As it is well known, a significant part of the manuscripts of the Zoroastrian long liturgy are accurate descriptions of the different variants of this liturgy. They include the Avestan text of the ceremony and directions about how to recite the liturgical texts and about the actions to perform. These ritual directions, known as nērangs, are written in different languages depending on the origin and date of the manuscript, mostly in Pahlavi and Gujarati. Whereas the Avestan text of the liturgies has got extraordinary attention, the nērangs have got no attention at all in the West. They have never been edited and only Darmester has reported about their contents in his French translation of the Yasna and Vīsperad. However, Darmesteter s sources for the nērangs were not the oldest and most reliable. He used Anklesaria s edition (1888) for the Gujarati nērangs and for the Pahlavi nērangs a collation made by West of the nērangs included in the Pahlavi Yasna manuscript Pt4, although the expect source for the nerangs is, of course, the Sade manuscripts. The main reasons for this lack of attention to the nērangs are two: 1. The rituals described in the manuscripts have been considered late and no importance has been attached to them. Even when the ritual character of the Avesta collection was clear as in the case of Spiegel, the text of the ritual was considered to consist just of a late ritual assemblage of fragments of the real Avesta described in the Dēnkard and today lost. 2. Furthermore, the neerangs, especially the Pahlavi, the oldest ones, have remained largely unknown, since they appear in the Iranian Sāde manuscripts. Only few manuscripts of this type were
known to Geldner and even less were available in European libraries and only one facsimile was published (the facsimile of K7). Regarding the second point, today we have discovered many new Iranian Sāde manuscripts that include the Pahlavi nērangs and we are publishing some of them in the webpage of the Avestan Digital Archive. In the handout you can see a list of the Iranian Sāde manuscripts of the long liturgy we know today (# 1). Thus we are now in better condition than ever before for studying the Pahlavi nērangs included in the Iranian Sade manuscripts. Regarding the consideration of the ritual presented in the manuscripts as a late creation without interest, Kellens article on the transmission of the Avesta more than 10 years ago has changed definitively our perspective. But already Darmesteter (1892: 1.xcii) had tried to prove the antiquity of the ritual described in the manuscripts on a different basis than Kellens. He presented three arguments: 1. Firstly, the information provided in the nērangs finds sometimes correspondence in the scarce ritual information that appears in the Pahlavi literature. He mentions an interesting parallel between DD 48.30-32 1 and the nērangs of Y27.4-5 (s. Handout 2). Very interesting is the fact, unnoticed by Darmesteter, that there is a significant lexical agreement between the DD and the nērangs. In both texts when the Haoma and the pomegranate are squeezed, the mortar is struck (with the pestle). As already noticed by West (1882 170 n. 4), the verb used in the DD for the 1 For my part, I don t find which the exact ritual correspondance of DD48.30.
action of striking the mortar with the pestle is škābīhēd whose actual meaning is is split and not is struck. Actually, the same verb, škāftan, appears in the nērang of the manuscripts. The DD does not only describe the same ritual, as already noticed by Darmesteter, but also uses the same idioms as the manuscripts 2. 2. Secondly, Darmesteter noticed that the number of the repetitions of the Old Avestan stanzas prescribed in the 10 th fragard of Vīdēvdād agrees with the number of repetitions made in the ritual. The practice of these repetitions is then at least as old as the 10 th fragard of Wīdēwdād. 3. The third point considered by Darmesteter is the connection between the nērangs of the manuscripts and the Nērangestān. He noticed that the Nērangestān includes some nērangs that are identical to some of the manuscripts. Furthermore, since the Nērangestān includes some nērangs in Avestan language that are the Avestan version of some Pahlavi nērangs, he concluded that there is continuity between the Pahlavi and Gujarati nērangs of the manuscripts and the Nērangestān and its lost Avestan original. Despite the three points made by Darmesteter are incontestable (although with some restrictions), they have remained totally unnoticed in the discussion about the nature and transmission of the ritual Avesta. In following, I m going to give additional arguments that confirm the close connection between the Pahlavi nērangs of the Iranian Sāde manuscripts and the Nērangestān. 2 Furthermore, Darmester didn t notice that also DD48.33 refers to a well known ritual action in the manuscripts of the long liturgy: the division of the consecrated milk in three 2 parts exactly like the nērang of VrS3.13 and N28.44 describe.
Darmesteter was indeed right when noticing that some nērangs of the Nērangestān describe identical actions as the ones in the manuscripts. Let me quote just one of the many examples. During the recitation of 4 Ahuna Vairiio in Y27.3 hōm is pounded (please from now on, check the comparison table of the handout for each passage I mention). The action described in the Nērangestān and in the manuscripts is exactly the same: The zōt pounds the hōm at the recitation of three words of the Ahuna Vairiio: s íiaōθananąm, aŋhəuš and mazdāi. By the first recitation it is pounded once each time, by the second twice and by the third, three times. At the recitation of xšaθrəmcā of the third Ahuna Vairiio, the pestle is to be raised to the height of the ears. The Nērangestān quotes an Avestan passage that contains the same direction: the pestle is to be raised to the height of the ears. This passage confirms the similarity between the nērang and the Nērangestān and provides a good example of the existence of an Avestan version of some Pahlavi nērangs. But there are further signs of the close relationship between the nērangs in the manuscripts and the Nērangestān. There is in the chapter 28 of the 2 nd fragard of the Nērangestān a description in Pahlavi of the ceremony celebrated during the ušahin gāh. The description is not complete: it begins with the paragna and describes the ceremony with detail till the beginning of the recitation of the Old Avestan texts (Y27.9) and finishes then with a description of the Āb-zōhr (Y66). The description of the ceremony from Y0 to Y27.9 goes in the new edition of Kotwal-Kreyenbroek from N28.31 to N28.47. A comparison of the description offered in the Nērangestān with the nērangs of the manuscripts shows that
both descriptions are based in the same tradition. It is even most likely that the Nērangestān s description is based in a similar description as we find it in the manuscripts. In the hand-out you find a table in which the Nēangestān is compared with the nērangs of three Iranian Sāde manuscripts: F3a for the Yasna, G18b for the Wīsperad and Ave977 for the Wīdēwdād. In the first column you see the complete text of N28.31-47 in its original order. In the other three columns all nērang that appear in these three manuscripts between Y0 and Y27.9 except the short nērangs indicating which priest have to recite certain texts. The first fact to notice is the high degree of agreement between the Nērangestān and the nērangs. They describe the same ceremony (with some very small variations): the Avestan texts are the same, the order is the same and the ritual actions are the same 3. The ritual shows some innovations in the manuscripts like the substitution of the different assistant priests like hāwan, frabardār, etc., by the raspīg who goes to the place of the corresponding priest and assumes his function 4. Furthermore, there are minimal changes. Just an example: whereas according to the manuscripts the main priest recites the two final Ahuna Vairiia of Y0.2 on the way to the zōtgāh and then enters it and recites there standing Y0.4, in the Nērangestān he recites too the two 3 As for the text, Kotwal-Kreyenbroek suppose that there is an Avestan text that a significant variation regarding the version extant in the manuscripts. In N28.40 appears the text nəmō. haomāi. mazdaδātāi. vaŋhuš. haomō. huδātō. They identify it with Y9.16 and suppose that it appears an additional mazdaδātāi in the Nērangestān. In fact it corresponds to Y11.8, the concluding text of the Hōm Stōd. 4 However, even in the manuscripts we find sometimes direct mention of other auxiliary priest. Thus F3a lets the frabardār place the parāhom into the right hand of the zōt during the recitation of Y11.10
Ahuna Vairiia on the way, but he does not enter the zōtgāh till staomī as əm (Y0.6). Before comparing the nērangs and the Nērangestān, we must be aware that, although there is a big homogeneity between the nērangs in the different manuscripts, nevertheless there are differences between them, even when they are supposed to go back to the same source. Not all the manuscripts offer the same quantity of information at each passage. Sometimes some nērangs do not appear in certain manuscripts, although mostly the nērangs appear in all of them 5, as a look to the table allows seeing easily 6. Besides, the contents of the nērangs are quite similar. The differences are minor and concern mostly a somehow different formulation of the same idea or eventually the addition or omission of minor details. The description of the ceremony of ušahina in N28 takes a similar position as a Sāde manuscript: there is high degree of agreement with the nērangs, although in this case the differences, especially in the formulation, are a bit bigger than among the different manuscripts. To most nērangs of the manuscripts corresponds indeed an indication in the Nērangestān and vice versa. In fact, I have found nērangs in 133 passages of the long liturgy between Y0 and Y27.9. Only in 18 passages there are not corresponding indications in the description of the Nērangestān. Often the 5 Just one example: the nērang indicating that at each recitation of the s íiaoθananąm of the four Ahuna Vairiio at Y13.7 the knot of the barsom must be moved appear identical in all manuscripts consulted, that is, the VS manuscripts of the Marzban family (Ave976, ave977, RSPA 230), in the YS F3a (a copy of Mf1) and in the VrS G18b. It is missing, however, in K4 that goes back to the same source of G18b. Actually G18b adds the indication that each time incense must be given to the fire too. 6 Thus in Y11.10 all manuscripts prescribe that the raspīg places the parāhōm in the right hand of the zōt, but only F3a indicates that the zōt has to hold his hand at the beginning of the barsom. The same nērang as in F3a appears in the N28.43.
omissions are simple indications like the prescription to offer sandal and incense to the fire (Y3.20, 21, 4.1, 24, 7.25, 8.1, Y10.5, Y11.11, Y13.7, Y15.1, Y22.8, Y24.12, Y26.11). Such trivial indications are actually sometimes missing in the manuscripts too 7. Notwithstanding, other more significant directions are missing too in the Nērangestān like the indication that during the recitation of Y15.2 the zōt has to put the frāgām twigs of the barsom and his hand on the knot of the barsom. The differences between the Nērangestān and the nērangs in the manuscripts regarding the number of nērangs and their positions are, however, comparable to the differences between different manuscripts. The only significant deviation is the presence in the Nērangestān of some digressions and discussions that usually are missing in the manuscripts as it is advisable in these practical scripts. Furthermore, the nērangs appear in the Nērangestān not only at the same position as in the manuscripts, but they are too largely the same. Actually, the phrasing is more similar in the different manuscripts than in comparison with the Nērangestān. Have a look, for example, to the directions for the beginning of the homāst (Y22.1). Evidently the directions are the same, but whereas the wording is identical in the three manuscripts, in the Nērangestān is a bit different, whenever sharing the same phraseology. A good example of the complex relationship between the nērangs of the different manuscripts and of the Nērangestān is provided by the beginning of the Srōš Drōn (Y3.1). The VS manuscripts of the 7 Thus in Y3.20 and 21,4.1 and 4.22 the indication appears in Ave977, but neither in F3a nor in G18b. In Y13.7 it appears in G18b, but not in the others.
Marzbān family do not include a nērang there, but F3a and G18b do as the Nērangestān does too. The beginning of the nērang is identical in F3a and G18b: dlwn W *gwšwdʾk ʾsm W bwd 3 tʾk HNHTWNtn drōn ud gōšudag ēsm ud bōy sē tāg be nihādan/be nihišn the priest has to place the drōn, the gōšudag and three pieces of sandal and incense They share even the same misspelling for gōšudag. But in the following, G18b agrees in the details with N28.36. Both mention that there is no need of frasast a kind of sacred bread and both use the same formulation for the direction of having a look to all elements that are going to be used in the ceremoy of the Srōš Drōn: AP-š PWN *ʾsyšn 8 KRA MNDWM-1 BRA nkylšn u-š pad āsāyišn har ciš-ē be nigerišn He should examine everything calmly where both have the same misspelling for āsāyišn. In F3a we find instead a different formluation which is more frequent in the manuscripts: hmʾk gywʾk sʾcšn KRA MNDWM-1 hwp BRA nkylšn hamāg gyāg sāzišn har ciš-ē xūb be nigerišn He should examine well all the preparatives there The Nērangestān inserts there a digression about mistakes in the recitation that clearly interrupts the nērang in the middle, but after this digression it continues exactly in the same way like G18b: 8 N28.36 adds HD
OD slwšdlwn OŠTEN -tn wls W *hwmyck <MN> blswm KN pʾhlycšn AMT-s Q DM YAMYTWN-yt ʾ-šʾyt tā srōš-drōn xwardan wars ud hōmīzag az barsom ōh pahrēzišn ka abar rasēd ā-šāyēd Till the partaking of the drōn, he should keep away the wars and the hōm from the barsom, (but) if they do come in contact, it is permissible. The formulation in F3a is again different and includes also the mention of urwarām. Such examples show the complicated interconnections between the nērangs of the Iranian Sāde manuscripts and the Nērangestān. But there is a connection that is clearly closer than the rest. This is the connection between F3a, a copy of Mf1, and N28. This connection is even more surprising, for F3a is a Yasna manuscript, whereas the ceremony described in the Nērangestān is a Wīsperad of ušahin. Nevertheless, there are many instances in which the nērangs of F3a are identical to the direction in the Nērangestān, at least for some parts. In the comparison table I have marked all passages where there are striking similarities between both descriptions. Now I m going to show you only the most relevant aspects of this connection, since I do not have time for a detailed comparison. The directions for Y15.1 are almost identical in F3a and N28.45. But the most interesting thing is that F3a and N28.45 share not only the same text, but also a peculiar orthography for the verb to raise. This verb appears in MacKenzie as abrāstan, abrāz-, with the present stem written with final c. This writing appears in this
passage in G18b and Ave977. The Nērangestān shows the three times that this verb appears in this passage always a final d instead a final c. Kotwal-Kreyenbroek correct this form all three times to abrāzišn. But F3a show also the same reading. Obviously this is not a mistake of both scribes, but represents the true SWIr form, while the writing attested in G18b and Ave977 corresponds to the NWIr form. Very interesting is Y27.4. We have already mentioned this passage because of his parallel in the DD. F3a shows the same nērang as the manuscripts G18b and Ave977: during the recitation four times of mazdā at mōi the hōm and the pomegranate has to be squeezed and the pestle struck against the mortar. This direction is missing in the Nērangestān, but instead we find another nērang that appears too exactly with same wording in F3a. This passage is very difficult and Kotwal-Kreyenbroek consider it faulty. They even believe that the author of the Nērangestān has recognized the faulty character of the passage and has marked it with the word ahōg error at the end, a procedure without parallels in the Pahlavi literature. In the Table you can see his edition of the Nērangestān passage and in #4 of the handout their translation: <If he pounds his own barsom, the zōt must pound (?) the fire, and he should do it in such a way that it (the pestle?) reaches all three [[including the hōm ud urwarām]]. Error. If Kotwal-Kreyenbroek s interpretation were right, it would have enormous consequences for our understanding of the history of the transmission of the Avesta. It would mean that the nērang of the Sāde manuscripts and the Nērangestān share the same written
sources or at least that the author of the nērang of F3a (that is, Mf1) used our manuscripts of the Nērangestān as source. Nevertheless, although the passage remains difficult, Kotwal- Kreyenbroek s interpretation is not the only possible. Since the last work is written in both manuscripts of the Nērangestān and in F3a as ēk <ʾywk > one and not as ahōg, I prefer to keep this reading. A possible interpretation of the passage could be: If he strikes his own barsom, then the zōt should strike too the fire. He should act so that he reaches the three (that is, hōm, barsom and fire). One is just for hōm and pomegranate. The passage remains in any case obscure and the agreement between the Nērangestan and F3a is even therefore more significant. Still more interesting are the nērangs in Y24.1 (please, see it in the comparison table). The beginning in F3a is again the same as in G18b and Ave977, sharing even some details with Ave977: with the words ahurāi mazdāi the priest should raise the cup with the hōm and pomegranate and hold it towards the cup with the consecrated milk, then N28.46 and F3a continue with an identical nērang which has been composed in the style of the exegetical literature: Z NE AYK cygwn W cnd YHSNN -šn ʾ-m LA lwšnk. dʾtwyh gwpt HWE-t gwš bʾlʾ YHSNN-šn prʾc OL pyš YHSNN-šn AP-š PWN BRA nkylšn ahurāi. mazdāi. haomą. āuuaēδaiiamahī. ēn kū ciyōn ud cand dārišn ām nē rōšnag dādweh guft hād gōš bālāy dārišn frāz ō pēš dārišn u-š <pad> be nigerišn.
It is not clear to me how and how much he should hold (the cup with the hōm and the pomegranate). Dādweh said that he should hold it to the height of his ears, hold it to the front and gaze at it Observe that both F3a and the Nērangestān repeat the statement in the first person: it is not clear to me. Both texts continue obviously the same tradition. Furthermore, the divergence between the author of the direction and the opinion of Dādweh shows that the composition of the nērangs was subject of discussion in a similar way as the interpretation of the Avestan texts. Only little material has come to us of these discussions: on one side, the Nērangestān and on the other, the Sāde manuscripts. The latter, conceived just as practical guides for the priestly instruction, dispense usually with the discussion and give just the right prescription. They pretend indeed to be authoritative guides for the ritual practice and not reflex of ritual controversies. In any case, the connection between some nērangs of F3a with the Nērangestān is closer than the connection between the nērangs of other manuscripts and the Nērangestān. At the present I m not able to choose between the numerous possible explanations for this relationship. Did the scribe of Mf1, the source of F3a, made use of the Nērangestān for the composition of his nērangs in the 18 th C.? We cannot exclude it, but the fact that the indications of the Nērangestān are recorded in F3a only occasionally and that agreements between the Nērangestān and the nērangs appear also
in other manuscripts, although in a minor degree, seems to indicate that we have to look for alternative explanations. Darmesteter recognized that the long liturgy as it appears in the manuscripts is not a late invention made up with the extant fragments of the lost Great Avesta, but a ceremony that was performed this way already in the Sassanian times and probably earlier. Kellens has shown recently that the structure of the long liturgy is reflected already in the Yašts. Darmesteter s assumption goes, however, further. Even the minimal details of the ceremony are the same in the Nērangestān and in the Pahlavi literature and in the description of the manuscripts. Since at least some of these details appear described in the Avestan nērangs included in the Nērangestān, this would imply that the ceremony had its actual shape already at times in which Avestan texts were still composed. Actually, the Avestan nērangs preserved are few and we don t know in which extent the Avestan ceremony did agree with the details of the ceremony described in the manuscripts. About the shape of the pre-sassanian long liturgy we can only state that it had the same structure and that it shares even some details with the long liturgy as it appears in the manuscripts. But for the Sassanian times we can state that the ceremony was essentially the same that appears in the manuscripts. Two are the most significant differences: the animal sacrifice described in the Nerangestan has disappeared from the long liturgy and usually one auxiliary priest, the raspīg, has assumed the function of the several auxiliary priests mentioned in tjhe Nerangestan and in the lityurgical texts of the Avesta. There are additionally changes in some details, but they are mostly minimal, especially if we
consider the gap of around thousand years between both descriptions. Besides, the descriptions of the ceremonies we find in the Nērangestān, the manuscripts and for some details in the Pahlavi literature are not independent from each other but continue the same tradition. The same parts of the ceremony are described in the different descriptions and sometimes even the wording is the same. It is obvious that the description of the ceremony of ušahina in N28 follows a description of the ceremony similar to the description we find almost thousand years later in the manuscripts. We don t know if such descriptions had been already written down at the time of the composition of the Nērangestān or were transmitted just orally. The small variations of the nērang even in manuscripts going back to the same source reveal indeed an important influence of the orality still in the 17 th C. The Sāde manuscripts appear thus under a new light. They are the primary sources of the Avestan ritual and they continue a tradition that goes back at least to the Sassanian times and might have its roots even in an Avestan metaritual literature describing the ritual actions that accompany the ritual text. Anklesaria, T. D. (1888). Avesta. The Sacred Book of the Parsis, Part 1. Yasna bâ Nîrang. Bombay, Fort Printing Press.
Darmesteter, J. (1892). Le Zend-Avesta. Paris. West, E. W. (1882 ). Pahlavi Texts. Part II. Oxford, Oxford University Press.