Navigation auf


Online Edition of the Paippalāda Recension of the Atharvaveda

About the project

General Information

The online edition presented here in principle follows the specifications established by Arlo Griffiths in his work “The Paippalādasaṃhitā of the Atharvaveda, Kāṇḍas 6 and 7. A New Edition with Translation and Commentary” (Griffiths 2009). This approach applies in particular to conventions concerning editorial policy (Griffiths 2009, XLVIII-LI), orthography and sandhi (pp. LI-LXXI), editorial signs (pp. LXXXIII-LXXXIV), and abbreviations of Sanskrit texts (pp. 453-456). The following special conventions should be noted:

  • Anusvāra is transcribed as ṁ instead of ṃ.
  • Intervocalic ch is written as cch (as in gacchati) in the text of the PS.

The data presented here consists essentially of the following formal elements:

  • Each sūkta is accompanied by a title which summarises its purpose as deducible from its content.

  • The introduction describes characteristic features of the sūkta in question. It usually gives an overview of the content and covers topics such as stylistic peculiarities, previous scholarly interpretations, the use of the sūkta in ritual (e.g. in the Kauśikasūtra), the history of its reception in later Sanskrit literature, as well as the metre and the number of stanzas (if divergent from the norm).

  • Our Re-edition of the Vedic text consistently takes into account the readings of the Kashmirian birch bark manuscript (Bhattacharya’s edition usually favours the Orissan tradition). In this way, we aim at reconstructing the common ancestor of the two branches, the so-called archetype *G (see Griffiths 2009, XLV-XLVI). Our access to photographs of a number of additional Orissan manuscripts (some of those listed in the table provided by Griffiths 2003, 337) has been a valuable resource in establishing the final form of many passages.

  • The text is provided with a metrical analysis, following established practice in Paippalāda research (see Griffiths 2009, LXXIX-LXXX).

  • Since no padapāṭha has been transmitted with the PS, we provide an artificial padapāṭha, i.e. a sandhi-free version of the Vedic text according to its interpretation by the present authors.

  • A core element of this edition is the translation. Our aim has not been to imitate the Vedic original, but to render its content in modern British English.

  • The critical apparatus is supposed to be “positive” in principle (see Griffiths 2009, LXXXII). As a result of balancing the efforts and benefits of including additional materials, the apparatus is limited to a small number of manuscripts. If we consider it desirable to cite additional readings, they are mentioned in the corresponding commentary. In the case of kāṇḍas 1 and 4, the apparatus contains only one Orissan manuscript besides those used by Bhattacharya (1997), viz. Ku (= Ku₁ from the village Kuruṁcaini, see Griffiths 2003, 354-355), the photographs of which are available at

  • We have considered the careful documentation of possible parallel texts found in other Vedic Saṁhitās to be crucial. The most common case is that there is a parallel version in the Paippalāda’s sister-śākhā, viz. the Śaunaka-Saṁhitā. These parallel versions are given with the full text. The discussion of the divergences between the parallel texts is an important part of the commentary.

  • The most comprehensive element of the edition are the stanza-by-stanza comments. They deal with text-critical issues as well as grammatical, metrical, and lexical peculiarities, and they try to clarify the meaning of the mantras by addressing aspects of the relevant cultural and historical contexts. Our comments critically discuss existing translations of the ŚS parallels and they integrate results from Vedic and Indological studies that have accumulated since the latter half of the 19th century. As the scientific reception of the Atharvaveda became largely dormant at the beginning of the 20th century, our integration of the most up-to-date scholarly research into our analysis of the PS also makes new and valuable contributions to the study of the ŚS.

Project-specific practices

  • We analyse the critically edited text on the morphological, lexical and semantic levels using a tagging software, the output of which is manually corrected and adjudicated by the team members. This data constitutes the linguistic level of our multi-layer annotation of the PS and is accessible as a lexicon on this website.
  • We detect textual parallels from Vedic and classical Sanskrit using corpus-based approaches, and evaluate their relevance for the PS. As the corpus of (digitally available) texts has grown enormously since the publication of Bloomfield’s Vedic Concordance, and modern corpus-based methods open up new ways for finding non-verbatim parallels, we expect that our project will make major contributions in this field.

  • We perform a manually validated word-semantic annotation of the Sanskrit text of each stanza using a semantic inventory derived from WordNet. This annotation forms the backbone of the content-based indexing of our edition.

About kāṇḍa 1

  • The authors of the present edition of the first kāṇḍa of the Paippalāda-Saṁhitā (PS 1) are Thomas Zehnder, Oliver Hellwig and Robert Leach (= Zehnder, Hellwig and Leach 2020).

  • PS 1 comprises 112 sūktas with a total of 483 stanzas and 1,873 pādas or pāda-like units.

  • PS 1 is divided into 22 anuvākas with five sūktas each, with the final two each having six sūktas.

  • Measured by the number of characters in the PS e-text, PS 1 makes up 6.34% of the whole Saṁhitā (approx. 51,000 of 804,000).

  • The standard length of the sūktas in PS 1 is four stanzas. This number is followed by 90 sūktas (90 × 4 = 360), while the other 22 have more than four. 14 sūktas have five stanzas (14 × 5 = 70), six have six (6 × 6 = 36), one (PS 1.86) has seven, and one (PS 1.84) has ten (360 + 70 + 36 + 7 + 10 = 483).