Footnotes on a Parthian sound change - CORE · Footnotes on a Parthian sound change Agnes Korn To cite this version: Agnes Korn. Footnotes on a Parthian sound change. Bulletin of

Footnotes on a Parthian sound change

Agnes Korn

To cite this version

Agnes Korn Footnotes on a Parthian sound change Bulletin of the School of Ori-ental and African Studies Cambridge University Press (CUP) 2013 76 (1) pp99 - 110lthttpjournalscambridgeorgactiondisplayAbstractfromPage=onlineampaid=8858903ampfulltextType=RAampfileId=S0041977X13000013gtlt101017S0041977X13000013gt lthalshs-01340647gt

HAL Id halshs-01340647

httpshalshsarchives-ouvertesfrhalshs-01340647

Submitted on 5 Oct 2016

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents whether they are pub-lished or not The documents may come fromteaching and research institutions in France orabroad or from public or private research centers

Lrsquoarchive ouverte pluridisciplinaire HAL estdestinee au depot et a la diffusion de documentsscientifiques de niveau recherche publies ou nonemanant des etablissements drsquoenseignement et derecherche francais ou etrangers des laboratoirespublics ou prives

Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives 40International License

Footnotes on a Parthian sound change1

Agnes KornUniversity of Frankfurt am Mainakornemuni-frankfurtde

To the memory of Jochem Schindler (1944ndash94)

AbstractThe treatment of Proto-Iranian θw (PIE tu) is one of the isoglosses dis-tinguishing Middle Persian from Parthian and thus important for WesternIranian dialectology The re-discussion of the Parthian development of thisconsonant cluster by Nicholas Sims-Williams presents a welcome oppor-tunity for some notes on the matter I will argue that there is someadditional evidence in favour of his suggestion that the Parthian result isnot -f- as previously assumed but a consonant cluster I will also suggesta modification of the steps that the development takes The Middle Persiandevelopment of θw as well as some related issues of historical phonologyand Pth orthography and Western Ir are likewise discussedKeywords Parthian Middle Persian Western Iranian Historicalphonology Isoglosses Iranian dialectology

1

According to the classical treatments by Tedesco (1921 199 f) and Henning(1958 96 f) Proto-Ir θw (PIE tu ) gives Middle Persian h (MP čahārlsquofourrsquo vs Avestan čaqβārō OInd catvārah MP čihil lsquofortyrsquo vs Avestančaqβarəsat- OInd catvārimśaacutet- abstract suffix MP -īh lt -iya-θwa-) but f inParthian as in čafār lsquofourrsquo čafrast lsquofortyrsquo and the abstract suffix -īf in inscrip-tional Parthian

This interpretation of the Parthian data needs to rely on explaining theadditional -t seen in -īft the variant of the abstract suffix seen in ManicheanParthian as an additional suffix (Tedesco 1921 200 suggests a derivationfrom -iya-θwa-tā-) However this approach does not offer an explanation forthe Pth verb (present stem) ltnydfʾr-gt niδfār- (past stem) ltnydfwrdgt

1 The present article is a revised version of a German paper For reasons of typographicalsimplicity θ w and y are used instead of q u and i for Proto-Iranian As per Iranologicaltradition italics represent the transcription (phonemical form) for Parthian and MiddlePersian but the transliteration (graphical form) for Sogdian Manichean MiddlePersian and Parthian are quoted from and in the form of DMD unless otherwise noted

Abbreviations Av = Avestan B = Buddhist Sogdian C = Christian Sogdian Ir =Iranian Manich =Manichean M = (Sogdian in) Manich script MP =Middle PersianOInd = Old Indic (Vedic and Sanskrit) OP = Old Persian PIE = Proto-Indo-EuropeanPth = Parthian S = (Sogdian in) Sogdian script For bibliographical abbreviations seethe references at the end of the article

I am indebted to Desmond Durkin-Meisterernst Jost Gippert Thomas JuumlgelNicholas Sims-Williams and Yutaka Yoshida for comments and discussion and to sev-eral colleagues for their contributions acknowledged in the notes

Bulletin of SOAS 76 1 (2013) 99ndash110 copy School of Oriental and African Studies 2013doi101017S0041977X13000013

niδfurd lsquoto hurry (intr and tr)rsquo which is likely to derive from ni-θwāraya- (cfOInd radictvar) and the noun ltnydfʾrgt niδfār lsquohastersquo (Henning 1958 97 n 2)2

While Henningrsquos etymology is certainly convincing his further suggestionsare less so he assumes that the word-internal result of Proto-Ir θw is Pth-f- while δf in ni-δfār- would show the result in word-initial position (forwhich there is no other example) and that δf would have been adopted fromthe (unattested) simplex δfār- This scenario is improbable not only becauseit implies the unlikely assumption that a cluster that is reduced to -f- inword-internal position would be retained word-initially but also because theparallel consonant cluster PIE du gt Proto-Ir δw is reduced to Pth b-word-initially (Sims-Williams 2004 540)

Sims-Williams (2004 540 545) thus suggests the alternative solution that δfis the regular result of θw in word-internal position For čafār lsquofourrsquo he assumesa dissimilatory loss of the dental elements of the consonant cluster ([tšaδfār] gt[tšafār]) a development that also occurred in this word in other Ir languages(eg Bactrian σοφαρο lsquofourrsquo vs regular λφ lt δf in αλφανζ- lsquoattainrsquo ltθwanǰa- abstract suffix -ιλ(α)wο Sims-Williams 2004 542) For the word-finalposition he posits a dialectal difference in the further development of -δf gt -f forinscriptional vs -ft for Manichean Parthian (Sims-Williams 2004 543 546)

2

This set of changes is so far based on one example of each but there seems to beadditional evidence confirming Sims-Williamsrsquo assumption that θw givesManich Pth -ft also implying that the abstract suffix -īft does not contain anadditional suffix

21The word ltpwrtgt lsquobridgersquo occurs in the Pth hymn cycle Angad Rōšnān VI57b3 Although this is a hapax legomenon its reading and meaning are reason-ably clear The existence of such a word in North-West Iranian is also confirmedby Gilaki purt purd and Zazaki pırd lsquobridgersquo4 Etymologically it is obviouslyrelated to Avestan pərətu- (cf Boyce 1954 194 ldquolt pərətu-rdquo) However a deri-vation from Proto-Ir prtu- would raise a problem on the phonological side

2 If the root had the shape PIE tu erH (as sometimes assumed) the past stem would beturH-to- gt θwarta- in which case Pth ltnydfwrdgt could be read niδford (thus DMD252b for the derivative ltnydfwrdggt lsquohurriedrsquo) However there are good argumentsagainst the laryngeal (EWAia I 685 de Vaan 2003 56 LIV p 655) so niδfurd lt-θwrta- lt -tu r-to- seems preferable (thus eg Ghilain 1939 74 Boyce 1977 64)

Weber (1994 111 n 11) interprets ltnydfʾr-gt as a compound related to MP dwār-lsquorun moversquo (according to Weber an Avestan borrowing) but MP dwār- differs fromthe Pth ltnydfwrdgt verb in its past stems (MP dwārist and dwārīd) Weberrsquos etymologyalso involves the problem that word-internal dw gives Pth ltdbgt δv (Sims-Williams2004 540)

3 Cf the edition of Boyce (1954 148) DMD 287a reads ldquopurt purdrdquo Boyce (1977)does not note the word

4 None of the contemporary varieties is a direct descendant of Parthian but they can hint atthe existence of otherwise unattested words and word forms in Middle West Iranian

100 A G N E S K O R N

Proto-Ir rt following a labial otherwise and expectedly5 gives Pth ltwrdgt-urd eg ltbwrdgt burd lt brta- (past stem of ltbr-gt lsquocarryrsquo) ltmwrdgt murd ltmrta- (past stem of ltmyr-gt lsquodiersquo) Proto-Ir prtu- should thus have givendaggerltpwrdgt purd

So it is worth considering whether Pth ltpwrtgt could derive from the obliquestem prθw- ie from the form that has always been seen as underlying the MPcognate puhl (prθw- gt purh gt puhl Huumlbschmann 1895 195 207 Hoffmann1986 171 181 n 20)6 The application of the change suggested bySims-Williams for Manich Parthian (see Section 1) yields prθw- gt purδf gtpurft Since a consonant cluster -rft is not permitted by Pth phonotactics7

purft could have been reduced to purt by a dissimilation vs the initial p-that is not unlike that in čafār

22A derivation of ltpwrtgt purt from purft lt prθw- suggests a parallel expla-nation for Pth ltmwrtgt murt lsquodeathrsquo8 from murft lt murδf lt mrθw-(Nicholas Sims-Williams personal communication) A dissimilatory loss of fin murft is surely as motivated as it is in purft On the other hand mrθw-would be the oblique stem of an as yet unknown Ir stem mrtu- besides theotherwise attested mrθyu- (Avestan mərəqiiu- Old Persian (uvā-)məršiyu-9OInd mrtyuacute-) but a stem mrtu- mrθw- lsquodeathrsquo is indeed reflected inSogdian mwrδw murθuacute10 This is likely to derive from the nominative andaccusative forms mrθuš and mrθum11 while a derivation from mrθyu- shouldeffect a palatalization of the vowel (Sims-Williams personal communication)Similarly the derivation of Pth ltmwrtgt from Proto-Ir mrti- suggested byHenning (1937 85) should probably give daggerltmyrdgt cf krta- gt ltkyrdgt (paststem of ltkr-gt lsquodorsquo) mrya- gt ltmyr-gt lsquodiersquo12

So far as the existence of mrθw- in Sogdian is concerned the word is foundin B pyšmwrδw13 lsquoafter deathrsquo and in the phrase M zʾδmwrδw lsquobirth-deathrsquo B zʾt

5 r gt Pth ur is the regular development in labial context (Rastorgueva and Molčanova1981 172) For Old Ir -t gt Pth d see Section 5

6 The word is likely to have had an ldquoamphidynamicrdquo paradigm PIE peacutertu- prtu -Eacute-(Hoffmann 1986 171) New Persian (classical) pul cannot come from prtu- sincethis would have given purd here as well

7 There are no Pth tautosyllabic clusters of three consonants (DMG 31123) in order toavoid them r in old sequences of rft does not yield ər but probably gives rə fromthe outset cf ltgryft gryptgt (deg)grift lt grfta- (past stem of ltgyrw-gt lsquoseizersquo)

8 Found in Angad Rōšnān VII 4a (cf Boyce 1954 154) in several copies9 On this word see Gippert 200110 In the alphabets used for the Manich (M) and Buddhist (B) Sogdian texts ltδgt is used

for δ and θ while the script of the Christian texts (C) has an extra letter ltθgt for θ11 mrθu- with generalized θ (from the oblique stem θw) is parallel to OP gāθu- from a

paradigm gātu- gāθw- (cf note 24)12 ltmyr-gt shows that the palatalizing effect of a following y is stronger than the labializing

effect of m-13 Two attestations in Benveniste 1940 (for the attestation ldquo8 52rdquo in Benveniste 1940 269

and Gharib 1995 337a read ldquo8 72rdquo) and one in the British Library Frag 6 line 5 (ratherfragmentary context) where Sims-Williams (1976 49) reads p[y](š)m(wr)δ but thereseems to be a final -w also in this attestation (p[y](š)m(wr)δ(w)) cf the photo of Or821282 on the webpage of the International Dunhuang Project (httpidpbluk)

F O O T N O T E S O N A P A R T H I A N S O U N D C H A N G E 101

(ʾt) mwrδw lsquobirth (and) deathrsquo (ie circle of reincarnation samsāra) where zʾδ-shows a change of Old Ir -t that otherwise does not occur in Sogdian SoBenveniste (1940 216) assumes a Pth origin while the original Sogdian phrasewould be ʾʾzy myry lsquobirth-deathrsquo14 Indeed Pth ltzʾdmwrdgt zādmurd is quitewell attested and Pth influence in the Sogdian Buddhist lexicon has beennoted for other words as well15 However while Pth influence in the use ofSogdian (deg)mwrδw and in the formation of zʾδmwrδw is possible the assumptionof a direct borrowing is faced with the difficulty that the attested Pth forms arein fact ltmwrtgt murt and ltzʾdmwrdgt zādmurd16 It would also be unlikely thatSogdian borrowed mwrδw from the stage of Pth murδf since one would expectPth δf to be rendered by Sogdian ltδβgt Such an output may be seen in Sogdpwtysδβ lsquoBodhisattvarsquo (besides variants such as pwtδystβ) which could owe itsltδβgt to Pth ltbwd(y)sdfgt bōdisaδf (thus Sims-Williams 2004 544 f see alsoSection 3)17

If mwrδw is thus an inherited Sogdian form Sogdian would show severalwords for lsquodeathrsquo (cf mrc B mwrtk(y)) The stem mrθu- underlying Sogdianmwrδw would derive from a paradigm mrtu- mrθw- the oblique stem ofwhich yields Pth ltmwrtgt18

3

Another item to be considered in the discussion of the Pth result of Proto-Irθw is the inscriptional Pth form ltnytprywtgt lsquohurriedrsquo corresponding toManich Pth ltnydfʾrgt etc19 It seems that the most straightforward interpretationof the lt-tp-gt is tf In this case one might consider a modification of the changesnoted in Section 1

Proto-Ir θw could have yielded Pth tf first which would be shown byinscriptional ltnytpr-gt nitfār- In lsquofourrsquo a dissimilation [tšatfār] gt [tšafār]

14 ʾʾzy myry lsquobirth-deathrsquo is found only in Benveniste (1940 56 line 1194) Gershevitch(1946 148) considers zʾδmwrδw as a loan translation Gharib (1995 453b) followsthis view which might be the reason for her reading Sogd zʾδmwrd [sic] zādmurdand pyšmwrδ(w) pišmurδ (the paragraphs referred to in Gershevitch 1954 only note(deg)mwrδw though) On the other hand she reads mwrδw murδu (Gharib 1995 221a)

15 Cf Sims-Williams (1983 139 2004 544) Sundermann 198216 The -d of the latter against the -t in ltmwrtgt can be explained by association (not only by

popular etymology) to the past stem murd perhaps additionally motivated by the final ofthe first member of the compound

17 However Yoshida (2008 344ndash53) who provides a list of variants and attestationsargues against Parthian influence in the Sogdian word for Bodhisattva

18 In Parthian other terms in this semantic field include ltʾwšgt ōš lsquodeathrsquo ltʿzgʾmgt izγāmlsquoflight exit (of the soul from the body)rsquo and Ind loanwords found in Buddhist contexts(ltmrngt maran ltprnybrʾngt parniβrān cf Sims-Williams 1983 140) MP shows marglsquodeathrsquo but nothing that would correspond to Pth ltmwrtgt Conversely marg is notfound in Parthian The MP hapax ltzydmrgyhgt (or ltzyrmrgyhgt (Sundermann 1984504) lsquo-deathrsquo is unlikely to be an error for ltzʾdmrgyhgt lsquobirth-deathrsquo (DesmondDurkin-Meisterernst personal communication) MacKenzie (apud Sundermann 1984504) considers a connection to Avestan ǰīti- lsquolifersquo Sundermann (ibid) a reading+ltzwddeggt (fast) or +ltzwrdeggt (force)

19 Paikuli inscription 21 d1 03 (cf Skjaeligrvoslash 1983I 49 II 79 f)


ltcfʾrgt would have taken place20 Word-final -tf would have undergone ametathesis to -ft in Manich Parthian and a reduction to -f in the dialect of thePth inscriptions thence the abstract suffix Manich -īft lt-yftgt inscriptional -īflt-pygt The output of Proto-Ir prθw- and mrθw- would have been reducedto ltpwrtgt purt and ltmwrtgt murt by the phonotactic ban on tautosyllabic clus-ters of three consonants (cf note 7) either at the stage of purtf and murtf or inthe metathized stage of purft and murft

The next stage would assimilate the tf to δf This would have concernedword-internal cases of tf other than lsquofourrsquo thence Manich ltnydfʾr-gt niδfār-and derivatives vs inscriptional nitfār- as well as borrowed tf which is likelyto be seen in ltbwd(y)sdfgt bōdisaδf lsquoBodhisattvarsquo and ltsdfgt saδf lsquobeing(sattva-)rsquo21

This approach appears to account for the data in an economic way and motiv-ate the dissimilation in čafār particularly well A development of word-finalθw gt tf gt -f(t) also seems to be more straightforward than θw gt δf (gt -fδ )gt -f(t) Pth θw gt tf is also quite parallel to Sogdian and Khwarezmian θw gt θf(Sims-Williams 2004 541 543) agreeing with these being ldquoclosely relatedlanguagesrdquo (Sims-Williams ibid) and Bactrian δf (gt λφ) would correspond tothe stage of Manich Pth word-internal -δf-

Alternatively we could consider an interpretation of both inscriptional lttpgtand Manich ltdfgt as θf (Jost Gippert personal communication) comparing it toAvestan fəδr- (oblique stem of pitar- lsquofatherrsquo) which is likely to reflect fθr-22

and to the development of word-internal δw gt Pth ltdbgt if this is δv as perSims-Williams (2004 540) However the assumption implies that one wouldneed to posit word-final developments of θf gt -f θf gt -ft plus -tf gt -θf forloanwords to account for ltbwd(y)sdfgt and ltsdfgt a set that is perhaps notaltogether compelling Hence a development θw gt tf gt -f(t) appears to bepreferable

4

There is another piece of evidence which is incompatible with the classical viewof the development of θw in Western Iranian MP nixwār- (Manich ltnyxwʾr- gtPahlavi ltnswbʾl-gt) lsquohurry hasten incitersquo is obviously a cognate of Pth niδfār-but čahār lsquofourrsquo and čihil lsquofortyrsquo would lead one to expect MP daggernihār-

20 -θw- gt Pth -tf- also seems to be assumed by Weber (1994 111 n 11 his only exampleltctfʾrgt lsquofourrsquo is not attested however) For word-final position Lentz (1926 253) andHuyse (2003 85 n 125) assume a development -θw- gt -ft (with dialectal variant -f) andinterpret this as a metathesis ie both also assume an intermediary stage -tfRastorgueva and Molčanova (1981 172) posit a development θw gt θf gt f for theword-internal position One could also consider a dissimilation of the dentals inčafrast lsquofortyrsquo lt čatfrast andor čafārdas lsquofourteenrsquo lt čatfārdas to which čafār ltčatfār could have been adjusted but such an explanation would only account forParthian not for the parallel developments in other Ir languages

21 Bactrian βωδοσατφο suggests that ltbwd(y)sdfgt was borrowed from a form with -tf(Sims-Williams 2004 544)

22 Cf Beekes (1988 73 86 and 235 sv ptar-) I am grateful to Michiel de Vaan for point-ing out this reference to me


In view of the discussion above and of the laconic note by Sims-Williams(2004 540) ldquo[nixwar-] lt [nihwar-]rdquo one may wonder whether the MP devel-opment of θw is not as traditionally assumed either and could posit theassumption that nixwār- lt nihwār- lt niθwār- shows the regular MP result ofθw in word-internal position The reduction seen in čahār and čihil wouldthen need to be due to a specific development here as well which could haveoperated at the stage of hw A reduction of the consonant cluster wouldseem particularly likely in the multiple clusters arising in čaθwrθatam (cfAv čaθβarəsatəm) gt čahwirhat23 (gt via čihwihl or čihird ) gt čihil lsquofortyrsquowhose -h- would have been transferred also to čahwār gt čahār In word-finalposition one would need to assume a reduction θw gt -hw gt -h which wouldoperate in the abstract suffix -īh (lt -iya-θwa-) and in prθw- lsquobridgersquo gtpurh gt puhl24 The adverbial suffix -īhā would need to have generalized hby paradigmatic levelling from -īh25

This approach implies ad hoc assumptions for čahār čihil and -īhā butaccounts for nixwār- which is otherwise left without explanation26

Moreover a development θw gt hw gt xw agrees quite well with other MPsound changes θ yields MP h generally (eg pahn lsquowide broadrsquo mēhanlsquohomersquo vs Avestan paqana- maēqana- Huumlbschmann 1895 203) Thesequence hw lt θw merges with old hw lt PIE su both resulting inMP xw27 Also parallel is the development of fw gtMP hw (kahwan lsquooldrsquolt kafwan Bailey 1979 62b 64b) But this development needs to be later thanthe change hw gtxw discussed above as the hw arising from fw does notyield xw28

5

51The interpretation of Pth ltpwrtgt purt lsquobridgersquo and ltmwrtgt murt lsquodeathrsquosuggested in Section 2 implies that Manich Pth lttgt and ltdgt encode two differ-ent phonemes also in the position after r Now there appear to be exceptionsexactly in this context according to Boyce (1975 17) lttgt otherwise encodest but ldquorarelyrdquo also d when in the position ldquoafter r (an archaic spelling) egwrt- besides wrd- (ward-)rdquo This raises the question whether ltrtgt and ltrdgt

23 Under any assumption (θw gt hw or directly gt h) r gives ir here in spite of the neigh-bouring w

24 MP čāh lsquospringrsquo and gāh lsquoplace thronersquo can be explained as deriving from -θu- (OldPersian gāθu- cf note 11) with θ generalized from the oblique case (cf Huumlbschmann1895 195 203 Brandenstein and Mayrhofer 1964 121) the same applies to Pth čāhand gāh

25 Gauthiot (1918 67) explains -īhā as ablative-instrumental -iya-θwāδā of the stem-iya-θwa-

26 Henningrsquos note (1939 105) about nixwār- as a ldquodeveloped form of niθvār-rdquo does notexplain anything and the borrowing from Parthian cautiously considered by Weber(1994 111 n 11) needs to assume an unprecented substitution of df by xw

27 On the possibly monophonematic status of MP ltxwgt see Weber 199428 The New Persian merge assumed by Weber (1994 113) for MP hw and xw (or rather xw)

is obscure to me in fact MP hw yields NP hu (kahun kuhan lsquooldrsquo ltMP kahwan) while xw

gives NP xu (saxun suxan lsquospeechrsquo ltMP saxwan)


are written indiscriminately and refer to the same pronunciation29 The data areas follows30

bull inflectional forms of the verb ltwrt-gt ltwrd-gt wartd- lsquoturnrsquobull its derivatives ltwrd(g)gt lsquoprisonerrsquo ltwrdy(y)wngt lsquowagonrsquobull its compounds and their derivatives inflectional forms of ltʾmwrt-gt ltʾmwrd-gt

am-wartd- lsquocollectrsquo with ltʾmwrdngt lsquoassembly (place)rsquo ltʾmwrdyšngt lsquocollec-tionrsquo ltʾmwrtʾdnyftgt lsquoassemblyrsquo one inflectional form of ltʿzwrt-gt iz-wartd-lsquoreturnrsquo with ltʿzwrdyšngt lsquoreturnrsquo one inflectional form of ltprwrt-gtpar-wartd- lsquoprevailrsquo31 vs fra-wartd- in ltfrwrdggt lsquoletter (roll)rsquo

bull ltʾrtgt (lt arta- Avestan aša- Old Persian artadeg OInd rtaacute-) besides ltʾrdʾwgt(ltartāu an- cf Avestan ašauuan- OP artāvan- OInd rtāvan-) both occuronly in connection with ltprwhrgt in a designation of the ether (one of theManich elements of light) ltʾrtgt could be an archaism of the religiouslanguage as is its cognate wrt- urta- in the Sogdian version of the prayerAšəm vohū (Nicholas Sims-Williams personal communication cfGershevitch 1976)

If one explains ltʾrtgt as an archaism or a borrowing from an older stage of thelanguage Pth wartd- lsquoturnrsquo with compounds and derivatives is the only case ofa variation ltdgt lttgt in Manich Pth orthography32 At the same time wartd- isthe only instance of Pth ltrtgt other than ltpwrtgt and ltmwrtgt33 The remainingcases are loanwords or unclear

bull lt sʾrtgt sārt lsquocaravanrsquo and lts(ʾ)rtwʾgt sartwā lsquocaravan leaderrsquo are borrowedfrom OInd sārtha- and sārthavāha- (as is Sogdian sʾrth Sims-Williams1983 133 135 140)

bull two items are unclear the hapax ltʾwrtʾdgyftgt (thus Sundermannrsquos reading oflt(ʾ)wr(tgy)ft gt cf DMD 70a) perhaps it belongs to ltwrtd-gt and ltʾmrtyngt(twice attested) for which Henning (apud Sundermann 1973 115) assumes aconnection to Avestan aša-34

29 This phenomenon needs to be distinguished from cases which show a variation ltdgt lttgt(cf Durkin-Meisterernst 2000 169 ff) These cases include ltbwtgt ltbwdgt būd (paststem of ltbw-gt baw- lsquobersquo) in a proportion 14 (Durkin-Meisterernst 2000 172) a similarproportion holds for pad lsquoto inrsquo (ltptgt ltpdgt) The variation ltdgt vs lttgt is found ininstances deriving from Old Ir t Conversely the Pth result from Old Ir d is always writ-ten ltdgt (eg ltkdgt kad lsquowhenrsquo Durkin-Meisterernst 2000 172 n 36) The remainingcases of lttgt are orthographic variants of lttgt (Boyce 1975 17)

30 Corresponding Manich MP words (where attested) have only ltrdgt31 Sims-Williams (1989 325) connects Pth ltprwrt-gt to Sogdian prwrt lsquoturn change

becomersquo (lt pari-wart-) and translates the attestation ltʾwd wʾd tftwʾdyg | ʾwwd nyprwrtydgt (verse) as ldquo( ) and the searing wind does not prevail thererdquo Perhaps onecould also consider a meaning within the semantic range of the other lt(deg)wrtd- gt eg ldquoand the searing wind does not swirl thererdquo or even ldquoand the wind does not turn sear-ing thererdquo interpreting ltprwrtydgt in the light of its Sogdian cognate

32 Boycersquos statement quoted at the beginning of this subsection and the note byDurkin-Meisterernst (2000 173) to the same effect thus need to be adjusted

33 Pth art is also found in names from other languages (Sanskrit Turkic)34 Another example might be the unclear hapax lthwʾwrt gt perhaps ldquohaving goodrdquo (but

maybe this is not a complete word cf DMD 192a) if lttgt here is a graphic variant oflttgt and not of ltdgt (cf note 29)


52The following points may be relevant in evaluating the orthography ltrtdgt OldIr t usually gives Pth ltdgt post-vocalically and after sonorants and also after reg ltmrdgt mard lsquomanrsquo (Av marta-) ltmrdyftgt mardīft lsquomanlinessrsquo ltsrdgt sardlsquocoldrsquo (Av sarəta-) ltsrdʾggt sardāg lsquocold (noun)rsquo ltwxrdgt wxard lsquoeatenrsquo (lthwar-ta-) ltwxrdyggt wxardīg lsquomealrsquo ltnbrdgt nibard lsquobattlersquo ltnbrdggt nibardaglsquowarlikersquo ltkyrdgt kird lsquodonersquo (Av kərəta-) ltkyrdgʾrgt kirdagār lsquomightyrsquo ltdyrdgtdird lsquoheldrsquo (Av dərəta-)35 The voiced counterpart Old Ir rd mostly yields Pthrδ eg ltzyrdgt zirδ lsquoheartrsquo (lt Proto-Ir zrdaya-)36 However Old Ir ard givesPth ār (Rastorgueva and Molčanova 1981 162) eg ltwʾrgt wār lsquoflowerrsquo (Avvarəδa-) ltsʾrgt sār lsquoyearrsquo (Av sarəδa-) So there is an opposition between -rdltOld Ir -rt and -rδ ltOld Ir -rd only for vowels other than a but no daggerarδ ltard vs ard lt art

Connecting the Pth data to developments in other Ir languages one mightwonder whether the mixed orthography ltrtdgt after a intended to mark aspecific pronunciation for which there was no orthographic convention ndash per-haps voiceless r + t as Durkin-Meisterernst (2000 173) assumes SimilarlyAv ltš gt which is the result of rt in certain contexts has been assumed to rep-resent voiceless r retroflex t or a fricative similar to Czech ř (Hoffmann 1986173 ff de Vaan 2003 602) Also noteworthy is the occasional lengthening ofAv a preceding ltš gt eg xvāša- lsquofoodrsquo lt hwar-ta-37 In Balochi Old Ir artgives ārt and ard gives ār (eg wār-t lsquoeatsrsquo vs war- otherwise gwārag lsquoblos-somrsquo38 vs Av varəδa-) while rt and rd after other vowels are preserved39

Pashto likewise has retroflex r from Old Ir rt and rd but this is independentof the preceding vowel40 So if the Pth orthography ltrtdgt did indicate aspecific sound or sound cluster the result of art would arrange itself with simi-lar phenomena in other Ir languages

It is not clear though why a variation ltrtdgt is only found with the familyltwrtd-gt and not with other words containing Old Ir art or why a ldquospecific pro-nunciationrdquo is only marked for wartd-41 Perhaps the variation ltrtdgt marks the

35 For examples of rt in labial context see Section 2136 The opposition between voiced stops (from Old Ir word-internal voiceless stops) and

fricatives (from Old Ir word-internal voiced stops) is not marked in the Manich scriptbut has generally been assumed at least for the older stages of Parthian Sundermann(1989 123) assumes a merge of both series for ldquoLate Middle Parthianrdquo (sixth c AD)thus also Rastorgueva and Molčanova (1981 160) See Korn (2010 424 f) for furtherdiscussion

37 Cf de Vaan (2003 54 f 104 596) Among the instances relevant here is qβāša- lsquoquickfirmamentrsquo (from the same root as Pth niδfār-) if this does not contain old ā (de Vaanibid)

38 Thus Sayad Ganj p 704 Barker and Mengal (1969II 463) note gwārig lsquowild yellowtuliprsquo

39 Cf Korn (2005 97 189 220)40 Cf Skjaeligrvoslash (1989 404) A change of r + dental to retroflexes is common cross-

linguistically (thus eg in Swedish and in Franconian dialects)41 Sogdian influence cannot be responsible for the orthography of Pth ltwrtd-gt the vari-

ation of ltδdgt and lttgt specifically after r noted by Gershevitch (1954 42 f sect 268 ff)does not exist rather a late stage of Sogdian probably had [d] as an allophone of t invoiced contexts thence some cases of C ltdgt for what is otherwise lttgt (Nicholas


word-internal development which is exclusively found in the only Pth presentstem with Old Ir art42 while the word-final position shows the expected ltrdgtard Inflectional forms and derivatives such as ltmrdʾngt mardān (plural)ltmrdyftgt mardīft etc were surely related to ltmrdgt mard lsquomanrsquo by the speakersand thus do not undergo word-internal development while a present stem mostlyoccurs with endings If ltrtdgt is the word-internal development it is perhaps lesslikely that ltrtdgt stands for a devoicing which would not have taken place inword-final position and a retroflex or fricative output would seem more likely

6

Summarizing the argument above Manich Pth lttgt and ltdgt encode two differ-ent phonemes also in the position after r and Pth ltpwrtgt purt lsquobridgersquo andltmwrtgt murt lsquodeathrsquo are to be read as purt and murt These words are likelyto go back to prθw- (the form from which MP puhl also derives) andmrθw- (while Sogdian mwrδw derives from mrθu- with generalized θ)These are the oblique stems of prtu- and mrtu- the former familiar fromAv pərətu- the latter otherwise only found in Sogdian

Pth ltpwrtgt lsquobridgersquo and ltmwrtgt are then additional evidence forSims-Williamsrsquo claim that Proto-Ir θw does not yield Parthian f as previouslyassumed but results in a consonant group which would be reduced in Pthpurft and murft By the logic suggested here -ft would be the Pth word-finaloutcome of θw in Manich Parthian (vs -f in inscriptional Parthian) vs -tf-(thus in inscriptional Parthian) gt -δf- (Manich) in word-internal position

Middle Persian may likewise show a consonant cluster as the result of θwyielding hw gt xw In čahār lsquofourrsquo and čihil lsquofortyrsquo specific processes must thenhave been at work to effect the simple h these would be parallel to clusterreductions in these numbers in other Ir languages

Table 1 presents the Pth sound changes of r and r + dental discussed in thispaper in comparison with some data of selected Western Ir languages

Examples for Zazaki include the cognates of Pth words mentioned above forard ser lsquoyearrsquo rd zeri lsquoheartrsquo (Paul 1998 169) vılıke lsquoflowerrsquo rt kerd-berd- (past stems of lsquodorsquo and lsquocarry awayrsquo) art serd lsquocoldrsquo rθw pırdlsquobridgersquo (cf Section 21) Since rt appears to give erd also in labial context(berd- lt brta-) one could perhaps consider vılıke a loanword (thus Paul1998 169) so that the regular output of rd in labial context could be er or per-haps ır (cf eg pır lsquofullrsquo which at least shows r in labial context although notrt)

Sims-Williams personal communication cf Sims-Williams 1985 163 n 1) Sogdiancompounds and derivatives corresponding to Pth ltwrdt-gt are well attested and alwayswritten with ltrtgt eg prw(ʾ)rt- lsquoturnrsquo zw(ʾ)rt- lsquoreturnrsquo wrtn lsquowagonrsquo the interpretationof wʾrδʾt (Frag Len 93 8) is not clear but it is unlikely to show wʾrδ- lsquoturnrsquo (PavelLurje and Nicholas Sims-Williams personal communication) Perhaps a denominativeverb wʾr- lsquorainrsquo is present here (Yutaka Yoshida personal communication)

42 Pth and MP (Pahlavi) nibard- lsquofightrsquo are probably denominative formations from nibardlsquobattlersquo cf the secondary past stems Pth nibardād (which is the only attested form of thePth verb) and MP nibardīd (not from the zero grade) cf OInd radicprt


In Balochi the contexts in which r yields ir and ur are not identical to thoseof MP and Pth ir ur While ir is the result in palatal contexts and ur in labialones the neutral context shows Balochi ur but MP Pth ltyrgt ir eg Balochiturs- vs MP Pth lttyrs-gt lsquobe afraidrsquo kurt vs MP Pth ltkyrdgt lsquodonersquo Otherexamples include rd zird lsquoheartrsquo ard gwārag lsquoblossomrsquo art sārt lsquocoldrsquoOwing to the absence of other examples for the context rθw it is impossibleto decide whether Balochi puhl lsquobridgersquo is a MP loanword or not (Korn2005 143ndash8 328 121)

Bibliography

Bailey Harold W 1979 Dictionary of Khotan Saka Cambridge Cambridge UniversityPress

Barker Muhammad A and Aqil Khan Mengal 1969 A Course in Baluchi MontrealMcGill University Press 2 vols

Beekes Robert SP 1988 A Grammar of Gatha-Avestan Leiden Brill

Benveniste Eacutemile 1940 Textes sogdiens eacutediteacutes traduits et commenteacutes (Mission Pelliot3) Paris Paul Geuthner

Boyce Mary 1954 The Manichaean Hymn-Cycles in Parthian (London Oriental Series3) London Oxford University Press

Boyce Mary 1975 A Reader in Manichaean Middle Persian and Parthian (ActaIranica 9) Leiden Brill

Boyce Mary 1977 A Word-List of Manichaean Middle Persian and Parthian (ActaIranica 9a) Leiden Brill

Brandenstein Wilhelm and Manfred Mayrhofer 1964 Handbuch des AltpersischenWiesbaden Harrassowitz

DMD =Desmond Durkin-Meisterernst 2004 Dictionary of Manichaean Middle Persianand Parthian (Dictionary of Manichaean Texts III Texts from Central Asia andChina 1 Corpus Fontum Manichaeorum Subsidia) Turnhout Brepols

Table 1 Development of r r + dental

Proto-Iranian Parthian Zazaki Balochi Middle Persian

rd ltyrdgt irδ er ird illtwrdgt urδ ıl urd ul

rt ltyrdgt ird erd irt irdltwrdgt urd urt urd

rθw [+ lab]_ ltwrtgt urt ırd uhl uhl

ard ltʾrgt ār ar ār Manich ltʾr gtrHd Pahl ltʾl gt 43

art word-internal lt-rtd-gt -a(r)d- erd ārt ārdrHt word final lt-rdgt -ard

43 The Pahlavi orthography is ambiguous and could also stand for ār (then identical withthe Pth output) cf Hoffmann (1986 183 n 38) At any rate New Persian has āl inrelevant words


DMG=Desmond Durkin-Meisterernst No date Grammatik des Westmitteliranischen(Parthisch und Mittelpersisch) auf Grund manichaumlischer Texte des Inschriftenmaterialsund auszugsweise der Pahlavī-Literatur Muumlnster (unpublished manuscript)

Durkin-Meisterernst Desmond 2000 ldquoErfand Mani die manichaumlische Schriftrdquo inRonald E Emmerick Werner Sundermann and Peter Zieme (eds) StudiaManichaica IV Internationaler Kongreszlig zum Manichaumlismus Berlin 14ndash18 Juli1997 Berlin Berlin-Brandenburgische Akademie der Wissenschaften 161ndash78

EWAia =Manfred Mayrhofer 1986ndash2001 Etymologisches Woumlrterbuch des Alt-indoarischen Heidelberg Winter 3 vols

Gauthiot Robert 1918 ldquoDe la reacuteduction de la flexion nominale en iranienrdquo Meacutemoiresde la socieacuteteacute de linguistique de Paris 20 (1916ndash18) 61ndash76

Gershevitch Ilya 1946 ldquoSogdian compoundsrdquo Transactions of the Philological Society1945 137ndash49 (= id Philologia Iranica (Beitraumlge zur Iranistik 12) WiesbadenReichert 1985 6ndash18)

Gershevitch Ilya 1954 A Grammar of Manichean Sogdian Oxford Oxford UniversityPress

Gershevitch Ilya 1976 Appendix in Sims-Williams 1976 75ndash82

Gharib 1995 = Badrozzamān Qarīb Sogdian Dictionary SogdianndashPersianndashEnglish Farhang-e soġdī soġdīndashfārsīndashenglīsī Tehran Farhangan 1374 hš

Ghilain Antoine 1939 Essai sur la langue parthe son systegraveme verbal drsquoapregraves les textesmanicheacuteens du Turkestan oriental (Bibliothegraveque du Museacuteon 9) Louvain Museacuteon(repr 1966)

Gippert Jost 2001 ldquoZum lsquoeigenenrsquo Tod des Kambysesrdquo in Heiner EichnerPeter-Arnold Mumm Oswald Panagl and Eberhard Winkler (eds) Fremd undEigen Untersuchungen zu Grammatik und Wortschatz des Uralischen undIndogermanischen in memoriam Hartmut Katz Vienna Edition Praesens 15ndash26

Henning Walter B 1937 ldquoA list of Middle Persian and Parthian wordsrdquo Bulletin of theSchool of Oriental Studies 9 79ndash92 (= id Selected Papers I (Acta Iranica 14)Leiden Brill 1977 559ndash72)

Henning Walter B 1939 ldquoSogdian loan-words in New Persianrdquo Bulletin of the Schoolof Oriental Studies 10 93ndash106 (= Selected Papers I 639ndash52)

Henning Walter B 1958 ldquoMitteliranischrdquo in Handbuch der Orientalistik I iv 1 20ndash130

Hoffmann Karl 1986 ldquoAvestisch šrdquo in Ruumldiger Schmitt and Prods O Skjaeligrvoslash (eds)Studia grammatica iranica Festschrift fuumlr Helmut Humbach (MSS-Beiheft 13)Munich 163ndash83 (= id Aufsaumltze zur Indorianistik 3 Wiesbaden Reichert 1992837ndash57)

Huumlbschmann Heinrich 1895 Persische Studien Strasbourg Truumlbner

Huyse Philip 2003 Le y final dans les inscriptions moyen-perses et la lsquoloi rhythmiquersquoproto-moyen-perse (Studia Iranica Cahier 29) Paris Association pour lrsquoavancementdes eacutetudes iraniennes

Korn Agnes 2005 Towards a Historical Grammar of Balochi Studies in BalochiHistorical Phonology and Vocabulary (Beitraumlge zur Iranistik 26) WiesbadenReichert

Korn Agnes 2010 ldquoParthian žrdquo Bulletin of the School of Oriental and African Studies73 415ndash36

Lentz Wolfgang 1926 ldquoDie nordiranischen Elemente in der neupersischenLiteratursprache bei Firdosirdquo Zeitschrift fuumlr Indologie und Iranistik 4 251ndash316

LIV = Helmut Rix et al 2001 Lexikon der indogermanischen Verben die Wurzeln undihre Primaumlrstammbildungen Wiesbaden Reichert (2nd ed)


Paul Ludwig 1998 ldquoThe position of Zazaki among West Iranian languagesrdquo inNicholas Sims-Williams (ed) Proceedings of the Third European Conference ofIranian Studies held in Cambridge 11th to 15th September 1995 Part I Old andMiddle Iranian Studies Wiesbaden Reichert 163ndash77

Rastorgueva Vera S and EK Molčanova 1981 ldquoParfjanskij jazykrdquo in Osnovy irans-kogo jazykoznanija 2 Moscow 147ndash232

Sayad Ganj = Sayad Hāšmī 2000 Sayad Ganj The First Balochi Dictionary SayadGanǰ Balōčīē awwalī bazānt balad Karachi Sayad Hashmi Academy

Schmitt Ruumldiger 1989 Compendium Linguarum Iranicarum Wiesbaden Reichert

Sims-Williams Nicholas 1976 ldquoThe Sogdian fragments of the British LibraryrdquoIndo-Iranian Journal 18 43ndash82

Sims-Williams Nicholas 1983 ldquoIndian elements in Parthian and Sogdianrdquo inKlaus Roumlhrborn and Wolfgang Veenker (eds) Sprachen des Buddhismus inZentralasien Vortraumlge des Hamburger Symposions vom 2 Juli bis 5 Juli 1981(Veroumlffentlichungen der Societas Uralo-Altaica 16) Wiesbaden Harrassowitz132ndash41

Sims-Williams Nicholas 1985 The Christian Sogdian Manuscript C2 (BerlinerTurfantexte 12) Berlin Akademie-Verlag

Sims-Williams Nicholas 1989 ldquoA new fragment from the Parthian hymn-cycleHuyadagmānrdquo in Charles-Henri de Foucheacutecour and Philippe Gignoux (eds)Eacutetudes irano-aryennes offertes agrave Gilbert Lazard (Studia Iranica Cahier 7) ParisAssociation pour lrsquoavancement des eacutetudes iraniennes 321ndash31

Sims-Williams Nicholas 2004 ldquoThe Parthian abstract suffix -yftrdquo in John HW Penney(ed) Indo-European Perspectives Studies in Honour of Anna Morpurgo DaviesOxford and New York Oxford University Press 539ndash47

Skjaeligrvoslash Prods O 1983 The Sassanian Inscription of Paikuli 3 Wiesbaden Reichert 2vols

Skjaeligrvoslash Prods O 1989 ldquoPashtordquo in Schmitt 1989 384ndash410

Sundermann Werner 1973 Mittelpersische und parthische kosmogonische undParabeltexte der Manichaumler (Berliner Turfantexte 4) Berlin Akademie-Verlag

Sundermann Werner 1982 ldquoDie Bedeutung des Parthischen fuumlr die Verbreitungbuddhistischer Woumlrter indischer Herkunftrdquo Altorientalische Forschungen 9 99ndash113 (= id Manichaica Iranica I (Serie Orientale Roma 89) Rome Istituto italianoper lrsquoAfrica e lrsquoOriente 165ndash81)

Sundermann Werner 1984 ldquoEin weiteres Fragment aus Manis Gigantenbuchrdquo inOrientalia J Duchesne-Guillemin emerito oblata (Acta Iranica 23) Leiden Brill491ndash505 (= id Manichaica Iranica II (Serie Orientale Roma 89) Rome Istituto ita-liano per lrsquoAfrica e lrsquoOriente 615ndash31)

Sundermann Werner 1989 ldquoParthischrdquo in Schmitt 1989 114ndash37

Tedesco Paul 1921 ldquoDialektologie der mitteliranischen Turfantexterdquo Monde Oriental15 184ndash258

de Vaan Michiel 2003 The Avestan Vowels (Leiden Studies in Indo-European 12)Amsterdam and New York Rodopi

Weber Dieter 1994 ldquoZur Problematik eines Phonems xw im Pahlavirdquo Studia IranicaMesopotamica amp Anatolica 1 107ndash18

Yoshida Yutaka 2008 ldquoDie buddhistischen sogdischen Texte in der BerlinerTurfansammlung und die Herkunft des buddhistischen sogdischen Wortes fuumlr bodhi-sattvardquo Acta Orientalia Academiae Scientiarum Hungaricae 61 325ndash58


Footnotes on a Parthian sound change1

Agnes KornUniversity of Frankfurt am Mainakornemuni-frankfurtde

To the memory of Jochem Schindler (1944ndash94)

AbstractThe treatment of Proto-Iranian θw (PIE tu) is one of the isoglosses dis-tinguishing Middle Persian from Parthian and thus important for WesternIranian dialectology The re-discussion of the Parthian development of thisconsonant cluster by Nicholas Sims-Williams presents a welcome oppor-tunity for some notes on the matter I will argue that there is someadditional evidence in favour of his suggestion that the Parthian result isnot -f- as previously assumed but a consonant cluster I will also suggesta modification of the steps that the development takes The Middle Persiandevelopment of θw as well as some related issues of historical phonologyand Pth orthography and Western Ir are likewise discussedKeywords Parthian Middle Persian Western Iranian Historicalphonology Isoglosses Iranian dialectology

1

According to the classical treatments by Tedesco (1921 199 f) and Henning(1958 96 f) Proto-Ir θw (PIE tu ) gives Middle Persian h (MP čahārlsquofourrsquo vs Avestan čaqβārō OInd catvārah MP čihil lsquofortyrsquo vs Avestančaqβarəsat- OInd catvārimśaacutet- abstract suffix MP -īh lt -iya-θwa-) but f inParthian as in čafār lsquofourrsquo čafrast lsquofortyrsquo and the abstract suffix -īf in inscrip-tional Parthian

This interpretation of the Parthian data needs to rely on explaining theadditional -t seen in -īft the variant of the abstract suffix seen in ManicheanParthian as an additional suffix (Tedesco 1921 200 suggests a derivationfrom -iya-θwa-tā-) However this approach does not offer an explanation forthe Pth verb (present stem) ltnydfʾr-gt niδfār- (past stem) ltnydfwrdgt

1 The present article is a revised version of a German paper For reasons of typographicalsimplicity θ w and y are used instead of q u and i for Proto-Iranian As per Iranologicaltradition italics represent the transcription (phonemical form) for Parthian and MiddlePersian but the transliteration (graphical form) for Sogdian Manichean MiddlePersian and Parthian are quoted from and in the form of DMD unless otherwise noted

Abbreviations Av = Avestan B = Buddhist Sogdian C = Christian Sogdian Ir =Iranian Manich =Manichean M = (Sogdian in) Manich script MP =Middle PersianOInd = Old Indic (Vedic and Sanskrit) OP = Old Persian PIE = Proto-Indo-EuropeanPth = Parthian S = (Sogdian in) Sogdian script For bibliographical abbreviations seethe references at the end of the article

I am indebted to Desmond Durkin-Meisterernst Jost Gippert Thomas JuumlgelNicholas Sims-Williams and Yutaka Yoshida for comments and discussion and to sev-eral colleagues for their contributions acknowledged in the notes

Bulletin of SOAS 76 1 (2013) 99ndash110 copy School of Oriental and African Studies 2013doi101017S0041977X13000013




2
























3















4









5





































6










Bibliography































































2
























3















4









5





































6










Bibliography












































































3















4









5





































6










Bibliography






























































3















4









5





































6










Bibliography

































































4









5





































6










Bibliography































































5





































6










Bibliography























































































6










Bibliography









































































6










Bibliography





























































6










Bibliography





























































Bibliography

























































































































Documents

Footnotes on a Parthian sound change - CORE · Footnotes on a Parthian sound change Agnes Korn To cite this version: Agnes Korn. Footnotes on a Parthian sound change. Bulletin of