Śrāvakabhūmi

A working bilingual edition of the Śrāvakabhūmi, in Sanskrit (Shukla 1973) and English.

Introduction

Śrāvakabhūmi means "The Stage of the Disciple" — from śrāvaka, literally "hearer," the term for those who heard the Buddha's teaching directly and trained for liberation as arhats, and bhūmi, "ground" or "stage." It is a section of the Yogācārabhūmi-Śāstra, the encyclopedic treatise on yogic practice traditionally ascribed to Asaṅga (4th century CE). Organized into four yogasthānas ("yogic foundations"), it lays out a detailed program of contemplative training: the nine stages of mental abiding (śamatha), the method of insight (vipaśyanā), the five temperament-matched practices (unattractiveness, lovingkindness, dependent origination, elements, breath), and the seven attentions of the mundane path.

The Sanskrit original was long thought lost in India until 1938, when the Indian scholar Rāhula Sāṅkṛtyāyana located and photographed a palm-leaf manuscript at the Ri phug hermitage of Zhalu (Źa lu) monastery in Tibet. Shukla's edition rests on those photographs; the manuscript itself was reproduced in facsimile in 1994, and that facsimile underlies the later critical editions.

Contents

This is an AI-assisted working translation. The English was drafted with a large language model from Shukla's Sanskrit, cross-checked by a second model, and then reviewed section by section, also with AI assistance, against the Sanskrit, Xuanzang's Chinese translation (T1579), and, where available, the critical editions of Deleanu (2006) and the Taishō Śrāvakabhūmi Study Group. Ryan Forsyth, who does not read Sanskrit, directed the work, set the editorial policies, and is responsible for the result. It is not a scholarly critical edition. Reports of errors or improvements welcome (see Contributing below).

Edition details

Sanskrit source
Karunesha Shukla, Śrāvakabhūmi of Ācārya Asaṅga (Patna: K.P. Jayaswal Research Institute, 1973; Tibetan Sanskrit Works Series 14), in the transcription hosted on SuttaCentral.
Translator
Ryan Forsyth (see the note above).
Format
Bilara-compatible segmented JSON (sources) rendered as side-by-side HTML.
License
CC0 1.0 Universal for the English and apparatus. The Sanskrit is Shukla's 1973 edition, in SuttaCentral's transcription; the dedication does not cover it, and no license is claimed for it here.

Textual status

The Sanskrit layer is based on Shukla 1973 (via SuttaCentral) but is not a fully diplomatic transcription. Obvious PDF/OCR extraction artifacts — chiefly intra-word whitespace breaks — have been normalized; the policy is in the emendation policy. Where the English departs from Shukla's printed text it says so inline ([Lacuna], [Damaged], [Conjectural: …], [Restored: …]) with a one-line note naming the witness; the markers are defined in the editorial conventions. For close philological argument, cite the segment ID and check the Sanskrit against Shukla and, where available, against the Tibetan and Chinese witnesses.

Known limitations

No Sanskritist has yet reviewed this translation, and its director does not read Sanskrit. Every section has been read in full, segment by segment, against Shukla's Sanskrit and Xuanzang's Chinese; sbh4.1 and sbh4.2, and parts of sbh3.2 and sbh4.3, were also collated against the critical editions of Deleanu (2006) and the Taishō Śrāvakabhūmi Study Group, which are the only ones in hand. The Sanskrit transcription was collated against Shukla's printed pages, and the places where it had dropped or duplicated text were brought back to the print. Shukla's text also has gaps, the largest about ten pages inside sbh3.2:55.16, which are marked and not filled. The section-by-section account, the critical-edition coverage, the sample results, and the gaps are set out under Known limitations in the README.

Contributing

Errors, alternative readings, and improvements are welcome via GitHub issues or pull requests on the source repository.