Introduction

i.1The Exemplary Tale of Sumāgadhā opens at Prince Jeta’s Grove, Anāthapiṇḍada’s Park, in Śrāvastī, where the Buddha is staying. At the time, Anāthapiṇḍada’s daughter Sumāgadhā is married off to Vṛṣabhadatta, the son a merchant who follows the nirgrantha or Jain tradition, in the distant city of Puṇḍravardhana. After arriving at the home of her in-laws, Sumāgadhā is disheartened on encountering the nirgrantha mendicants. When her mother-in-law asks why she seems despondent, Sumāgadhā tells her about the Buddha. At her mother-in-law’s request, she invites the Buddha and the saṅgha of monks for a meal, and she does so by preparing an offering and calling out from the rooftop. When Ānanda inquires about this invitation, the Buddha announces that all monks with miraculous powers must take a tally stick and travel to Puṇḍravardhana. As the śrāvakas arrive on different extraordinary vehicles that they have created with their miraculous powers, Sumāgadhā relates a brief story about each of them, with these stories alluding to other tales about the śrāvakas.

i.2Finally, the Buddha arrives in the company of a retinue of gods and gandharvas and converts the people of Puṇḍravardhana with his own miraculous display. When the monks ask how it is that Sumāgadhā’s marriage has benefited so many beings, the Buddha relates the story of her past life as the princess Kāñcanamālā during the time of the Buddha Kāśyapa and, in turn, Kāñcanamālā’s past life as the virtuous wife of a farmer, explaining that she has performed buddha activity in the past and continues to do so. This sūtra also contains the popular account of the ten dreams of King Kṛkin, which are interpreted by the Buddha as foretelling the future decline of the Dharma.

i.3The story of Sumāgadhā and the subsequent conversion of the nirgranthas has been a popular narrative in the Buddhist world, going back to its apparent circulation in ancient Gandhāra as early as the second century, if not earlier. It has been told and retold many times, also appearing, for example, in Kṣemendra’s (ca. 990–ca. 1070 ᴄᴇ) Bodhisattvāvadāna­kalpalatā, a poetic retelling of a number of Buddhist avadānas and jātaka stories. The account of King Kṛkin’s ten dreams, too, has also circulated independently of this sūtra, appearing in various formulations in a variety of texts, and even finding its way into Persian, Arabic, and Slavic literature. Gö Lotsāwa Shönnu Pal (’gos lo tsA ba gzhon nu dpal, 1392–1481), in his Blue Annals (deb ther sngon po), mentions the ten dreams at the beginning of his account of the division of the Buddhist community into eighteen schools after the death of the Buddha Śākyamuni, during the reign of King Aśoka. In particular, Gö Lotsāwa highlights the passage in The Exemplary Tale of Sumāgadhā concerning the dream of the cloth remaining untorn though being pulled by eighteen men, explaining that although the teaching of the Buddha was divided, each of the eighteen schools provided a viable path to liberation. According to Noriyuki Kudo, King Kṛkin’s story was likely added to the story of Sumāgadhā at the end of the second century, shortly after the story of Sumāgadhā and the conversion of the non-Buddhists had spread in Gandhāra.

i.4Tracing the origins of this sūtra is a complex matter and has been studied in painstaking detail. It will have to suffice here to offer a condensed summary of our understanding of the history of the text to date as it relates to the Tibetan translation found in the Kangyur. In short, there are several fragmentary Sanskrit manuscripts from Gilgit that predate the Tibetan translation, probably by three or four centuries, and present a text that appears quite similar to the hypothetical manuscript used for the Tibetan translation. Other Sanskrit manuscripts from Nepal postdate the Tibetan translation and include additional material possibly culled from other sources such as the Divyāvadāna.

i.5The sūtra was first studied by Tsurumatsu Tokiwai in a dissertation at the University of Strasburg published in 1918. This was followed in 1968 by Yukata Iwamoto’s edition of several Sanskrit manuscripts from Nepal dated to the seventeenth century, which he compared with a thirteenth-century “Calcutta” manuscript. This publication also included English translations of four Chinese versions. In 1993, Markus Görtz completed an (unpublished) MA thesis containing a new edition of the Sanskrit, which was not accessible to us.

i.6More recently, in 2011, a new color facsimile of the Gilgit Manuscripts discovered in 1931 was published, allowing for the identification of three Sanskrit fragments of The Exemplary Tale of Sumāgadhā among them. Kudo estimates these manuscripts to date to the sixth or seventh century at the latest. With three manuscripts (A, B, and C) of the Delhi collection, and fragments from the Srinagar collection, Kudo was able to reconstruct in almost its entirety a Sanskrit source that appears to be similar to that which would have been used for the Tibetan translation.

i.7According to Kudo’s account, the story of Sumāgadhā and the conversion of the non-Buddhists circulated in the Gandhāra region in the second century ᴄᴇ, and by the end of that century had been compiled, along with the story of King Kṛkin’s dreams, as The Exemplary Tale of Sumāgadhā. One of these manuscripts presumably made its way to Kashmir and then to Tibet. Much later we have the seventeenth- or eighteenth-century “Nepalese” manuscript, to which material‍—which Kudo suggests may date to the first half of the third century and includes material that also appears in the Divyāvadāna ‍— had been added. This Nepalese manuscript was edited by Iwamoto along with the thirteenth-century “Calcutta” manuscript. We also have Kṣemendra’s poetic retelling from the eleventh century, which, like the Tibetan, was probably based on a Kashmiri version.

i.8In Pali literature, too, particularly in the Pali commentary on the Dhammapada, we find the story of Cullā Subhaddā, the virtuous daughter of Anāthapiṇḍika, which shares several features of the story found in the Sanskrit and Tibetan versions of The Exemplary Tale of Sumāgadhā. The story of Visākhā from the same commentary also narrates a story about Subhaddā, but this version lacks the elements found in the Sanskrit versions of the story. Iwamoto also notes a parallel version in the Manorathapūraṇī, Buddhaghosa’s commentary on the Aṅguttara Nikāya.

i.9Several versions of the sūtra are also found in the Chinese canon: San mo jie jing 三摩竭經 (Taishō 129), translated by Zhu Lü-yan in 230 ᴄᴇ; Xu mo ti nu jing Xu mo ti nu jing (Taishō 128), translated by Zhi qian in 240 ᴄᴇ; and Gei gu chang zhe nu de du yin lu jing 給孤長者女得度因綠經 (Taishō 130), translated by Dānapāla. There is also a version of the sūtra found in the Chinese translation of the Ekottara Āgama (Zeng yi a han jing 增壹阿含經, Taishō 125). Iwamoto’s study includes English translations of these four Chinese versions. He also cites a fifth version that is probably an extract from the Ekottara Āgama.

i.10There is also a version of The Exemplary Tale of Sumāgadhā found in the Mongolian Kangyur, titled Sayin magada-yin domuγ-i ögülegči kemekü.

i.11According to the colophon in the Tibetan Kangyur, the Tibetan translation of the sūtra was produced by Dharmaśrībhadra and Tsültrim Yönten (tshul khrims yon tan) and later corrected by Rinchen Sangpo (rin chen bzang po, 958–1055). Given that Dharmaśrībhadra is also said to have lived sometime during the late tenth to the mid-eleventh century, one may tentatively date the translation to the early eleventh century. In support of this dating is the fact that the sūtra is not listed in the Denkarma (ldan/lhan dkar ma) or Phangthangma (’phang thang ma) imperial catalogs from the ninth century.

i.12Since there is no single complete Sanskrit source that corresponds precisely to the Tibetan translation‍—at least one that is not based on reconstruction‍—we have translated the sūtra based primarily on the Degé edition of the Tibetan Kangyur. We have, however, consulted the Sanskrit as edited by Kudo from the Gilgit manuscripts and the edition by Iwamoto reconstructed from the later manuscripts. Through this process we found that the Tibetan aligns quite closely but by no means perfectly with what is found in the Gilgit manuscripts.