The Heart of the Perfection of Wisdom, the Blessed Mother

The short version of The Heart Sūtra was first compiled‍—most likely as an abstract of longer Prajñāpāramitā sūtras‍—in China sometime in the middle of the seventh century ᴄᴇ.4 Shortly thereafter, it appears that the sūtra was translated from Chinese into Sanskrit. The material used to compile the sūtra was extracted from key passages in Prajñāpāramitā scriptures in Chinese translation, especially Kumārajīva’s (344–413 ᴄᴇ) translation of The Perfection of Wisdom in Twenty-Five Thousand Lines (Taishō 223). These central statements were then framed by an introduction describing the bodhisattva Avalokiteśvara’s role as the teacher, and then a concluding mantra, which functioned to encapsulate the blessings of the Prajñāpāramitā teachings.5 In this way, the short version of The Heart Sūtra combines established Buddhist Prajñāpāramitā scripture with dhāraṇī practice in an abridged format that is eminently suited for recitation. Notably, at that time in China such scriptural digests summarizing the words of the Buddha (buddhavacana) were common, and The Heart Sūtra is certainly not unique in this way.6 What is remarkable, however, is the immense importance and popularity the short version of The Heart Sūtra came to enjoy in China and the rest of East Asia. When viewed from the perspective of its roots in Prajñāpāramitā literature and its transformative and liberating efficacy in the perception of its many devotees, The Heart Sūtra does fulfill important criteria for being a genuine Buddhist scripture. For example, the Mahāyāna sūtra Inspiring Determination (Adhyāśayasañcodana, Toh 69) defines authentic buddhavacana as any discourse7 that fulfils the following four criteria: (1) it must be meaningful, (2) it has to be consistent with the Dharma, (3) it should reduce mental defilements, and (4) it should present the qualities of nirvāṇa as opposed to saṃsāra.8 In this sense The Heart Sūtra is perhaps not so different from many other Mahāyāna scriptures, in particular those of the Prajñāpāramitā family, that likewise were transmitted across a variety of time periods, regions, cultures, and languages‍—constantly in flux, yet ever encapsulating the transcendent aspects of authentic buddhavacana.

Traditionally, the earliest version of The Heart Sūtra in Chinese was thought to be Taishō 250, which in traditional sources is described as a translation of Kumārajīva. However, this attribution is generally no longer accepted, and some scholars have placed it later in time‍—some even in the beginning of the eighth century ᴄᴇ.9 Apart from Taishō 250, the earliest witness for The Heart Sūtra’s existence is a famous stone stele, which was discovered in the early twentieth century at the Yunju Temple in the Fangshan region near Beijing. The inscription on this stele is a rendering of the short version of The Heart Sūtra, and the colophon is dated to 661. The colophon further states that this is the translation prepared by Xuanzang (i.e., an equivalent of Taishō 251).10 Of the three shorter versions of the text in Chinese,11 this is the one that has been commonly read and recited across East Asia down to the present day.

In terms of Sanskrit sources, the first rendering of the short sūtra in Sanskrit was likely produced shortly after the compilation of The Heart Sūtra in Chinese in the middle of the seventh century. However, the earliest witness we have of a Sanskrit text is not an actual Sanskrit manuscript, but rather a version with the Sanskrit transliterated into Chinese characters (Taishō 256). This text was likely produced by the famous translator and teacher Amoghavajra (705–74) and can therefore be dated to the eighth century.12 While multiple Sanskrit manuscripts of the shorter version of The Heart Sūtra exist, any dating of such manuscripts is typically fraught with much uncertainty.13 Scholars have, however, dated a Sanskrit manuscript of the short version (currently kept in Tokyo National Museum) to the ninth or tenth century, and this appears to be the earliest extant Sanskrit manuscript.14 Tibetan translations of the short sūtra were never included in any of the Kangyurs, but several have been identified among the Dunhuang manuscripts.15

In sum, we can conclude that the Chinese version of the short sūtra must have been compiled sometime between 404 (when Kumārajīva completed his translation of Taishō 223) and 661 ᴄᴇ (the date on the stone stele), and that the Sanskrit translation was produced soon thereafter.16 However, in addition to these mutually distant historical markers, we also have data that place the likely time of compilation much more precisely within just a two-year period from 654–56 ᴄᴇ. First, 654 ᴄᴇ marks the year when the Dhāraṇī­samuccaya (Taishō 901), which is the likely source of the mantra in The Heart Sūtra, was translated into Chinese.17 Second, in Xuanzang’s biography (Taishō 2053, written in the late seventh century) it is mentioned that in 656 Xuanzang offered a gold-lettered version of The Heart Sūtra to the imperial family as a gift. Although the biography is a later source than the stone stele, the date provided in the biography is the earliest placement of The Heart Sūtra in the historical record and this may possibly be the very occasion on which The Heart Sūtra was first compiled in Chinese.18

The history of the longer version of The Heart Sūtra presents us with an equally complex picture of textual transmission. Notably, there are at least two distinct longer versions of The Heart Sūtra. Both longer versions add to the short sūtra by presenting a more extensive, traditional introduction (beginning with the statement “Thus did I hear at one time”) and a conventional concluding section as we generally know them from the Mahāyāna sūtras. As a result, the longer versions appear much more like a typical sūtra than does the short version.

The earliest version of the long sūtra, traditionally dated to 738 ᴄᴇ, is a text that only exists in Chinese (Taishō 252). While the text presents itself as a translation from Sanskrit, it is most likely an early attempt to present the short sūtra in a traditional literary format by adding an introduction and a concluding section, which appear to have been composed directly in Chinese. This version does not seem to have gained much popularity, as its unique introduction was never replicated or copied in other sources.19

The four remaining Chinese versions of the long sūtra (Taishō 253, 254, 255, and 257) are all translations that were produced either from Sanskrit or Tibetan manuscripts.20 Although they each feature certain idiosyncratic elements, all of them clearly belong to the same family in text-historical terms. Unfortunately, there are no surviving Sanskrit manuscripts from the period when these four translations were produced, from the late eighth to the early eleventh century.21 The earliest Sanskrit witnesses we have of the longer version of The Heart Sūtra are Nepalese manuscripts that were produced several centuries later.

Just as with the Sanskrit translation of the short sūtra, we also do not know the circumstances in which the Sanskrit text of the longer version first appeared. It must have taken place, however, prior to 788 ᴄᴇ, when the first Chinese translation from Sanskrit was produced. At around the same time, the two Tibetan translations, which were subsequently included in the Kangyur (see more below), must also have been produced. We can say that with some confidence, since the translation is recorded in the Denkarma catalog of completed translations, which is dated to 812.22

In the Tengyur collections seven commentaries on the long sūtra by Indian scholars and one by a Chinese scholar are included in Tibetan translation. Some of these commentaries were composed by such well-known names in Tibetan history as Kamalaśīla (740–95), Vimalamitra (eighth/ninth c.), and Atīśa (982–1054).23 The authors of these commentaries were all active from the eighth to the eleventh century and most of them are known to have visited Tibet. Interestingly, the different commentators view The Heart Sūtra through very diverse hermeneutical lenses and interpret it variously through the systems of sūtra, tantra, and even ritual practice (sādhana). This great variety of interpretations, combined with the fact that there is no mention of The Heart Sūtra in any other source linked to India, makes it seem likely that the commentaries were all composed in Tibet during the sojourns of these Indian masters there.24

It therefore appears that the history of the long sūtra can be told largely as a Sino-Tibetan exchange of ideas and manuscripts, which unfolded during the eighth to the eleventh century. In this regard, there are strong indications that the Dunhuang region was an important center for the exchange of such ideas and manuscripts. Among the scriptural treasures discovered in the Dunhuang caves in the early twentieth century, all of which must date prior to the turn of the eleventh century when the caves at Dunhuang were sealed off, there are numerous manuscripts of The Heart Sūtra in both Chinese and Tibetan. In addition to the shorter versions mentioned above, there are a number of longer versions, and apparently even a few “hybrid” manuscripts that are neither of the standard short or long sūtras.25 Like the Chinese and Sanskrit versions, these Tibetan Dunhuang manuscripts provide a variety of readings, but so far no comparative analyses of these texts, or critical editions of the manuscripts, have been published.

In Tibet, only the longer version of The Heart Sūtra was included in the various Kangyur collections, where it has been preserved in two recensions, which Jonathan Silk has designated Recension A and Recension B.26 These two recensions represent individual manuscript traditions and appear to have been translated into Tibetan by different translators, although the exact details remain unknown.27 The two recensions vary on several significant points, such as the title of the text, the name of the absorption in which the Buddha rests, the mantra that Avalokiteśvara reveals, and the presence or absence of a translator’s colophon.

The fact that The Heart Sūtra incorporates elements of both sūtra and tantra, and therefore has been variously classified throughout its history, is evident from its placement in the different Kangyur collections, where it often appears twice in both the Tantra section and the Perfection of Wisdom section.28 Generally speaking, all Kangyurs include Recension A in their Tantra sections, whereas some Kangyurs, mostly those of the Thempangma (them spangs ma) branch, also include Recension B in the Perfection of Wisdom section.29 The Kangyurs that belong to the Tshalpa (tshal pa) branch mostly include the text only once, in the Tantra section.30 In the Degé Kangyur, however, the text appears in both the Perfection of Wisdom and the Tantra sections, but, unlike the Thempangma Kangyurs, the Degé Kangyur includes Recension A both times, thus omitting Recension B.31

Our translation is based primarily on the version of Recension A that is included in the Degé Kangyur. We have made Recension A our primary focus in order to present in English one important version of The Heart Sūtra as it has been transmitted in Tibet since the early ninth century, rather than attempting to edit this version in search of an elusive urtext. However, to give the reader a taste of the differences between the two Tibetan recensions‍—differences that the Tibetans found worthy of preservation in their Kangyurs‍—we have included in the form of notes the most significant variations between the Degé text and the version of Recension B that is included in the Perfection of Wisdom section in the Stok Palace Kangyur.32 We have also attempted, whenever possible, to let our translation be informed by the published Sanskrit editions of The Heart Sūtra, the eight commentaries on the text in the Tengyur collections, and the growing scholarly literature on The Heart Sūtra.