Introduction

i.1In the span of a human life, it can sometimes feel as if we have all the time in the world. As time goes by and one day bleeds into the next, the months and years pass away. Grave concerns over the inherent dis-ease of existence are put out of mind as more pressing matters arise and we become concerned with what appear to be more immediate goals. The Sūtra on Impermanence ( Anityatāsūtra) is a discourse that steers the listener or reader away from such notions.

i.2The Sūtra on Impermanence is a short work, which may be separated into fifteen sections. It begins with the very brief opening half of its narrative frame 1.­1. This opening frame gives little information apart from indicating that the Buddha was staying at the Jeta’s Grove, Anāthapiṇḍada’s Park, in Śrāvastī. There is no interlocutor. The Buddha addresses the monks in his presence by declaring that all conditioned states are impermanent and therefore should be rejected 1.­2, noting that “life indeed concludes with death, has its limit in death, for there is nothing that is born that will not die,” the refrain of the sūtra 1.­3. He then goes on to present various examples of types of beings, starting with wealthy people of high social status who, despite their status, will ultimately die 1.­4. He continues with further examples of beings ascending in importance: kings 1.­5, sages 1.­6, gods in the realm of desire 1.­7, gods in the form realm 1.­8, and gods in the formless realm 1.­9. All such beings will die. Moving beyond this general hierarchy within Buddhist cosmology, the Buddha continues to the three vehicles and declares that even those who have realized awakening and are free from further births, that is, arhats 1.­10, pratyekabuddhas 1.­11, and buddhas 1.­12, have bodies that will eventually come to an end. The Buddha then reiterates his refrain that there is nothing that is born that will not die, using a simile of clay pots that are created and eventually destroyed 1.­13. Following this, the Buddha recites a series of verses on the transitory nature of life designed to inspire one to soteriological pursuits 1.­14. The sūtra ends with the concluding half of the narrative frame, which is even more brief than the opening half, stating the delight of those who heard the Buddha’s discourse 1.­15.

i.3A number of Sanskrit witnesses of The Sūtra on Impermanence survive. These extant manuscripts might be classified into at least two separate transmissions. The first is what we may call the “Potala Transmission,” which consists of two Indic manuscripts copied in Dhārikā script. These were both collected into the Sanskrit manuscript library at the Potala Palace in Lhasa and are still housed there to this day. While these two manuscripts are unavailable for inspection, photostats have been made that are held by the China Tibetology Research Center (CTRC) in Beijing. To call this a transmission itself is somewhat debatable. The two manuscripts, while sharing the same script, were copied by different hands and often show divergences from one another. However, they do share similarities that are not seen in the other witnesses, which suggests the possibility of a shared transmission. Nonetheless, it is unclear when each manuscript came into the possession of the Potala or from where they were produced. Additionally, because of the extensive use of the Dhārikā script over a number of centuries, it is not possible to provide a satisfactory estimate for the dates of these two manuscripts. Both witnesses appear as component works in larger multitext sūtra manuscripts, but the exact nature of these two distinct multitext manuscripts remains unclear. They are without known titles and are not known to have circulated in South Asia or beyond, apart from the witnesses within the Potala Palace Collection.

i.4The second extant transmission of The Sūtra on Impermanence may be referred to as the “Nepalese Transmission.” This transmission consists of a number of manuscripts that have been uncovered in collections throughout the Kathmandu Valley and are now spread throughout collections in Nepal, Europe, and Japan. There are seven individual witnesses known to scholars: two witnesses in the National Archives of Nepal, one witness in a private collection in Lalitpur (Patan) in Nepal, one witness in the collection of the Société Asiatique in Paris, one witness in the collection of the Royal Asiatic Society in London, one witness in the collection of the Tōyō Bunko in Tokyo, and one witness in the collection of the Tokyo University Library. All of the known manuscript witnesses of The Sūtra on Impermanence in the Nepalese Transmission are rather late, dating from the eighteenth to the nineteenth century ᴄᴇ. All of the manuscripts we have examined were copied in variations of the so-called Nepalese akṣaras (Newari script, Pracalit, etc.) and this is doubtlessly also the case for the manuscripts that have not been checked. Without fail, each witness within the Nepalese Transmission is found as a component work within larger Dhāraṇīsaṃgraha manuscripts. The classification of The Sūtra on Impermanence as a component work of the Dhāraṇīsaṃgraha appears to have been a Nepalese innovation, as the earlier Sanskrit witnesses from the Potala Transmission and the Tibetan and Chinese translations do not classify the sūtra as having any association with dhāraṇī. It may be that the repetition of the sūtra’s refrain or the verses were the factors that caused this text to be associated with dhāraṇī collections. However, it is perhaps more likely that The Sūtra on Impermanence became associated with the Dhāraṇīsaṃgraha due in part to its short length. Dhāraṇīsaṃgraha collections are made up of shorter texts, and short sūtras are included as well as dhāraṇī texts. It should be noted that while The Sūtra on Impermanence only appears as a component work within Dhāraṇīsaṃgraha manuscripts, it is not included in all Dhāraṇīsaṃgraha manuscripts, the traditional contents of which appear to have been somewhat fluid. Of particular note within the Nepalese Transmission are the witnesses from the Royal Asiatic Society (RAS) and the Tokyo University Library (TUL). It is certain that these two manuscripts were produced by the same scribal tradition. The TUL witness was either copied from the RAS witness or, probably more likely, was copied from an intermediate witness that is no longer extant.

i.5The translation of The Sūtra on Impermanence within the Degé Kangyur contains only a very brief colophon mirroring the Sanskrit colophon. The colophon simply states that the sūtra has ended, providing no details on the translation. This brief colophon is seen in the majority of witnesses to The Sūtra on Impermanence within the Tibetan Kangyurs. However, there are three witnesses with expanded colophons: two witnesses in the Langdo (lang mdo) collection and one in the Namgyal Kangyur. These colophons state that The Sūtra on Impermanence was translated by the team of Kamalagupta (tenth–eleventh century) and Rinchen Sangpo (958–1055), who were frequent collaborators. Kamalagupta, a Kashmiri paṇḍita, was an immigrant to Tibet, and Rinchen Sangpo was a native Tibetan translator. This places the date of the translation within the tenth and eleventh centuries in the beginning of the second transmission of Tibetan translations. Beyond these colophons there are no major variations among the Tibetan versions of The Sūtra on Impermanence within the various Kangyurs. The Sūtra on Impermanence is always included in the General Sūtra Section in all Kangyurs. There is another work titled The Sūtra on Impermanence (1) (Anityatāsūtra), Toh 309, appearing directly before Toh 310, the sūtra translated here. While these two works share a title and theme, their content differs, and they are each unique works. It should be noted, however, that the opening and concluding narrative frame of Toh 309 is nearly the same as that of Toh 310. This may suggest that the two works developed in connection with one another and are possibly parallels of the same work from different Buddhist textual traditions.

i.6There is one Chinese translation by Fatian 法天 (aka Dharmadeva), Foshuo zhuxing youwei jing 佛說諸行有為經, completed in 984 ᴄᴇ and found in the Collected Sūtras (Jingji bu 經集部) section of the Taishō Tripiṭaka. The content of this Chinese translation generally agrees with that of the Sanskrit and Tibetan. However, as is so often the case with Chinese translations, sections are sometimes abbreviated and blurred together.

i.7There are no direct Pali equivalents to The Sūtra on Impermanence. There are multiple works with the same name, Aniccasutta, found in the Saṃyuttanikāya, but none of these are directly related to the sūtra translated here. While there are no direct equivalents in the Pali canon, there are multiple instances of passages and phrases that directly parallel the content of The Sūtra on Impermanence, often but not always in the Saṃyuttanikāya.

i.8There are also a number of parallel passages to be found within surviving Sanskrit Buddhist works beyond the verses shared with the Udānavarga, which themselves appear across a spectrum of texts. That the present sūtra is not found in Pali, but that its modular pieces may be found in Pali works, may be telling. This, taken with the fact that the Chinese translation is rather late and is in the Collected Sūtras section of the Taishō, not associated with any āgama, may suggest that The Sūtra on Impermanence belongs to an āgama that was not translated in its entirety into Chinese. This would possibly also suggest a (Mūla-)Sarvāstivāda āgama affiliation for this work. This may be bolstered, too, by the fact that Kamalagupta, the primary translator of the work into Tibetan, was from Kashmir, where the Mūlasarvāstivāda tradition would have been dominant during his lifetime. Another source pointing toward a (Mūla-)Sarvāstivāda affiliation is the inclusion of the verses found in the Udānavarga, which was likely a Sarvāstivāda work. Nonetheless, we should not conclude any doctrinal affiliation with certainty. It is also possible, although less plausible, that this work was translated later because it was composed later and was not included in any āgama/nikāya collection.

i.9The primary source texts used for this English translation were the Tibetan translation in the Degé Kangyur and a critical edition of the Sanskrit created from the witnesses of the Royal Asiatic Society, Tokyo University Library, and private collection of Mr. Padmajyoti Dhakhwa of Patan. The Tibetan of the Degé and the Sanskrit are generally consistent but differences between the two are noted when encountered. While there are no major differences between the Tibetan and Sanskrit, when differing readings are noted, translations of more substantial differences are provided in the notes. There is some disagreement found between the Tibetan and the Sanskrit on the number of verses at the end of the sūtra, with two recorded in the Degé and four in the Sanskrit, sharing one verse between them. All verses from both the Tibetan and Sanskrit are translated, totaling five verses.

i.10The Sanskrit edition was created for this translation and may be found in the appendices. In addition to the RAS, TUL, and PDP manuscripts that make up the edition, variant readings from the first of the two CTRC manuscripts are also always noted. The edition follows the general orthography of the three Sanskrit manuscript witnesses. Therefore, some variations in the spelling of words are not emended to conform to classical Sanskrit standards. For example, gemination is always reported. Sandhi is not always standardized because the formations used would not have been considered incorrect when the manuscripts were copied, and they can provide important information about inherent punctuation of statements.

i.11In addition to the Degé and Sanskrit critical edition, multiple Kangyurs were consulted. The Degé was checked throughout against the Peking Kangyur with substantial variants, which do not occur often, noted when present. The Choné and Stok Palace versions of the Tibetan were also consulted. The Chinese translation was also consulted throughout. Variants between the Chinese and the Tibetan and Sanskrit are occasionally noted. Instances where something appears in the Tibetan but not the Sanskrit and/or Chinese are always noted.