; Schippmann 1971a, 17782). Althoughthe exact chronology of this building activity is unclear, Tiliasexcavations revealed extensive remodeling and spoliation ofthe site (Tilia 19721978, 316; 1977, 745).

Fig. 2. Reverse of a tetradrachma of the Fratarakid WadfradadI, ca. 150 B.C.E., diam. 28 mm, wt. 12.77 g (Alram 1986, no.533; courtesy T.K. Mallon-McCorgray).


8/34


Achaemenid stone architectural members, such as adoorjamb and lintel, all taken from material at Perse-polis.36The Fratarakids removed a doorway from thetaara(private palace) of Darius I. With its depictionsof beardless eunuch servants in profile wearing Per-sian robes and carrying personal articles of the king

(as in other relief sculpture there), incorporated intoits new context, it is possible the Fratarakids gave thesefigures a new interpretation or identity. A window jambassociated archaeologically with the sacred area of thecomplex carries the simple, low-relief images of twofigures in profile. They hold ritual paraphernalia intheir hands in a contemporary Middle Iranian gestureof reverence.37The window that the jamb decoratedcommunicated with the antechamber to the innersacred area, linking their actions to the sacred areainside.38Although they were Fratarakid creations, theiconography on these reliefs responds to and rein-terprets aspects of the Achaemenid reliefs, adapting

their striding profile and outstretched, raised arms tocontemporary post-Achaemenid, Persian visual cul-ture. If Islamic accounts can be believed, the Templeof Anahid in Staxr, of which the Sasanians took overthe hereditary priesthood, similarly integrated ele-ments of Achaemenid architecture such as bull capi-tals and reliefs.39

More remarkable for their absence, the Arsacidsapparently never sponsored any activity in the Achae-menid ruins of the province, nor did they carve a rockrelief in Pars near the Achaemenid tombs. Arsacidkings appeared in monuments in other provinces intheir empire, such as the rock relief of Mithridates I atKhong-e Nowruzi, deep in Elymas.40One can conjec-ture that this dearth of Arsacid evidence in Pars is theresult of the Sasanians particularly thorough job ofobliterating their monuments, as occurred at Bisotun,or simply because by this time, Persepolis and Naqsh-eRostam no longer held any special significance beyondthe province. Bisotun, the site of Darius Is monumental

rock relief and inscription, preserves limited evidenceof Arsacid engagement with the Achaemenid site. Al-though they do not match the scale or intricacy of theSasanian material in Pars, Mithridates II, Gotarzes II,and a king named Vologases carved reliefs several me-ters distant, on Bisotuns lower rock face or in the field

to the north, though, given their orientation, thesewere intended to engage with Bisotuns walled sanctu-ary below rather than with Darius Is relief.41

All told, Achaemenid visual culture and architec-tural traditions enjoyed a longer afterlife in Pars thananywhere else, both as spolia and reinterpreted newcreations. By incorporating Achaemenid iconogra-phy into their own visual culture and Achaemenidsites into their ritual life, the Fratarakids implied thatsome sort of meaningful, if not lineal, relationshipexisted between them and the Achaemenid empire.Without further evidence, we are unable to elabo-rate on the nature of this relationship; however, the

Sasanians predecessors in Pars presaged much ofthe intricacy, albeit not the scale, of many of the laterSasanian practices. These include ritual engagementwith Achaemenid sites, creative reuse of Achaemenidarchitectural elements in newly created structures, andselective integration of Achaemenid visual culture intheir own images. Although carried out on a smallscale, this suggests a regional set of memory practicesthat challenged the early, pre-imperial Sasanians asthey clawed their way up to provincial power in theearly third century.

early sasanian memory practices in pars

The founder of the Sasanian empire, Ardashir I,led his familys rise from obscure, local garrison com-manders to provincial kings by systematically assas-sinating neighboring chieftains and annexing theirdomains.42The family overthrew the king of Pars in212 C.E., setting up an eventual conflict with the Ar-sacid king of kings, Ardawan IV. During the Sasanians

36 Schmidt 1953, 556.37 Schmidt 1953, 51, figs. AC; Shahbazi 1986; Choksy

1990a, 305; 1990b, 2015.38 Schmidt 1953, 50, fig. 16, no. 4.39 al-Masudi Muruj al-dahab (Barbier de Meynard 1865,

para. 1403 [4.77]). Although Boyce (1975, 45965; 1997) ar-gued this structure was built by the Achaemenids, without anyarchaeological evidence of the Achaemenid fire temple (orany other Achaemenid temple for that matter), the weight ofthe evidence suggests that it was created with spolia in Staxr orin a reoccupied part of Persepolis by the Fratarakids, or by theSasanians themselves.

40Vanden Berghe 1983, 478; Vanden Berghe and Schipp-mann 1985, 328.

41 Canepa (forthcoming). One of the five Arsacid kings of

kings named Walax (Vologases) carved the so-called Parthi-an stone, located several meters away from the cliff (Luschey1990, 293; von Gall 1996a, 1996b; Kleiss 1970, 1996). On the

wider possibilities and problems relating to Parthian engage-ment with the Achaemenids, see Shayegan 1999; Fowler 2006.Harper (2006, 6 n. 9) cites Ghirshman (1976, pl. 41.6) as evi-dence of Parthian reuse of Achaemenid architectural mem-bers at Bard-e Neshanda; however, in his text, Ghirshman(1952, 8; 1976, 199200) is clear that these were the same sixcolumn bases he discovered in the painted Sasanian room inSusa, not, as it appears in Harpers text, from the Parthianlevel of Bard-e Neshanda.

42 Daryaee 2009, 34. Al-Tabari provides the most detailedstory of the Sasanians rise and likely reflects official propa-ganda (Shahbazi 2005a).


9/34


bloody two-decade rise from local dynasts to kings ofkings, Pars monumental patrimony again played animportant role as raw material for expressing a visionof Iranian kingship for a new regime. Once in power,the Sasanians took possession of Staxr and its sur-rounding symbolic landscape, being driven to match

and supercede their predecessors engagement withthe Achaemenid structures.Persepolis contains the earliest evidence for a grow-

ing repertoire of Sasanian memory practices from thisearly pre-imperial period (205/6224).43 The earlySasanians marked the site subtly but permanently,sponsoring a number of small, detailed portraits ofthemselves and their court incised into doorways andwindow frames in the Harem of Xerxes and the taaraof Darius I (fig. 3).44The chronology and identity ofthe figures in these remarkable miniature reliefs arenot entirely clear; however, comparisons with theircoinage suggest that they represent Ardashir Is father,

Papag, and brother, Shapur.45Despite their small size,the basic ideologicial operations that they undertakeare not only quite complex, they also set a conceptualprecedent for later Sasanian endeavors carried outon a truly monumental scale. These images claimedPersepolis for the dynasty, putting their mark onto aprestigious site occupied by successive generationsof Macedonian governors, Fratarakids, and likely theSasanians own deposed rivals. More importantly, byincising their images into the very substance of theAchaemenid structures, the early Sasanians began aprocess of fusing their fledgling dynasty tangibly andpermanently to the remains of the ancient Persianpast. This fusion provided physical and visual evidenceof the otherwise unfounded antiquity and dubiouslineage of the Sasanians. It should be noted that thistechnique did not depend on exact knowledge of theoriginal patron of the ruins. It capitalized primarily onthe inherent antiquity of the structures, grafting theSasanians onto the contemporary stature of the ruinsin Pars collective experience.

Ritual performance, though ephemeral, estab-lished a similarly meaningful link between the newkings and the ancient site. The graffiti themselvesrecord a number of ritual activities, all in some way

related to offering respect or homage. These cohere

well with later records of Sasanian activities at thesite. They portray the kings in stately procession withtheir courtiers; standing, offering incense at altars, orproffering honorary diadems, presumably intended

to relate to an actual image or altar or simply to thesite itself. Of a different tenor, Persepolis even plays arole in Ardashir Is accession. Upon the death of hisfather (Papag), Ardashirs older brother Shapur be-came king, a fact that Ardashir, then a petty prince,disputed.46 If al-Tabari can be believed, on his wayto confront Ardashir, Shapur stopped at Persepolis,where part of a building fell on him and killed him.Whether the masonry fell by accident or with a pushfrom one of Ardashirs agents, Ardashirs remainingbrothers subsequently met him at Persepolis, wherethey elected and crowned him king.47Like the incised

portraits, it suggests that the Sasanians inherited from

43 The beginning of the Sasanian era occurred in 205/6,according to the Middle Persian inscription on the Stele ofBishapur dedicated to Shapur I (Frye 1984, 29196; Weise-hfer 1987, 372; Schippmann 1990, 1020).

44 Herzfeld 1941, 3079; Schmidt 1953, fig. 99a, b; Calmey-er 1976, 638, figs. 3, 4; Trmpelmann 1992, 53; Callieri 2003;2007, 13234.

45Based on the numismatic evidence, the scholarly consen-sus is that these represent the early Sasanians; cf., e.g., these

images with the headdress on the coins of Shapur (Alram1986, pl. 22.65355; Huff 2008, 323). For an alternate inter-pretation, see Wiesehfer 2007, 47. See Callieri (2003) for anaccount of earlier identifications of these figures.

46 The fact that Shapur I honors his uncle Shapur, the sonof Papag, as a king need not cause us pause (al-Tabari 816 [Bo-sworth 1999, 89]).

47al-Tabari 816 (Bosworth 1999, 89); Daryaee 2009, 4.

Fig. 3. Sasanian graffiti portraying a king offering a diadem,ca. 212218 C.E., from Persepolis, Harem of Xerxes.


10/34


their pre-imperial Persian forebearers a tendency toexpress their ideas of power in Achaemenid contexts,not only visually but through ritual practice. Takenas a whole, the early Sasanians memory practices onthe platform of Persepolis provided a precedent andinspiration for their later, more ambitious activities

once Ardashir I took supreme power.

Innovation in Urban and Natural Topographies ofMemory at Ardashir-Xwarrah

Once he had consolidated power in Pars but beforehis final victory, Ardashir I withdrew to a site endowedwith exceptional natural defensive capabilities about125 km south from the less easily defended city ofStaxr. There he built Ardashir-Xwarrah (Royal Gloryof Ardashir), known from the 10th century as Firu-zabad (fig. 4).48The city not only served as his strong-hold against an anticipated Arsacid reaction but alsoas the king of kings great experiment in city planning

and memorial practices.49Reflective of his develop-ing ideas of kingship, it presents an urban plan of asovereign who intended the city not only to functionas the center of an empire but also to symbolize thecentrality of the king of kings.50The king of kings laidout the city in a perfect circle with a diameter of 1,850m.51Twenty radial roadways extend well past the for-tifications up the cliffs and out into the surroundingcountryside, and an inner fortified area contained acentral watchtower and a fire temple provisioned withcisterns. Far from an unbroken, ancient Near Easterntradition, a precisely circular city on such a scale waswithout precedent.52

While the city and buildings of Ardashir-Xwarrahwere radically unique in design, Ardashir I anchoredit in the ancient Persian past by adapting practicesdeveloped by the Fratarakids and his family at Perse-polis. The city and its official structures repeatedlypresent self-consciously innovative designs but usearchitectural forms and even imported architecturalmembers from Persepolis to cloak the new creations

in the august architectural forms of the ancestorsof the king of kings. The stucco ornament around

the niches of both of Ardashirs palaces, the Qalaye-Dokhtar and the Great Palace, carefully translate theshapes of the stone cornices crowning Persepolitandoorways. Traces of stucco on the tower (the Terbal)indicate that it bore such motifs as well (figs. 5, 6). 53The Qalaye-Dokhtar, a mountaintop fortress thatcontrolled the northern entrance into the valley, em-ployed such experimental architecture that its walls

48 Location: 28518.82N, 523155.20E.49 Gyselen 1989, 44.50This idea is explored most thoroughly in the work of Huff

(Huff 19691970, 1972, 1976, 2008; Huff and Gignoux 1976).It has been argued that these early ideals of a centralized em-pire contrast with the later balance of power that developedbetween Parthians and Persians (Pourshariati 2008).

51 Huff 2008, 45. For the best overview of Huffs work atFiruzabad and the citys urban fabric, see Huff 1999. For thefire temple, see Huff 1972; see also Schippmann (1971a, 10022) for earlier scholarship and literary attestations.

52 Oft-cited earlier circular cities were irregular sprawlsrather than carefully planned circles. Under Ardashir I,

Darabgerd was triangular, only converted to a circular lay-out under the Arab governor of Fars in the late seventh/earlyeighth century when circular cities, such as Baghdad, beganto come back into vogue with Sasanian inspiration (Novk1999, 21429; Huff 2008, 512; cf. Dettmann 1969, 203). Ar-dashir I founded the city of Weh-Ardashir in Ctesiphon on acircular plan with Ardashir-Xwarrah as the likely precedent(Krger 1992).

53Huff 1971; Huff and Gignoux 1976, 13436; Krger 1982,19798; cf. Roaf 1998, 34. In earlier scholarship, the GreatPalace was erroneously referred to as the Ateshkade (the FireTemple), a name sometimes still encountered.

Fig. 4. Map of Ardashir-Xwarrah (present-day Firuzabad,Iran) and surrounding features.


11/34


began to shift and crack even before it was finished.54The Great Palace, built without fortifications in the

valley below after Ardashirs final victory, used muchless risky designs but still employed complex domeand vault construction (fig. 7).55Contrasting with itsinnovative design, Achaemenid cavetto cornices trans-lated into stucco articulate the niches in the domedinterior spaces (fig. 8). The main fire temple (theTakht-e Neshin), a domed, cruciform structure withradiating ayva-ns, incorporated Achaemenid columnshafts, which were brought to the site; these, and the

54

Although the Qalaye-Dokhtar likely did not serve longas the official residence, considering the cracks and the muchmore convenient palace on the plain below, its presence onthe canyon cliffs high above would have been dishearteningto any invading army and awe inspiring to any entering thevalley. Yazdgerd III did, however, fortify it again against theArabs at the end of the empire (Huff 2008, 44).

55 Huff 2008, 424.56According to Huff (19691970; 1972, 53640; 2008,

479), the masonry techniques match contemporary Romantechniques, indicating it was not an Achaemenid construc-tion as some have conjectured. An ayva-nis a New Persian termfor entranceway or balcony that scholarship has adopted as a

technical term to refer specifically to the monumental barrel-vaulted entranceways favored particularly by the Sasanians,although the architectural form enjoyed a long history in thedevelopment of Iranian architecture.

57 Huff 2008, 54. The double Persepolitan entrancewayalso occurred in the Sasanian constructions at Kangavar(Azarnoush 1981, 2009).

58 Fragments excavated by Schmidt in the Sasanian build-ings clustered around the Kaba indicate this same type ofAchaemenid ornament was applied in structures at Naqsh-eRostam, too (Schmidt 1970, 75, fig. 30, no. 23 [field 2.39];Krger 1982, 19697).

Fig. 5. Fragmentary cavetto cornice inspired by Persepolismolded out of stucco at Ardashir Is fortress of Qalaye-Dokhtar, from Ardashir-Xwarrah, in July 2001 (now de-stroyed by exposure to the elements).

Fig. 6. Remains of the masonry platform of the Takht-eNeshin with the Terbal in the background, before 224 C.E.,from Ardashir-Xwarrah.

ashlar masonry, evoke Achaemenid platforms suchas at Pasargadae.56The area before Ardashir Is pal-ace even hints at a double stairway in imitation of thePersepolitan stairway leading up to the main entrance,though without further excavation, this cannot be con-clusively determined.57

At this point, the Achaemenidizing architecturalmotifs signified Persian antiquity in a general sense andsucceeded in encapsulating and exporting the prestigeof the ruins at Staxr.58The city and its grand structuresreinforce in architectural and visual terms the idea thatArdashir I not only was the rightful heir of the Persian

heritage but also was responsible for renewing it. Likethe reoccupation of Persepolis in previous generations,the exact identity and history of their Persian predeces-sors who initially created these forms was not impor-tant for these ornamental and architectural motifs tobe powerful; rather, it was the general experience oftheir antiquity and value that was significant.

After he defeated Ardawan IV, Ardashir I carvedtwo monumental rock reliefs into the cliffs of the


12/34


Tang-e Ab, a narrow gorge that served as the north-ern entrance to the valley. They depict Ardashir I un-horsing the Arsacid king and an investiture scene in

which the king of kings receives the royal diadem fromOhrmazd.59The Sasanian practice of carving monu-mental reliefs, the most enduring of their technologiesof memory and deserving of a separate study in its ownright, reflects a knowledge of and engagement withthe collective remnants of their predecessors (Lullubi,

Elamite, Assyrian, Achaemenid, Seleucid, or Arsacid)that were visible then as now.60The Sasanians not onlysurpassed all other dynasties in the volume of rock re-liefs, they also used them in a way that was ultimatelyquite distinctive, stemming from their early experi-ences of the region around Staxr.61The impact of theAchaemenids monumental rock-cut tombs and Elam-ite reliefs at Naqsh-e Rostam would have directly intro-duced the king of kings to the powerful effect that thisgenre could have. These complementary techniquesof subtle ornamental appropriation and the creationof an environment marked with rock reliefs made Ar-dashir Is political claims visible as an innate part of the

architectural and natural order of things. Ardashir Isearly experimentations at Ardashir-Xwarrah, in turn,provided precedents for Naqsh-e Rostam, the supremesite of memory established by the dynasty.

Naqsh-e Rostam: From Achaemenid Necropolis toSasanian Site of Memory

Out of all the sites that the early Sasanians favoredfor commemorative activity, the site of Naqsh-e Rostamstands out as preeminent (figs. 911). It lies about 8km north of the ruins of Persepolis, 25 km south ofCyrus capital, Pasargadae, and only 2 km from Staxr.The New Persian place-name refers to a cluster of re-lief sculptures carved onto the southern rock face ofone of several flat-topped rock hills that project intothe plain of Persepolis, not far from a ford in thenearby river. The site had some sort of importancein the Elamite era but was transformed by Darius I toserve as the main royal necropolis for the Achaemeniddynasty. The Achaemenids and the Sasanians bothcarved more rock reliefs at Naqsh-e Rostam than at

59Vanden Berghe 1983, 616.60 Canepa (forthcoming). Following conventions estab-

lished by the sites early surveyors, Iranian rock reliefs areidentified by their New Persian place-name and a numberthat corresponds to a topographical place, not the order inwhich they were created. The most useful catalogues of rockreliefs in Iran are Vanden Berghe 1983; Vanden Berghe andSchippmann 1985. Vanden Berghes numbering and mostof his attributions reflect present scholarly consensus. Theexceptions are Taq-e Bostan I (possibly Ardashir II becauseof internal elements such as the late fourth-century style ofclothing of the middle figure [Shahbazi 1985; Canepa 2009,1089; cf. Azarnoush 1986]), Bishapur VI (which, as I believeGhirshman [1950] correctly observed, contains more peo-ples than just the Romans in this triumphal scene [Vanden

Berghe 1980; earlier bibliography in Herrmann 1981, 328]),and Naqsh-e Rostam 3 and 4, for which there is no real con-sensus. For recent research on the iconographic conventionsof Sasanian rock reliefs, see Soudavar 2009.

61Vanden Berghe 1983, 57. As well as Vanden Berghes cat-alogue overview, the main documentary publications of theSasanian rock reliefs include Fukai and Horiuchi 1969, 1972;Schmidt 1970; Herrmann and Howell 1977; Herrmann 1980,1981, 1983; Huff 1984; Vanden Berghe 1986; Herrmann et al.1989. Most other literature published between 1978 and 2003is collected in Vanden Berghes Bibliographie analytique and itssupplements (Vanden Berghe et al. 1979; Vanden Berghe andHaerinck 1981, 1987; Haerinck and Stevens 1996, 2005). Forpost-2003 material, see Canepa 2009.

Fig. 7. Exterior view of the Great Palace of Ardashir I atArdashir-Xwarrah, after 224 C.E. View of the central barrel-vaulted entranceway (partially restored) leading to threedomed chambers.

Fig. 8. Detail of Achaemenid-inspired cavetto cornices in cen-tral domed chamber, ca. 224 C.E., from Ardashir-Xwarrah,the Great Palace of Ardashir I.


13/34


any other site in the province or empire. The wealthof sculptural, architectural, and inscriptional featuresfrom these two temporally disparate dynasties suggeststhat it served as a ritual and symbolic center unparal-

leled elsewhere in the empire. The main constituentsof Naqsh-e Rostam are traces of Elamite reliefs, fourAchaemenid funerary reliefs, one mysterious Achae-menid tower of ashlar masonry known as the Kaba-yeZardosht (the Kaba of Zoroaster), and eight com-pleted Sasanian bas-relief sculptures (see fig. 11).62The Sasanians carved rock reliefs at Naqsh-e Rostamfrom the third to the early fourth centuries; however,numerous minor features in the surrounding area,such as rock-hewn mountaintop or cliffside ossuariesfor the deposition of the bones of Zoroastrian faith-ful, continued to be created until the Arab invasion,

62 One of the Achaemenid tombs is securely attributed toDarius I by its inscription. Schmidt (1970, 79107) attributedthe others to Xerxes I, Artaxerxes I, and Darius II on the ba-sis of relative wear and style. The Elamite reliefs date to 1700B.C.E. and the eighthninth centuries B.C.E. (Vanden Ber-ghe 1983, 267, 66; Seidl 1986). In 1971, when the groundwas leveled to accomodate the shahs anniversary celebra-tions, two minor reliefs of a lion and male figure were re-vealed (Roaf 1974). It is possible that more minor reliefs liebelow ground level. The site also contains an unfinished re-

lief, possibly started during Khosrow IIs reprisal of the genrein the seventh century, on which an inscription was carvedduring the Qajar era. Nearby, there lies an Islamic shrine, theEmamzade Saf-e Mohammad (Stronach 1966, 217).

63Several have inscriptions dating them to the late Sasanianperiod (Trmpelmann 1992, 1724; Huff 1998, 2004).

64The walls have been dated tentatively to the third centuryC.E. (Kleiss 2001; Gropp 2004).

65 Schmidt 1970, 548; Schippmann 1971b; Kleiss 1976,14250; Trmpelmann 1992, 4951.

Fig. 9. View of the northeastern portion of the site of Naqsh-e Rostam, Iran (from left toright): the Achaemenid tombsof Artaxerxes I(?), Darius I, and Xerxes I(?); below, the Sasanian relief of Hormozd II with unfinished enthronementscene; above, reliefs of Shapur I and Kerdir, double joust relief of Bahram II(?), investiture relief of Narseh, and thelarge unfinished relief with ossuaries.

indicating that the site was in use and important forthe local nobility until the end of the dynasty.63Sasa-nian dirt ramparts extended from the cliffside andenclosed an extensive area around the Kaba.64 Al-

though the area within the fortification wall has notbeen fully excavated, Schmidts tests indicated that adense concentration of Sasanian buildings clusteredwithin the walls.65

What is remarkable about the Sasanian dynastysadditions to the site is not simply their monumental-ity but the extent to which they sensitively, seamless-ly, and unrelentingly incorporated the Achaemenidmaterial into their larger vision. Here, the first twogreat kings of kings of the dynasty adapted and ex-panded pre-imperial Sasanian practices such as incis-ing their identities visually or epigraphically directly


14/34


into Achaemenid features and using them as a stagefor ritual practice. This engagement went far beyonda superficial interest in the Achaemenid remains: theintense building activity and rituals performed at thesite created a coherent experience of a single cultural,dynastic, and historical whole. Furthermore, parallelsbetween the visual, discursive, and ritual expressionsof the two dynasties are so startlingly close that it is

much more likely that some sort of causative, if notlineal relationship, lay behind the pronounced simi-larities than to posit that they were random, as hasrepeatedly been implied.66

Although it is probable that the Fratarakids and pre-imperial Sasanians used the site, Ardashir I was respon-

sible for forging the first enduring link between theAchaemenid remains and the Sasanian dynasty, thusbeginning the full transformation of the site into a spe-cifically Sasanian site of memory. At the site, which noking had modified since the Achaemenids, Ardashir Ibegan the permanent fusion of the remains of the twodynasties by carving the first rock relief on the westernend of the rock face (fig. 12).67This was the last reliefhe executed in Pars, and it combines the themes of hisearlier reliefstriumph and investitureinto a single,harmonious image.68It depicts a symmetrical compo-sition of equestrian investiture and triumph, one thatseveral of his successors would emulate.69Inscriptions

on the horses identify the two figures as Ardashir Ion the left, receiving the diadem from the great godOhrmazd on the right.70They each trample the pronebodies of their defeated enemies, the Arsacid king Ar-dawan IV and Ahreman (Zoroastrianisms demon ofdemons). The symmetrical composition presents anaudacious rhetorical statement, with a clear logic ofequivalency between the historic earthly achievementand future apocalyptic act.

As well as grounding his recent defeat of the Ar-sacid king of kings in the inevitability of Zoroastrianeschatology, Ardashir I presented this novel statementas intrinsically ancient by selectively appropriatingAchaemenid sculptural forms and finish.71Details suchas the high polish, treatment of horses, the humanbody, and drapery indicate that the craftsmen closelystudied Achaemenid sculpture, with the relief sculp-

66 Studies with this implication are summarized in Daryaee20012002, 34.

67 Interestingly, his relief was not situated under the Achae-menid tombs in the area enclosed by the earthenwork walls(see fig. 11 herein).

68 Five reliefs are securely attributed to the king of kings(listed chronologically): Firuzabad (Ardashir-Xwarrah) I andII, Naqsh-e Rajab III, Naqsh-e Rostam I, and Salmas (VandenBerghe 1983, 62; Levit-Tawil 1993). Darabgerd coheres mostclosely with the themes of Shapur Is reliefs; however, thecrown matches one he shared with his father during their co-regency rather than his normal regnal crown.

69 One of Shapur Is two reliefs at Naqsh-e Rajab (IV) por-trays Ohrmazd investing the king of kings using the same sym-metrical composition, although without the defeated enemiesbelow. Shapur Is eldest son, Bahram I (271274 C.E.), creat-ed a relief at Bishapur (V) very similar to this one. The earli-est of Shapur Is three reliefs at Bishapur (I) again uses the

composition, this time with figures underneath the horses. Init, the Roman emperor Gordian III takes the place of the de-feated Arsacid sovereign, Ardawan II, and the relief adds theadditional figure of Philip the Arab genuflecting before theking of kings. See Canepa (2009, 5968) for in-depth analysisof this relief in the context of Roman-Sasanian competition.

70 On the left: This is the visage of the Mazda-worshippinglord Ardair, the King of Kings of Iran, who is of the radi-ant image of gods, son of the lord Papag, the king. On theright: This is the visage of the god Ohrmazd (Back 1978,28182). It is possible that this could have been added laterunder Shapur I. This would present an interesting scenarioof further reinscription of the past by this sovereign. MiddlePersian Ohrmazd/New Persian Hormozd(from Old PersianAuramazda; Avestan, Ahura Mazda) is the name of both thegreat god and several Sasanian kings.

71 Herrmann 1981, 15160; Trmpelmann 1992, 62; Harp-er 2006, 15.

Fig. 10. Map of Hosayn Kuh and Naqsh-e Rostam (adaptedfrom Kleiss 1976, fig. 13).


15/34


ture of Persepolis and Naqsh-e Rostam providing readyexamples to king and artisan alike.72Both king and godride horses whose proportions, thick neck, and head

72 For valuable close analysis of sculptural forms and techniques at Persepolis, see Roaf 1983, 42 (for hands).

Fig. 11. Map of Naqsh-e Rostam. Arabic numberals indicate the Sasanian rock reliefs, Roman numerals indicate theAchaemenid tombs.

Fig. 12. Rock relief of Ardashir I, after 224 C.E., from Naqsh-e Rostam (Relief 1) (from leftto right): a page, Ardashir I,and the god Ohrmazd; underhoof, Ardawan IV (left) and Ahreman (right).

position recall the fine steeds brought by Persepoli-tan tribute bearers (figs. 13, 14). The artisan sculptedOhrmazds right hand similarly to the hands of Persian


16/34


or Median soldiers and dignitaries who grasp theirspears or staffs (fig. 15). The thumb, which tapers toa graceful point, rests on top, yet slightly behind, theindex finger, while the backs of fingers repeat in paral-lel lines, visible from the knuckle back, with fingertipsextending over the hollow of the palm.

The eyes of the god and king follow the design andsculptural forms of the Persepolis reliefs. A slightlyraised border, with a subtle incision delineating ei-ther side, outlines the edges of the wide, almond-shaped eyes. Like the eyes at Persepolis, they protrudeconvexly from the sculptural plane. A similar sharp,raised border delineates the figures brows. Althoughpartially destroyed, the remnants of their noses indi-

cate the artisans gave the god and king profiles delib-erately similar to the Persepolis reliefs. The folds ofOhrmazds cloak ripple in regular zigzagging foldsacross his thigh, echoing the standard treatment ofdrapery of Persian robes in the Persepolis reliefs (seefigs. 13, 15). The relief adapts the standard mode of

representing beards at Persepolis, with repeated cir-cular curls filling an upper zone that extends from themoustache across the upper cheeks and alternatingcircular curls and wavy lines, forming a more exten-sive lower zone (see figs. 15, 16). Finally, the Sasanianreliefs intentionally used motifs inspired by the rep-resentations of the Achaemenid kings to image thegod Ohrmazd. Like those of the Achaemenid kingsof kings, the gods beard is long, squared-off, andclosely curled, in contrast to Ardashir Is contemporaryfashion with the beard drawn into a fillet and endingin a bulbous bunch of curls. Ohrmazds crenellatedcrown recalls the crowns of the Achaemenid kings and

winged figure on the majority of the tombs overhead.73

73 Henkelman 19951996, 28486.

Fig. 13. Detail, Ohrmazds horse and drapery folds. Rockrelief of Ardashir I, from Naqsh-e Rostam (Relief 1).

Fig. 14. Detail of the Armenian delegation leading ahorse, from the south wing of the east side of the apada-na,Persepolis.

Fig. 15. Achaemenid king enthroned with a beard of a stylerepresentative of the Achaemenid beards of the king ofkings visible at Naqsh-e Rostam. Relief originally from theapada-nabut moved to the treasury in antiquity during thereign of Artaxerxes I.


17/34


The reverses of Ardashir Is coinage regularly depictan enthroned fire altar, wherein altar and throneelements are drawn from Achaemenid portrayals atNaqsh-e Rostam and Persepolis (see figs. 15, 17).74Very similar to Ardashir Is use of Persian ornamen-tal motifs at Ardashir-Xwarrah, the sculptural formsof his rock relief elegantly integrate and depart fromthe ancient precedents to present the viewer with a

new expression of Persian kingship that was innova-tive yet rooted in the past.

In contrast to Ardashir Is single relief at Naqsh-eRostam, his son, Shapur I, was responsible for fullyconverting the site into the complex site of memoryit remained for the life of the empire. In order tomanipulate the wider cultural memory surroundingNaqsh-e Rostam, Shapur I engaged with the site visu-ally, inscriptionally, and ritually. He boldly carved arock relief directly in the center of the four Achae-

menid tombs, in the space underneath the tombof Darius I and the tomb attributed to Artaxerxes I(fig. 18). Unlike Ardashir Is relief, it lies within thebounds of the ramparts, across and slightly northeastfrom the Kaba. Shapurs relief is decidedly trium-phal. It clearly commemorates his capture of the Ro-

man emperor Valerian and the supposed submissionof Philip the Arab.75Compared with his other, moreelaborate reliefs, this one lacks the slain Roman em-peror underhoof and an operatic cast of nobles andsubjects from across his empire.76Unlike other sitessuch as Bishapur, where the king projected his visualpropaganda to a wider audience and whose imagesand interests often appeared therein, the reliefs ofboth Ardashir I and Shapur I at Naqsh-e Rostam aresimplified, with the major elements reduced to thecommemoration of royal deeds and identities. Paral-leling this, Shapur Is inscription takes care to instructhis successors on the benefits of piety, an important

concern of the reliefs as well:

74 Pfeiler 1973; Shahbazi 2001, 66; Alram 2008, 1822.75 This refers to the claims of the king of kings that Philip

was forced to sue for peace, and, as part of the agreement,pay a lump sum of 500,000 denarii on the spot and a large an-nual indemnity (Rubin 1998, 178). In my opinion, MacDer-mots (1954) argument for identifying these emperors is stillthe most convincing. For a recent alternative view, see Over-laet 2009.

76 The king of kings executed triumphal rock reliefs withsimilar compositions and ideas at several other sites, includ-ing Bishapur II and III and, to a lesser extent, Darabgerd.

These reliefs include numerous subordinate figures such asRoman captives, members of Shapurs court, or other subjectpeoples. The image of the king of kings triumphing over de-feated Roman emperors became the standard, official imageof Shapur I and was repeated in a wide variety of media. In ad-dition to rock reliefs, the idea appears in a cameo gemstoneportraying an equestrian duel between Shapur and Valerianas well as in Sasanian coinage: the reverse of a specially issuedgold double dinar portrays the king of kings holding Valeriancaptive (Alram et al. 2007; Canepa 2009, 6871).

Fig. 16. Detail of Ohrmazds beard and crown. Rock reliefof Ardashir I, from Naqsh-e Rostam (Relief 1).

Fig. 17. Reverse of a silver drachm of Ardashir showing afire altar supported by an Achaemenid-inspired, lion-leggedthrone, diam. 25 mm, wt. 4.55 g (Gbl SN 3.1.2)(courtesyT.K. Mallon-McCorgray).


18/34


Now just as we now are conscientious in the concernsand cult of the gods, and are the creature [dastgerd]of the gods, and just as we, with the help of the gods

sought out and conquered these lands, and did thingsof fame and daring, so let him too who shall be rulerafter us beconscientious in the concerns and cult ofthe gods, so that the gods may make him their crea-ture [dastgerd] too.77

Epigraphic and Ritual Alterations of the Kaba-ye ZardoshtShapur I introduced a complex alteration to the

Kaba-ye Zardosht, the most prestigious structure atthe site (fig. 19).78Darius I originally constructed the

Kaba-ye Zardosht in careful imitation of a tower builtby Cyrus the Great at Pasargadae, known as the Ze-ndan-e Solayman.79The ashlar masonry tower, whose

faces measure 12.60 m high by 7.25 m wide, rests ona triple-stone plinth. It gives the impression of havingthree stories, but the lower half of the tower is solid,while the upper half accommodates a single room 5.58m high and 3.70 m wide.80An imposing flight of stepson the north of the structure, now partially destroyed,led to its elevated cella. Although theories abound,the original Achaemenid function of the Kaba-yeZardosht is unknown.81Excavations in the vicinity of

77KZ 51. For summary of the scholarship on the word dast-

gerd informing my translation, see Gignoux 1994.78For a foundational study on the architecture of the struc-

tures, see Stronach 1967.79 Nylander (1966) concluded that the Zendan was con-

structed before the Kaba, since the Kaba made use of ironclamps to join its blocks and the Zendan did not. The Zendanwas likely built during the same period as the palaces of Pasar-gadae (ca. 540 B.C.E.), and the Kaba, during the reign ofDarius I, sometime after 520 B.C.E. (Gropp 2004).

80 Stronach 1967, 28788; Schmidt 1970, 3449. As dis-cussed above, the reverses of most of the Fratarakid coinsportray coffered towers that appear strikingly similar to the

profile of the Kaba (Stronach 1966, 226; Boyce 1989, 8; Potts

2007).81 Numerous scholars, from the 19th-century explorer

Ker Porter to Schmidt, have suggested the towers served theAchaemenids as fire altars or fire temples (theories reviewedin Schippmann 1971a, 19495; Potts 2007, 28285). Severalscholars, including Herzfeld (1908), Demandt (1968), andBoyce (1975, 45758), have argued that the Zendan andKaba were tombs. Frye (1974, 386) and Sancisi-Weerdenburg(1983) viewed the towers as treasuries for royal parapherna-lia or as coronation towers. For an architectural study ofthe towers, comparing them with other structures in the an-cient Near East, see Stronach 1967. For useful reviews of the

Fig. 18. Rock relief of Shapur I holding Valerian captive and receiving the submission of Philip the Arab, fromNaqsh-e Rostam (Relief 6).


19/34


Naqsh-e Rostam have revealed an early royal pavilionconstructed about 500 m from the Kaba-ye Zardosht,recalling Pasargadae.82This could indicate that thetowers at Naqsh-e Rostam and Pasargadae hosted ac-tivities that the king of kings could publicly view, ei-ther from a distance or in short procession from these

palaces or pavilions.Whatever their exact function, it is clear that theKaba and Zendan were uniquely important for theAchaemenids. The towers enjoyed a certain prestigein Pars after the Achaemenids fall, as indicated bythe Fratarakid coins that portray an attendant ven-erating a tower, often with a divine symbol hoveringoverhead.83The area around the tower at Naqsh-e Ros-tam, in particular, shows signs of building and refur-bishment through the Hellenistic era. Schmidts testtrenches revealed a dense concentration of buildingaround Kaba-ye Zardosht, which he dated variouslyto the Achaemenid, Helllenistic, and Sasanian eras.84I

believe that Potts succeeded in proving that towers aredepicted on the Fratarakid coinage; however, withoutsome sort of external corroboration, we can do littlemore than speculate about the towers function underthe Fratarakids. A fire altar, fire temple, a receptaclefor a dormant sacred fire, a treasury, and a tomb haveall been argued for the Fratarakid period as for thepreceding and succeeding periods.85While they didnot necessarily retain their original Achaemenid func-tion, this suggests that the towers inspired some sortof reverence, and perhaps even cult activity, duringthe Seleucid era and beyond.

Whether he drew from its earlier significance orcompletely reinvented it, Shapur I made the Kaba-yeZardosht the centerpiece of his newly refashioned siteof memory. Shapur I carved a monumental trilingualinscription on the lower portion of the east, west, andsouth faces of the tower.86The inscription commemo-rates the king of kings lineage, empire, and deeds andfounds an elaborate ritual protocol. Like Darius Is OldPersian, Elamite, and Akkadian inscription nearby onhis tomb, the trilingual inscription itself was an inher-ently imperial and powerful statement in the ancientNear Eastern and Iranian worlds. Even if a viewer was

illiterate, or simply not allowed to draw close enoughto the structure to read it, the mere presence of theinscription on this prestigious structure would havebeen a powerful visual statement. Shapur Is inscrip-tion was in Middle Persian, Parthian, and Greek. Thelatter two languages were those of his defeated en-emies and had the power to carry his message to theEast and West. Unlike Darius Is inscription high up onthe rock face, that of Shapur I would have been easilylegible and thus expected to do more than merely im-press from a distance. A visitor would also occasionallyexperience these inscriptions aurally; those who read

literature and the various interpretations of the structure, seeGropp 2004; Potts 2007.

82 Tilia 1974.83 Choksy 1990a, 1990b; Potts 2007, 29697.84As he and others have observed, the remainder of largely

unexcavated soil around Naqsh-e Rostam will provide the bestevidence as to what occurred there for the Sasanian period aswell as the Achaemenid era. Any attempts to defend one sin-gle interpretation should be reserved until the area has been

fully excavated, even though the present generation mightnot live to witness it (Stronach 1967; Schmidt 1970, 538, fig.23).

85See Potts (2007) for an overview of the early literature. Itshould be noted that the Fratarakidcoins do not attempt toportray fire emanating from the towers crenellations, withinthe structure, or anywhere else nearby.

86 The Middle Persian version is on the east wall, with theParthian on the west and the Greek on the south.

Fig. 19. Kaba-ye Zardosht, Naqsh-e Rostam. View of thenorth face with stairs and east face with the Middle Persianinscriptions of Shapur I and Kerdir on the lower masonrycourses.


20/34


such inscriptions were expected to read them aloud,effectively re-enlivening them and the kings memoryat every recitation.87

The first portion of the inscription proclaimsShapur Is royal genealogy, the extent of his empire,and his conquests.88This section corresponds closely

to the major themes of the king of kings rock reliefsand reflects a centrally planned propaganda cam-paign. Similar in strategy to the Persepolis graffitiand the monumental rock reliefs, the mere presenceof Shapur Is inscription boldly claimed the structurefor the king and dynasty. Their content, however,goes quite a bit further, echoing the general orderand even phraseology of the Achaemenids Old Per-sian inscriptions. The Middle Persian inscriptionsclosely follow their Old Persian predecessors not justin structure but with successions of close thematicand lexical correspondences, ranging from the titlesand descent of the king to the relative importance of

his provinces to the methods by which he creates andholds his empire.89

The linguistic and thematic parallels betweenShapurs Kaba-ye Zardosht inscription and DariusIs inscription on his tomb are startlingly similar, es-pecially considering the seven centuries that separatethem.90In fact, they are so close that positing some sortof lineal connection between the two, either throughoral tradition or epic poetry, makes much more sensethan arguing that they arose completely at random.91Much like their reuse of Achaemenid visual and archi-tectural material, the incredible linguistic and ritualparallels between the content of the inscriptions ofShapur I and the Achaemenids suggest that the Sasani-ans valued and drew on a deep cultural well of Persiantraditions, even if they did not fully understand theexact history of their ultimate source.92While ShapurIs lineage and deeds were fresh in the memory of hiscourt and people, ancient Persian cultural precursorsinspired the language in which he celebrated them.

The final two-fifths of Shapur Is Kaba-ye Zardoshtinscription describes a complex ritual protocol. ShapurI founded five sacred fires (a-dur) and established elab-orate daily rituals based on them that celebrated hisimmediate family, ancestors, and court.93According tothe inscription, the five fires were maintained for the

benefit of the soul and memory (Middle Persianpadama-ruwa-n ud panna-m) of Shapur himself, his queenof queens, and three of his sons.94Every day, a sheepand a portion of bread and wine were to be offeredfor the benefit of the soul of each. The inscription listshis ancestors, including Sasan, Papag, and Ardashir, aswell as other less-prominent members of his family. Itthen commemorates his vast court hierarchy, from thechief scribe and priest down to minor bureaucrats andthe court jail warden, microcosmically replicating thecourt hierarchy in ritual terms. Not only did Shapur Iphysically anchor this newly founded protocol on theAchaemenid tower, but he constructed the ritual itself

with inspiration from ancient Persian royal traditions.Much like the Achaemenid-inspired language of theinscription, the Sasanian protocol calling for sheep,bread, and wine has a remarkably close ancient coun-terpart. Elamite tablets from the Persepolis Fortifica-tion Archive refer to the funerary monuments (umar)of the Achaemenid kings and the cult rendered atthem for the benefit of the kings soul, including ra-tions for the upkeep of the officials in charge andanimals for sacrifice at the tombs.95Paralleling thisprimary source material, Arrian reports that withinthe precincts of Cyrus tomb, there was a small housecreated for the Magi who cared for Cyrus tomb sincethe time of Cyrus son, Cambyses, and who receivedthe stewardship from father to son. Every day they weregiven by the king a sheep, fine white flour, wine and,each month, a horse, to sacrifice for Cyrus.96I wouldargue that these Elamite texts not only corroborateArrians report but, more importantly, provide yet an-other piece of primary source evidence that proves the

87 Kerdirs inscription (KKZ), which the high priest carvedon the Kaba, under Shapur Is Middle Persian inscription,asks that, whoever [may see] this [memorial and] read it out,

[let] him be more confident in his own [soul?] about [this wor-ship and] rites (KKZ 21) (MacKenzie 1999, MR 251 [empha-sis and underlining original]).

88 Skjaerv 1985, 59396.89 Skjaerv 1985, 59398.90 Skjaerv 1985, 603.91 Skjaerv 1985, 603. Further parallels manifest between

Darius Is inscriptions and many of Ardashir Is statementsstemming from later Pahlavi literature (Shahbazi 2001, 668).For orality in Sasanian culture general, see Huyse 2008. Onthe continuity and change in the concept of Iran, see Shah-

bazi 2005b.92 Evocatively laid forth in Skjaerv 1985.93 KZ 3350.94 KZ 334. See Huyses commentary (1999, 106) for the

tradition of founding fires for the sake of the soul in laterZoroastrianism.

95 For studies of primary sources attesting to the institutionof the royal tombs, cult, and their caretakers in Achaemenidtimes, see Henkelman 2003; 2008, 28791, 42932, 546; Tup-lin 2008; Canepa 2010b.

96Arr. Anab. 6.29.78; cf. Strabo 15.37. Food offeringswere, and continue to be, an important part of the Zoroas-trian staomiritual carried out for the spiritual benefit of souls,living and deceased (Kotwal and Choksy 2004).


21/34


Sasanian technologies of memory were deeply rootedin Achaemenid practices.

All available evidence suggests that Shapur I estab-lished these fires somewhere within the precincts ofNaqsh-e Rostam, thus using both an ancient site andarchaizing rituals to give his new creation the weight of

antiquity. After Shapur Is death and during the reignof Bahram II (276293), the empires high priest,Kerdir, carved a Middle Persian inscription of his ownon the Kaba located directly below the Middle Persianversion of Shapur Is inscription. Kerdir also carvedan abbreviated version on the cliff face, adjacent toShapur Is rock relief.97In both inscriptions,98Kerdirproudly proclaims that Shapur I bequeathed to himcontrol of the fires and rituals (a-dur ud kerdaga-n)established in the king of kings inscription and evengoes so far as to quote Shapur Is bequest:99

ud e-n a- dur ud kerdaga- n e- pad nibit, a- n-im o-h-

go-nagdar a

-buhr a

-ha

-n a

-h pad wa

-spuhraga

-n pa

-yma

-rkune- d, ku- -t bun-xa-nag e-n e-w bawe- d, ud iyo-n da-ne-

ku- kerd yazada-n ud ama- weh, owo-n kun.

And he, Shapur, the king of kings, assigned to me thesefires and rituals in the inscription in this way: Let thisbe your property, and as you know this ritual protocolfor the gods and for us is good, act accordingly.

Few phrases in the Sasanian inscriptions have re-ceived such scrutiny as the phrasebun-xa-nag. I followSkjaerv and Huyses translation of the term as prop-erty or estate and, along with Shaki, their under-standing that the phrase refers to the site of Naqsh-e

Rostam, rather than the Kaba itself.100Kerdir left an-other copy of his long Kaba-ye Zardosht inscriptionat a separate site in Pars called Sar Mashhad, as well asan additional short inscription at a site called Naqsh-eRajab, located only about 2.75 km from Naqsh-e Ros-tam. The inscriptions share roughly the same content,and Sar Mashhad replicates the text on the Kaba-yeZardosht almost verbatim. However, only the two in-scriptions carved at the site of Naqsh-e Rostam contain

the lines that refer to Shapur Is inscription (nibit),the fires and rituals, and the king of kings bequest.This suggests that Kerdir was eager to showcase whatShapur I had entrusted to him. In using bun-xa-nag, aphrase that Zoroastrian law used regularly in transac-tions transferring the ownership of a fire temple, and

the demontrative pronoun, Kerdir deictically refers toNaqsh-e Rostam as the site of these fires and rituals,founded by Shapur I.101

The buildings discovered in Schmidts west testtrench yielded neither normal domestic materialnor military equipment, suggesting they served someother, less utilitarian function.102The 3.50 x 5.80 mroom that Schmidt excavated in Building I containedrectangular niches, 73 cm wide x 5356 cm high and2630 cm deep, located about 1.10 m from the groundand set into the center of the west and east walls, andinto the western section of the north wall.103Thoughdestroyed, Schmidt speculated that niches covered

the south wall as well. Building II contained two con-nected rooms, both of which were provided withsingle niches of roughly similar size and height fromthe ground.104Their exact function is unknown. It isquite possible that the building served as a library orarchive, although some sort of auxiliary cultic func-tion cannot be dismissed.105Remnants of the deco-rative treatments on the structures suggest that theSasanians covered their new creations in Achaemenidornamental features reinterpreted in new mediums,as at Ardashir-Xwarrah. Plaster covered in a gray wash,similar in color to the Persepolis treasury walls, cov-ered the mudbrick buildings clustered around theKaba, and some were decorated with stucco inspiredby Achaemenid architecture.106

The exact role that the Kaba-ye Zardosht playedin Shapurs ritual protocol is uncertain and shouldremain open until new evidence emerges, as shouldthe function of the post-Achaemenid buildings sur-rounding the Kaba. Until that time, a number ofgeneral observations are possible. The very presenceof the inscribed record (nibit) of the foundation of

97 KNRm (Huyse 1998, 112).98 KKZ 23; KNRm 67.99 KKZ 3.100 Shaki 1974; Skjaearv 1989, 1993. For a review of the lit-

erature, see Huyse 1998, 11016.101 ce passage montre que cest une expression technique

justement employee lors de la transmission de la direction destemples du Feu. . . . Cest la mme operation entre ahpuhret Kardir qui est dcrite dans linscription (Humbach 1974,204). Similar phrases occur in Bactrian contracts document-ing the sale of secular property (Huyse 1998, 116 n. 41).

102 Schmidt 1970, 57.103 Schmidt 1970, 54.

104 The north room measured 4.7 x 3.0 m and had at least

one niche per wall, 85 cm from the ground, measuring 6070cm wide x 50 cm high x 2530 cm deep. The south room,measuring 2.98 x 1.80 m, had a single niche 80 cm from theground in the west and east walls and two niches in the southwall. They measured 5062 cm wide x 4550 cm high x 2530cm deep (Schmidt 1970, 54).

105While they likely did not house altars or cult statues, with-out knowing the nature of the rest of the site, I believe theirfunction should be held open (Huyse 1998, 11516).

106 Schmidt 1970, 548, 75, fig. 30, no. 23; Krger 1982,19697.


22/34


the cult and its transfer to Kerdir conveys the Kabasgeneral prestige and, at least indirect, connection withcult.107Unless evidence of cult furniture or a largerfire temple is discovered, one should not dismiss outof hand the possibility that these ritual activities tookplace on or around the Kaba.108Similarly, any subsid-

iary functions that the tower might have served mustremain open until new evidence emerges. Much likethe Achaemenid and Fratarakid periods, the threefunctions that have often been put forward for theSasanian era are a tomb, treasury, and religious struc-ture.109 Although a treasury cannot be discounted,there is no evidence to support Hennings specificinterpretation of the Kaba as a treasury containinga master copy of the Avesta.110The fact that the inte-rior of the Kaba presents an unventilated space sug-gests that it would be difficult to contain a sacred fireinside it. Of the more likely suggestions, it has beenarguedbut cannot be provedthat Shapur reused

the Kaba as his tomb.111Such a close, fused, cultic-monumental-funerary function would cohere quitewell with Achaemenid funerary practices revealed inthe Persepolis Fortification Tablets as well as the over-all pattern of Sasanian-Achaemenid correspondencesthat Shapur Is activities at the site present.112Even ifShapur Is bones were interred elsewhere at the siteor a different site entirely, his visual, inscriptional, andritual legacy ensured that the Kaba could at least serveas a cenotaph. With luck, the unexcavated portions ofNaqsh-e Rostam will provide additional informationat some point in the future.

BishapurDuring roughly the same period Shapur I remade

Naqsh-e Rostam into a dynastic sanctuary, he foundedthe city of Bishapur, located only 300 km to the west ofArdashir-Xwarrah.113The excavated remains of the citydemonstrate how the technologies of memory estab-

lished at Ardashir-Xwarrah and Staxr inspired ShapurIs own creations. In just one generation, the memorialpractices that Ardashir instituted became an integralpart of Sasanian city building: although only a smallportion of the city has been excavated, Bishapur evinc-es a use of Achaemenid architectural forms on certain

prestigious and innovative structures. Like the Takht-eNeshin and palaces in Ardashir-Xwarrah, several struc-tures in Bishapur carried architectural ornament thatdeliberately appropriated Achaemenid visual culture.Rather than transporting ancient architectural mem-bers or translating them into the more ephemeralmedium of stucco, however, at Bishapur, contempo-rary stonemasons translated into local stone thoseAchaemenid architectural forms that had emerged asthe most privileged in third-century Sasanian reuse.114An Achaemenidizing cornice, for example, marksthe main entranceway into the grand chamber of thesubterranean Anahid temple, and the colossal bull

protomes there hearken to Persepolitan bull capitals(figs. 202).115

Shapur I combined these older traditions with ex-perimentation in imaging his expanding Sasanianideology of kingship. In this city, he used captives andselect artistic techniques from Antioch in memory ofhis western triumphs.116The Achaemenidizing cor-nices are set atop doorframes that derive fr

Canepa_Achaemenid Sites Sasanian Identity.pdf

Documents

Transcript of Canepa_Achaemenid Sites Sasanian Identity.pdf