On the Peripheries of Civilizations:The Evolution of a Visual Tradition in Gandhāra
Juhyung Rhi*
The region of Gandhāra is mostly known for its distinctive visual tradition, which is commonly called Gandhāran art (Fig. 1). More recently, Gandhāran Buddhism and Gāndhārī, the language of Gandhāra, have also become increasingly familiar terms in Buddhist historical and philological studies.1 Gandhāra, in the most restricted definition, means the Peshawar valley to the western side of the Indus River, but in the usage common in art historical and Buddhist studies it encompasses wider areas including Taxila, Swāt, and eastern Afghanistan (Fig. 2).2 In this paper, Gandhāra will be used according to this broader definition, which is also identified as “Greater Gandhāra” by many scholars.
In many ways, Gandhāra is on the peripheries—in ethnic groups, political dominions, and cultural divisions. Geographically, it is located at the point where three major regional units, i.e., South, West Asia, and Central Asia meet each other. Historically, it continually passed through diverse political powers ruling from the neighboring areas. It was also on the fringes of prominent civilizations during the ancient period, including Iranian, Indian, and Hellenistic civilizations. Cultural achievements of the region were thus often perceived as peripheral or derivative, and its distinctive
Figure 1. Standing Buddha (detail). From Sahri-Bahlol, Peshawar valley. Peshawar Museum. (J. Rhi)
Figure 2. Map around Greater Gandhāra. (Modified from The Crossroads of Asia, 1992, 1-2.)
On the Peripheries of Civilizations 3
visual tradition that evolved in the first several centuries of the Common Era has been commonly treated as the most remarkable cultural phenomena that unfolded in this marginal part of the Eurasian continent. Frequently neglected in this conventional conception is the initiative taken by the local population of the region in the achievement of this extraordinary cultural feat. This paper addresses the problem of the Gandhāran initiative, by tracing the evolution of its visual tradition with a special note on the role of the community that employed the Kharoṣṭhī script.
In the period crucial to the development of Gandhāran visual tradition, from about 500 BCE to 500 CE, the majority of the occupants of the region were ethnically Indians. They were probably descendants of the Gandhāri, an offshoot of Indo-Āryans, who migrated to the subcontinent throughout the so-called Vedic age.3 Gandhāra, the land where they settled, is listed in Aṅguttara Nikāya, a Pāli Buddhist scripture, as one of the sixteen great states (mahājanapadā) of ancient India that flourished during the Buddha’s time.4 Much later, the Chinese monk Faxian, who visited India in the first decade of the fifth century CE, considered Uḍḍiyāna (Swāt), Gandhāra, and Nagarahāra (Jalalabad valley in eastern Afghanistan) all parts of Tianzhu (India).5 Another Chinese monk, Xuanzang, who reached India around 630, had the definite idea that the northwestern boundary of Indu (India) lay between Kāpiśa (north of the Kabul valley) and Lampāka (Laghman valley in present Afghanistan).6 The precise meaning of Tianzhu or Indu in these Chinese sources could be problematic, for there was obviously no concept of India as a single political entity at the time. These terms most likely reflected the ethnic or linguistic boundary of the areas inhabited by Indians. Still, the most compelling evidence for the ethnicity of the residents of Gandhāra during this period is the regional dominance of the Kharoṣṭhī script as well as Gāndhārī language.
The Kharoṣṭhī script was predominantly used in Gandhāra from the fourth century BCE to the fourth century CE (Fig. 3).7 As is well known, it derives from Aramaic, a script officially used in the Achaemenid Empire in Persia; it was most likely created by adapting Aramaic for the local language of Gandhāra, a northwestern dialect of the Middle Indic. The earliest extant documents in Kharoṣṭhī are two sets of royal edicts by Aśoka (of the Indian empire Maurya based in the middle Gangetic valley) found in Shāhbāz-gaṛhī and Mānsehrā in and near the Peshawar valley.8 It is noteworthy that three inscriptions Aśoka left in Kandahar, further to the west in Southern Afghanistan, were not in Kharoṣṭhī but in Greek and Aramaic. One of them is bilingual and biscriptal in Greek and Aramaic; another is in Greek, and
4 Juhyung Rhi
the third one is in Aramaic-Māgadhī (Māgadhī in Aramaic transcription and its Aramaic translation).9 Apparently, the Greek inscriptions were intended for the Greek population, and the Aramaic for the Iranian population in the region. The particular use of Kharoṣṭhī in Gandhāra indicates the dominant presence of an Indic-speaking population in the area. It is also notable that all the remaining inscriptions of Aśoka within what eventually became India proper were written in Brāhmī, another prominent ancient Indic script. This clearly indicates the cultural independence of Gandhāra from the rest of India.
Although we do not know the date and circumstances of the inception of Kharoṣṭhī in Gandhāra, we can easily presume that the Achaemenid rule over the region during the fifth and fourth centuries BCE was instrumental in its creation. Georg Bühler, a great pioneer in Indian epigraphy, suggested long ago:
Figure 3. Kharoṣṭhī script. (Salomon, Indian Epigraphy, table 2.5.)
On the Peripheries of Civilizations 5
. . . it appears natural to assume that the Persian Satraps carried with them also into India their staff of subordinates, who were accustomed to the use of the Aramean letters and language. And this would fully explain how the Hindus of the Indo-Persian provinces were driven to utilize the characters, commonly employed by the scribes and accountants of their conquerors . . . The Kharoshṭhī Alphabet would appear to be the result of the intercourse between the offices of the Satraps and of the native authorities, the Indian chiefs and the heads of towns and villages, . . .10
Most specialists have accepted this plausible hypothesis.11 However, this raises the question of the role of the local Indian population. Can we assume that they simply played a subservient role in the appropriation and establishment of Kharoṣṭhī? We know that in this period a highly sophisticated and distinctive culture was evolving among the local population through dynamic —and less inhibited—interactions with external elements. A cogent example is the work of the famous grammarian Pāṇini, who was active in Gandhāra around the fourth century BCE or earlier. Linguistic specialists believe that Aṣṭādhyāyī, his monumental work, was too complex to have been written without the support of literacy, and have proposed Aramaic or Kharoṣṭhī as the candidates for Pāṇini’s lipi (script).12 Whichever was Pāṇini’s lipi, it is apparent that the need for literacy was felt strongly in Gandhāra not just for administrative purposes but also for intricate cultural products; the desire for literacy was probably stronger in Gandhāra than in the Gangetic valley, where orality continued to enjoy greater prestige. Though Kharoṣṭhī and Brāhmī are generally acknowledged to have been created independently—while the latter was also probably derived from Aramaic or other Semitic scripts—Brāhmī shows occasional borrowings from Kharoṣṭhī and possibly was preceded by Kharoṣṭhī.13 Compared with the Gangetic valley, Gandhāra was obviously no backwater during this period, especially considering that Indian civilization was still seeking its distinctive form in many sectors through vital interactions with civilizations from the West.
Around 200 BCE, Kharoṣṭhī was adopted along with Greek script for legends in the coinage of Greco-Bactrian kings; later kings of Śaka and Parthian dynasties, as well as the Kuṣāṇas, who took turns in ruling Gandhāra until the third century CE, maintained this practice. At least by the first century BCE, scribes started using Kharoṣṭhī to write down Buddhist scriptures, as found in numerous manuscriptal remains discovered within the past two decades in Gandhāra (especially in such places as Haḍḍa and Bāmiyān) and vigorously worked on in Buddhist philological scholarship.14 They are mostly written in northwestern Prakrit, now commonly called Gāndhārī. A large number of Buddhist dedicatory inscriptions also appear
6 Juhyung Rhi
with relics and images dedicated to Gandhāran monasteries.15 They were invariably written in Kharoṣṭhī, except for a handful of aberrations in Brāhmī or the biscriptal Kharoṣṭhī/Brāhmī.16 Evidently, the Buddhist visual tradition of Gandhāra was inseparably tied with Kharoṣṭhī. Around the third to fourth centuries, i.e. the time when Kharoṣṭhī disappeared only to be replaced by Brāhmī in Gandhāra and thus in the Indian subcontinent, Gandhāran visual tradition also lost its momentum as well as its idiosyncratic character.17 Thus the Kharoṣṭhī/Gāndhārī community played a critical role in its evolvement.
The Kharoṣṭhī/Gāndhārī community must have been mainly comprised of ethnic Indians, as Gāndhārī was essentially an Indic language. Donors with Greek names are rarely seen in dedicatory inscriptions, but they also recorded their dedications in Gāndhārī.18 It is also notable that in the period when the visual tradition flourished in Gandhāra, there were practically no Buddhist epigraphical documents in scripts and languages other than Kharoṣṭhī/Gāndhārī. The foreign immigrants—such as Iranians, Greeks, and Central Asian nomads—who probably formed minorities in the region seem to have been largely assimilated into the Kharoṣṭhī community. Still, this community had a significantly distinct identity and cultural orientation compared to those of the Gangetic valley or elsewhere in the south.
However, choices made at the earliest stage of the visual tradition in Gandhāra were more complex than the simplification above permits. When the desire arose for an iconic representation of the Buddha, which provided a critical impetus for the rise of a Buddhist visual tradition, Gandhāra obviously had more advantages than the Gangetic valley in that it already possessed diverse prototypes transmitted from the Hellenized West. This desire itself could have been brought on by such prototypes and practices of external origin.19 The apparent Hellenistic ancestry of the Gandhāran visual tradition made earlier scholars seek its origin in the Greco-Bactrian or Indo-Greek community.20 However, no tangible evidence has been found in excavations conducted in limited scales on sites that may be associated with the Greek dominion, while no Buddhist image inscribed with Greek has ever been found.21 Most likely, the desire for iconic images had not yet ripened enough by the time the Indo-Greeks were in political ascendancy.
The earliest extant examples of the Gandhāran visual tradition are known to have come from Swāt and are datable to the beginning of the Common Era.22 They show considerably different choices from what became the mainstream creations of Gandhāra slightly later. These images seem to reflect influence from Hellenized Iran and its adjacent areas in Central Asia. Thus they nicely correspond with the Iranian theory on the origin of
On the Peripheries of Civilizations 7
the Gandhāran visual tradition, which was first seriously proposed by Daniel Schlumberger and has since attracted a wide following among specialists.23 Simultaneously, scholars have pointed out the similarities between the finds from Swāt and the earliest Buddhist images from Mathurā in the upper Gangetic valley. Based on these similarities, J.E. van Lohuizen-de Leeuw, a staunch proponent for Mathurā’s precedence over Gandhāra in the creation of Buddha images, saw influence from Mathurā in these Swāt images and forceful evidence supporting her theory.24 We must admit, however, according to the hard evidence so far available to us, that there have been no Buddhist images from Mathurā datable as early as the Swāt images. I do not wish to conclude by this fact alone that the finds from Swāt set precedents for those from Mathurā; rather the affinities between the images of the two regions may stem from the common stimulus they appropriated. In the Gangetic valley, visual creations emerged prominently by the third century BCE under the Mauryas, although this fact itself is somewhat enigmatic. However, they were still largely in the formative stage until the first century CE, and we need not assume that India proper was more advanced than Gandhāra in visual creations.
While Swāt maintained its regional idiosyncrasy through later periods, what may be termed as the mainstream creations in the Gandhāran visual tradition were centered on the northern side of the Kabul River in the Peshawar valley. The finds from this region are more refined than contemporary Iranian works and closer to Hellenistic/Roman models, although considerably austere and stiff (Fig. 1). Overall, compared with earlier specimens, they seem greatly advanced in monumentality, execution of details, and expression. Several visual types are discernible, which are attributable to different sub-regional bases, workshops, and periods of duration.25 This phase in the Gandhāran visual tradition is hard to date with any certainty, but has been commonly assigned to the second Kuṣāṇa dynasty initiated by Kaniṣka (r. ca. 127-150 CE). Two dated images, one in the year 89 (most probably of the Kaniṣka era) and the other in the year 5 (of the Kaniṣka era or of its second century), have been considered to support this presumption.26
The establishment of this phase of the Gandhāran visual tradition has sometimes been linked to the dynastic involvement of the Kuṣāṇas. However, no evidence corroborates such dynastic patronage. No Buddhist dedicatory inscription has been found written in Kuṣāṇa’s native Bactrian language (at least since the second century BCE) or Greek script, as one finds in the commemorative inscription carved at the famous dynastic fire
8 Juhyung Rhi
temple of Surkh Kotal in northern Afghanistan.27 A reliquary (or perfume box) dedicated at Shāh-jī-kī Ḍherī, the Kaniṣka vihāra, is also inscribed in Kharoṣṭhī/Gāndhārī, and its epigraphical content does not show any clear indication of dynastic patronage.28 Furthermore, relatively few stone carvings were discovered in and around the dynastic capital Puruṣapura, or modern Peshawar, while the majority was concentrated on the northern side of the Kabul River where numerous monasteries and workshops were present.29 The Kharoṣṭhī community based in the northern Peshawar valley most likely played a crucial part in the emergence of this flourishing phase.
Beginning in the early second century CE, monumental Buddhist images were produced in Mathurā in a distinctive manner. It is remarkable that the two regions made such different choices, although both were inhabited by Indic-speaking populations and were under the same Kuṣāṇa rule. These choices could have been deliberate ones based on the perception of separate identities, which is most symbolically expressed in the use of two different scripts, Kharoṣṭhī and Brāhmī.
The only source of inferiority for the Gandhāran Buddhist community would have been that the Buddha never set foot on their land in his lifetime. To compensate this deficiency, various places claimed to be the loci of incidents of the Buddha’s previous incarnations.30 Tangible objects were also installed as sacred relics of the Buddha, such as the Buddha’s alms-bowl in Peshawar and the Buddha’s skull bone in Haḍḍa.31 The bowl in Peshawar was
Figure 4. Veneration of the Buddha’s Bowl. On the pedestal of a standing Buddha. Peshawar Museum. (J. Rhi)
On the Peripheries of Civilizations 9
declared as to have been brought from Pāṭaliputra or somewhere in Central India, by Kaniṣka. It became the most sacred relic in the Peshawar valley (Gandhāra proper), and thus the veneration of the relic became a popular theme carved on numerous images in the area (Fig. 4).32 Apparently, however, the claim was not acknowledged by Buddhists in Mathurā, who believed that the bowl was worshipped in a celestial realm—as I have argued elsewhere.33 A story cited in the Harivaṃśa, a final addition to the great epic Mahābhārata, composed by a resident of Mathurā around 300 CE, depicts the quelling of Kālayavana’s attack on Mathurā.34 The villain is referred to as Kālayavana, “Black Greek,” but this figurative appellation most likely designates the northwesterners of the subcontinent in general. This story can be read as exhibiting the inherent antagonism between the inhabitants of the two regions, which was aggravated by a series of invasions on Mathurā by alien rulers.
The latest dated inscriptions in Kharoṣṭhī from Gandhāra are those dated to the 300s series of an unknown era. Among Buddhist images, there are two Buddhas, each dated to the years 318 and 384, as well as a Hārītī dated to the year 399.35 These dates have been commonly attributed to the so-called “Old Śaka” era, for which diverse opinions have been put forward, including dates from 248/7 BCE to 84/3 BCE.36 Richard Salomon recently proposed an Indo-Greek era for these dates, which he places in 186/5 BCE.37 According to Salomon, the dates of these three images thus correspond to ca. 132, 198 and 213 CE. However, in my understanding of the visual evolvement of Gandhāran Buddhist images, these dates seem too early. For instance, the Buddha of the year 318 shows significantly degenerated features; it is hard to believe that it precedes the Indraśailaguhā Buddha in the Peshawar Museum by some eighty years. The Indraśailaguhā Buddha is dated to the year 89 of the Kaniṣka era, which can be converted to 215/6 CE if we agree (as many scholars including Salomon do) that the beginning of the era can be dated to 127 CE. It seems to me that the three images are datable as no earlier than the third century CE. Salomon considers the image of the year 399 among the last instances of Kharoṣṭhī inscriptions and, based on his new calculation, assigns the disappearance of Kharoṣṭhī from Gandhāra to around the early third century CE.38 However, I would prefer to date this to approximately a century later.
Around the third or fourth century, Kharoṣṭhī suddenly became obsolete in Gandhāra. Salomon believes that this is due to the demise of the Kuṣāṇa empire—where Kharoṣṭhī was mainly used by administrative clerks and Buddhist clerics—and does not connect it with the loss of the cultural
10 Juhyung Rhi
identity of the community.39 Intriguingly enough, however, this phenomenon coincided with the disappearance of the idiosyncratic character of Gandhāran visual traditions. From this time on, while Brāhmī replaced Kharoṣṭhī in Buddhist inscriptions and manuscripts, the distinctive visual style f rom Mathurā and Sārnāth of the Gupta period (c. 320-550) was prevalently adopted in dedications and decorations in Buddhist monasteries, or sometimes intermingled with the new Sassanian styles transmitted from Iran (Fig. 5). The Hindu presence in religious images also became conspicuous. These phenomena seem to have been not merely coincidental but intricately related. Although the Kuṣāṇas may not have brought about the unique culture of Gandhāra, they could have provided an environment that fostered its development independent from
other neighboring civilizations. The demise of the Kuṣāṇas thus meant the dissolution of cultural independence forged in the region for the previous several centuries and the loss of a long-held identity for a cultural unit. The community represented by the Kharoṣṭhī script and a distinctive visual tradition was not able to survive the advance of a powerful civilization that finally established itself in consummate form in the Gangetic valley and disseminated its influence to neighboring areas. And this marked the end of a remarkable civilization in Central Eurasia, which was neither entirely Indian nor Iranian nor Classical.
Acknowledgements
I would like to thank Professor Richard Salomon for kindly providing information on the Kharoṣṭhī community in Gandhāra, including his own scholarship, although my assessments ended up to be different from those of
Figure 5. Standing Buddha. Bronze. British Museum. (Masterpieces of Buddhist and Hindu Sculpture from the British Museum, Tokyo: Asahi shinbun, 1994, pl. 69.)
On the Peripheries of Civilizations 11
his work.
NOTES
1 See for example such notable recent publications as Richard Salomon, Ancient Buddhist Scrolls from Gandhāra: the British Library Kharoṣṭhī Fragments (Seattle: University of Washington Press, 1999); Pia Brancaccio and Kurt Behrendt, ed., Gandhāran Buddhism: Archaeology, Art, Texts (Vancouver: University of British Columbia Press, 2006). 2 The regional boundary of Gandhāra varied through history. It sometimes included Taxila on the eastern side of the Indus river as its capital. N.N. Bhattacharyya, The Geographical Dictionary: Ancient and Early Medieval India (Delhi: Munshiram Manoharlal, 1991), 130-31 (citing Mahābhārata, XII.207.43 and Rāmāyaṇa, VII.113.11, VII.114.11); B.C. Law, “North India in the Sixth Century B.C.,” in The History and Culture of the Indian People, vol. 1: The Age of Imperial Unity (Bombay: Bharatiya Vidya Bhavan, 1951), 14. The Chinese pilgrim Faxian, who traveled to Gandhāra around 400 CE, referred to both Jiantuoweiguo (kingdom of Gandhāra) and Fuloushaguo (kingdom of Puruṣa, i.e. Puruṣapura, modern Peshawar) for the Peshawar valley. He seems to have meant by Gandhāra to the central and northern part of the Peshawar valley, centered around Puṣkalāvatī, modern Charsada. T2085, 51: 858b, cf. James Legge, A Record of the Buddhistic Kingdoms (Oxford: Clarendon Press, 1886), 31-36. Xuanzang, who arrived in Gandhāra some two hundred years later, used the name Gandhāra to indicate the area restricted to the Peshawar valley. Samuel Beal, Si-yu-ki, Buddhist Record of the Western World (London: Trübner & Co., 1884), vol. 1, 97-118.3 The Ṛgveda (i, 126, 127) refers to the Gāndhāris for the good sheep wool they produced. They are also mentioned in the Atharvaveda (v. 22, 14). A.A. MacDonell and A.B. Keith, Vedic Index of Names and Subjects (London: John Murray, 1912), vol. 1, 218-19; N.N. Bhattacharyya, 130-31. 4 Aṅguttara Nikāya (ed. Richard Morris, rev. 2nd ed., London: Luzac & Co., for PTS, 1958-1961), I:213; IV:252. Cf. Law, pp. 1-3, 14.5 Faxian says that one reaches North India by crossing the Congling (the Onion Mountains). This is followed by the accounts of such regions as Tuoli (Dārel), Wuchang (Uḍḍiyāna), Suheduo (Swāt), Jiantuowei (Gandhāra), Zhuchashiluo (Takṣaśilā), Fulousha (Puruṣapura), and Najie (Nagarahāra). T2085, 51: 857c-858a, cf. Legge, 24-40. 6 T2087, 51: 875b, cf. Beal, vol. 1, 68.7 Richard Salomon, Indian Epigraphy (New York and Oxford: Oxford University Press, 1998), 42-55.8 Harry Falk, Aśokan Sites and Artefacts: A Sourcebook with Bibliography (Mainz: Verlag Philipp von Zabern, 2006), 127-29 (Mānsehrā), 132-35 (Shāhbāz-gaṛhī).9 Falk, Aśokan Sites and Artefacts, 242-46. 10 Georg Bühler, “The Origin of the Kharoshthi Alphabet,” Indian Antiquary 24 (1895), 288, cf. idem, Indian Paleography (1905; reprint, New Delhi: Oriental Books Reprint, 1980), 36-38.11 Salomon, Indian Epigraphy, 52-54, cf. Salomon, “On the Origin of the Early Indian Scripts,” Journal of the American Oriental Society 115-2 (1995): 271-79.12 Oskar von Hinüber thinks that Pāṇini’s lipi could be Kharoṣṭhī or Aramaic (von Hinüber, Der Beginn der Schrift und frühe Schriftlichkeit in Indien [Mainz: Akademie der Wissenschaften und der Literatur, 1990], 56-57), whereas Harry Falk considers it to be Aramaic (Falk, Schrift im alten Indien [Tübingen: Gunter Narr Verlag, 1993], 257-59). Nicely summarizing and critically reviewing the ideas of von Hinüber and Falk, Salomon finds it more likely that the lipi was Kharoṣṭhī (“On the Origin of the Early Indian Scripts,” 274-75; Indian Epigraphy, 46).
12 Juhyung Rhi
13 Salomon, Indian Epigraphy, pp . 19-30, especially 23 (for Brāhmī’s relationship with Kharoṣṭhī); Salomon, “The Origin of the Early Indian Scripts.”14 Richard Salomon, “New Manuscript Sources for the Study of Gandhāran Buddhism,” in Gandhāran Buddhism: Archaeology, Art, Texts, ed. Pia Brancaccio and Kurt Behrendt (Vancouver and Toronto: University of British Columbia Press, 2006), 135-150.15 Sten Konow, Kharoshṭī Inscriptions with the Exception of Those of Aśoka (Calcutta: Government of India, 1929), which is evidently dated. For the most updated comprehensive database created by the Early Buddhist Manuscripts Project, Stefan Baums and Andrew Glass, “Catalog of Kharoṣṭhī Inscriptions,” in The Early Buddhist Manuscripts Project, http://www.ebmp.org/a_inscriptions.php (accessed December 31, 2009).16 Salomon, Indian Epigraphy, 70-71.17 Mainly based on the observation of the masonry of image shrines in Gandhāra, Kurt Behrendt has proposed a long, and fairly late, chronology—the third to eighth century CE—for the production of Gandhāran Buddhist icons. Behrendt, The Buddhist Architecture of Gandhāra (Leiden and Boston: Brill, 2004), especially, 4-11, 244-54, 268-87. However, I am skeptical about his suggestion, as apart from the masonry, evidence seems to indicate that most images cannot be that late. Even when we admit that the dating of buildings may be based on the masonry—which itself is heavily dependent on the information induced from monuments in Taxila—it is possible that image shrines currently extant were constructed for reinstallation of images, later than their initial dedication. 18 Konow, nos. I (reliquary inscription of the meridarkh Theodoros from Swāt), II (copper plate inscription of a meridarkh recording the dedication of a stupa), XXIV (inscribed stone of Thaidora recording the dedication of a tank), LXX (inscription of Miṇaṃdra on a stone panel depicting two wrestlers). 19 This remark may remind readers of the colonialist preconceptions of early Indian Buddhist art. With no doubt, one should not anachronistically rejuvenate the viewpoints of by-gone era. However, we should also keep it in mind that chronological precedence is a problem concerning what happened in history, not what should have happened in light of our politically correct standpoints. 20 For the most prominent example, see Alfred Foucher, “The Greek Origin of the Image of Buddha,” in his Beginnings of Buddhist Art (Paris: Paul Geuthner, 1917), 111-38.21 It is also revealing that virtually no Buddhist artifact or remains dating prior to the Common Era has been located to the west of the Jalalabad valley in Greater Gandhāra. I believe that the Bactrian-Greek community in Afghanistan played little role in the rise of the Buddhist visual tradition of Gandhāra.22 Domenico Faccenna, “Excavations of the Italian Archaeological Mission (IsMEO) in Pakistan: Some Problems of Gandharan Art and Architecture,” in Central Asia in the Kushan Period (Moscow: Nauka, 1974), vol. 2, pp. 126-79; idem, Butkara I (Swāt, Pakistan), 1856-62, IsMEO Reports and Memoirs 3, pt. 1 (Rome: IsMEO, 1980), 47-75.23 Daniel Schlumberger, “Descendents non-méditerranéens de l’art grec,” Syria 37 (1960), pp. 132-66, 253-318; Chantal Fabrègues, “The Indo-Parthian Beginnings of Gandhara Sculpture,” Bulletin of the Asia Institute 1 (1987): 33-43.24 Johanna E. van Lohuizen-de Leeuw, “New Evidence with Regard to the Origin of the Buddha Image,” in South Asian Archaeology 1979, ed. Herbert Härtel (Berlin: Dietrich Reimer Verlag), 377-400.25 Juhyung Rhi, “Identifying Several Visual Types in Gandhāran Buddha Images,” Archives of Asian Art 58 (2008): 43-85.26 Harald Ingholt and Islay Lyons, Gandhāran Art in Pakistan (New York: Pantheon Books, 1957), pl. 131; Stanislaw Czuma, Kushan Sculpture: Images from Early India (Cleveland: Cleveland
On the Peripheries of Civilizations 13
Museum of Art, 1985), pl. 109. 27 W.B. Henning, “The Bactrian Inscription,” Bulletin of the School of Oriental and African Studies 23-1 (1960): 46-55; J. Harmata, “The Great Bactrian Inscription,” Acta Antiqua (Academiae Scientiarcum Hungaricae) 12-1/2 (1964): 373-471.28 The proper reading of the inscription of the reliquary as well as its dating has been a controversial issue. See Gérard Fussman, “Numismatic and Epigraphic Evidence for the Chronology of Early Gandharan Art,” in Investigating Indian Art, ed. M. Yaldiz and W. Lobo (Berlin: Museum für Indische Kunst, 1987), 67-88, especially 77-82; B.N. Mukherjee, “A Note on the Shāh-jī-kī-ḍherī Casket Inscription of Kanishka I,” Berliner indologische Studien 4/5 (1989): 375-79; Harry Falk, “The Inscription on the So-called Kaniṣka Casket,” Silk Road Art and Archaeology 8 (2002): 111-13, cf. Elizabeth Errington, “Numismatic Evidence for Dating the “Kaniṣka” Reliquary,” Silk Road and Archaeology 8 (2002): 101-20. 29 D.B. Spooner, “Excavations at Shah-ji-ki-Dheri,” Archaeological Survey of India Annual Report 1908-09 (1912): 38-59; H. Hargreaves, “Excavations at Shah-ji-ki-Dheri,” Archaeological Survey of India Annual Report 1910-11 (1914), 25-32; Rhi, “Identifying Several Visual Types,” 74-77.30 Dīpaṅkara-jātaka in Nagarahāra (FX, XZ); Śibi-jātaka in Swāt (FX)/Uḍḍiyāna (XZ); Kṣāntivādin enduring tortures in Uḍḍiyāna (XZ); giving up life for listening to half a gātha in Uḍḍiyāna (XZ); breaking bones to hear dharma and writing down what he heard in Uḍḍiyāna (XZ); transforming himself into a gigantic snake and feeding starving people in Uḍḍiyāna (XZ); a peacock king finding a spring for his retinue in Uḍḍiyāna (XZ); King Maitrībala feeding five yakṣas with blood in Uḍḍiyāna (XZ); offering eyes in Gandhāra (FX); (XZ); Viśvantara-jātaka in Gandhāra (XZ); Candraprabha offering the head in Taxila (FX, XZ); Mahāsattva feeding starving tigers near Taxila (FX)/in Siṃhapura (XZ). FX: Faxian (Legge, pp. 30-39); XZ: Xuanzang (Beal, vol. 1, 91-147). In the Chinese pilgrims’ records, jātakas are not told as prevalently as other areas in India. 31 Legge, 34-38; Beal, vol. 1, 95-99. 32 Shōshin Kuwayama, “The Buddha’s Bowl in Gandhāra and Relevant Problems,” in South Asian Archaeology 1987, ed. M. Taddei and P. Callieri (Rome: IsMEO, 1990), 945-78.33 Juhyung Rhi, “The Fate of a Bowl (or Bowls): Representations of the Buddha’s Bowl in Early Indian Buddhism,” in The Art of Central Asia and the Indian Sub-Continent in Cross-Cultural Perspective, ed. Anupa Pande (New Delhi: National Museum Institute and Aryan Book International, 2009), 61-77.34 Norvin Hein, “Kālayavana, A Key to Mathurā’s Cultural Self-Perception,” in Mathurā, The Cultural Heritage, ed. Doris Meth Srinivasan (New Delhi: American Institute of Indian Studies, 1989), 223-35.35 Ingholt and Lyons, Gandhāran Art in Pakistan, figs. II.1, II.2, II.3.36 For the Old Śaka era, consult Salomon, Indian Epigraphy, 181.37 Richard Salomon, “The Indo-Greek Era of 186/5 B.C. in a Buddhist Reliquary Inscription,” in Afghanistan: ancien carrefour entre l’Est et l’Ouest, ed. Osmund Bopearachchi and Marie-Françoise Boussac (Turnhout: Brepolis, 2003), 359-401.38 Richard Salomon, “Whatever Happened to Kharoṣṭhī? The Fate of a Forgotten Indic Script,” in The Disappearance of Writing Systems: Perspectives on Literacy and Communication, ed. John Baines, John Bennet, and Stephen Houston (London and Oakville, CT: Equinox Publishing, 2008), 139-55.39 Salomon, “Whatever Happened to Kharoṣṭhī?”
