A Dictionary of Gāndhārī

Stefan Baums and Andrew Glass


The present publication in progress is the first attempt to compile a comprehensive dictionary of the Gāndhārī language. The editors started their work on the Dictionary in the year 2002, when they were at the University of Washington, from a digital text of the index to John Brough’s edition of the Khotan Dharmapada. Between November 2002 and February 2003, a student assistant (Tho To) programmed the first version of the database system that forms the foundation for the Dictionary as well as its sister publications, the Gāndhārī Catalog and Bibliography, and he continued improving it until his graduation in March 2005. In September 2003, Glass took up a research fellowship at Bukkyo University, and over the course of the following four years he single‐handedly programmed a new version of the database software that, with ongoing improvements, continues to be in use. During this time, Baums continued work on the Dictionary at the University of Washington, supervising the student helper and adding items to the Catalog and Bibliography. By September 2007, the database contained digital texts for the six Gāndhārī manuscripts then published and for a total of 486 inscriptions.

A two‐year grant from the National Endowment for the Humanities for the years 2007–09 made it possible for Baums and Glass to rejoin forces at the University of Washington, and they were able to employ a number of student assistants (Haley Doan, Douglas Querl, Robert Mullen, Glory Choe, Matthew Nelson, Shelly Horn and Donald Craig) to help with data‐entry, image‐formatting and programming tasks. The year 2008 saw the addition of texts for the 782 Niya documents edited in Boyer, Rapson and Senart and Burrow 1937 (digitized by Mariner Padwa for his dissertation work at Harvard University) and for 18 manuscripts from the Senior Collection (transcribed by Mark Allon of the University of Sydney). At the end of 2008, Glass took up employment with the Microsoft Corporation and Baums continued to coordinate work on the Dictionary at the University of Washington.

As it became clear that the Dictionary would require long‐term funding in order to reach completion, the editors together with Ingo Strauch of the Freie Universität Berlin began to explore the possibility of an additional base for Gāndhārī studies under the auspices of the Union of the German Academies of Sciences and Humanities. In April 2009, a first application was submitted to the Union with the support of Jens‐Uwe Hartmann of the Ludwig‐Maximilians-Universität München and Harry Falk of the Freie Universität Berlin. Another major addition to the Dictionary database occurred in the summer of 2009, when Ingo Strauch contributed his complete transliteration of the 18 manuscripts in the Bajaur Collection. In March 2010, Baums left the University of Washington to take up a research fellowship at Bukkyo University, and in September 2010 he joined the University of California, Berkeley, where he taught Sanskrit, Pali, Gāndhārī and Buddhism while continuing to work with Glass on the Dictionary, Catalog and Bibliography.

Funding for the long‐term project Buddhist Manuscripts from Gandhāra, under the auspices of the Bavarian Academy of Sciences and Humanities and based at the University of Munich, was approved in November 2011. This project contributes to the edition of the currently known Gāndhārī manuscript material (with a focus on the British Library, Bajaur and Split Collections) and provides institutional support to the Dictionary of Gāndhārī. During the first three years of the project’s runtime (2012–14), the technical infrastructure of the Dictionary is being reimplemented on the basis of the Text Encoding Initiative’s guidelines, and lexicographic coverage is being completed for all Gāndhārī texts published to date. Coverage for newly published Gāndhārī texts is added to the Dictionary as they appear. During the last seven years of its runtime (2026–32), the Buddhist Manuscripts from Gandhāra project will support the final redaction of the Dictionary as well as the compilation of a historical grammar of Gāndhārī, a history of Gandhāran literature, and a comprehensive Kharoṣṭhī paleography. The editors of the Dictionary continue their joint endeavor, with Baums coordinating the lexicographic work undertaken at the University of Munich, and Glass working with the research team at the University of Washington.

Completeness and Access

The Dictionary database contains full transliterations for almost all published and unpublished Gāndhārī texts currently known. The remaining gaps are British Library fragments 5C, 8, 11, 19, 22, 23, 24, 26 and 27; Senior fragments 1, 4B and 25; the five fragments of the Split Collection; some pottery inscriptions from Southern Uzbekistan; and some recently‐discovered Niya documents. The digital source texts of the Dictionary are based on the best published editions where such are available. Where necessary, the Dictionary editors correct and improve the published texts; all such changes are justified in footnotes to their digital versions. For the numerous unpublished manuscripts, the editors could rely on the generosity of their Gāndhārī colleagues who made available their unpublished transcriptions for early integration into the Dictionary: Ingo Strauch of the Freie Universität Berlin provided digital texts for all 18 manuscripts in the Bajaur Collection; Mark Allon of the University of Sydney for 18 unpublished manuscripts from the Senior Collection; and Timothy Lenz of the University of Washington for the second text on British Library fragment 4 and for British Library fragment 29. Transcriptions of other unpublished texts are owed to meetings of the Early Buddhist Manuscripts Project between 2001 and 2010 in which one or the other of the Dictionary editors participated and acted as notekeeper: British Library fragment 10, half of British Library fragment 15, the University of Washington Scroll, the Library of Congress Scroll, and the Bamiyan palm‐leaf fragments in the Schøyen, Hirayama and Hayashidera Collections. A third group of digital texts are original transcriptions by the Dictionary editors: the first text on British Library fragment 4, the second text on British Library fragment 13, half of British Library fragment 15, British Library fragments 17 and 20, and several inscriptions and Niya documents.

Grammatical information, Sanskrit and Pali equivalents, and English translations are currently only available for a small minority of word forms in the database, and lemmatization likewise remains a task for the future. Completeness and accuracy in these regards can only be achieved in conjunction with the ongoing editorial activity, which will continue for at least another two decades even if no further discoveries increase the currently known number of manuscripts. The Dictionary website makes all published Gāndhārī material available to the general public, both through its seach interface and in the form of full texts. Unpublished source material that is the subject of active editorial work is restricted to internal use and not available to the general public until its edition is complete. While the Dictionary itself (i.e., all lexical and grammatical information, the arrangement of the data, and the textual corrections and improvements made by the Dictionary editors) is the work of Baums and Glass, many of the Dictionary’s source texts remain under copyright by their original transcribers, editors or publishers.

Conventions and Usage

The abbreviations used for the source texts of the Dictionary follow the established conventions of the Critical Pāli Dictionary, the Abkürzungsverzeichnis zur buddhistischen Literatur in Indien und Südostasien and the Gandhāran Buddhist Texts series. They aim for brevity as well as for clarity within the range of texts covered. Because they had to take into account the established abbreviation systems, and because several new abbreviations had to be added, two different abbreviatory styles coexist: (1) Abhidh, Bps, Dhp, etc., and (2) ASP, EĀ, KaVā, etc. Complete consistency of abbreviation style has not been an aim, since it would come at the expense of breaking with established conventions and would make the Dictionary less accessible for its readers, particularly those who wish to use it in conjunction with dictionaries of other Buddhist languages.

Our knowledge of Gāndhārī literature continues to expand and change at a rapid pace, but remains incomplete. To accommodate this state of affairs the Dictionary employs three different types of abbreviations: (1) abbreviations for specific texts (e.g., Dhp, SaṅgCm), (2) abbreviations for genre classifications rather than texts (e.g., Av, MSū), and (3) abbreviations for unclassified manuscripts, inscriptions, wood documents or coin legends (e.g., CKM 89, CKI 249). Abbreviations of the first two types use superscript letters and numbers to distinguish between different manuscript instances of the same text or genre. As more literary texts are identified and new manuscripts come to light, abbreviations for existing items will be made more specific by switching from type (3) to (2) or (1) or by adding distinguishing superscripts, and new abbreviations will be added to the list. The overall abbreviation system of the Dictionary will, however, remain stable, and the resolution of individual abbreviations is provided in the digital version of the Dictionary on mouseclick. Non‐literary texts are always referred to by their stable catalog numbers. In the following overview, those published sources that are already covered in the Dictionary (whether in full or in the available samples) are given in normal font color; sources that are in the process of being edited and that will be covered in the Dictionary when they are published are given in grey.

Abhidharma (BC 9.2)
Abhidharma (BC 12)
Abhidharma (BC 14)
Abhidharma (BC 16)
Abhidharma (BC 18)
Abhidharma (BL 17)
Abhidharma (BL 20+23)
Abhidharma (BL 28)
Anavataptagāthā (BL 1)
Anavataptagāthā (RS 14)
Arapacana verses (BC 5)
Arthapada (SC 1)
Aṣṭasāhasrikā Prajñāpāramitā (SC 5)
Avadānas (BL 1.2)
Avadānas (BL 2)
Avadānas (BL 3A)
Avadānas (BL 4.2)
Avadānas (BL 12+14.2)
Avadānas (BL 16+25.2)
Avadānas (BL 21.2)
Avadānas (SC 4)
Bahubuddhasūtra (LC)
BC x
unclassified text in the Bajaur Collection
Bhadrakalpikasūtra (Bamiyan)
BL x
unclassified text in the British Library Collection
Bodhisattvapiṭakasūtra (MS 17)
coin legend
unclassified wood document
unclassified manuscript
Commentary (BL 3B.2)
Commentary (BL 10)
Commentary (UW)
Dharmapada (Khotan)
Dharmapada (BL 16+25)
Dharmapada (SC 3)
Ekottarikāgama sūtras (Bamiyan)
Ekottarikāgama sūtras (BL 12+14)
HG x
unclassified text in the Hayashidera Collection
HI x
unclassified text in the Hirayama Collection
Karmavācanā (BC 7)
Khaḍgaviṣāṇasūtra (BL 5B)
Madhyamāgama sūtra (BC 1)
Madhyamāgama sūtras (RS 1+3)
Madhyamāgama sūtra (RS 10)
Madhyamāgama sūtra (RS 12)
Manasvināgarājasūtra (BC 3)
Mahāparinirvāṇasūtra (Bamiyan)
MS x
unclassified text in the Schøyen Collection
Mahāyānasūtra (BC 2)
Mahāyānasūtra (HI 8)
NC x
unclassified text in the New Collection
Nirdeśa (BL 4)
Nirdeśa (BL 7, 9, 13, 18)
Nirdeśa (BL 13.2)
Pratyutpanna­buddha­saṃmukhāvasthita­samādhi­sūtra (NC 4)
Prātimokṣasūtra (BC 13)
Prātimokṣasūtra (BC 13.2)
Rājanīti (BC 9)
RS x
unclassified text in the Senior Collection
Saṃyuktāgama sūtras (RS 5)
Saṃyuktāgama sūtras (RS 11)
Saṃyuktāgama sūtra (RS 13)
Saṃyuktāgama sūtras (RS 17)
Saṃyuktāgama sūtra (RS 19)
Saṃyuktāgama sūtras (RS 20)
Saṃyuktāgama sūtras (RS 22)
Saṅgītisūtra commentary (BL 15)
SC x
unclassified text in the Split Collection
Sarvapuṇyasamuccayasamādhisūtra (MS 89)
Śrāmaṇyaphalasūtra (RS 2)
Stotra (BC 8)
Stotra (BC 10)
Stotra (BL 5C)
Stotra (MS 8)
Sucitti sūtra (NC 1)
Udāna (CKD 204)


The Dictionary uses the following system of text‐critical marks in its source texts:

[ ]
uncertain reading
(* )
editorial restoration of lost text
⟨* ⟩
editorial addition of omitted text
⟪ ⟫
scribal insertion
{ }
editorial deletion of redundant text
{{ }}
scribal deletion
lost part of an akṣara
illegible akṣara
lost akṣara
textual loss at left or right edge of support


The headwords of the Dictionary can be searched by means of the search box in the menubar in the upper right portion of the screen. Search strings can be entered in either the most commonly used Indological transliteration scheme or in the Kyoto‐Harvard scheme. Alternatively, double‐clicking the search box will provide a row of buttons to enable the input of special characters for those who do not have an Indological input method installed on their computers and do not with to use the Kyoto-Harvard scheme. By default, the Dictionary will display headwords that begin with the search string. More complicated patterns can be searched for by means of the following operators:

Zero or more unspecified characters (*dama). A * is implicit at the end of search strings.
One unspecified character (a_pa).
( | )
Two or more alternative substrings (a(vh|bh|s)ira).
The end of a headword.
When prefixed to a search string, headwords (‘lemmata’) are searched (L:agara). This is the default.
When prefixed to a search string, source spellings (‘forms’) rather than headwords are searched (F:akara).
When prefixed to a search string, Sanskrit etyma rather than headwords are searched (S:agāra).
When prefixed to a search string, Pali cognates rather than headwords are searched (P:agāra).


Headword, source‐spelling, Sanskrit and Pali searches also allow the use of regular expressions. The headwords of articles matching a search string are displayed on the left‐hand side of the screen. Clicking on any of the headwords displays the corresponding article in the right‐hand portion of the screen. Articles currently provide Sanskrit and Pali equivalents and English translations (where available), spelling variants (determined automatically on the basis of Sanskrit equivalents) and a list of references. Clicking on any line number in the list of references displays this line under the article. Clicking on the abbreviation for a text in the list of references opens up the complete text in a separate window or browser tab.

To increase the value of the database as a research tool for Buddhist philology, four Sanskrit dictionaries and one Pali dictionary are provided under the same search interface: Monier‐Williams’s Sanskrit‐English Dictionary, Edgerton’s Buddhist Hybrid Sanskrit Dictionary (currently only headwords), Oguibénine’s supplements to Edgerton’s dictionary, Bechert, Röhrborn and Hartmann’s Sanskrit‐Wörterbuch der buddhistischen Texte aus den Turfan‐Funden (only headwords), and the Pali Text Society’s Pali‐English Dictionary. When switching between these dictionaries by means of the menubar, the database remembers the last search term. This makes it possible to look up a word in one dictionary and then compare the definition of the same word in another dictionary without having to reenter it.


The editors have been fortunate to receive support through their employment by the following institutions: the University of Washington, the Danish Research Agency, the Society for the Promotion of Buddhism, Bukkyo University, Leiden University, the University of California, Berkeley and the University of Munich. The employment of student helpers for the initial programming and data‐entry tasks was made possible by a subvention from the Early Buddhist Manuscripts Project (2002–05) and by funding from the National Endowment for the Humanities and a private sponsor (2007–10); special thanks are due to Richard Salomon and Collett Cox for securing this funding, and to Youngie Yoon for administrating it. Boris Oguibénine and the Council of the Pali Text Society gave their kind permission to include digital versions of their lexicographic works in the Dictionary of Gāndhārī database. The establishment of the Buddhist Manuscripts from Gandhāra project and the funding that it provides for the Dictionary were made possible by the active and ongoing support of Jens‐Uwe Hartmann and Harry Falk; the editors would like to express their gratitude to them, to Ingo Strauch, and to the anonymous reviewers of the project proposal to the Union of the German Academies of Sciences and Humanities. Last not least, the editors remain indebted to their Gāndhārī colleagues around the world who so generously contribute unpublished manuscript transcriptions to the database, and whose comments as users of the Dictionary have led to countless improvements.

Munich and Seattle
22 November 2014