Navigation – Plan du site

The Zoroastrian Long Liturgy

Alberto Cantera
p. 51

Texte intégral



Manuscript 4055, folio 157v, beginning of the Widwdd 9

3The long liturgy is the archetypal ritual of the Mazdean community and is very likely the foundation of its identity.

4This complex ceremony has many variants: Yasna, Yašt ī Wisperad, Widēwdād or even Wištāsp Yašt. Yet only the daily liturgy has been published in full; the others have been published only in parts. Unfortunately, the latter do not allow us to reconstruct the liturgy in all of its variety as it was celebrated from the thirteenth to the seventeenth century. Moreover, the liturgical manuscripts include ritual instructions that have never been published (e.g. in Pehlevi, Pāzand or Gujarati), even though they are essential for understanding the ceremony.

5These gaps can be explained by looking into the history of Avestan texts and studies, which developed from the second half of the nineteenth century. Spiegel and Westergaard, who were the first editors of the Avesta, did not consider the Avestan texts to be liturgical texts. The original manuscripts were therefore considered to be exegetical manuscripts that included a Pehlevi translation. Their analysis overturned the original situation, as it showed that the original manuscripts were in fact the liturgical texts from which the exegetical excerpts had come. The only complete text the exegetical texts present is the one for the daily liturgy. As for the rest of the variants, the exegetical texts include only the sections that needed a translation because they did not appear in the daily liturgy. These are the fragments which are included in modern editions.

6According to late nineteenth-century authors, such as Darmesteter and Geldner, the situation was more complicated: when the Pehlevi texts describing the Great Sasanian Avesta were published, they brought to light the fact that the Avestan texts that we have today are not identical to the Great Sasanian Avesta. The extant texts were recognized to be the recitatives of a liturgy that already existed during the Sasanian era. However, this discovery had no bearing on Geldner’s edition, as it came about when the latter was virtually completed. During the twentieth century, until Jean Kellens’ 1998 article in the Journal Asiatique, this fact was ignored and the ritual use of the Avestan texts of the long liturgy was considered to be of secondary interest. Nevertheless, the antiquity of the long liturgy has been established.

7Kellens has recently shown that the conceptual backbone of the long liturgy was already present in the Yašt or the Hādōxt Nask. An analysis of ritual instructions in the Avestan language of the Nērangestān has allowed us to show that the Avestan recitative of the long liturgy existed prior to the Sasanian era, at a time when some were still capable of writing texts in Younger Avestan. Moreover, the intercalation ceremonies also belonged to the older time period, and they developed a central aspect of the long liturgy: the sacrifice and the following journey of the priest’s soul in the hereafter allowed it to meet with the divinity and afforded it the means to communicate the long liturgy to the sacrificial community.

8The long liturgy’s recitative is not a late collection of the texts that were saved from the disintegration of the Great Avesta. This liturgy was a living ceremony which existed in a very similar form to that represented in the manuscripts long before the Sasanian era, but which continued to evolve since the time when it was composed and first set in writing, in all probability during the Sasanian Empire or shortly thereafter. It is not only the ritual that has evolved, but also the linguistic form of the Avestan recitative. It has reached us in a south-western variant which was fixed a few centuries after the beginning of the Sasanian era. After that, however, the recitation of the Avestan texts continued to change: there are clear differences between the Iranian and the Indian recitations. The manuscripts, which were designed as guides for learning how to perform the ritual correctly, are not the more or less exact copies of a Sasanian or later original, but reflect the variants of the long liturgy as they were celebrated at the time that the manuscripts were produced. Today, editors have to choose the temporal horizon of their edition and simultaneously show the synchronic and diachronic variety of rituals in the long liturgy.

• Alberto Cantera was invited by the Faculty upon Prof. Jean Kellens’ proposal.

• The videos of the lectures are available on the Collège de France’s website ( on Prof. Jean Kellens’ page.

• Over the past few years, Mr Alberto Cantera has digitalized over 150 Avestan manuscripts, a fair number of which have been published on his website: Avestan Digital Archive (

9La Lettre du Collège de France 37 • November 2013

Haut de page

Table des illustrations

Légende Manuscript 4055, folio 157v, beginning of the Widwdd 9
Fichier image/png, 1,4M
Haut de page

Pour citer cet article

Référence électronique

Alberto Cantera, « The Zoroastrian Long Liturgy », La lettre du Collège de France [En ligne], 8 | mars 2014, mis en ligne le 12 août 2015, consulté le 23 mars 2017. URL :

Haut de page


Alberto Cantera

Professor at the University of Salamanca (Spain)

Articles du même auteur

Haut de page

Droits d’auteur

© Collège de France

Haut de page