A N N O A C C A DE M IC O 2018-2019 C DL : Filologia,letterature e storia del mondo antico ( LM ) - 29955 ( L M - 15)
GLOTTOLOGIA MAGISTRALE
P r o f. M a r co Mancini
Programma d’esame
C. Ciancaglini-A. Keidan, Linguistica generale e storica per studenti di lingue orientali e classiche, vol. II, Firenze, Le Monnier, 2018 N. Sims Williams, Le lingue iraniche, in A. Giacalone-Ramat-P. Ramat (a cura di), Le lingue indoeuropee, Bologna, Il Mulino, 1993, pp. 151-179.
Le dispense sono costituite dai seguenti saggi: – Huyse, Philip (1999) Some Further Thoughts on the Bisitun Monument and the Genesis of the Old Persian Cuneiform Script. In: Bulletin of the Asia Institute 13:45-66. – Schmitt, Rüdiger (2004) Old Persian. In: Roger D. Woodard ed.., The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: 717-741 – Testo di DB – Fotocopie distibuite a lezione
Bulletin of the Asia Institute, a Non-Profit Corporation
Some Further Thoughts on the Bisitun Monument and the Genesis of the Old Persian Cuneiform Script Author(s): PHILIP HUYSE Source: Bulletin of the Asia Institute, New Series, Vol. 13 (1999), pp. 45-66 Published by: Bulletin of the Asia Institute, a Non-Profit Corporation Stable URL: http://www.jstor.org/stable/24048957 Accessed: 14-12-2015 03:38 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Bulletin of the Asia Institute, a Non-Profit Corporation is collaborating with JSTOR to digitize, preserve and extend access to Bulletin of the Asia Institute. http://www.jstor.org
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions Some Further Thoughts on the Bisitun Monument and the
Genesis of the Old Persian Cuneiform Script1
PHILIP H U Y S Ε
While many studies dealing with the origins of no Babylonian translation; nor do we have an the Old Persian cuneiform script have appeared Aramaic version of it, or a translation in any in the past few decades,2 a new edition by other language employed in the Achaemenid Schmitt, published in 1991, calls for more work empire. Another reason is that the Old Persian on the subject. Now based on a thorough verifi- text is seriously damaged at some crucial points, cation of a series of good photos taken in the which—until Schmitt 1991—prevented a more early sixties, the text of the Old Persian version accurate restoration of the text, while the corre of the Bisitun inscription is at last on a firm ba- sponding Elamite text, though completely pre sis.3 In addition, many facts closely connected to served, contains several key words which are the matter which seemed in doubt some thirty or either hapax legomena or of which the exact forty years ago have become certainties. meaning is contested. There is presumably no other passage in an Following are the texts of both versions in full Old Persian inscription that has been debated length according to the current standard edi about so intensely as the last lines of the fourth tions, accompanied by the translations given by column of the great Bisitun inscription (DB IV their respective editors: 88-92 = par. 70), also known as the "Schrifter findungsparagraph." While some scholars would Old Persian (DB IV 88-92): [88]. . . θ-α-t-i-y : d-a-i-y ; : v-s-n-a : a-u- r-m-z-d-a-h : like to see in it proof that Darius I (522-486 b.c.) v-u-s x-s-a-y-0-i-y [89] 1"m ·' ,·1 : a-d-m : a-k"-u-n-v-m : introduced the Old Persian cuneiform script by d'-i-p-i-'c^-i-'ç-m t-y P-t-i-S-m : a-i-i-y-a : u-t-a : p-v-s-t- [90]a-y-[a] : u-t-a : means of that very inscription,4 others think c™ ; i"m] ; ; ti-tU-s-m-lc-i-y :[n-a-m that the script already existed in the time of 5 n-a]-f-m : a-k"-u-n-v-m : p-[t]-i-s-[m · u]-v-a-d-a- [91] OyrusΓνrus tnethe Treat (ca.(ca 558558-580 58U b.c.]β c I anaand stilistill . [t.m) [a_ku.u.n].v.[m] : u.t.a : n-i-y-p-i-W]-i-[y : u}-t-a others claim that its reach as far back as ...... origins p+s-i-y-a p.s.a.[v] pm Median times. Much on a correct un- depends [92] i-p-i-'c-i-ç]-m : f-[r]-a-s-t-a-y-m : v'-i-\s]-p-d-ci: ■ derstanding of this par. 70 in DB (part l).7 In fact, a.t.T d-h-y-a-[v\ : k-a-r : h-m-a-[t]-x-S-t-a it plays a central role in the question of the gen esis of the Old Persian cuneiform script (part 2), /0âti Dàrayavaus xsàyaOiya: vasnà Auramazdâha ima adam utà and it also has consequences for the reconstruc- dipiciçam, taya akunavam, patisam ariyâ, utà carmà nâma tion of the consecutive stages in the creation of pavastâyà8 grftam âha, patisamci nàfam akunavam, patisam uvàdàtam akunavam, utà the Bisitun monument (part 3). niyapaiOiya utà patiyafraOiya paisyà mâm, pasava ima dipiciçam frâstàyam vispadâ antar dahyàva, kâra hamâtaxsatâ/ 1. DB, Paragraph 70
"Proclaims Darius, the king: By the favour of Aura The reasons for the re- continuing controversy mazdâ this (is) the form of writing, which I have of the Bi- garding the interpretation of par. 70 made, besides, in Aryan.11 Both on clay tablets and situn inscription are manifold. One problem is on parchment it has been placed. Besides, I also that there exists only an Elamite parallel ver- made the signature; besides, I made the lineage. And before me. sion of that passage in the Old Persian text, but it was written down and was read aloud
45
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Afterwards I have sent this form of writing every- adverb /patisam/ followed by the enclitic em where into the countries. The people strove12 (to use phatic particle /ci/; Kent 1953, 130, and others after Schmitt 45 it)." (Quoted 1991, [transliteration] ha(j them erroneously understood as [p-t]-i-s-m and 73f. [transcription and translation]) i-y /patisa(m)-mai/, that is, as the adverb /pati sam/ followed by the enclitic pronoun of the 1st and pers. sg. /mai/. Bef°re translati°n and Elamite [DBl]13: [1]
46
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script material object, but rather as referring to the something like "in front of," when used as an "contents" of the text). As long as the exact form adverb, and "opposite, facing," when used as an of the Old Persian equivalent was not deter- adjective. And since he thought that /patisam/ mined, it seemed the best course to find an as an isolated adverb did not really make sense acceptable translation for Elam. htup-pi-me. But in the context of DB IV 89ff., he tried to associ now that Schmitt 1990, 57-59, has firmly estab- ate it with the verb /akunavam/, which indeed lished the Old Persian form, a solution seems occurs three times in the neighborhood of /pa nearer. tisam/, once (DB IV 89) immediately before it, Whatever the precise function of Elam. -me twice (DB IV 90f.) following it and separated by in btup-pi-me may be, the suffix clearly corre- at least one other word. An expression OP /pati sponds to OP /-ciça-/ in /dipi-ciçam/, so that sam (adv.) kr-/ thus seems quite within the one may assume that it is the "kind, sort, type" bounds of probability. Taken literally, in Lazard's or "form, shape, appearance" (OP */ciça-/, cf. MP view it would mean "mettre devant," that is, /cihr/ with the same meaning) of the "inscrip- "mettre en regard (des autres)," "graver en face ~ tion" (Elam. htup-pi OP /dipi/) that is discussed ou à côté," as in DB IV 89, or "mettre devant" in here. But Schmitt's assumption that "Darius the sense of "mettre en tête, apposer dessus," as first speaks ... of the 'outer form' of his 'new in DB IV 90f. writing' [i.e., /dipi-ciçam/ in DB IV 89, my addi- This ingenious analysis is attractive,21 but tion] and adds then a remark on its 'inner form' some details remain problematic. On the one in that thus it became possible to write a text hand, Lazard is forced to accept a (slight) change 'in Aryan' [i.e., /patisam Ariyâ/ in DB IV 89, my in meaning of /patisam kr-/ between DB IV 89 addition]" is perhaps just a little too subtle to and 90f., and his translation results in a rather be true. trivial statement. In Lazard's interpretation, what I am inclined to think that the solution Darius would have wanted to say in par. 70 is adopted by Harmatta 1966, 262, was somewhat that he has engraved the inscription DB in such nearer when he took Elam. -me to be a collec- a way that it now faces the reader, and that he tive suffix and suggested that Elam. htup-pi-me has put his name and genealogy at the begin "means such an inscription or document, which ning of the inscription. On the other hand, all has several specimens, versions, or eventually the examples given by Lazard to prove the like parts, but forms a unity." But in the light of lihood of a locution /patisam kr-/ are composed what I just said, I would change Harmatta's pro- with a substantive in the accusative, like OP posai to the effect that Elam. btup-pi-me does /zura kr-/ "to deceive" (lit. "to make evil") or not apply to such an "inscription" (coll.) as a Avest. /skandam kar-/ "to destroy" (lit. "to make whole, but on the contrary to a "version, copy, destruction"), not with an adverb or an adjec part" of it (that is, one particular "form" of the tive. Still, the association of /patisam/ to /kr-/ is "inscription"). Such an interpretation not only an excellent solution, though one single mean fits best in all other instances where the word ing for the word would seem to be sufficient for occurs but would also offer an obvious explana- all three attestations; it could then either be— ~ tion of why the complex form OP /dipiciçam/ preferably—interpreted as an adverb with loca Elam. htup-pi-me was needed in DB IV 89 and tive function ("to put opposite") or alternatively ~ 91 f. instead of the simple Elam. htup-pi OP explained as an adjective ("to make (as) opposite, /dipi/. But for that point, see below, after com- facing") in the accusative singular neuter, used ments on the next difficulty in the Old Persian predicatively and in congruence with the neuter text contained in the term /patisam/. substantive /dipiciçam/ (but the phrase can of Previously, most scholars have taken that word course still be translated adverbially).22 to be an adverb standing on its own and mean- I conclude the first part of my study with my ing "moreover, besides." Unfortunately, it is not own translations and some final comments: attested elsewhere in the Old Persian inscrip its Avestan counter- tions, but the meaning of old Persian (DB IV 88-92): "Proclaims Darius, the part paitisa(-) (AiW 836) has been reexamined by king: By the favor of Auramazdâ this version, which I Lazard 1976, 182-84. He came to the conclusion put opposite in Aryan, has been placed both on clay that the basic signification of the word must be tablets and on parchment. I also put opposite (my)
47
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script signature (and) I put opposite (my) lineage. And it concern the entire inscription instead of the Old was written down and was read aloud before me. Persian text only) and not enough (it would Afterwards I have sent this version into everywhere concern exclusively the inner form, but not the the countries; the people strove (to use it)." outer appearance) I have made one long sentence from /vasnâ a Auramazdàha/ through /grftam àha/ instead of , , , dividing it up into two phrases as is the case Elamite DBl]: Darius the king, proclaims: By the in Schmitt,g translation fand in all the others be_ will ol Uramazda, I made this version otherwise, m f . . . . f , , , , , fore>' JOUI of the two phrases has the advan Aryan, which did not exist before, both on clay and f that we do not need to fal1 back on an on parchment.23 And I made (my) name and (my) lin- taSe of the verb 'to be an el eage, and it was written and read aloud before me. ellipsis (even though Afterwards I sent (that) same version into all coun- lipsis of that kind is nothing unusual of itself ). tries,· the people strove (to use it)." As an immediate consequence of that change, it was the new Old Persian version which was To understand the full implications of this copied on clay tablets and on parchment to be "new" translation of the two versions, see figs. 1 sent out to all countries of the Empire.26 Unfor and 2. What Darius is trying to tell us in the Old tunately, only few fragments in Babylonian and Persian version, is—I think—that he had that Aramaic of such copies have survived. The Ara text added on the opposite side of (which here maic translation27 (dated around 420 b.c.) found means: beneath) the relief and the Babylonian on papyrus fragments at the Jewish military col and the (original) Elamite versions. Darius also ony in Elephantine resembles the Babylonian had his "signature" and "lineage" added sepa- version more closely28 and would thus contra rately in the inscription DBa over the relief, diet my above assumption. But the larger of two which thus faces on its part the Old Persian fragments of the Babylonian inscription29 found version of the inscription DB.24 That the new in the city palace of Babylon seems altogether interpretation disturbs the usual parallelism be- closer to the Old Persian and the Elamite texts, tween the Elamite and the Old Persian texts albeit that it contains casualty figures like the should not concern us: probably for the reasons Babylonian version preserved on the rock at evoked by Trumpelmann, the scribe of the Elam- Bisitun. ite text found it necessary to add a short addi- As far as I can see, the only problem—though tional remark (absent in the Old Persian version) of major concern—regarding my attempt at a and to stress that the Old Persian text had not "new" interpretation is the position of OP /pati existed at first.25 In the Elamite version he did sam/ at its first occurrence in DB IV 89. If /pati not translate the word OP /patisam/ "opposite" sam/ should be connected as an adverb or a in any of the three instances in which it oc- predicate adjective to the verb /akunavam/ pre curred, because it was applicable neither to the ceding it as I suggested above, the word order position on the monument of the first Elamite is contradictory to the expected normal Old Per version of DB nor to the position of DBl (Elam.). sian word order of subject-predicate adjective Finally, the new interpretation would also clar- verb (on which see Kent 1953, 95 §307). On the ify why the usual term OP /imam dipim/ (DB IV other hand, for reasons of expressiveness the ~ 43, 48, 70, 73, 77) Elam. btup-pi hi (DB III 66f., word order in Old Persian was quite free, and 70, 84, 85, 88) was inappropriate in the context the fact of being able to write in his own ~ of par. 70: In all those cases, OP dipi- Elam. "Aryan" language was certainly of sufficient htup-pi relates to the inscription as a whole, importance for Darius to deserve special em irrespective of its single versions. In par. 70, phasis as an adjunct at the end of the sentence, however, all that matters is the concrete reali- after the verb (see some other examples in Kent zation of the inscription in the Old Persian re- 1953, 96 §310 VII). daction only, but with its inner (i.e., contents) While my interpretation, like all previous in and outer forms (i.e., layout) alike. Therefore, a terpretations, remains a hypothesis, it gives a translation "text" or "contents"—as proposed by better explanation for the divergence between Harmatta and others—is not pertinent either, the Old Persian and the Elamite texts, and it un since it is at the same time too much (it would questionably has the advantage that Schmitt's
48
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Bisitun,
Mt. at
rock
the on
DB
inscription
trilingual
major
the of
versions
Babylonian
the
and
versions,
Elamite
two
the
text,
Persian
Old 4. pi. the of 1991,
position Schmitt The on 1.
Fig.based
49
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
relief,
Bisitun
the
on
"a-1")
labelled
(here 1
DBa
inscriptions
("Bab.")
Babylonian
and
("Elam."),
Elamite
("OP"],
Persian
Old
minor5.
the pi. of 1991,
position Schmitt
The on 2.
Fig.based
50
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script essential improvements in the readings of the that the k"-sign is the only one composed of Old Persian text have been fully taken into ac- merely two elements, but if the ru-sign was also count. If the way in which I understand the text among the "first" signs created, why was it not is correct, there would be no question of the equally composed of two instead of three ele "invention" of the Old Persian cuneiform script ments? And for what reason would signs like by Darius I in par. 70. To all those who would ka, ja, ba, da, etc., have been composed of three consider the word order in DB IV 89-91 an insur- elements only, whereas the repeatedly needed mountable obstacle to my interpretation, how- sign a was composed of four elements and the ever, the translation suggested by Schmitt 1991, useful sign va even of five?33 On the other hand, 73f., and quoted above still appears to be prefer- if the Old Persian cuneiform script had already able by far to all others.30 A translation "besides" existed as a coherent and complete writing sys for OP /patisam/ cannot be excluded (though it tem at the time of Cyrus, what would have pre needs confirmation), and Schmitt's translation is vented him from using it more extensively, and in perfect accordance with the "rules" of Old why would Darius not have planned an Old Per Persian syntax, but it does not explain why OP sian version of DB from the very start of his un /patisam/ was left untranslated three times in dertaking in Bisitun?34 the Elamite text. Apart from Hallock's note on the simplicity of the signs ku and iu, three further external argu ments have been advanced in favor of the ex 2. The Introduction of the istence of the Old Persian script by the time of Old Persian Cuneiform Script Cyrus: (1) the alleged Old Persian inscription on the tomb of Cyrus in Pasargadae; (2) the (trilin The result of my examination so far may appear gual) inscriptions CMa-c in Pasargadae, tradi rather disappointing to all those who support tionally attributed to Cyrus; (3) the Old Persian the view that Darius I was the one who intro- inscriptions of Darius' great-grandfather Aria duced the new Old Persian script. But does the ramnes (AmH) and of his grandfather Arsames new interpretation of the preceding pages mean (AsH), which turned up in Hamadân on gold tab that we finally have to return to the old thesis lets. In contrast to this, only one argument ap that it was, after all, Cyrus the Great who "in- pears to support the introduction of the new vented" the cuneiform signs or at least initiated writing system by Darius; it is embedded in the the use of them? I believe not. 21st (apocryphal) letter of Themistocles. Initially, The observation by Hallock 1970 that the the balance seems to favor Cyrus, but there has signs for ku and ru are among the simplest in the been considerable progress in research since the Old Persian cuneiform syllabary, even though important studies by Nylander 1967 and Lecoq they are not among the most frequently used, has 19 74. been misleading. Since both of these signs occur In Greek literature, we find five references to in the name of the founder of the Empire (OP ku- the presence of inscriptions on the tomb of u-ru-u-s /Kurus/), it seemed so obvious that Cyrus II, but since the other three reports are Cyrus II could have drafted a small number of based on either Aristoboulos (FGiHist 139 F signs in order to be able to write his name—sup- 51a), who must have seen the tomb at least plemented at a later time with the signs needed twice, or Onesikritos (FGiHist 134 F 34), the to write some simple words like "I," "king," number of sources can actually be reduced to "empire," etc.—and that Darius I could have these two historians who had accompanied Al commissioned a few people to finish the job.31 exander on his campaign in the East. Both claim This theory of a gradual creation of the Old Per- to have seen an inscription, in Persian but en sian cuneiform script over a period of at least one graved with Greek letters according to Onesikri to almost four decades and sign by sign whenever tos, in Persian and with Persian signs according the need was felt cannot claim to be overly con- to Aristoboulos. Reservations regarding the trust vincing, however.32 Moreover, the alleged sim- worthiness of these accounts from an archaeolog been made Stronach plicity of the two signs ku and iu is entirely ical point of view have by relative and its relevance questionable. It is true 1978, 26, who emphasized that "it was certainly
51
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script never a standard Achaemenian practice to place distic reasons, since Darius was badly in need inscriptions on tombs." A thorough examina- of integrating Cyrus into his own Achaemenid tion of the language and style as well as of the family. And, being an eloquent writer, where contents of the "inscription(s)" finally confirms else if not in Pasargadae, "the hub of Cyrus' these doubts entirely; Schmitt 1988, 18-25, has former rule" (Stronach 1990, 201), could Darius convincingly shown (pace Lecoq 1974, 98f. and have attempted this fusion of the two houses 1997, 78-80), that we cannot have the slightest with a better chance of success?38 doubt about the inauthentic character of the The Old Persian inscriptions of Ariaramnes inscriptions quoted, which, on the contrary, ap- (AmH) and Arsames (AsH), which came to light pear to have been concocted by the Alexander in Hamadân in 1930 and 1945 respectively, also historians. have been adduced as proof that the Old Per The most serious debate no doubt concerns sian cuneiform script must have existed before the three short inscriptions CMa-c in Pasar- Darius.39 Because of the numerous grammatical gadae,· surely, no serious scholar will still chal- faults, serious doubts were cast on their authen lenge the idea that the inscription CMb should ticity from the beginning, however, and even a actually be attributed to Darius (hence later modern falsification has been taken into consid called: CMb-DMa),35 since his name occurs in eration. The faults mostly concern the incorrect it, and most will also agree that Darius may well use of cases (esp. of hybrid forms of the genitive have been the author of CMc (it reads: "Cyrus, case) and other barbarisms of the same kind as the great king, an Achaemenid"). For the other occur in the late Achaemenid inscriptions of the inscription CMa, which reads "I, Cyrus, the fourth century b.c. For this reason, they cannot king, an Achaemenid," the situation is different: have been fakes from the time of Darius,40 since At the issue of a long discussion, Nylander 1967 all inscriptions of that time, including Darius' would not go beyond attributing at best the Old duplicate inscription DH from Hamadân on a gold Persian version of this brief trilingual inscrip- and a silver plate from Hamadân, are grammati tion to Darius, while he assigned the Babylonian cally correct. According to Schmitt 1999, 105-11, and the Elamite versions to Cyrus. Yet, I think who has systematically restudied both inscrip that Stronach 1990 has proved conclusively tions and their faults, the fourth century b.c., (pace Lecoq 1974, 52-56, and 1997, 80-82) that and in particular the time of Artaxerxes III (359/ CMa with all its three versions should be as- 358-338/337 b.c.), is the most likely time for cribed to Darius. I cannot repeat all of the argu- their fabrication, since Artaxerxes III is the only ments here,36 but apart from the fact that no Achaemenid king who traced his lineage back to stylistic differences in the engraving of the Old Arsames in the inscription A3Pa (ibid., p. 105). Persian version versus the other two can be dis- In the 21st letter of Themistocles (ed. Hercher, covered (see Stronach 1990, 197), the most un- p. 752), the Greek politician requests his Argive equivocal element that betrays Darius as the friend Temenidas to send him some silver era author of CMa is the use of the term "Achae- ters, iron armor plates, and golden censers, "in menid" in the three parallel versions (ibid., 200). which the Old Assyrian signs are carved and If the brilliant thesis of Rollinger 1998 is cor- not those which Darius, the father of Xerxes, rect—as I am convinced it is—that there were as had recently ordained." A fresh investigation of a matter of fact two families, the Teispids, to all attestations for the expression "(As)syrian," which Cyrus and the first "Achaemenid" kings "Persian," and "Chaldaic letters" by Schmitt belonged, and the Achaemenids, the house of 1992b leaves no doubt as to the interpretation Darius, who was consequently not in line of of the passage: even if we accept that the let succession, then Darius had everything to gain ter itself was not written by Themistocles and by faking a couple of inscriptions in which he may instead have been a product of Roman had Cyrus present himself as an Achaemenid.37 Imperial times, it contains reliable information There can be no question of a posthumous sa- and real knowledge concerning the Old Persian lute by Darius in honor of the founder of the cuneiform script, which must be meant here.41 empire Cyrus as a last token of piety; his motive In the end, a close examination of the ex is far less noble and dictated by mere propagan- ternal criteria possibly in favor of the existence
52
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script of the Old Persian cuneiform script at the time largely dominate those of four and five elements of Cyrus turns out to be quite devastating: an as far as frequency of use is concerned, and the inscription on Cyrus' tomb in Pasargadae never same holds good for C°-signs versus C2- and Cu existed; the author of the inscriptions CMa-c signs; he therefore suggested that some basic (better: DMa-c) was no doubt—the usurper— text—which contained the name of Cyrus (see Darius; and the inscriptions of Ariaramnes and above)—might have served as a model for the Arsames were certainly fabricated at a later creator(s) of the Old Persian script. As long as it time. Thus, everything seems to point to Da- was believed that Cyrus was the inventor of the rius, and now even the Bisitun inscription it- cuneiform script, it seemed a hopeless task, self may give us a hint of this. however, to try to reconstruct such a basic text, It can safely be considered an established fact and only sheer luck could have brought it to that the Old Persian cuneiform script is indeed light if it really existed. But since all external the result of an invention at one precise mo- evidence points to Darius as the creator of the ment and not of a process of creation that took Old Persian script and to the Bisitun inscrip many years. Even if one would not wish to go tions as the first texts in which the new writing as far as Windfuhr 1970, 124, as to accept four system was ever used, it is rather strange that basic principles of sign composition,42 it never- no one has investigated those inscriptions in theless remains true that the Old Persian cunei- this respect. form script is clearly not derived from any of the As was stated above, the Old Persian script existing cuneiform models (see Lecoq 1974, 38- was most likely a new creation sui generis, and 48, and 1997, 59-72).43 This independence from the theory forwarded by Mayrhofer 1979, passim, other contemporary systems in use and the es- and 1989, 180f., 182f., to explain the strange and thetic "Stilprinzip"44 formulated by Hoffmann seemingly illogical selection of the signs of the 1976, 621, are almost sure indications of a de- types C2 and Cu versus the "complete" series of liberate invention. The final result is probably signs of the type Ca is in my view extremely not as perfect as a modern linguist would have helpful and would seem to stop only one step been able to realize, but, notwithstanding a short of the final solution. Reduced to its es number of structural flaws, it was fairly well sence, Mayrhofer's assumption is that, in order adapted to rendering the Old Persian language not to break the stylistic principle, only those (ibid., 645). signs were created that were absolutely indis If the Old Persian cuneiform writing system pensable. Apart from the vowels a, i, u and the were entirely consistent, one would expect the Ca-series, they include such C2- and Cu-signs, corresponding consonantal sign (Ca, C\ Cu) to which were needed in the names of prominent each vowel [a, i, u). One probable reason why persons or in unusual geographic names. That not every theoretically possible sign was actu- would explain why a sign like tu was needed so ally invented is that 24 more signs would have that a name such as k-t-p-tu-u-k /Katpatuka-/ been necessary in addition to the existing 36, "Cappadocia" could be written in an unam and in that case, more complex signs going biguous way; otherwise, *k-t-p-t-u-k might have against the stylistic principle would have be- been read as */Katpatauka-/. Similarly m1 and come unavoidable. Another explanation is that mu were invented to render /Armina/ "Armenia" not all signs were needed; phonemic combina- and /Mudrâya/ "Egypt" unequivocally,· but the tions like */cu/, */ju/, */yi/, and */vu/ are not to signs *bJ and *0U were not needed, as there was be expected in Old Persian. But these cannot no danger that well-known geographic names have been the only reasons; at any rate, no com- like b-a-b-i-ru-u-s /Bâbirus/ and α-θ-u-r-a /AGura/ pelling phonemic rationale can have been at the could have been misunderstood as */Babairus/ origin, since a sign like t2 is lacking, although and */A0aura/ respectively. For the same reason, no there was an urgent morphological need for it signs like *pi or *zu were never created, since to distinguish between the active verbal ending native speaker of Persian would misread words z-u-r /-tiy/ (*
53
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Mayrhofer 1989, 180—to the fact that it occurs form of the third person singular being θ-α-t-i-y only "in muttersprachlichen Phonepiketten." I /Gati/ (DBa 4, 9, 13f.|. Thus, the explanation that think that Mayrhofer's theory is not only accept- the need for a distinctive sign t1 in the active able, but conclusive, although—to his great dis- ending *
54
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Stage Relief Major Inscription Minor Inscriptions
I the whole relief without Skunkha no inscription DBa-j (Elam.) II pars. 1-69 (Elam., 1st copy) III pars. 1-69 (Bab.) DBb-j (Bab.) IV pars. 1-70 (OP) DBa-j (OP) DBl (Elam. [= OP, par. 70))
V addition of Skunkha "erasure" (?) of 1st Elam. copy; DBk (Elam. and OP) engraving of 2d Elam. copy pars. 71-76 [= col. V] (OP)
Fig. 3. The five construction phases on the Bisitun monument according to Hinz 1968.
because he would not have had enough time to portant discoveries made by Heinz Luschey develop it. This statement may be somewhat (1968) concerning the composition of the relief, exaggerated, but—as shown above—haste clearly and particularly the conclusions reached by Leo played a negative role in the creation process. Triimpelmann (1967) and Walther Hinz (1968) On the other hand, this was not accepted by on the arrangement of the inscriptions and the Lecoq 1983, 37, who states that "on pourrait tout genesis of the complete Bisitun monument.52 aussi bien dire que la création d'une vingtaine As far as I know, since Luschey's discovery de signes cunéiformes aussi simples que ceux that the last rebel Skunkha (at the right) must de l'écriture v[ieux] p[erse] ne demandait pas un have been engraved later, only Nagel 1983 has effort ni un temps considérables."51 But two dif- proposed a modification with regard to the re ferent issues appear to have been confused here. lief. However, this discussion is only of mar While it is certainly true that the actual creation ginal relevance in the present context.53 Of more of the signs could have been realized within only direct concern is the criticism by Lecoq 1997, a few days, once that the stylistic principle had 86, that the theory of a genesis in numerous been specified, the "test phase" of the new script phases with scribes and artists constantly chang may well have taken several weeks or even ing their minds is in sharp contrast to the care months. If the creation procedure were to be re- fully planned Achaemenid monuments of later constructed, the following scenario might per- times.54 Even if it was Darius himself (most haps claim some degree of probability: First, the likely) and not the scribes and the stonemasons vocalic signs a, i, u were created—is it mere on their own authority who decided to alter the chance that they are all three composed of four composition of the monument, the critique is elements?—together with the "complete" series important enough to deserve a serious reply, of consonantal/syllabic signs of the type Ca. This According to Hinz 1968 five construction first series of signs was then complemented with phases can be discerned (fig. 3). Triimpelmann Ci-/Cu-signs according to the principle "as many 1967 recognized six phases (fig. 4), which largely as necessary, but as few as possible." While the coincide with the ones discovered by Hinz result is reasonable, the final touch is lacking be- (divergences are marked in italics). Considering cause the inventors had not been accorded suffi- that Achaemenid art shows a predilection for cient time to try out the script. symmetry, it is certain that the original plans for the Bisitun monument must have been more final result. I can to ^ harmonious than the agree 3. The Genesis of the a large extent with the stages reconstructed by Bisitun Monument Triimpelmann and—even more—by Hinz, but a few precisions are required in the light of the The pioneering research on the genesis of the conclusions reached above. Bisitun monument carried out in the sixties has If Darius had really changed his mind as often been almost completely accepted; note the im- as was assumed by Hinz and Triimpelmann,
55
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Minor Stage Relief Major Inscription Inscriptions
I the whole relief without Skunkha no inscription DBa (Elam.) II pars. 1-69 (Elam., 1st copy) DBb-j (Elam.) III pars. 1-69 (Bab.) DBb-j (Bab.) IV pars. 1-70 (OP) DBb-i (OP) DB1 (Elam. [= OP, par. 70])
V addition of Skunkha "erasure" (?) of 1st Elam. copy; DBk (Elam. and OP) engraving of 2d Elam. copy VI pars. 71-76 [= col. V] (OP) DBa (OP)
Fig. 4. The six construction phases on the Bisitun monument according to Trtimpelmann 1967.
that would imply that the workers would have sions, and the first four columns of the Old Per needed to prepare a larger surface every few sian version of DB, are concerned with that first months to carry out his orders. To "blast" the year. It would have lasted well into Darius' sec rocks, I presume that the stonecutters used tech- ond regnal year of 520/19 b.c., until the Baby niques similar to those employed by other peo- Ionian texts were all engraved. After that, we pies in antiquity, in which wooden wedges were must assume a break of several months, during driven into small holes in the rock surface and which the Old Persian cuneiform script was in then moistened until they swelled so much that vented. Work may have resumed by the end of they made the rock burst. This technique could 519 b.c., after the creation of the new writing hardly have been so perfect that one could have system was completed. While the Old Persian determined in advance the precise spot where version of DB was being engraved, news of the the rock would crack; in other words, would successful repression of Skunkha and his fol Darius have given orders to enlarge the surface lowers must have reached Darius, who insisted for the monument at regular intervals if there on having the events of his third regnal year was a risk that these actions might destroy a (519/518 b.c.) documented on the Bisitun mon part of the relief or of one of the inscriptions? ument. Activities in Bisitun presumably came Instead, the surface must somehow have been to an end during the course of 518 b.c. prepared from the very beginning. Each of those two major stages can be subdi To return to Lecoq's critique, the number of vided into two further phases (I 1-2 and II 1-2), work processes can indeed be reduced to two, each of which took place within the same cam even if the chronological order of the stages paign of construction works, but should clearly worked out in the two tables above is correct for be distinguished chronologically (indicated by a the most part. Everyone appears to have over- full line in the table of fig. 5). Every one was the looked an important remark by Luschey 1968, result of a new decision on the part of Darius 92f., who had discovered three different layers after a change of mind: in stage I, the break be of debris at the foot of the inscription, of which tween phases 1 and 2 was brought about by Da the upper and the middle layers only are directly rius' resolution to add Babylonian versions of related to works on the Bisitun monument.55 the existing inscriptions,· in stage II, the break Thus, we must indeed distinguish between two between phases 1 and 2 was occasioned by Da major coherent stages between which works rius' intention to have the image of the rebel were interrupted (see stages I and II in fig. 5, Skunkha added. Lastly, a further partition of marked by a double line). The first major stage some of these chronologically distinct phases in may have started soon after the defeat of the a, b, and c seems possible. But, in contrast to the Margian rebel Frâda on 28 December 521 b.c.,56 categories I—II and 1-2, this last type of subdivi or shortly after the end of Darius' first full reg- sion does not so much indicate different periods nal year on 3 March 520 b.c., since the events of time since a-b-c must have taken place al described in the Elamite and the Babylonian ver- most simultaneously as parts of only one in
56
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Stage Date Relief Major Inscription Minor Inscriptions
la the whole relief without Sku. ca. "lb" DBa-j (Elam.) 520/19 lc 1st B.C. pars. 1-69 (Elam., copy) ~2~ pars. 1-69 (Bab.) DBb-j (Bab.) la DB1 (Elam. [= OP, par. 70]) ca. 519 lb" DBa-j "(OP) B.C. lc pars. 1-70 (OP) II 2a addition of Skunkha "erasure" (?) of 1st Elam. copy DBk (Elam. and OP) ca. 518 2b engraving of 2d Elam. copy B.C. 2c' pars.~71-~76 [= col.~vj (OPf
Fig. 5. The two construction phases (with subdivisions) on the Bisitun monument according to Huyse.
struction on the part of the king. Rather, they the Median highlands near Ekbatana/Hamadân. would suggest the precise order in which the First preparations for the Bisitun monument different elements of one single sequence were may have begun soon after the defeat of the executed one after the other (since a, b, and c Margian rebel Frâda on 28 December 521 b.c., belong together each time, the distinction is or—at the latest—shortly after the end of Da marked by a broken line only in the table of fig. rius' first full regnal year on 3 March 520 b.c. 5). In other words, my proposal is more or less Since the place was a sacred site for the Elam a refined model of the one elaborated by Hinz ites (see Luschey 1968, 66), Darius chose to have (see fig. 3), and it differs to a larger extent from his inscription engraved in the Elamite language Trtimpelmann's model (especially the position- (which had also been the official Achaemenid ing of DBa [OP] in his last phase VI is unaccept- court language until then), while the relief was able). Some final explanations in the conclusion obviously inspired by the model of the local will clarify this. Elamite ruler Anubanini at Sar-e Pol-e Zohâb.58 That the relief was conceived as a harmonious unity with all Elamite texts becomes evident 4. Conclusion when we have a look at the surface prepared for both.59 A spot some 66 m over the lake appeared It is time to attempt a reconstruction of the to fit best because some 20 m underneath it, genesis of the Bisitun monument. After his there was a natural platform which was broad wrongful seizure of power, rebellions broke out enough (ca. 10 m) to have a scaffolding put up on everywhere in the empire, and the Achaemenid it.60 A surface large enough for the relief and usurper Darius was anxious to show his fellow the Elamite text was dressed as well as the ir countrymen as quickly as possible that he really regular shape of the rock allowed it,61 and the was the next legitimate king. As soon as he had surface just below presumably was superficially regained control over the situation, he decided smoothed out as well,62 in order to prevent will to make his position clear in words and images. ful vandalizing.63 While the stonemasons were A good place had to be found57 in the neighbor- working on the relief,64 the legends accompany hood of the fortress Sikayuvati, where Gaumàta, ing the images of the rebels (DBb-j) and the short Darius' most dangerous opponent, had been slain but self-confident inscription of Darius himself on 29 September 522 b.c., and the steep east (DBa), the text of DB was being prepared. The face of Mt. Bisitun seemed ideal. At the foot of king "dictated" the text ("Proclaims Darius, the the mountain there was a lake with spring wa- king") in his Persian mother tongue,65 and two ter, where caravans and the army used to rest on bilingual scribes wrote it down simultaneously their route from the Mesopotamian lowlands to in Elamite (on clay tablets) and in Aramaic66 (on
57
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
and seem to parchment). Then, both texts were worked out67 first, the stonemasons engravers and refined in the royal chancellery to meet the have been willing to use the same "niching" standards of official administrative language. technique as was used for the relief and the Thus, the personal tone of DBa (Elam.) was re- original Elamite version (see n. 59), but then placed by a more matter-of-fact style in DB had to abandon that idea, as it turned out to (Elam.), pars. 1-4 (see Hinz 1968, 96); and the be impossible to cut away the projecting rock Aramaic scribes added precise numbers of the on which the inscription now stands.70 Another casualties. point which the Babylonian and the Elamite This work done, the final text was read aloud texts have in common is the division of DB into in front of Darius and submitted for approval, 55 (respectively 54) exactly corresponding sec and, immediately thereafter, the stonemasons tions,71 whereas the—original—Old Persian text started engraving the text on the rock. Mean- has 69 paragraphs. while, the king had come to the conclusion that After this first stage, a break of a few months an Elamite text alone would not be sufficient; was necessary to give the scribes enough time he now also wanted a translation in the lan- to invent the Old Persian cuneiform script. But guage of his Babylonian predecessors and—last the continuation of the rebellions taking place but not least—in his own language. The Babylo- in the empire caused Darius to become impa nian translation was prepared from the Aramaic tient, and he asked the committee of inventors text immediately (see η 66), but for the Old to break off their work and write the text with Persian version there was no ready solution. A the script as it stood. This he explained in the new script had to be invented, and Darius com- last paragraph of the fourth column (par. 70), missioned a committee to carry out this order of which an Elamite—but no Babylonian72— as soon as possible, since his second regnal year version was also made. The Elamite text DBl (520/19 b.c.) was already well advanced. At the was engraved over DBa (Elamite), and then the same time, in Bisitun, the stonemasons added Old Persian versions of the legends DBb-j were the Babylonian legends DBb-j beneath the im- added.73 In the meantime, the Old Persian re ages of the rebels, and, since there was not much translation of the main inscription DB from space available, the legends for Gaumâta (DBb) the Elamite text,74 but with some updates,75 and Àçina (DBc) had to be squeezed in under was finished and could also be engrave.d, It was the Elamite legend DBb. No Babylonian version puf on the dressed surface beneath the relief, of DBa was joined, since the free space under- although this face was not really suited for that neath and over the king had to be foreseen for purpose, since water regularly poured down from the—as yet unwritten—inscriptions DBb (OP) several gaps in the rock (see n. 62). Since the and DBa (OP) respectively,· as to the Babylonian Bisitun monument, however impressive, was version of DB, it was added to the left of the barely visible from below, at the foot of the relief on the prominent slope, where it seems mountain (but see Voigtlander 1978, 1), Darius awkward.68 ordained that copies of (both? the relief76 and) Although the Babylonian texts apparently the inscription DB be made and that they be were not part of the original plan, they cannot sent out everywhere in the empire, have been engraved that much later and must As work was being done on the Old Persian have been executed during the first major stage text, news arrived that the followers of Darius of works on the Bisitun monument. As was had gained the upper hand over the Scythian noted by Cameron 1960, 60, the style in which rebel Skunkha. This required a last modification the Babylonian version was carved is very simi- to the monument, which Darius, by some ca lar to the style employed in the first Elamite price, wanted to see realized: Skunkha's image version and clearly differs from the style em- was finally added at the right of the relief,77 ployed not only in the Old Persian text but also which partially disturbed the Elamite text78 so in the second Elamite version: in both the orig- that an—almost exact—copy had to be engraved inal Elamite version and the Babylonian version again under the Babylonian version, to the left the cuneiform signs were "less skilfully and of the Old Persian text.79 As a final change, a less deeply incised" than in the other two.69 At fifth column, relating the events of Darius' sec
58
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
ond and third regnal years was added at the right 1974, 66-77; Herrenschmidt 1989; Malbran-Labat of the Old Persian text before construction 1992; Grillot-Susini, Herrenschmidt, and Malbran Labat 58f. closed after well over two years of work. A few 1993, 38, 8. As has been shown Benveniste months later, Darius paid a visit to the long by 1951, 42-47, this word means the thin clay envelope used to pro awaited monument (see n. 55). tect unbaked clay tablets (see also Harmatta 1966, 275-77 and Lecoq 1974, 8If.). 9. This restoration of the word was proposed by Notes Harmatta 1966, 280 (see also Schmitt 1991, 74). 10. The incorrect transliterated form */niyapin 1. This article is a revised and expanded version of 0iya/ of Schmitt 1991, 45, has here been tacitly re a paper presented in German at a conference on "Ori placed by the rectified transliteration /niyapai0ya/; ent und Okzident im Kontext althistorischer Quellen Schmitt 1992a, 153 n. 50, himself had already noticed forschung," held at Innsbruck University (6-9 July and remedied the mistake. 2000). I would like to thank Prof. Rtidiger Schmitt 11. "Aryan" simply stands for "(an) Iranian (lan (Saarbrucken) and Prof. Nicholas Sims-Williams (Lon guage)," as is shown in line 3 of the Rabatak inscrip don) for giving me some useful comments on a first tion (see Sims-Williams 1998, 81), where Bactr. αριαο draft of the article, but the responsibility for the views "(in) Aryan" actually means "(in) Bactrian." expressed here remains entirely my own. 12. The meaning of this word has been much dis 2. For the oldest literature, see the comprehensive puted, but the significations "to learn" (viz. the writ survey by Lecoq 1974; papers published up to 1989 ing) or "to copy/to study" (viz. the inscription) seem have been summarized by Mayrhofer 1989. Among rather improbable, as was correctly put by Harmatta more recent studies, see Herrenschmidt 1989; Stronach 1966, 270f. (Elam.), 8If. (OP); differently Lecoq 1974, 1990; and Malbran-Labat 1992. 75-77. On this matter see also Schmitt 1991, 74. 3. In various respects this text edition means enor 13. Due to lack of space, the Elamite text was not mous progress as compared to the previous standard added at the end of the main text but engraved as a edition by Kent 1953, 116-34 (see the "Addenda to separate text (DBl) over the Elamite version of the Kent's 'Lexicon'" and the "Corrigenda and delenda to short inscription DBa and to the left of the Old Per Kent's 'Lexicon'" in Schmitt 1991, 8If.); Kent's edi sian version of DBa (for more details see part 3). tion was based largely on the observations made by 14. While Grillot-Susini, Herrenschmidt, and Mal George G. Cameron in the forties and fifties, which at bran-Labat 1993 give the text as one block (par. 55), that time had enabled considerable improvements in I—like others before me—have divided the text into the edition of the text (on this point see Schmitt 1991, meaningful entities (indicated by bold numbers), 25f.). The new readings and corrections by Schmitt are whereas the actual length of the lines as engraved explained and argued for in greater detail in Schmitt into the rock is indicated by slashes. 1990. Still, even Schmitt 1991, 74, stresses that "I my 15. Although admittedly they are not completely self regard neither the restoration nor the translation satisfying, I adopt the supplements DB IV 90 given here for this last paragraph as final." m\ and [n-a-m-n-a\-f-m from Schmitt 1991, 74; I 4. Thus for example Schmitt 1990, 56-60, and share his reservations on the correctness of the recon Schmitt 1991, 73f. The idea had already been put forth structed forms but have nothing better to propose. earlier on several occasions by Walther Hinz (see be The problem with these two words differs from the low n. 19). other difficulties in the text, however; whereas it is 5. Such is the final conclusion of Lecoq 1974, 103, true that the exact forms of the original cannot (yet) and others. be reconstructed, there can be absolutely no doubt 6. Among the most prominent supporters of this about the rough meaning of those words, and at any theory are Igor M. D'jakonov and Ilya Gershevitch; rate their precise meaning is not necessary for a gen for further details see the summary in Mayrhofer eral understanding of the phrase. 1989, 178f. 16. It was Lazard 1976, 182, who correctly drew the 7. Since many difficulties of both the Elamite and attention of his colleagues to this point, which was the Old Persian versions of par. 70 have already been then further elaborated by Schmitt 1990, 59f. word has often been translated adequately dealt with, there is no need for a detailed 17. Although this the adverbial as study of every element of those passages here. For the as "an other one," interpretation Old Persian text, see Harmatta 1966, 272-83; Schmitt "otherwise" seems preferable, as has been argued by 1991, 73f.; and esp. idem, 1990, 56-60; for the Elamite Harmatta 1966, 263-65, and Herrenschmidt 1989, version of the passage, Harmatta 1966, 258-72; Lecoq 198-201. Still another solution has been adopted by
59
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Malbran-Labat 1992, 67: "sur un autre (matériau) . . . course to DBl that the phrase refers, and not to DB (à savoir) sur argile et sur peau." [OP], par. 70). Strangely enough, this detail must have 18. As an immediate consequence of this and no escaped Malbran-Labat's notice. ~ matter how OP /dipiciçam/ Elam. htup-pi-me should 23. I am not sure whether the short phrase [5] in the be translated, it can be safely excluded that Darius Elamite version DBl relates to the preceding sentence wanted to stress the simple existence of such a long or, on the contrary, to the following. Both answers royal inscription in itself as something quite new (for seem grammatically possible, but in my translation I more details see Nylander 1967, 143 n. 20). have opted for the former solution, which would bet 19. This interpretation, which was probably put ter preserve the parallelism with the Old Persian text. forth for the first time by Willy Foy (in ZDMG 52 24. The extra mention of this point looks more log [1898], 564), has been supported many times by ical and less trivial, if Darius is now talking about an Walther Hinz,· see, for example, Hinz 1942, 343-49; other inscription (DBa), rather than the one of which and, much later, e.g., Hinz and Koch 1987, vol. 2, 365. he was just speaking (DB). 20. Suffice it here to refer to Harmatta 1966, 255 25. According to Triimpelmann 1967, 298, this fact 63; Nylander 1967, 143 n. 19; and Lecoq 1974, 67-73, can be explained thus, that the innovation of the Old who carefully studied all other instances where the Persian script was obvious to the Iranians, but not to word occurs in Old, New, and achaemenid Elamite. the Elamites, who were familiar with their own script 21. Herrenschmidt, 197f., has also taken up this only. idea of Lazard's, but her attempt to connect /patisam 26. When Darius says "afterwards I sent out this kr-/ (which she understands as "reproduire") with OP (OP)/(that) same (Elam.) version," that can only mean /patikara-/ "image" (litt. "counter-feit") does not con "after the Old Persian text was redacted," but before vince me. it was engraved on the rock. This interpretation is the 22. For similar Middle Persian and Parthian verb only logical one, for why would Darius have spoken phrases of an adjective in predicative function + kar of copies of the Old Persian text that had already been dan/kirdan functioning as a denominative verb see sent out into the empire, if he had not yet written the Brunner 1977, 22-24. It should be mentioned here text? that the Elamite version differs in another point from 27. On this Aramaic text, which must be an the Old Persian text. After hai-ii-ia-ma, the Elamite abridged version of an older Aramaic copy, see text inserts a short phrase beginning with ap-pa, Greenfield and Porten 1982. which is absent from the Old Persian text. Since the 28. As is well known, the Elamite and the Old Per detailed explanations of Harmatta 1966, 265-69, no sian texts are closely connected, while the Aramaic serious scholar will challenge his statement that the and the Babylonian versions agree. The divergence subordinate clause introduced by Elam. ap-pa in DBl between the two pairs mainly concerns the use of 4 relates to the phrase preceding it, that is, DBl 3f. 'ύ Semitic month names and of "Median" forms of htup-pi-me da-a-e-ik-ki hu-ut-tà hai-ii-ia-ma. Accord names, and the including of precise numbers of the ingly, the phrase used to be translated for a long time enemies killed or captured in the latter pair (see, as ". . ., which formerly [sâ-is-sà] was not (in-ni)." among others, Schmitt 1998, 162). Herrenschmidt 1989, 201-4, was the first to contest 29. These fragments were given the inventory this translation; unlike all others before her, she numbers BE 3627 (later renumbered Berlin VA Bab. joined the two words sà-is-sà and in-ni into one and 1502) and Bab. 41446 (now lost?),· for further details considered in-ni to be a suffix forming adjectives see Voigtlander 1978, 63-66. (". . ., which existed before"). Malbran-Labat 1992, 30. I have not yet seen the article by P. Lecoq, 67, and also in Grillot-Susini, Herrenschmidt, and "Le chapitre 70 de l'inscription de Bisotun: Nouvel Malbran-Labat 1993, 59, with n. 162—obviously in essai d'interprétation," which the author announced disagreement with her two colleagues—goes even fur as "à paraître" in Lecoq 1997, 286, but I presume that ther in that she accords a locative ("supérieur," "du his new interpretation will be the same as the one dessus") value instead of a temporal ("antérieur") to proposed ibid, on pp. 212f. (where it was presented sà-is-sà (". . ., which is above"). A weak point of her without any detailed commentary to justify his trans is that this kind of hypothesis "forme à valeur ad lation; for this reason, I refrain from commenting on as it is called jective" appears to be quite frequent in it in a detailed manner here). Lecoq's translation of Medio-Elamite, but more or less exceptional in achae the Old Persian text runs thus: "Le roi Darius déclare: menid Elamite,· and even if a suffix in-ni were still in 'Grâce à Ahuramazdâ, voici le texte que j'ai traduit use in Darius' time, the main difficulty remains that [this apparently renders /patisam kr-/, my remark] en this does not interpretation make sense in the present aryen; et sur tablette et sur cuir, il avait été traduit since the Old Persian context, |= Aryan) version of DB aussi; j'ai traduit ma généalogie; je l'ai approuvée is not over DBl but far below it is of engraved (it [sic]; et cela a été écrit et lu devant moi; ensuite, j'ai
60
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
envoyé ce texte partout parmi les peuples [sic]; 43. The only exception appears to be the sign for " l'armée [sic] y a collaboré/ /la/, which was taken over from the corresponding 31. This idea was proposed by Hallock 1970 and Neo-Babylonian or Elamite signs, as first suggested by supported by Mayrhofer 1979, who still adhered to it Paper 1956 and now generally accepted. in 1989, 181. 44. No oblique wedges were used, with the excep 32. The same objection was made by Lecoq 1983,38. tion of the word divider; angle-shaped wedges have 33. The sign ν" would have been required for exam their openings exclusively to the right. ple in x-s-a-y-d-i-y v-z-r-k /xsàyaOiya vazrka/ "great 45. The complete list of words and names in King," a title of which the Babylonian equivalent was which C: and Cu-signs are used is as follows: v'-s-t employed by Cyrus in the foundation cylinder. a-s-p /Visptàspa/ and declined forms (DBa 3, 5, 5f.); 34. I shall return to this point in the final part of my a-'v'-h-y-r-a-d'-i-y /avahyaràdï/ "for that reason" (DBa paper. Suffice it here to say that this sort of criticism 9f.); p-ru-vi-i-y-t /paruviyata/ "from [hacâ] ancient was already advanced by Windfuhr 1970, 125 ("Ele times" (DBa 11); du-u-\i-i-t-a-p-r-n-m /duvitàparanam/ mentary estimates of probability exclude the possibil "now as ever" (DBa 17); m-gu-u-s /magus/ "magus" ity that such a system is naturally or historically (DBb 2); <3-d"-u-ru-u-j'-i-y /adurujiya/ "lied"(DBb 2f.; developed") and elaborated by Stronach 1990, 197. DBc 2f.; DBd 2; DBe 3f.; DBf lf.; DBg 2f.; DBh 2f.; 35. On this badly damaged inscription see Nylander DBi 2f.; DBj 2f.); b-r-ti-i-y /Brdiya/ (DBb 4f.; DBh 5); 1967,170-73, or—for a quick orientation—Lecoq 1974, a-m'-i-y /ami/ "(I) am" (DBb 5, 7; DBc 8f.; DBd 4f.; 57f. For a hypothetical reconstruction of the fragmen DBe 6, 9; DBf 4f.; DBg 7f.; DBh 5f., 9; DBi 6, 10; DBj 5); tary text see Borger and Hinz 1959. ku-u-r-u-s /Kuraus/ (DBb 5f.; DBh 6f.); n-i-d'-i-t-b-i-r 36. Stronach has summarized his arguments in /Nidintabaira/ "Nidintu-Bël" (DBd 1); n-b-u-ka-u-d-r Stronach 1997, 351-54. Against the objection that c-t /Nabukudracara/ "Nebuchadnezzar" (DBd 3f.; DBi Cyrus must have felt it necessary to place prominent 5f.); s-ku-u-x /Skunxa/ (DBk If.). labels on his constructions, Stronach 1990, 198, re 46. These countries are [b]-a-b-i-tu-u-s /Bâbirus/ plies that "on the basis of the extant monuments at "Babylon" (DB 14), ma-u-d-r-a-y /Mudràya/ "Egypt" Pasargadae it can in fact be argued that Cyrus' main (DB 15), a-i-n^-i-n /Armina/ "Armenia" (DB 15), k-t concern was to find and combine certain traditional p-ta-u-k /Katpatuka/ "Cappadocia" (DB 15f.), u-v-a-r symbols of all-encompassing kingship which would z-m'-i-y /Uvârazmî/ "Chorasmia" (DB 16), [s-u-gu]-u-d suffice to underline the newly secured, imperial di /Suguda/ "Sogdiana" (DB 16), and 0-t-ga-u-s /Datagus/ mensions of Persian rule." The use of the simple title "Sattagydia" (DB 17). of "king" in CMa is also in sharp contrast with the 47. There seems to be no urgent reason for the complex royal titles of the Babylonian foundation cyl sign nu; was it created through inadvertence? Or inder; this equally points towards the authorship of was it needed to avoid confusion between *a-n-u-s Darius (1990, 199f.). */anausa-/ "immortal" and a-n"-u-s-i-y /anusiya/ "fol 37. On this point see also Rollinger 1998, 195 with lower" (DB II 95, etc.), as was suggested by Mayrhofer n. 202. 1979, 295f. and 1989, 182 n. 30? This hypothesis 38. For the same reason, Darius finished construc seems more plausible than the one suggested by tion begun by Cyrus at the site of Pasargadae sev Werba 1982, 206, according to whom nu could have eral years after completion of the Bisitun inscription been used in deleted inscriptions of Brdiya-, whose (Waters 1996, 16). full name would have been *brdi-9anuvan-. But such 39. For a detailed survey of research on these in a reconstructed full name is "pura fantasia" (Schmitt scriptions until 1974 see Lecoq 1974, 48-52. 1992c, 131); the only other name that ever existed for 40. This thesis recently has been put forth by Bardiya/Smerdis was his nickname *Tanu-vazrka Lecoq 1997, 124-26. (Gr. Τανυοξάρκης [Ctesias], Ταναοξάρης [Xenophon]) 41. The "Old Assyrian signs" must then refer to "having a giant's body" (= physique). And apart from the Aramaic script, since the Greeks apparently were the fact that all external evidence seems to exclude not able to distinguish between the different forms of the existence of Old Persian cuneiform inscriptions cuneiform script. The alternative solution, that the before Darius, only throne names—no "real" names, "new" Old Persian cuneiform script is here opposed let alone nicknames—appear to have been used in to the "old" Babylonian and Elamite scripts, can thus official royal inscriptions. be safely excluded (see Schmitt 1992b, 28-31). 48. See Mayrhofer 1989, 181 nn. 25f. for further 42. These four basic principles of sign composition details. to are: derivation of four basic components; addition of 49. As Nicholas Sims-Williams points out me, were one would a— 1, 2, or 3 wedges; threefold rotation; and mirroring. A if this explanation true, expect like in order to between number of justified critical remarks have been united lacking—sign p1 distinguish by Lecoq 1974, 86. a nominative form c-i-s-p-i-s /Cispis/ and a genitive
61
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script form c-i-s-p-i-s /Cispais/ (as in DBa 8); on the other strated by Borger 1982, 123f. And since in DB (OP), hand, the inventors of the script had soon clearly rec par. 52 [= IV 5], Darius adds the meaningful sentence ognized the problem themselves, since in the—later /pasâva ya0â xsâya0iya abavam/ "after that I became (see above and part 3)—main inscription DB I 5f., the king," Triimpelmann 1967, 296, rightly points out genitive form is henceforth written c-i-s-p-a-i-s /Cis that Gaumàta remained the ruler in power as long as pais/ and was thus clearly distinguished from the he lived, and it was only after his death that Darius nominative form! could make a claim to the throne; to the reader of the 50. Schmitt 1990, 25-28, ascribes this inconsis inscription it is thus obvious that the events within tency within the same inscription to the fact that one year after he came to power to which Darius different stonemasons had been at work. Since the refers can only have happened after the murder of stonemasons were presumably illiterate (see Schmeja Gaumâta. 1992, 190f.), it is more likely that these discrepancies 54. Instead of the traditional five or six phases, go back to the scribes. Lecoq 1997, 85-87, proposed to reduce their number 51. Less than a decade before, the reasoning of to two or three, but some of the details of his recon Lecoq 1974, 66, sounded quite different; while criti struction are incompatible with the archaeological cizing Hinz's theory that the inventors of the Old Per data: the original conception would include the relief sian script were pressed by time, he objected that and the Old Persian and Elamite texts, while the "même si Darius en est l'inventeur, on ne peut nier Babylonian version would have been realized either que cette écriture suppose un long tiavail de réflexion more or less at the same time or shortly thereafter,· [my italics, Ph. H.) et de comparaison entre les autres the second (or: third) phase is constituted by the ad systèmes graphiques." dition of the Scythian rebel to the relief and of the 52. Luschey and Triimpelmann had both studied minor inscriptions. the relief from a scaffolding during the winter 1963/ 55. According to Luschey 1968, 94, the lower layer 1964. For a first short survey of the results, see may coincide with a visit of the monument by Darius "Bïsotûn" in Eh, vol. 4 (1990), 289-305 (esp. 300b at a later time when the monument was completely 302b and the synoptic table on p. 301 [after Triimpel finished. mann 1967]); for a quick orientation, see Lecoq 1974, 56. Since Darius pretends to have realized all of his 64f. (after Hinz 1968). exploits within "one" year (first in DB [OP], par. 52, 53. Nagel, seeking an explanation for the diverg and then four more times in pars. 56, 57, 59, 62), it ing order of the rebels in the inscription and on the re suffices when he gives the days and the months when lief, suggested that the second from the last figure in dating the end of the revolts; depending on whether the row of rebels may also have been a later addition one takes this one year to be his year of accession to (between phases IV and V of Trtimpelmann's recon the throne (522/521 b.c.) or his first full regnal year struction,· see the table in fig. 4). Nagel's paper is not (521/520 b.c.), two conversion dates are possible for convincing: according to the description of Skunkha the 23d day of the IX month, on which Frâda was and others by Luschey 1968, 79f., there is clearly a beaten, namely 10 December 522 and 28 December stylistic difference between the—flatter—image of the 521, of which the last is far more likely (see Borger Scythian with the pointed cap and the—bareheaded— 1982, 119f. Otherwise, the repression of Frâda would other rebels. If the image of Frâda (= AGamaita accord have followed soon after the slaying of Gaumâta, and ing to Nagel) had been created at the same time as as Borger correctly states (ibid., 119), at that time that of Skunkha, one should expect a similar stylistic what was happening in far-off Bactria and Margiana difference between his image and the preceding fig could hardly have been a major concern to Darius, ures, but I cannot see this, whereas the difference be and, conversely, Dâdarsi, the then Bactrian satrap, tween Skunkha and the others is quite obvious. could hardly have been interested in a war against Fràda's strange outfit remains a problem, but the so Margiana, as long as it remained uncertain as to lution offered by Nagel is probably not the right one whether Darius' doubtful claims to the throne would (regarding Frâda see also the pertinent remarks by be accepted. Borger 1982, 120-22, and esp. n. 24). That AGamaita 57. On the choice of the place see Luschey 1968, is not portrayed also remains a riddle, but since the 66f. number nine an plays important "magic" role—Da 58. This was already recognized by Herzfeld; for rius claims to have been the ninth king in a row of further details see Luschey 1968, 68 (with n. 19) and his family—one might imagine that Darius did not 84f. On the (artistic) provenance of the sculptor of the want to have more than nine rebels portrayed, and relief see Luschey, 87-89. preferred to include the image of the last rebel 59. See Luschey 1968, 91 ("eine klare Komposi Skunkha. Gaumâta is a case apart at any rate, since tion"), and Triimpelmann 1967, 286 ("unmittelbar his defeat does not to belong the "one" year of Darius zusammengehôrig"). The "niching" technique used which see also below n. (on 56), as was demon by the sculptors is described by Cameron 1960, 60
62
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
η. 4; at some points, the surface was in such poor Trùmpelmann 1967, 287, and esp. Borger 1982, 115). shape that bad portions had to be cut out and replaced What determines the chronology on the relief is the by new insets (see Cameron, ibid. , and the technical date at which each of the rebels was defeated, not the remarks in Luschey 1968, 83). time of their death (ibid., 128). According to Borger 60. Large blocks, blasted from the rock above, 1982, 127, a plausible explanation for this divergence could still be seen below on the platform in modern between relief and inscription might be that the chro times (see Trtimpelmann 1967, 284); and holes to nological overlapping of so many events within a short blast or break up the rock were also discovered on the period in many parts of the empire made an account platform (see the drawing in Luschey 1968, 65 fig. 2). according to geographical order look more advisable. According to Trtimpelmann (ibid.) the stonemasons 65. Both the Elamite and the Babylonian versions did not work from this lower platform, as Luschey show strong influences from Old Persian in their syn says, but from another natural platform which must tax and lexicon (see Schmitt 1991, 19b, with further have been at the same level of the horizontal fissure literature). under the relief and which was broken away once 66. In DB, par. 52 (OP IV 5f.), there is talk of 19 bat the construction works were done. Luschey's theory tles that Darius claims to have fought, but the Ara seems far more credible. maic version is the only one in which precisely 19 61. The transition between relief and text is very battles are summed up. According to Borger 1982, smooth, and the Elamite legend engraved over Fràda 130, this may be an indication that the Aramaic ver is slightly bent downwards (the curve is clearly visi sion was indeed one of the first to be written down. ble in Schmitt 1991, pi. 34). This means—as was And even the very fact of including battlefield statis rightly noticed by Borger 1982, 121 n. 24, in another tics is an unusual feature in neo-Babylonian inscrip context—that the legend DBj (Elam.) can only have tions and chronicle texts (see Voigtlander 1978, 8), so been carved into the rock after the preparation of the that this too may be a further sign that the Babylo surface for the inscription DB (Elam.) was finished. nian text is based on an "older" Aramaic version (and Finally, the upper margins of both the relief and the not vice-versa), as are the frequent Aramaisms in the first column of the DB (Elam.) coincide (see Triimpel Babylonian text (see Voigtlander 1978, 7, who alterna mann 1967, 287). Trtimpelmann 1967, 287, 294f., and tively ascribes the Aramaisms to the idiom of the dic Hinz 1968, 97, argued that the surface on top of the tating Darius, who might have acquired a working relief was enlarged at a later stage, in order to make knowledge of Aramaic while he was a young army place for the inscriptions DBl (Elam.) and DBa (OP), officer in Egypt [cf. Hdt. 3.139]). but Luschey 1968, 67f., assumed—more plausibly to 67. Some minor adaptations to local circumstances me—that this space was smoothed from the begin can be seen, for example, in the Elamite version, ning and that it was on the contrary the irregular where Ahura Mazda is characterized in addition as form of the rock over the figure of Ahura Mazda "God of the Aryans," for this and further examples which was accepted as such by the stonemasons. see Schmitt 1980, 120, and 1998, 162f. 62. That the surface was smoothed out over the 68. The stonemasons somewhat miscalculated the whole length can be guessed from a close look at space needed for the Babylonian text, so that they had Schmitt 1991, pi. 2, but, to be certain, an investiga to dress a larger surface; they continued to write in tion on the spot would be required. That the surface one column, but from line 36 to the end in line 112, below was originally not intended to be inscribed at the lines are much longer (over 4 m). all, since it was totally unsuited for it, can be de 69. The signs were also smaller in the first Elamite duced from the detailed description in Schmitt 1991, version (2 cm) than in the second Elamite and the 22 (see also Voigtlander 1978, 4). From the natural Old Persian versions (2.8 cm) (see Cameron 1960, 60); horizontal fissure—"actually the exit of a kind of un in the Babylonian version, the line height is "two 4 derground river which flows in the spaces between fingers," that is, varying between 3 and cm (see the transverse geological layers of the whole rock Voigtlander 1978, 3). massif" (Schmitt 1991, 22b)—underneath the relief 70. According to Voigtlander 1978, 2f., the original and the Elamite text of DB streams of water gushed intention to cut away the buttress-like rock can still idea was soon aban out after every rain shower, which in winter froze to be seen in photographs, but that ice. It must have been obvious to the workers that doned because of the unstable nature of the rock, and behind it. serious damage was to be expected if the surface un because of the discovery of a deep cavity der the fissure was ever inscribed. 71. The Elamite version of DB has one paragraph 63. From DB (OP), pars. 65-67 it is clear that less, as DB (Babyl.) section 55 is part of DB (Elam.) Darius was worried about eventual acts of vandalism. section 54 (see Voigtlander 1978, 62 n. 2). reads in litera 64. The rebels are arranged in chronological order 72. Unlike what one occasionally on the relief, while the order in the inscription is ture (see, e.g., Herrenschmidt 1989, 195; Lecoq 1974, lack of clearly geographical first and then chronological (see 65, and 1997, 86; Hinz 1968, 97; etc.), space
63
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
was probably not the reason why there is no Babylo 79. See the description by Luschey 1968, 91, who nian version of par. 70 (see Voigtlander 1978, 1 and 3 notes that the new Elamite text looks like a huge clay [cf. Schmitt 1980, 112 n. 21]). It is true that the Baby tablet put against the rock. It is possible, though less lonian engravers had spaced out the letters of the last likely, that stages II 2a and II 2b (see the table in fig. line 112, so as to fill out the whole line, but there still 5) must be inverted as suggested by Triimpelmann was enough space underneath it for a few more lines. 1967, 294; if there existed no copy at all of the original 73. That this Elamite text DB1 was inscribed before Elamite text on clay tablets—which I find hard to the Old Persian versions of DBa (and the legends DBb imagine—then the second version had to be copied by j) can be deduced from the fact that it was nicely the first one, and that could of course only have hap aligned with the left side of DBa (Elam.); the stonecut pened before the first column was destroyed by the ter of the Old Persian text DBa could not possibly sculptors of Skunkha's figure. have known how much space was to be provided for the Elamite text, if this text had been engraved first. From the "squeezed-in" positions of especially DBd-f References it is evident that the Old Persian legends were in scribed still later than the ones Lu Babylonian (see AiW C. Bartholomae, Altiranisches and Hinz but that DBa and schey 1968, 91, 1968, 96), Worterbuch. Strassburg, 1904. were before the main DB DBb-j engraved inscription Benveniste 1951 E. Benveniste. "Etudes sur le becomes clear when one recalls that the invention of vieux-perse." Bulletin de la the cuneiform was based on DBa script presumably Société de Linguistique de and the legends and that the orthographic conven Paris 47:21-51. tions were not yet entirely matured. Borger 1982 R. Borger. Die Chronologie des From the fact that the 74. stereotype introductory Darius-Denkmals am Behis formula OP is used less /9àti Dàrayavaus xsâyaBiya/ tun-Felsen. Gôttingen. often in the Elamite and Babylonian versions, Voigt Borger and R. Borger and W. Hinz. 6f. lander 1978, (cf. Schmitt 1980, 107), concludes Hinz 1959 "Eine Dareios-Inschrift aus that the Old Persian text may have been retranslated Pasargadae." ZDMG from one of these two or from a third unknown (Ara 109:117-27. maic?). A strong indication that the Old Persian text Brunner 1977 C. J. Brunner. A Syntax of was indeed a retranslation of the Elamite text be may Western Middle Iranian. seen in DB (OP) IV 33, where the unusual word order Delmar, New York. in the Old Persian text is normal in Elamite quite (see Cameron 1960 G. G. Cameron. "The Elamite Schmitt 1991, 19a and 69 1998, [and 161]). Version of the Bisitun Inscrip 75. For further details see, for example, Hinz 1968, tions." Journal of Cuneiform 97f.; Trumpelmann 1967, 291; and Schmitt 1991, 19a Studies 14:59-68. and 69 on DB IV where the [and 1998, 161] (OP) 39f., Ghirshman 1965 R. Ghirshman. "A propos de form can be "original" imperative */astu/ (which l'écriture cunéiforme vieux deduced from the Elamite text) was replaced by the perse." JNES 24:244-50. subjunctive /ahati/. Greenfield and ). v.. ureenneia ana ts. 1'orten. 76. Seidl 1976 attempted to identify some blocks Porten 1982 The Bisitun Inscription of found in as of a of the Bisitun Babylon fragments copy Darius the Great: Aramaic relief; a fragment of a colored brick which was at Version. Cllr, pt. 1, vol. 5, some time of as to another such thought belonging Texts 1. London. copy turned out to be something else, as was shown Grillot-Susini, F. Grillot-Susini, C. Herren by Jacobs 1997. Herrenschmidt, and schmidt, and F. Malbran-Labat. 77. The most detailed of the description technique Malbran-Labat 1993 "La version élamite de la ("a sharp inward slant") used for sculpting Skunkha trilingue de Béhistun: Une (not of the figure itself, for which see Luschey 1968, nouvelle lecture." JA 75 and and is 79f.; Nagel 1983, 185f.) provided by 281:19-59. Cameron 1960, 60 n. 5. Hallock 1970 R. T. Hallock. "On the Old 78. While Cameron 60 n. that the 1960, 4, says Persian Signs." JNES 29:52-55. original Elamite text was simply abandoned, Luschey Harmatta 1966 J. Harmatta. "The Bisitun 1968, 91 ("getilgt"), and Trumpelmann 1967, 294, are Inscription and the Intro affirmative that it was erased for the really (except duction of the Old Persian three Elamite lines under the figure of Skunkha, Cuneiform Script." AAnASH which were apparently overlooked). 14:255-83.
64
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Herrenschmidt 1979 C. Herrenschmidt. "La Perse Assyriologiques Brèves et rien que la Perse: Essai sur Utilitaires, notice 86, 66f. la royauté d'Ariaramnès et Mayrhofer 1979 M. Mayrhofer. "Uberlegungen d'Arsamès." In Pad nâm i zur Entstehung der altpersi yazdân: Etudes d'épigiaphie, schen Keilschrift." BSOAS de numismatique et d'histoire 42:290-96. de l'Iran Ancien, 5-21. Paris. Mayrhofer 1989 . "ËJber die Verschrif Herrenschmidt 1989 . "Le paragraphe 70 de tung des Altpersischen." l'inscription de Bisotun." In Historische Sprachforschung Etudes irano-aryennes offertes 102:174-86 (repr. in Ausge à Gilbert Lazard, ed. C.-H. de wàhlte Kleine Schriften, vol. 2, Fouchécour and Ph. Gignoux, Festgabe fur Manfred Mayrho 193-208. Stir, cahier 7. Paris. fer zum 70. Geburtstag, 387 Hinz 1942 W. Hinz. "Zur Behistun 99 [Wiesbaden, 1996]). Inschrift des Dareios." ZDMG Nagel 1979 W. Nagel. "Pasargadae: Ein 96:326-49. Lagebericht zum Problem des Hinz 1968 . "Die Entstehung der Beginns Achâmenidischer altpersischen Keilschrift." Kunst und altpersischer AMI, N.F. 1:95-98. Schrift." Mitteilungen der Hinz 1973 . Neue Wege im Altper Deutschen Orientgesellschaft sischen. Wiesbaden. 111:75-88.
Hinz and Koch 1987 W. Hinz and H. Koch. Elam Nagel 1983 . "Frada, Skuncha und isches Wôrterbuch. Vol. 2. der Saken-Feldzug des Darius Berlin. I." In Kunst, Kultur und Hoffmann 1976 K. Hoffmann. "Zur altpersi Geschichte der Achàmeni schen Schrift." In Aufsàtze zur denzeit und ihr Fortleben, ed. Indoiranistik, vol. 2, 620-45. H. Koch and D. N. MacKenzie, Wiesbaden. 169-89. Berlin.
Jacobs 1997 Β. Jacobs. "Eine weitere Kopie Nylander 1967 C. Nylander. "Who Wrote des Brsutun-Reliefs? Zu einem the Inscriptions at Pasar Reliefziegel aus Susa." gadae?" Orientalia Suecana Archàologische Mitteilungen 16:135-80. aus Iran und Turan 29:303-8. Paper 1956 H. H. Paper. "The Old-Persian Kent 1953 R. G. Kent. Old Persian: /l/-Phoneme." JAOS 76:24-26. Grammar. Texts. Lexicon. Rollinger 1998 R. Rollinger. "Der Stamm AOS 33. 2d ed. New Haven. baum des achaimenidischen
Lazard 1976 G. Lazard. "Notes de vieux Kônigshauses oder die Frage perse." Bulletin de la Société der Legitimitât der Herrschaft de Linguistique de Paris des Dareios." Archâologische 71:175-92. Mitteilungen aus Iran und Lecoq 1974 P. Lecoq. "Le problème de Turan 30:155-209. l'écriture cunéiforme vieux Schmeja 1992 H. Schmeja. Review of Schmitt perse." In Actlr 3, 25-107. 1990. Kratylos 37:190-91. Téhéran and Liège. Schmitt 1980 R. Schmitt. "Zur babyloni Lecoq 1983 . "Observations sur schen Version der Blsutun l'écriture vieux perse." In Inschrift." AfO 27:106-26. Iranian Studies, ed. G. Gnoli. Schmitt 1988 . "Achaimenidenin Orientalia Romana, vol. 5, schriften in griechischer lite 31-39. SOR 52. Rome. rarischer Uberlieferung." In
Lecoq 1997 . Les inscriptions de la Actlr 28,17-38. Téhéran and Perse achéménide. Paris. Liège. Luschey 1968 H. Luschey. "Studien zu dem Schmitt 1990 . Epigraphisch Darius-Relief von Bisutun." exegetische Noten zu AMI, N.F. 1:63-94. Dareios' Bîsutun Inschriften. Malbran-Labat 1992 F. Malbran-Labat. "Note sur le Sitzungsberichte der Ôster §70 de Behistoun." Nouvelles reichischen Akademie der
65
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions η υ y s ε: The Bisitun Monument and the Genesis of the Old Persian Cuneiform Script
Wïssenschaften, Phil.-Hist. Taktu in Chinese." In Pro Kl., 561. Vienna. ceedings of the Third European Schmitt 1991 . The Bisitun Inscrip Conference of Iranian Studies. tions of Darius the Great: Old Pt. 1, Old and Middle Iranian Persian Text. Cllr, pt. 1, vol. 1, Studies, 79-92. Wiesbaden. Texts 1. London. Stronach 1978 D. Stronach. Pasargadae: A Schmitt 1992a . "Zum Schluss von Report on the Excavations " Dareios' Inschrift 'Susa e.' Conducted hy the British AMI, N.F. 25, 1992 [1993], Institute of Persian Studies 147-54. from 1961 to 1963. Oxford. Schmitt 1992b . "Assyria grammata Stronach 1990 . "On the Genesis of und âhnliche: Was wufiten die the Old Persian Cuneiform
Griechen von Keilschrift und Script." In Contribution à Keilinschriften?" In Zum l'histoire de l'Iran. Mélanges Umgang mit fremden offerts à Jean Perrot, 195-211. Sprachen in der griechisch Paris. rômischen Antike, ed. C. W. Stronach 1997 . "Darius at Pasargadae: Muller, K. Sier, and J. Werner, A Neglected Source for the 21-35. Stuttgart. History of Early Persia." Topoi, Schmitt 1992c . "Note sul soprannome suppl. 1, Recherches récentes di Smerdis ΤΑΝΥΟΞΑΡΚΗΣ." sur l'empire achéménide, Archivio glottologico italiano 351-63. Lyon. 67:122-32. Trùmpelmann 1967 L. Triimpelmann. "Zur Entste Schmitt 1998 . "Ubersetzung im hungsgeschichte des Monu Dienst der Politik: Die mehr mentes Dareios' I. von Bisutun
sprachigen Kônigsinschriften und zur Datierung der Ein im Achaimenidenreich." In fiihrung der altpersischen Wort, Text, Sprache und Schrift." AA 82:281-98. Kultur: Festschrift far Hans Voigtlander 1978 E. N. von Voigtlander. The Schmeja zum 65. Geburtstag, Bisitun Inscription of Darius ed. P. Anreiter and H. M. the Great: Babylonian Version. Olberg, 157-65. Innsbrucker Cllr, pt. 1, vol. 2, Texts 1. Beitrage zur Kulturwissen London. schaft, Sonderheft 103. Waters 1996 M. W. Waters. "Darius and the Innsbruck. Achaemenid Line." Ancient Schmitt 1999 . Beitrage zu altpersi History Bulletin 10:11-18. schen Inschriften. Wiesbaden, Werba 1982 C. Werba. "Die arischen 1999. Personennamen und ihre Seidl 1976 U. Seidl. "Ein Relief Dareios' I. Trâger bei den Alexanderhis in Babylon." AMI, N.F. 9:125 torikern (Studien zur irani 30. schen Anthroponomastik)." Sims-Williams 1998 N. Sims-Williams. "Further Vienna, unpub. diss. Notes on the Bactrian In Windfuhr 1970 G. L. Windfuhr. "Notes on
scription of Rabatak, with an the Old Persian Signs." Indo Appendix on the Names of Iranian Journal 12:121-25. Kujula Kadphises and Vima
66
This content downloaded from 130.216.158.78 on Mon, 14 Dec 2015 03:38:31 UTC All use subject to JSTOR Terms and Conditions chapter 5 Old Persian rudiger¨ schmitt
1. HISTORICAL AND CULTURAL CONTEXTS
Old Persian is one of two Old Iranian languages which are attested in the Achaemenid royal inscriptions (see below), members of that branch of the Indo-European language family called Indo-Iranian, or Aryan (the Persians designate themselves and their language by the term ariya-). The Iranian languages began to take shape when the ancestors of the Indo- Aryans left the common homeland in the steppes of Central Asia in the first half of the second millennium BC. The Western Iranian peoples, the Medes who settled in Media and the Persians in Fars¯ (speaking a Northwestern and Southwestern Iraniandialect respectively), step into the light of history in the ninth century BC, when Median names are first attested in Assyrian documents. While “Old Persian” was certainly the language of Fars,¯ the variety which is attested in the Achaemenid inscriptions appears to be a rather artificial idiom, peppered with dialectal and archaic words, unlike any dialect actually spoken (characteristics of a distinct spoken Old Persian may be discerned from certain spontaneous phonetic developments, and from Old Persian words and names as rendered in other languages). The language called Old Persian was thus restricted to royal usage (as was the cuneiform script in which Old Persian was recorded). Even so, Old Persian was neither the lingua franca nor the administrative language of the Achaemenid Empire, roles fulfilled by Aramaic and, to a limited extent, various regional languages spoken within the empire. As a consequence, the linguistic situation of the empire was a quite complex one; and epigraphical Old Persian was itself influenced by theseotherlanguages,particularlyinitsvocabularyandevensyntax(e.g.,intheoccurrenceof a postpositive genitive, as in xˇsaya¯ ϑiya xˇs¯ayaϑiyan¯ am¯ “king of kings” or vaˇsn¯a Auramazd¯aha “by the favor of Auramazda”).¯ The language of the Old Persian inscriptions is dialectologically homogeneous in princi- ple. Only some lexical items (technical terms, etc.) prove to be borrowed from other Iranian languages, mainly the Northwestern Iranian dialect of the Medes (see §6), the political predecessors of the Persian Achaemenids. The only direct and authentic sources available for the Old Persian language are the cuneiform inscriptions on durable objects (rock, stone, metal, rarely clay tablets) ranging over the period from Darius I (522–486 BC) to Artaxerxes III (359/8–338/7 BC), but dating in the main from the reigns of Darius I and Xerxes I (486–465 BC). In this short period the inscriptions, for the most part, are trilingual (in Old Persian, Elamite, and Babylonian), but even the oldest text, the one of the B¯ısutun¯ monument of Darius I (see below), has sections which are only in Old Persian, or in Old Persian and Elamite. With Artaxerxes I
76 old persian 77
(465–425/4 BC) the number, size, and significance of the texts begin to decrease rapidly, and they consist almost exclusively of stereotyped formulae, which, in part, seem to have been poorly understood at the time of composition. On the other hand, however, apart from their trilingualism, it is just this monotonous stereotyped style of the texts, along with the great number of parallel texts with their often-repeated invocations of the supreme god and with the regularly quoted royal titles, that has facilitated an understanding of the language and texts and which has allowed reconstruction of fragmentary texts. The abbreviatory system of citing texts is presented at the end of the chapter. The decreasing number of Old Persian texts after the reign of Xerxes I may be attributed to a loss of fluency with the royal language. By that period, spoken Persian had evolved into a somewhat different form, so discrepancies between everyday speech and the traditional language of inscriptions had arisen. Only upon that basis can the serious grammatical faults which appear in the texts of later Achaemenid kings (mainly of Artaxerxes II and III) be understood. Most of those “corrupt” forms (incorrect endings, hybrid genitive forms, etc.) can be found in the monolingual inscription A3Pa of Artaxerxes III; but they also occur in most of the inscriptions of Artaxerxes II and in the monolingual texts claiming to have been com- posed by Ariaramnes and Arsames in the sixth century BC (that these texts were produced under Artaxerxes III instead, is suggested by the fact that among the later Achaemenids it is only this king who derives his lineage from Arsames, and not only from Darius’ father Hystaspes). The use of a form like b¯um¯am in lieu of the expected accusative singular feminine b¯um¯˘ım “earth” can best be explained by positing an actually spoken monosyllabic [bu:m] (like Middle Persian b¯um) and a scribal attempt to “transform” the spoken form into an Old Persian one (an attempt which was rendered detectable by its lack of success, as it used the a¯-stems as the normal class of feminine nouns). A similar archaizing process is seen in the pseudo-Old Persian accusative singular ˇs¯ayat¯am for expected ˇsiy¯atim “happiness,” where the later form ˇs¯at has been changed into ˇs¯ayat- by reversing the regular sound change of Old Persian aya¯ to Middle Persian a¯ (though being inappropriate here) and adding again the ending -am¯ of the feminine a¯-stems.
2. WRITING SYSTEM
2.1 Graphemic shape and inventory Old Persian texts are recorded only in a cuneiform script. This script does not, however, directly continue the Mesopotamian cuneiform tradition (see WAL Ch. 8, §2), being similar to the other cuneiform systems only in the employment of “wedge-shaped” characters. In other words, the Old Persian script is not the result of an evolution of the Mesopotamian system, but a deliberate creation of the sixth century BC. It remains unclear why the Persians did not take over the Mesopotamian system in earlier times, as the Elamites and other peoples of the Near East had, and, for that matter, why the Persians did not adopt the Aramaic consonantal script (Aramaic being the lingua franca of the Persian Empire; see §1). Old Persian cuneiform was used only by the Achaemenid kings for two centuries and only for their own language – that is, the rather artificial literary language of their royal inscriptions. The use of this script was thus in effect a royal privilege. It was a splendid and imposing script best suited for hard surfaces, and apparently used neither for poetic texts nor for administrative nor historical writings. 78 The Ancient Languages of Asia and the Americas
Table 5.1 The Old Persian cuneiform script Syllabic symbols ai u ai u bcC dfghjklm b(a) c(a) c¸(a) d(a) f(a) g(a) h(a) j(a) k(a) l(a) m(a) npr sStˇvxyz n(a) p(a) r(a) s(a) sˇ(a) t(a) (a) v(a) x(a) y(a) z(a) ÎJμ V di ji mi vi DG N M0RT du gu ku mu nu ru tu Logograms 86 7 45
XSDHˇ 1 DH2 BG BU xsˇaya¯ qiya- dahyu- dahyu- baga- bum¯ ¯˘ı- “king” “land” “land” “god” “earth” 123
AM1 AM2 AMha Auramazda˜ Auramazda˜ Auramazdaha˜ (genitive singular)
The total number of phonetic characters (which consist of two to five single elements) is thirty-six. These are naturally divided into four groups:
(1) A. Three pure vowel (V) characters: a, i, u B. Twenty-two syllabic characters whose vowel component is a (Ca), but which can also be used to represent a consonant occurring before another consonant or in word-final position (C): b(a), c(a), ¸c(a), d(a), f (a), g(a), h(a), j (a), k(a), l(a), m(a), n(a), p(a), r(a), s(a), ˇs(a), t(a), ϑ(a), v(a), x(a), y(a), z(a) C. Four syllabic characters with inherent i vowel (Ci): di , j i , mi , vi D. Seven syllabic characters with inherent u vowel (Cu): du, gu, ku, mu, nu, ru, tu
In addition, there are eight logograms for commonly used words such as “king,” “god” or “land”; these are not obligatory and are not used consistently. The logograms are of a more complex shape, contain up to twelve elements and even show angles placed above angles (as is the case with the numerals). Further, a word-divider is used as well as number symbols (vertical wedges for the units, angles for the tens, and a special symbol for 100 (found in a single inscription). One of the remarkable stylistic features of Old Persian cuneiform is that the wedges and angles which make up the cuneiform symbols never cross. The attested characters (excluding the numerals and the word-divider) are presented in Table 5.1. Within the relatively short period of its use this writing system shows a few changes in character shapes – an attempted standardization of the height of those wedges which at first old persian 79
(i.e., in the B¯ısutun¯ text) took up only half the height of the line. However, the mechanics of the writing system (see below), with all its “imperfections,” remain unchanged.
2.2 Orthographic conventions As the set of CV characters with inherent i or u vowel shows, the inventory as a whole is inconsistent and asymmetric in its structure, for no ascertained reason (phonetic or otherwise):
(2) da ga ja ka ma na ra ta va di ji mi vi du gu ku mu nu ru tu Beyond this, there are no Ci and Cu characters of the form bi/u, ci/u, ¸ci/u, f i/u, hi/u, li/u, pi/u, si/u, ˇsi/u, ϑi/u, xi/u, yi/u, zi/u. Even if the writing system were not plagued by such omissions, the ambiguity of many spellings would not be eliminated; the entire group of Ca graphemes has its own affiliated spelling difficulties, which reveal that this writing system is neither phonemic nor phonetic. As a consequence of the preceding graphemic problems, a number of orthographic con- ventions had to be employed when particular phonemic sequences are written. The most important of these “rules” (to the extent that they can be identified with certainty) are the following: 1. Long vowels are not distinguished from short ones except for a¯ in medial position. 2. Proto-Iranian final ∗-a is written with an additional (i.e., as <-Ca-a>), though in all probability this indicates an actual lengthening of the vowel. 3. The vowels¯˘ı and u¯˘ are written with the vocalic characters and , and medially with an additional preceding
11. Early Iranian ∗h (from Indo-Iranian ∗s) is omitted in writing before Old Persian u¯˘, m, and r (cf.
Given the cumbersome nature of the writing system, clear, one-to-one correspondences between graphs and phonemes do not exist. Some of the above spelling rules result in critical morphology being hidden, particularly rule 5 (e.g., the absence of a distinction between tai and ti means that third singular, indicative present endings, active -ti and mediopassive -tai cannot be distinguished) and rule 8 (the omission of preconsonantal n blurs, for exam- ple, the distinction between the third-person singular and plural endings -ti, -tu and -nti, -ntu). The ambiguous nature of Old Persian spelling means that there is normally some set of possible interpretations of a word. In any particular case then a correct reading is dependent upon careful philological and linguistic (in particular, etymological) analysis – chiefly by comparisonwithcognatelanguages(Avestan,Vedic,etc.)orwithlaterPersiandevelopments. Inthecaseofnamesandtechnicalterms,theformswhichtheytakeinElamiteandBabylonian versions of an Old Persian inscription plays a decisive role. For example, the Old Persian spelling
2.3 Origin of the script The problems of the origin of the Old Persian cuneiform script, of the date and process of its introduction, have been treated again and again without general agreement having yet been reached concerning the controversial issues. There are several factors that one must take into account:
1. The passage DB IV 88–92, in which a new “form of writing” (Old Persian dipici¸cam) is mentioned that Darius has made and is said to be ariy¯a “in Aryan.” 2. A number of archeological and stylistic observations regarding the B¯ısutun¯ monu- ment, by which several subsequent stages in its genesis may be established. 3. Those Old Persian inscriptions that are supposed or claimed to predate Darius I. 4. The structural analysis of the script itself. old persian 81
Though the oldest attested inscriptions in Old Persian language are the B¯ısutun¯ texts (first the minor captions, then the major inscription), the creation of a new type of writing for recording the king’s mother tongue seems to have begun already under Cyrus II. This assumption is based not least on the observation that the characters ku and ru needed for writing the royal name Kuru˘s must belong to some initial set of characters, for their shapes have a quite simple pattern, even though the phonemic sequences expressed by them are not very common. A similar observation reveals that this writing system was created for the Old Persian language and not for some other Iranian dialect like Median: the fricative ¸c, which is the Old Persian reflex of Proto-Iranian ∗ϑr and which was foreign to Median, likewise is represented by one of the simplest characters, which must have been among the earliest of signs created. A number of striking features appear to suggest that the invention of the script indeed began under Cyrus, but that Darius was the first to employ it. An original strategy seems to have aimed at a consistent and unambiguous system of marking short and long vowels and diphthongs by means of a complete set of three CV characters – for each consonant – used in conjunction with three V signs; for example: ∗ ∗ ∗ (3)
2.4 Decipherment Because Old Persian cuneiform fell into disuse with the fall of the Achaemenid Empire, and thus knowledge of that script and of the values of its individual characters was lost already in antiquity, this writing system had to be deciphered in the modern era. Old Persian texts first came to the attention of the West during the seventeenth century. A solid basis for the decipherment was laid by C. Niebuhr, who in 1778 published the first precise copies of Achaemenid trilingual texts and who recognized that the first and most simple system was written from left to right. Following the identification of the word-divider and the attribution of the texts to the Achaemenids, G. F. Grotefend, in 1802, began the process 82 The Ancient Languages of Asia and the Americas
of decipherment. By assuming that the inscriptions were records of the ancient Persians and might therefore contain the names, titles, and genealogies of some of their kings, he succeeded in determining the approximate phonetic values of about ten signs. From this starting point, other scholars, progressing step by step, brought the decipher- ment to its conclusion. In 1826 R. Rask identified the n(a) and m(a) signs in the genitive plural ending -¯an¯am (corresponding to Avestan-anam ) and thus produced the first evidence for a close relationship with the Avestan language. In 1836 E. Burnouf and C. Lassen undertook a more systematic comparison with Avestan. Lassen, in 1845, made the very important dis- covery that the consonant characters of the Old Persian script could have an inherent vowel, as in the ancient Indian scripts. The work was completed in 1846/1847 by H. C. Rawlinson with his publication, translation, and interpretation of the entire DB text. A final touch was added in 1851 by J. Oppert, who established the value of the last (and most rarely used) of the phonetic signs, l(a), which even now is attested only in four foreign names for the marginal phoneme /l/ (not belonging to Old Persian proper).
3. PHONOLOGY
3.1 Phonemic inventory Identifying the complete system of Old Persian phonemes is a rather difficult task, since only a minimal set of phonemes is revealed by the attested graphemes. In order to advance beyond that set, the data must be analyzed and evaluated on a language-internal basis and by methods of historical-comparative linguistic analysis.
3.1.1 Consonants The following consonantal phonemes can be confidently identified for Old Persian:
(4) Bilabial Labiodental Interdental Dental Velar Stop Voiceless ptk Voiced bdg Fricative f q x Nasal mn (the velar nasal [ŋ] is only a positional variant with allophonic status). In addition, Old Persian possesses two so-called “palatal” affricates c and j, which in all probability were palato-alveolar /c/ˇ and //. There also occur six fricatives – /s/, /z/, /c¸/, /s/,ˇ /z/,ˇ and /h/, the liquids /r/ and /l/, and the glides /y/ and /w/. The actual pronunciation of those phonemes is not as secure as is suggested by the con- ventional representation. Thus, regarding the voiced stops /b, d, g/, it has been hypothesized that they were – at least in intervocalic position (if not more generally) – voiced fricatives [β, ð, γ]. The sibilant /z/,ˇ which is not represented graphically by a separate character, but is written with the j sign, must be postulated for reasons of historical phonology: DB II 64 n-i-j-a-y-m = [niz-ˇ ayam]¯ “I departed, went off” presents evidence for the Proto-Aryan verbal root ∗ay + prefix ∗niˇs-/niˇz- (with a j sign denoting the reflex not of Proto-Aryan ∗, but of ∗ˇz, the voiced counterpart of ∗ˇs in the position before a voiced sound). For the time be- ing, however, the question of whether ˇz and are two distinct phonemes or only allophones of one and the same archiphoneme remains unresolved. old persian 83
The fricative phoneme identified as the palatal /c¸/ is the Old Persian reflex of the Proto- Iranian cluster ∗ϑr (which is preserved in [nearly] all other Old Iranian dialects). Its pho- netic realization remains unclear, however. It can be said with certainty only that the sound was pronounced as a voiceless sibilant (certainly not as a palato-alveolar sibilant [s]ˇ and not as an affricate [c]);ˇ in Middle Persian its reflex has merged with that of Old Persian /s/. Old Persian has a syllabic [ r], which is only a contextually conditioned allophone of the liquid /r/ (between stops), however, and not an independent phoneme. The lateral /l/ has a marginal position in the phonemic inventory of Old Persian, since it is attested only in four foreign names.
3.1.2 Vowels Old Persian possesses three short and three long vowel phonemes, presented in Figure 5.1:
FRONT CENTRAL BACK
HIGH 1 / iu / u Figure 5.1 Old Persian vowels LOW a / a
Whether the long vowels are somewhat lower than the short ones cannot be established. In addition, there are two “short” and two “long” diphthongs, which are not phonemes, but only biphonematic combinations of the short or long low-central vowel with a subsequent short high-front or back vowel; since the first is the syllable nucleus, those diphthongs result in
(5) Short diphthongs Long diphthongs ai ai¯ au au¯
Those four diphthongs, inherited from Proto-Iranian, are preserved in Old Persian as such at the time of the origin of the Old Persian cuneiform script and during the reign of Darius I and Xerxes I, as can be deduced from their regular orthographic representation (see §2). From a later period, there is evidence of a monophthongization of ai and au to ¯e and o¯ respectively – seen in the development from Old to Middle Persian and revealed by transcriptions of Persian words in other languages (the “collateral” tradition; see §6). The only transcription evidence of any linguistic weight for Old Persian proper is provided by the Elamite language, which has no diphthongs itself (see WAL Ch. 3, §3.2). The Elamite script therefore lacks a regular means of spelling such sounds and so offers little possibility of documenting an early (pre-460 BC) monophthongization. Even so there are, in fact, unmistakable Elamite attempts to render Old Persian diphthongs: for example, ti-ig-ra-ka- u-da for Old Persian tigra-xauda- “with pointed caps.” It should be noted that not every graphic sequence seemingly pointing to ai and au actually records a diphthong. Spellings like a-i-ˇs-t-t-a “he stood” (from Proto-Iranian ∗a-hiˇstao), the theonym a-u-r-m-z-d-a (from Proto-Iranian ∗Ahura Mazd¯a) or the country name h-r-u- v-t-i-ˇs (from Eastern Iranian ∗Harahwat¯ı- “Arachosia”) record sequences of two syllables, [-a$i-] and [-a$u-] (i.e., A-uramazd¯a, not Au-ramazd¯a, etc.). 84 The Ancient Languages of Asia and the Americas
3.2 Phonotaxis Vowels and diphthongs are not subject to any phonotactic restrictions, and likewise all single consonants appear in initial and intervocalic position. For the final position, however, only singleconsonants(neithergeminateconsonantsnoranyotherconsonantclusters)arefound, and only -m, -r, and -ˇs are written. Those final consonants which are omitted in writing were perhaps still pronounced but in some manner phonetically reduced. Note that original Proto-Iranian ∗-a is written as Old Persian <-a> (i.e., [-a:]), but original ∗-an or ∗-ad is written as -
3.3 Syllable structure It is difficult to make specific observations about the syllable structure of Old Persian. Most syllables appear to be open: [$(C)V]; more rarely [$C1C2V$] (e.g., xˇsa-¸ca- “kingdom”) or even [$C1C2C3V$] (e.g., xˇsn¯a-s¯a-ti “he may know”). In the case of consonant clusters the syllable boundary may fall within the cluster or before it; the position of the boundary may depend on various criteria: the relative sonority of the particular elements of the cluster; the presence and position of a morpheme boundary; whether or not the cluster concerned is permissible in word-initial position; and so forth. Syllables also occur with the structure [$VC$], [$CVC$], and [$C1C2VC$] (e.g., u-fraˇs-ta- “well punished”), and perhaps also those with two consonants following the syllabic nucleus (e.g., ϑans-ta-nai “to say”).
3.4 Accent Accent is not marked in the Old Persian writing system; consequently both the nature and the position of the accent are quite uncertain. In the development from Old to Middle Persian, final syllables disappear, suggesting that the accent was fixed in the manner of Classical Latin or later Old Indo-Aryan. There may be (indirect) evidence for the hypothesis that the inherited free accent (perhaps a pitch or tonal accent), of which there are traces in Avestan and in modern Iranian languages (especially Pashto), survived until the reign of Darius I.
3.5 Diachronic developments In this section, only the most interesting and significant diachronic phonological develop- ments will be presented (and only vis-a-vis` Proto-Iranian).
3.5.1 Consonants Among consonantal developments, the most distinctive concerns the Old Persian reflexes of the Proto-Iranian continuants (presumably affricates ∗ts and ∗dz), which are themselves old persian 85 reflexes of the Proto-Indo-European palatals ∗k, ∗g, ∗gh: in contrast to the other Iranian languages Old Persian shows ϑ in, for example, viϑ- “house, royal house” = Avestan v¯ıs- = Vedic v´ı´s- from Proto-Aryan ∗w´ı´c-, and d (if not [ð]; see §3.1.1) both in, for example, yad- “to worship” = Avestan yaz- = Vedic yaj- from Proto-Aryan ∗ya´-, and in adam “I” = Avestan azəm = Vedic ah´am from Proto-Aryan ∗a´ham´ . There are also certain distinctive Old Persian consonantal changes of a conditioned or syntagmatic type. These changes show an Old Persian development which has progressed beyond that seen in the other Old Iranian languages. Thus, the Proto-Iranian cluster ∗ϑr develops into Old Persian ¸c in, for example, pu¸ca- “son” = Avestan puϑra- = Vedic putr´a-. That this change is of a rather late date is suggested by the fact that Proto-Persian ∗ϑr,whereϑ is a reflex of Proto-Indo-European ∗k, Proto-Iranian ∗ts , has also undergone the change: thus, one finds Old Persian ni-¸c¯araya- “to restore” = Avestan ni-sr¯araiia- from Proto-Aryan ∗´crai- and Proto-Indo-European ∗klei-. Before ∗n or ∗y Proto-Iranian ∗ϑ became Old Persian ˇs: for example, a-r-ˇs-n-i- ([arasni-])ˇ “cubit” from Proto-Iranian ∗araϑni- = Vedic aratn´ı-; h-ˇs-i-y- ([hasiya-])ˇ “true” = Avestan haiϑiia- from Proto-Iranian ∗haϑya- = Vedic saty´a-. Old Persian ˇsiy develops from Proto-Iranian ∗ˇcy (i.e., from a Proto-Indo-European ∗kw that was palatalized before ∗y): for example, ˇs-i-y-a-t-i- ([siya:ti-])ˇ “happiness” = Avestan ˇs´a¯iti- from Proto-Aryan ∗ˇcy¯ati- = Latin qui¯eti-, nominative qui¯es. A completely independent development of Old Persian, setting it apart from all the other Iranian languages (and thus one of its chief innovative characteristics), is the simplification of the Proto-Iranian clusters ∗tsv and ∗dzv, producing Old Persian s and z (not sp and zb): for example, a-s- ([asa-]) “horse” = Avestan aspa- = Vedic a´´sva-; v i-i-s- ([visa-]) “all” = Avestan v¯ıspa- = Vedic v´ı´sva-; h-z-a-n-m (acc. sg. [hiza:nam]) “tongue” (for the spelling h-z- see §2.2, 13), evolving from Proto-Iranian ∗hidzv¯ao as do Avestan hizuu¯a- or Parthian zb n ([izβa:n]) from earlier ∗hizb¯ano.
3.5.2 Vowels The vowels and diphthongs of Proto-Iranian remained unchanged in Old Persian at least until the period of Darius I and Xerxes I (on the later monophthongization of the short diphthongs see §3.1.2). The reflex of Proto-Iranian word-final short ∗-a is usually written as <-Ca-a> = -¯a,asinu-t-a ([uta:]) “and” (Avestan uta, Vedic ut´a); it appears probable that this lengthening was a linguistic reality and not only a graphic phenomenon. Vowel contraction seems to play a minor role in Old Persian. The most obvious example is that of ∗-iya- producing -¯ı-,asinn-i-ˇs-a-d-y-m ([ni:sa:dayam])ˇ from uncontracted ∗ni-a-ˇs¯adayam “I have put down” (cf. the alternative form n-i-y-ˇs-a-d-y-m), and in m-r-i-k- ([mari:ka-]) “young man” from ∗mariyaka- (with a secondary -Ciya- from∗-Cya-, from Proto-Aryan ∗maryaka- (= Vedic maryak´a-). Proto-Iranian sonorants, ∗m, ∗n, ∗y, ∗w, and ∗r (including Proto-Iranian ∗ar from Proto- Aryan ∗ rH as in darga- “long” = Old Avestan dar ga- = Vedic d¯ırgh´a-, etc.), remain un- changed in Old Persian. Proto-Aryan ∗Cy and ∗Cw developed into Old Persian Ciy and Cuw respectively, regularly written as
was pronounced as [ər]. Since in Old Persian orthography this [ər] can be rendered only in a makeshift fashion (like the sequence [ar]) by <(C)a-r-C>, other unambiguous evidence is required to confirm the value [ər] – either morphological (e.g., k-r-t- “made, done” = [kərta-] with the zero-grade of the root like Avestan kər ta- and Vedic k rt´a-), or etymological (e.g., a-r-ˇs-t-i- “spear” = [ərsti-],ˇ revealed by Vedic rs.t.ı´-). A special case is the development of Proto-Iranian ∗ r to Old Persian u in the present and aorist stems of the root kar “to do” u i (e.g., k -u-n-u-t-i-y [kunauti] “he does” = Avestan kərnao ti = Vedic k rn. oti´ ); these are usually explained as allegro forms originating in (and spreading from) the imperative. Two phonetic phenomena, which have given such a strange appearance to many Avestan words (see Ch. 6, §§3.3; 3.4.2; 3.4.10), are without significance for Old Persian. Epenthesis (i.e., the insertion of i or u into an existing syllable) is completely foreign to Old Persian, and anaptyxis (i.e., the development of a vowel between two consonants) is nearly unknown. The Avestan epenthesis, which is triggered by an ensuing i/y or u/w (as in Avestan haiϑiia- “true” from ∗haϑya-,see§3.5.1), is not attested in Old Persian inscriptions (transcription of Old Persian words in other languages may reveal that a late process of this sort characterized colloquial Old Persian). Anaptyxis is found only in the case of the clusters dr and gd when followed by u: for example, one finds du-u-ru-u-v- ([duruva-]) “firm” = Avestan druua- ([druwa-]) = Vedic dhruv´a-; present tense stem du-u-ru-u-ji -i-y- ([duruiya-]) “to lie” = Vedic dr´uhya-; s-u-gu-u-d- ([Suguda-]), as well as s-u-g-d- ([Sugda-]), “Sogdiana.”
4. MORPHOLOGY
4.1 Morphological type Typical of ancient Indo-European, Old Persian is an inflectional language with synthetic morphological patterns. Owing to lack of evidence, both the nominal and pronominal and, still more, the verbal paradigms are known only partially in most instances. Therefore it is not possible to give a fully formed account of the formation, function, and actual use of nominal, pronominal, and verbal forms. The same is true, by and large, with regard to nominal and verbal stem formation.
4.2 Nominal morphology The grammatical categories marked on the Old Persian noun are case (seven), gender (three), and number (three). Whereas the three genders (masculine, feminine, and neuter) andthethreenumbers(singular,dual,andplural)inheritedfromProto-Indo-Europeanhave preserved their usual significance and function, the case system has been reduced by one in Old Persian. Likewise gender and number show the expected and customary grammatical agreement (see §5.6), though there are some instances in which two singular subjects occur not (as would be expected) with a dual, but with a plural form of the verb. The seven attested nominal cases are the following: (i) nominative (for subject); (ii) vocative (for direct address); (iii) accusative (for direct object and direction); (iv) genitive (used as possessive, subjective, objective, and partitive genitive); (v) locative (for indication of place or goal); (vi) instrumental (for indication of means, cause, and extension); and (vii) ablative (only combined with prepositions). The functions of the Proto-Indo-European dative (as the case of the indirect object) have been absorbed by the Old Persian genitive (e.g., haya s˘iy¯atim ad¯a martiyahy¯a “who created happiness for man”). Moreover, the case old persian 87 system has also been reduced and simplified by abandoning formal distinctions; thus, for example, there are only three separate forms in the singular of the a-¯ stems: nom., voc. -a¯; acc. -¯am; gen.(-dat.), abl., loc., instr. -¯ay¯a.
4.2.1 Stem formation Old Persian has inherited from Proto-Indo-European its two chief means of nominal stem formation: (i) derivation (by means of primary or secondary suffixes attached to the un- derlying [verbal] root itself or to an already derived nominal stem), and (ii) composition of two word stems (with or without a particular [compositional] suffix). Also playing a role in stem formation are ablaut (see WAL Ch. 17, §3.2) and, for derivation, the vowel-lengthening process known as v rddhi. Only some subset of the numerous inherited nominal suffixes of Old Persian can be treated here, since the scanty evidence available does not allow one to judge whether some particular formation is only a traditional relic within Old Persian or actually remains a living and productive process. One of the productive suffixes is undoubtedly the “locatival” suffix -iya-, forming – adjectives, especially ethnics such as Armin-iya- “Armenian” (from Armina-), Uj-iya- – “Elamite” (from Uja-), Mac-iya- “inhabitant of Makran”¯ (from Maka-), and so forth. The Proto-Iranian suffix ∗-hwa-/∗-ˇswa- forming fractions (see §4.6) seems to be similarly productive. A distinctive phenomenon of derivation which Old Persian has inherited and which, as several indisputable examples show, is still productive in this language, is the lengthening of the first vowel of a word, a process traditionally called v rddhi (a term coined by the ancient Indian grammarians). The clearest examples attested are the ethnic M¯argava- “inhabitant of Margiana,” derived from Margu- “Marv, Margiana”; and the month name B¯agay¯adi-, based on ∗baga-y¯ada- “worship of the gods.” Other apparent cases are not without problems: for example, the month name aigraci-¯ ; a form which – could v rddhi be confirmed – would be essential for settling the question of whether Old Persian derivatives of words with i or u vowels have the v rddhi form ai¯ and au¯ like Old Indo-Aryan or the short diphthong ai, and au, as it is found in Avestan.
4.2.2 Nominal declension Old Persian nouns have been traditionally grouped into declensional classes, though with regard to the origin of the nominal system at an earlier stage of the Indo-European parent language, a number of other criteria are of relevance, chiefly accent placement and ablaut variation and their distribution over the root, the (optional) suffix, and the ending (see WAL Ch. 24, §4.1.1.3). Old Persian evidence is available for stems ending in -a-, -¯a-, -i-, -¯ı-, -¯ı/y¯a-, -u-, -¯u-, -h- or -ˇs-, -r-, -n- and in several stops and fricatives. The only productive stems, however, are those ending in vowels, and in particular those of the a-class, as those lexemes suggest which show forms of different declensions side by side: most clearly tunuvant- “strong” (in nom. sg. tunuv¯a) versus tunuvanta- (in gen. sg. tunuvantahy¯a); compare the “bridge” accusative singular tunuvantam. The only paradigms which are known somewhat extensively are those of the stems in a- and a-¯ ; their singular and plural forms may be given in (6) and (7) (for the dual see below); all other case forms and declensional patterns are presented only in the larger summary of (8) and (9): 88 The Ancient Languages of Asia and the Americas
(6) The Old Persian a-stems Singular Plural Example Ending Example Ending Animate Nom. martiya “man” -ø < ∗-s martiya-¯ a¯ < ∗-as¯ bagaha¯ “god” -aha¯ < ∗-asas¯ Voc. martiya¯ ∗-ø — Acc. martiyam -m martiya-¯ a¯ < ∗-ans¯ Gen. martiyahya-hy¯ a¯ martiyan¯ am¯ -an¯ am¯ Abl. Pars¯ a-¯ a¯ < ∗-at¯ Sakaibisˇ =instr. Instr. kar¯ a“army”¯ -a¯ martiyaibisˇ -aibisˇ Loc. Parsai¯ -i Madai¯ suvˇ a¯ -aisuˇ + -a¯ dastay-a¯ “hand” -i + -a¯ Neuter Nom.-acc. xsacˇ ¸am “kingdom” -m ayadan¯ a-¯ a¯ < ∗-a¯ “place of worship”
(7) The Old Persian a¯ -stems Singular Plural Example Ending Example Ending Animate Nom. tauma¯ “family” -ø stun¯ a¯ “column” -a¯ < ∗-as¯ Voc. —— Acc. taumam¯ -m [hamic¸i]ya¯ “rebellious” -a¯ < ∗-ans¯ Gen. taumay¯ a-y¯ a¯ < ∗-yas¯ ◦zanan¯ am¯ “with . . . races” -an¯ am¯ Abl. Same as genitive — Instr. framan¯ ay¯ a¯ “order” -ya—¯ Loc. A uray¯ a-i¯ + ama¯ skˇ auv¯ a¯ “skin” -u < ∗-su + -a¯
The set of case endings attested in Old Persian may be summarized in (8) and (9) without differentiating them by declensional class and without a detailed historical-comparative interpretation:
(8) Summary of Old Persian singular case endings Animate Nom. -ø, -sfromˇ ∗-s; -ø from ∗-ø Voc. -ø from ∗-ø Acc. -m, -am from ∗-m, -m Gen. -a from ∗-as; -ø, -sfromˇ ∗-s; -hyafrom¯ ∗-sya; -yafrom¯ ∗-yas¯ Abl. -afrom¯ ∗-at;¯ -ø from ∗-t; or identical to the genitive Instr. -afrom¯ ∗-a;¯ -yafrom¯ ∗-ya¯ Loc. -i from ∗-i; -ø from ∗-ø, both with or without postpositive -a¯ Neuter Nom.-acc. -m from ∗-m; -ø from ∗-ø old persian 89
(9) Summary of Old Persian plural case endings
Animate Nom. -a from ∗-as; -afrom¯ ∗-as;¯ -aha¯ from ∗-asas¯ Voc. Identical to the nominative, but not attested Acc. -afrom¯ ∗-ans;¯ -ø, -sfromˇ ∗-ns ∗ – Gen. -an¯ am,¯ -un¯ am¯ from -Vnam¯ Abl. Identical to the instrumental Instr. -bis,ˇ -aibisfromˇ ∗-bisˇ Loc. -aisuvˇ a,¯ -suvˇ afrom¯ ∗-sw-ˇ a;¯ -uvafrom¯ ∗-sw-a,¯ attested only with postpositive -a¯ Neuter Nom.-acc. -afrom¯ ∗-a¯
Several dual forms are securely attested in Old Persian texts, such as nom. u-b-a ([uba:]) “both”; acc. g-u-ˇs-a ([gausa:])ˇ “both ears”; gen. g-u-ˇs-a-y-a ([gausa:ya:]);ˇ instr. d-s-t-i-b-i-y- a ([dastaibiya:]) “with both hands,” all belonging to stems in -a-. In addition, the following occur: nom. u-ˇs-i-y ([usi:]),ˇ as well as u-ˇs-i-y-a ([usiya:]),ˇ three times each, and instr. u-ˇs-i-b-i-y-a ([usi:biya:]),ˇ from neuter uˇsi- “intelligence” (literally “ear” and therefore in dual number). Adjectives behave like the nouns with regard to stem formation and declension. The comparative is formed by means of the Proto-Indo-European suffix ∗-yes-/-yos- and the superlative by ∗-is-to-. As examples, consider Old Persian nom. masc. sg. t-u-vi -i-y-a ([taviya:]), from ∗tau-yah- “stronger,” and m-ϑ-i-ˇs-t ([maqista])ˇ “greatest.”
4.3 Pronominal morphology A variety of pronouns is attested in Old Persian: (i) personal pronouns (including the so-called anaphoric pronoun); (ii) several demonstrative pronouns; (iii) relative; and (iv) interrogative-indefinite pronouns.
4.3.1 Personal pronouns The personal pronouns are characterized (i) by an absence of grammatical gender; (ii) by a remarkable heteroclisis between the nominative and oblique cases; and (iii) by the exis- tence of frequently used enclitic forms. All these characteristics have Proto-Indo-European ancestry. The following personal pronouns are attested in Old Persian:
(10) Accented forms First Second First Plural Nominative adam tuvam vayam Accusative mam¯ quvam¯ — Genitive mana—¯ amaxam¯ Ablative -ma — — Enclitic forms Accusative -ma—¯ — Genitive -mai -tai — 90 The Ancient Languages of Asia and the Americas
The dual forms are not attested at all; the genitive has taken over the function of the dative. Ablative -ma, though being attested only in combination with the preposition “by,” h-c-a-m ([haca-ma])¯ “by me,” is not enclitic (demonstrated by accented Vedic m´at). The anaphoric pronouns “he, she, it” share the characteristic features of the personal pronouns, though there are no nominative forms and no heteroclisis. Old Persian exhibits enclitic forms built from the stems -ˇsa-/-ˇsi- and -di-:acc.sg.-ˇsim “him,” gen. -ˇsai “his,” acc. pl. -ˇsiˇs “them,” gen. -ˇs¯am “their”; acc. sg. -dim “him” and acc. pl. -diˇs “them.”
4.3.2 Demonstrative pronouns Other pronominal stems exhibit grammatical gender distinctions and, in part, are charac- terized by a declension differing from that of nominal stems in -a- and -¯a-.Includedin this group are three demonstrative pronouns. The pronoun iyam (nom. sg. masc./fem.) “this” combines forms of the stems i-, ima-, and a-: for example, ima (nom.-acc. sg. neut.), an¯a (instr. sg. masc.), ahy¯ay¯a (loc. sg. fem.). The remaining two are aita- “this here” (more emphatic), and hau- (nom. sg. masc./fem.) “that”; the paradigm of the latter is supplemented in the oblique cases by the stem ava-: for example, ava (nom.-acc. sg. neut.), avai (nom.-acc. pl. masc.), avaiˇs¯am (gen. pl. masc.), av[¯a] (nom. dual masc.).
4.3.3 Relative and interrogative pronouns The relative pronoun, which has also acquired the function of an article (see §5.5), is an Old Persian innovation. Its stems haya- (nom. sg. masc./fem.) and taya- (elsewhere) “who, which” emerged from the fusion of the Proto-Aryan correlating demonstrative and relative pronouns ∗s´a-/∗t´a- + ∗y´a- “the one, who.” The interrogative pronoun is not attested in Old Persian texts and can be recovered only from the indefinite pronouns kaˇs-ci (nom. sg. masc.) “somebody,” ciˇs-ci (neut.) “something,” which are derived by means of the generalizing particle -ci,asinya-ci (nom.-acc. sg. neut.) “whatever.”
4.3.4 Pronominal adjectives The declension of certain adjectives, which are semantically close to the pronouns, shares also the special declensional forms of pronouns. Old Persian attests only aniya- “other” (e.g., nom.-acc. sg. neut. aniya, abl. sg. masc. aniyan¯a); haruva- “all” (e.g., loc. sg. fem. haruvahy¯ay¯a); and hama- “the same” (in gen. sg. fem. hamahy¯ay¯a).
4.4 Verbal morphology The grammatical categories of the Old Persian verbal system were inherited from Proto- Aryan, the consequent and consistent structure of which can still plainly be observed in the earliest Vedic texts. But with regard to both function and form, a great number of funda- mental innovations and reorganizations have occurred which leave the distinct impression that Old Persian, like Young Avestan (see Ch. 6, §1), has begun to part company with the Proto-Aryan system and already represents a kind of transitional stage from Old to Middle Iranian. This is revealed by phonetic developments and innovations in nominal morphology, but especially by changes in the system of verbal morphology: (i) the aspectual opposition of aorist versus imperfect has been lost; (ii) aorist and perfect tense forms are attested only rarely; (iii) a periphrastic “neo-perfect” has emerged (see §4.4.6); and (iv) present stems in -aya- begin to gain prominence. Old Persian verbal forms are marked for tense (originally aspect), voice, mood, and the usual three persons and three numbers. The Old Persian evidence is, however, rather old persian 91 unbalanced, owing to the nature of the contents of the inscriptions: thus, for example, the only dual form found in the texts is the third plural imperfect active aj¯ıvatam “they both (still) lived.” Together with the three persons and numbers, two of the three voices (i.e., active and middle) find expression in two sets of personal endings: the so-called primary endings in the present indicative (which alone denotes a real present time) and subjunctive (which may do the same, at least in the speaker’s view), and the secondary endings otherwise, apart from the imperative, which has distinctive endings. 4.4.1 Voice The voices usually have their customary functions (inherited from the Indo-European parent language). A particularly striking exception is provided by certain third plural middle forms whichlackmiddlefunctionandaretobeinterpretedashavingarisenonlytoavoidambiguity. Passive morphology is more innovative, with the following attested: (i) forms built from the passive stem in -ya- (e.g., imperfect a-ϑanh-ya “it has been said”), common to Indo- Iranian for the present stem; (ii) middle forms like a-naya-t¯a “he was led”; and (iii) phrases consisting of a verbal adjective in -ta- plus the copula (which usually is omitted, however, in the third person: see §4.4.6). 4.4.2 Mood The five moods attested in Old Persian are indicative, subjunctive, optative, imperative, and, as an Indo-European relic, injunctive (see below). Typical of Iranian is both the use of the perfect optative for the irrealis of the past, and (even more so) the use of the present optative with the temporal augment a- (thus looking like an imperfect optative) to express a repeated action of the past (e.g., av¯ajaniy¯a from ∗ava-a-jan-y¯a-t “he used to slay”). The Old Persian moods exhibit the same functions as their counterparts in YoungAvestan. The indicative is used to express factual statements – present indicative (formed with the primary endings) for those in present time, and imperfect indicative (the augment a- and secondary endings being added to the present stem) for those in past time. The subjunctive expresses the eventual or potential realization of actions in the present or future; the present subjunctive is formed with primary endings, which are added to the present stem enlarged by -a- (e.g., ah-a-ti “it may be”). The optative is used for wishes and prayers and is formed with a stem in -iy¯a- (in the athematic singular) or -¯ı- (otherwise) – suffixes descended from ∗ ∗ Proto-Indo-European -yeh1-/ -ih1-; the optative takes secondary endings (e.g., 2nd sg. mid. yadaiˇs¯a “you may worship”). The imperative is the mood of command and prayer and makes use of distinctive imperative endings which are added to the present or aorist stem. The injunctive (with secondary endings) is found in Old Persian only in prohibitive constructions introduced by the particle m¯a “not!” but even in preventive clauses never combined with forms of the aorist tense stem. Together with the loss of the aorist (see §4.4.3) Old Persian obviously has lost the inherited distinction between the inhibitive present injunctive and the preventive aorist injunctive. Moreover, if combined with the optative present, the prohibitive particle m¯a denotes a corrective notion with regard to a present action: for example, daiv¯am¯a yadiyaiˇsa “the Daivas shall not be worshiped any longer!”
4.4.3 Tense The tenses find expression in stem formations which had originally been used to distinguish aspect (imperfective vs. perfective) and still did so in Proto-Aryan and Proto-Iranian. Several doubletsofsuchformsmakeitclear,however,thattheimperfect(whichisbuiltonthepresent stem and thus expressed the imperfective aspect of a past action) and the aorist (being the counterpart in the perfective aspect) are used in Old Persian without any obvious difference 92 The Ancient Languages of Asia and the Americas
in function, suggesting that aspectual distinctions were no longer being productively made. The “sigmatic” aorist adarˇsi “I took possession of” (1st sg. indic. aor. middle of the root dar-) alone seems to point to a living use of the aorist indicative (i.e., for conveying the perfective aspect of an action). The one perfect form attested is an optative expressing past irrealis, caxriy¯a “he might have done.” Regarding perfect morphology, therefore, all that can be said is that Old Persian inherited stem reduplication (ca-xr- from Proto-Aryan ∗ˇca-kr- and Proto-Indo-European ∗kw e-kw r-), but nothing can be discerned about the particular endings of the perfect indicative active.
4.4.4 Verbal stems The stem formations occurring in Old Persian are essentially those inherited from Proto-Aryan and in the end often from Proto-Indo-European. This includes the inheri- ted distinction between the thematic and the athematic stems marked by the presence or absence of the thematic vowel -a- (from Proto-Indo-European ∗-e/o-;seeWAL Ch. 17, §3.4) preceding the personal endings (e.g., athematic as-ti “he is,” but thematic bav-a-ti “he becomes”). The present and aorist stems (and likewise the only perfect stem attested; see §4.4.3) are formed either from the verbal root to which one of a set of suffixes is attached, or from the unsuffixed root itself (root presents and root aorists). Most numerous and to a certain degree productive are the present stems in -aya- like t¯avaya- “tobeable,”m¯anaya- “to wait, expect,” and so forth. Ancestral formations of Proto-Indo-European origin are the stems in -sa- (= Avestan -sa-) like p rsa- “to ask, interrogate” (= Avestan pərsa-), t rsa- “to be afraid” (= Avestan tər sa-), xˇsn¯asa- “to know.”
4.4.5 Verbal endings The various sets of verbal endings are only partially attested in Old Persian; these are pre- sented in (11)–(16) together with their Proto-Aryan preforms:
(11) The Old Persian primary endings: active Singular First -mi from ∗-mi (also in the thematic verbs); -ni from ∗-ni (subjunctive) Second -hi from ∗-si (attested only in subjunctive) Third -ti from ∗-ti Plural First -mahi from ∗-masi Second — Third -nti from ∗-nti
(12) The Old Persian primary endings: middle Singular First -ai from ∗-ai; -nai from Proto-Iranian ∗-nai (subjunctive) Second -hai from ∗-sai Third -tai from ∗-tai Plural Not attested old persian 93
(13) The Old Persian secondary endings: active Singular First -m from ∗-m; -am (athematic) from Proto-Aryan ∗-am replacing Proto- Indo-European ∗-m Second -ø from ∗-s Third -ø from ∗-t; -sˇ after ai, au (in imperfect and optative forms like akunauˇs “he did” = Avestan akər naot) ˜ Dual Third -tam = Avestan -təm (see §4.4) Plural First -mafrom¯ ∗-ma Second — Third -ø from ∗-nt; -h after a and -sˇ after ai (in imperfect and optative forms like abaraha “they brought” or yadiyaiˇsa “they shall not be worshiped”) from ∗-s
(14) The Old Persian secondary endings: middle Singular First -i from ∗-i Second -sˇafrom¯ ∗-sa Third -tafrom¯ ∗-ta Plural First — Second — Third -ntafrom¯ ∗-nta
(15) The Old Persian imperative endings: active Singular Second -afrom¯ ∗-a (thematic) and -di from ∗-dhi (athematic) Third -tu from ∗-tu Plural Second -tafrom¯ ∗-ta Third -ntu from ∗-ntu
(16) The Old Persian imperative endings: middle Singular Second -uva¯ and -suvˇ afrom¯ ∗-swa Third -tam¯ from ∗-tam¯ Plural Not attested
4.4.6 Nonfinite verbal forms Old Persian exhibits only one type of infinitive: a construction with the formant -t-n-i-y ([-tanai] or [-tani]?), being an oblique case, dative (or locative) singular, of an action noun in -tan-, and built on the full-grade verb root: for example, cartanai “to do”; bartanai “to bear;” 94 The Ancient Languages of Asia and the Americas
ϑanstanai “to say.” In the case of kantanai “to dig” and nipaiˇstanai “to engrave, write,” the passive interpretation “to be dug,” “to be engraved” cannot be ruled out. The only reliably attested active participles are tunuvant- “strong” (literally “being able”; nom. sg. masc. tunuv¯a,from∗-w¯ant-s) and yaudant- “being in turmoil” (only acc. sg. fem. y-u-d-[t-i]-m ([yaudant-i(:)m]). Present middle participles are formed by means of the suffix -mna- = Avestan -mna-,asinxˇsaya-mna- “being in control of.” The commonly occurring verbal adjective or perfect passive participle in -ta- is inherited from the Proto-Indo-European formation in ∗-to-, which usually is added to the zero-grade verbal root: for example, k rta- “done, made”; jata- “slain”; p¯ata- “protected”; but also basta- “bound” like Young Avestan basta- (in contrast to Vedic baddh´a-) and the like. In addition, there are also some formations in -ata- (like ϑak-ata- “passed” or han-gm-ata- “assembled”; cf. Avestan gmata-) which go back to Proto-Indo-European ∗-eto-. The verbal adjective in-ta- is used in Old Persian particularly for creating the new pe- riphrastic perfect of the type man¯ak rtam “(it was) done by me” (cf. Middle Persian man kard) replacing the inherited Proto-Aryan active perfect for expressing an accomplished action and/or a situation achieved by it. In origin this “neo-perfect” was formed by com- bining the copula “to be” with the -ta-adjective, though the third singular asti “she/he/it is” normally has been deleted. Moreover, the agent of transitive verbs is expressed in the gen- itive case (though the sense of the construction is not a possessive). Examples include the following: ima, taya man¯ak rtam “this [is], what [has been] done by me”; taya B rdiya avajata “that Smerdis [had been] slain”; yadi k¯ara P¯arsa p¯ata ahati “if the Persian people shall be protected.”
4.5 Compounds In principle, Old Persian exhibits all the types of compounds known from the other ancient Aryan languages (see Ch. 2, §4.4.2) and inherited from Proto-Indo-European (see WAL Ch. 17, §3.5.1). Compounds contain two elements, the last of which is inflected. Attested are determinative and possessive compounds (including those which have an inseparable prefix like a(n)- “without, un-”; u- “well-”; or duˇs- “mis-, dis-” as first element), but no copulative compoundsareattestedasyet.Especiallyremarkablearethecompoundshavingaverbalstem as the first element; Old Persian exhibits a number of such formations in anthroponomastics: for example, the throne names of Darius and Xerxes, D¯araya-vauˇs “holding the good” and Xˇsaya- rˇsan- “having command of heroes.” These forms reveal that Old Persian does not share in the Aryan recasting of the first element as a participial form in-at-, as one finds in Avestan and Old Indo-Aryan (cf. Avestan D¯araiiat.raϑa- “holding the chariot,” xˇsaiiat.vac- ˜ ˜ “having (a good) command of speech”; Vedic dh¯aray´at-ks.iti- “sustaining the creatures,” ks.ay´ad-v¯ıra- “having command of heroes”).
4.6 Numerals Since the cardinals are normally indicated by numeral signs and not written phonetically, hardly anything can be said about them. The number 1 is aiwa-, which like Avestan a¯euua- goes back to Proto-Indo-European ∗oi-wo- “one, alone” (= Greek oˆı(w)os (o²( )ov)). One hundred must have been ∗ϑata- (= Avestan satəm = Vedic ´sat´am) and in all probability is attested in the name of the province Sattagydia, Θata-gu-. Other cardinals are reflected in the “collateral” linguistic traditions (see §6), especially in Elamite garb, in compounded titles like ∗daϑa-pati- (Elamite da-sa-bat-ti-iˇs) “chief of ten, decurion” or ∗ϑata-pati- (Elamite sa-ad-da-bat-ti-iˇs) “chief of hundred, centurion.” old persian 95
Of the ordinals there are attested in the Old Persian inscriptions: fratama- “first” = Avestan fratəma-; duvit¯˘ıya- “second” = Old Avestan daibitiia-, Young Avestan bitiia- (= Vedic dvit¯´ıya-); ¸cit¯˘ıya- “third” = Avestan ϑritiia-; navama- “ninth” = Avestan naoma- (from ∗nawəma-). A quite interesting Iranian innovation is found in the fractions formed by addition of the Proto-Iranian suffix ∗-swa- (realized as Avestan -huua- or -ˇsuua-). The Old Persian reflexes are attested in Elamite renderings only and can be reconstructed as ∗¸ciˇsuva- “one- third” (Elamite si-iˇ s-maˇ s;ˇ cf. Avestan ϑriˇsuua-); ∗ca¸cuˇsuva- and (with haplology) ∗ca¸cuva- “one-quarter” (Elamite za-aˇs-maˇs, za-iˇs-ˇsu-maˇs, za-iˇs-ˇsu-iˇs-maˇs; cf. Avestan caϑruˇsuua-); ∗pancauva- “one-fifth” (Elamite pan-su-ma-iˇs; cf. Avestan paŋtaŋhuua-); ∗aˇstauva- “one- eighth” (Elamite aˇs-du-maˇs; cf. Avestan aˇstahuua-); ∗navauva- “one-ninth” (Elamite nu- ma-u-maˇs); ∗daϑauva- “one-tenth” (Elamite da-sa-maˇs) and ∗v¯ıstauva- “one-twentieth” (Elamite mi-iˇs-du-ma-kaˇs, with an additional ka-suffix).
5. SYNTAX
5.1 Word order The word order found in the Old Persian inscriptions is on the whole rather free, as is common among the ancient Indo-Iranian languages. The “unmarked” order, however, is Subject–Object–Verb (SOV):
(17) Auramazda-mai¯ upastam¯ abara Auramazda-me¯ aid he brought “Auramazda¯ brought me aid” For enclitic -mai,see§5.3. Other complements, especially those indicating place, may follow the verb. There are attested, however, a number of cases showing varying order of the sen- tence constituents: for example, (i) of copula and predicate noun (cf. DNb 42f. ϑanuvaniya uϑanuvaniya ami “as a bowman I am a good bowman” vs. DNb 44 rˇstika ami uv rˇstika “as a spearman I am a good spearman”); or (ii) of two coordinated constituents (DB IV 72f. yadi im¯am dipim vain¯ahi imaiv¯a patikar¯a “if you shall look at this inscription or these sculptures” vs.DBIV77yadi im¯am dipim imaiv¯a patikar¯a vain¯ahi). Nevertheless some peculiarities of word order must be noted, mainly “marked” sentence- initial or sentence-final position of words for reasons of emphasis. Here belong, for example, the initial position of the object (OSV) when expressed by a deictic pronoun
(18) ima hadisˇ adam akunavam this palace I I have built “I have built this palace” or the nonfinal (medial) position of verbs expressing an urgent plea. Notable is also the uncommon initial position of the verb in the formulaic expression ϑati¯ NN xˇs¯ayaϑiya “proclaims NN, the king.” When two or more coordinated elements form the subject or the object of a sentence, only the first element is placed before the verb, and the remaining elements follow, for example:
(19) mam¯ Auramazdap¯ atu¯ utamai xsacˇ ¸am me Auramazda¯ may he protect and my kingdom “May Auramazda¯ protect me and my kingdom!” 96 The Ancient Languages of Asia and the Americas
Within phrases the word order is more fixed. A noun or pronoun (in the genitive case) which is dependent upon a noun precedes that noun: for example, Kurauˇspu¸ca “son of Cyrus”; man¯a pit¯a “my father.” Exceptions which are attested in royal titles (cf. xˇs¯ayaϑiya xˇs¯ayaϑiy¯an¯am “king of kings” in contrast to Middle Persian ˇs¯ah¯an ˇs¯ah) or religious formulae (vaˇsn¯a Auramazd¯aha “by the favor of Auramazda”)¯ are caused by foreign influence.
5.2 Topicalization A striking feature of Old Persian syntax and stylistics is the frequent use of a sentence- initial (so-called) casus pendens (usually an absolute nominative), which is resumed by a demonstrative pronoun (20A) or adverb (20B):
(20) A. Vistˇ aspa¯ mana¯ pita,¯ hau Parqavai aha¯ Hystaspes my father that one in Parthia he was “Hystaspes my father, he was in Parthia” B. P rga nama¯ kaufa, avada...¯ P rga by name mountain there “There is a mountain, P rga by name, there . . . ” This phenomenon is often combined with another stylistic peculiarity found in the Old Persian inscriptions, the origin of which must be sought, as Vedic parallels in prose texts show convincingly, in colloquial Proto-Aryan and not, as has been previously presumed, in Aramaic influence. This concerns parenthetical (more exactly, prosthothetical) construc- tions taking the form of nominal (i.e., verbless) clauses which introduce less common personal or geographical names: for example, D¯ad rˇsiˇsn¯ama Arminiya, man¯a bandaka, avam . . . “[There is] an Armenian, Dad¯ rsiˇ by name, my vassal, him . . . ” It should be noted that nominal sentences are very frequently used in Old Persian, mainly because the third singular form of the copula is normally omitted; consider DB I 27:
(21) ima, taya manak¯ rtam this what by me done “This [is], what [has been] done by me” with relevant examples in both the main and relative clauses.
5.3 Clitics Old Persian attests a number of enclitics (atonic lexemes which in Old Persian form a graphic unity with the preceding word); chiefly the following: (i) the oblique cases of the personal pronouns (including the anaphoric pronoun); (ii) the copulative and disjunctive conjunc- tions (-c¯a “and,” -v¯a “or”); and (iii) various emphatic particles. According to Wackernagel’s Law the enclitics are attached to the first accented word of the sentence or clause in Old Persian, as in Proto-Aryan and, still earlier, in Proto-Indo-European. This becomes partic- ularly clear from examples like (17), Auramazd¯a-mai upast¯am abara “Auramazda¯ brought me aid,” when contrasted with
(22) pasava-mai¯ Auramazda¯ upastam¯ abara afterwards-me Auramazda¯ aid he brought “Afterwards Auramazda¯ brought me aid” Enclitics which are construed with single words only and not with an entire sentence do old persian 97 not follow Wackernagel’s Law, but are attached to that particular word: for example, yaϑa¯ paruvam-ci “just as [it was] previously.” For a special treatment of enclisis see Schmitt 1995.
5.4 Coordination and subordination In the Old Persian inscriptions both coordination and subordination are used for expressing complex statements. It is not uncommon to find short simple sentences following one another, either accompanied by a connector (a coordinating conjunction like ut¯a “and” or atemporaladverblikepas¯ava “afterwards, then”), or without such (asyndeton). In other cases (and, in part, in closely parallel passages), subordinate clauses occur introduced by a relative pronoun or by some appropriate conjunction. Most conjunctions used in Old Persian are derived from the (original) stem of the relative pronoun (as is the case in the cognate languages, too): for example, yaϑa¯ (often correlated with avaϑa¯ “thus”) “when, after, so that” (introducing temporal, modal, and consecutive clauses); yadi “if” (normally with a subjunctive verb), “when” (with an indicative; introducing temporal and conditional clauses). While both of these are inherited, y¯at¯a “until, when, as long as” is a new formation, as is taya “that, so that” (acc. sg. neut. of the relative pronoun) which introduces causal, explicative clauses, indirectly reported speech, and so forth. Relative clauses are commonly attested, positioned both before and after the main clause. There are also some passages that show a subordinate infinitive. Typical is that construc- tion after a main clause containing verbs like “to order,” “to be able,” “to dare” (e.g., adam n¯ıˇst¯ayam im¯am dipim nipaiˇstanai “I ordered to engrave this inscription”); another likewise typical use of an infinitive construction is that expressing purpose after verbs like “to go,” “to send” (e.g., parait¯apatiˇsD¯ad rˇsim hamaranam cartanai “went forth against Dad¯ rsiˇ to fight a battle”).
5.5 Relative constructions The relative pronoun haya-/taya- functions as a definite article in expressions indicating various attributive complements to nouns, with case attraction if appropriate; for example: (23) A. Gaumata¯ haya magusˇ (nominative) Gaumatam¯ tayam magum (accusative) “Gaumata¯ the magus” B. karam¯ tayam Madam¯ (accusative) “The Median army” C. vi am tayam¯ amaxam¯ (genitive plural) “Our [royal] house” D. xsacˇ ¸amtayaBabirau¯ (locative) “The kingship in Babylonia” Those constructions have similar counterparts in Avestan, but have spread considerably in Middle Persian and are ultimately the source of the Modern Persian iz.afat¯ construction.
5.6 Agreement Grammatical agreement in Old Persian is of the sort common to the older Indo-European languages: (i) appositive and attributive adjectives and nouns agree in gender, number, and case; (ii) predicate nouns and adjectives agree at least in case, but now and then there are particular conditions for gender and number; (iii) relative, resumptive, and anaphoric 98 The Ancient Languages of Asia and the Americas
pronouns agree in gender and number, whereas their case is dependent upon their syntactic use (examples of case attraction not being attested); (iv) verbs agree with their subject in person and number. The existence in Old Persian of the Proto-Indo-European use of a singular verb with a neuter plural subject cannot be demonstrated, both for lack of evidence and for orthographic reasons. The only evidence is found in the usual dating formulae (see §6), and there the copula aha¯ (with ϑakat¯a nom. pl. neut.) may be third-person singular as well as plural.
5.7 Stylistics A comprehensive and systematic study of the stylistic features that may be detected in the Old Persian inscriptions (which show clear traces of stylization), is an urgent desideratum. There is found evidence for the stylistic figures of the asyndeton, of chiasmus, parallelism, and so forth; see the discussion in Kent 1953 (pp. 99f. §§ 316–317 in the relevant paragraphs). Some additional stylistic features can be briefly noted here. Epiphora (repetition of the same words at the end of each of a set of sentences) occurs several times: for example, in DPd 22 and 24 had¯a visaibiˇs bagaibiˇs “with all the gods.” Examples of personification are attested: for example, with dahyu- “land” (which “does not fear anybody else”) or duˇsiy¯ara- “crop failure” (which “may not come”). But attempts to demonstrate rhyming phrases in Old Persian texts or to detect metrical passages (especially in DB) are not convincing in this author’s view.
6. LEXICON
The Old Persian vocabulary is known only in part owing to the limited corpus of the texts and to their stereotyped character. On the whole it corresponds closely to the vocabulary of the other attested ancient Aryan languages, Avestan and Old Indo-Aryan (especially Vedic). A striking characteristic feature of Old Persian is the considerable quantity of foreign words and names which it uses. Such foreign influences, however, are only to be expected in such a multinational state as that of the Persian Empire. Among those foreign elements, borrowings from the Median language take a special place, and they can be justified historically without difficulty. The fact that particular terms are of Median origin can sometimes be established by phonetic criteria, even if the non-Persian phonetic developments observed are not unique to the Median language, but also belong to other Old Iranian dialects. Medisms occur more frequently among royal titles and among terms of the chancellery, military, and judicial affairs (vaz rka- “great,” z¯ura- “evil,” z¯urakara- “evil-doer,” etc.); they are found not least in the official characterizations of the empire and its countries (uvaspa- “with good horses,” vispazana- “with all races,” etc.). From a dialectological perspective, one notes some peculiar developments. Particularly striking is the case of the verb “to say, speak”; Old Persian continues neither Proto-Iranian ∗waˇc- nor ∗mrau-, both of which are attested in Avestan, but has gaub-. A similar case is found with “to hear”: Old Persian has lost Proto-Iranian ∗srau- (Avestan srauu-), and has instead the root a-xˇ¯ snau- (literally “to grasp, understand”). In addition to the shared isogloss of Old Persian gaub- “to say, speak” and Sogdian γwβ- ([γo:b-]) “to praise,” there are a number of remarkable features common to Old Persian (Southwest Iranian) and Sogdian (East Iranian). For example, to both belong ∗kun- “to do” (from Proto-Iranian ∗kar-, pres. ∗k rnau-) in Old Persian kunau- = Sogdian kwn- ([kun-]). Both share the meaning “to have” for the Iranian root ∗dar- “to hold, keep” (Old Persian dar-, pres. d¯araya-), and the dating formulae of the type Old Persian NN m¯ahy¯a X raucabiˇs ϑakat¯a old persian 99 aha¯ “in the month NN X days had passed” and Sogdian pr ’tδrtykYRH’pr10sγth “in the third month at/after ten passed [days].” In other cases, borrowings from some East Iranian language have been assumed: for example, k¯asaka- “semiprecious stone.” In addition, the influence of the other languages spoken by the indigenous peoples of the Ancient Near East can be detected in the Old Persian lexicon. Thus, the Persians seem to have acquired dipi- “inscription” from Elamite, maˇsk¯a- “[raft of] skin” from some Semitic language, and p¯ıru- “ivory” likewise from some Near Eastern source. A considerable portion of the Old Persian lexicon has simply not survived (because of the nature of the texts). However, the possibility exists of reconstructing Old Persian lexemes, provided they are inherited from Proto-Aryan (and from Proto-Indo-European), by comparing the Proto-Aryan vocabulary (which can be reconstructed from the very rich records available in Old Indo-Aryan) with Middle and Modern Persian words, since such later attested lexemes necessarily must have passed through an Old Persian stage. In addition, a great many Old Persian lexemes, including proper names, are preserved in a borrowed form in non-Persian languages – the so-called “collateral” tradition of Old Persian (within or outside the Achaemenid Empire). The main sources of that tradition are Elamite (especially the Persepolis tablets), Late Babylonian (with numerous administrative texts), Aramaic (as the lingua franca of the official imperial administration), Hebrew, Egyptian, and Greek authors (from Aeschylus and Herodotus) and inscriptions. It must be borne in mind, however, that not every purported Old Iranian form attested in this manner is an actual lexeme of Old Persian. Thus, for example, the title “satrap,” best known in its Greek form satrphv, in fact mirrors Median ∗xˇsaϑra-p¯a-, whereas the first element of the Old Persian form was xˇsa¸ca- and the form attested epigraphically is xˇsa¸ca-p¯a-van-. A collection of the complete material attested in the various branches of the collateral tradition is not available; Hinz 1975 offers the most comprehensive collection, though is far from being complete (e.g., by omitting even Median ∗xˇsaϑra-p¯a-) and is often unreliable.
7. READING LIST
The most comprehensive treatment of Old Persian (containing a full descriptive as well as historical grammar, the transcribed texts with English translation, and a lexicon with full references) is found in Kent 1953; for a traditional grammar see also Meillet and Benveniste 1931. A more structured outline of morphology and an etymological lexicon (including, in part, the collateral tradition) is presented by Mayrhofer in Brandenstein and Mayrhofer 1964 (pp. 55–82 and 99–157). Mayrhofer 1979: II (pp. 11–32) provides a special treatment of the personal names attested in the inscriptions. A brief account of the Old Persian language (with the most essential bibliography) is also presented in Schmitt 1989. A complete corpus of all Old Persian Achaemenid inscriptions is not available; there are only partial collections outdated by later discoveries or limited to certain groups or types of texts. The Old Persian texts alone can be found in Kent 1953: 107–157 (with an English translation); this has been supplemented by Mayrhofer 1978, who also provides a full inventory list of the Old Persian texts (pp. 37–47); though even this list is not up to date.
Abbreviations The most important Old Persian texts are listed below. Texts are usually cited utilizing a system of abbreviations, in which the king’s name normally appears first (D = Darius I, X = Xerxes I, A1−3 = Artaxerxes I–III, etc.), followed by the place of origin (B = B¯ısutun,¯ 100 The Ancient Languages of Asia and the Americas
P = Persepolis, N = Naqs-iˇ Rustam, S = Susa, etc.). Several texts by the same king at the same place are distinguished by additional small letters: DB: the major inscription of Darius I at the rock of Mt. B¯ısutun,¯ the most extensive and most important trilingual inscription, with five columns and 414 lines of Old Persian text (newly edited by Schmitt 1991). DNa, DNb: two major trilingual inscriptions at the tomb of Darius I at Naqs-iˇ Rustam, the lower text DNb being some kind of guide for the ideal ruler (new edition by Schmitt 2000:23–44). DPd, DPe: two monolingual Old Persian inscriptions which form part of an ensemble of texts at the southern wall of the Persepolis terrace and in all probability are the oldest Persepolitan inscriptions (new edition by Schmitt 2000:56–62). DSab: the trilingual cuneiform text on the Egyptian-made statue of Darius I excavated in Susa in 1972. DSe, DSf: two major trilingual building inscriptions from the palace of Susa, which are preserved, however, only in a great number of fragments. DZc: the longest of the cuneiform inscriptions from the Suez Canal. XPf: a bilingual (Old Persian and Babylonian) foundation document of Xerxes from Persepolis, which is of special historical importance owing to some details reported about the king’s succession. XPh: the trilingual, so-called Daiva-inscription describing a revolt and praising the cult of Auramazda¯ (rather than the Daivas). XPl: an Old Persian text on a stone tablet, which is essentially parallel to DNb, but associated with the name of Xerxes I.
Bibliography Brandenstein, W. and M. Mayrhofer. 1964. Handbuch des Altpersischen. Wiesbaden: Otto Harrassowitz. Hinz, W. 1975. Altiranisches Sprachgut der Neben¨uberlieferungen. Wiesbaden: Otto Harrassowitz. Kent, R. G. 1953. Old Persian: Grammar, Texts, Lexicon (2nd edition). New Haven: American Oriental Society. Mayrhofer, M. 1978. Supplement zur Sammlung der altpersischen Inschriften. Vienna: Osterreichische¨ Akademie der Wissenschaften. ———. 1979. Iranisches Personennamenbuch. Vol. I. Vienna: Osterreichische¨ Akademie der Wissenschaften. Meillet, A. and E. Benveniste. 1931. Grammaire du vieux-perse (2nd edition). Paris: Edouard´ Champion. Schmitt, R. 1989. “Altpersisch.” In R. Schmitt (ed.), Compendium Linguarum Iranicarum, pp. 56–85. Wiesbaden: Dr. Ludwig Reichert. ———. 1991. The Bisitun Inscriptions of Darius the Great: Old Persian Text. London: School of Oriental and African Studies. ———. 1995. “Zur Enklise im Altpersischen.” In H. Hettrich, W. Hock, P. Mumm, et al. (eds.), Verba et Structurae: Festschrift f¨ur Klaus Strunk, pp. 285–301. Innsbruck: Institut fur¨ Sprachwissenschaft der Universitat¨ Innsbruck. ———. 2000. The Old Persian Inscriptions of Naqsh-i Rustam and Persepolis. London: School of Oriental and African Studies. Prods Oktor Skjærvø
An Introduction to Old Persian
revised and expanded 2nd version
Copyright © 2002 by Prods Oktor Skjærvø
Please do not cite in print without the author’s permission.
This Introduction may be distributed freely as a service to teachers and students of Old Iranian. In my experience, it can be taught as a one-term full course at 4 hrs/w.
My thanks to all of my students, who have actively noted typos, inconsistencies of presentation, etc., and especially to Matthew Stolper, for his 7 pages (so far) of miscellaneous notes and corrections.
Please help improve the Introduction by reporting typos and whatever comments you may have to [email protected]
APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN
DB 1.17-20 DARIUS’S INSCRIPTION AT BEHISTUN. ƒ˝tiy D˝rayavau x˝yaƒiya COLUMN 1 im˝ dahay˝va tay˝ man˝ *patiy˝ia van˝ Auramazd˝ha man˝ badak˝ ˝hat˝ DB 1.1-3 man˝ b˝jim abarat˝ adam D˝rayavau x˝yaƒiya vazarka *taya˝m hac˝ma aƒahaya xapav˝ raucapativ˝ x˝yaƒiya x˝yaƒiy˝n˝m ava akunavayat˝ x˝yaƒiya P˝rsaiy x˝yaƒiya dahay¨n˝m DB 1.20-24 Vit˝spahay˝ puça ƒ˝tiy D˝rayavau x˝yaƒiya A˘r˝mahay˝ nap˝ atar im˝ dahay˝va martiya haya agriya ˝ha avam Hax˝maniiya ubartam abaram haya ar^ka ˝ha avam ufrastam aparsam DB 1.3-7 van˝ Auramazd˝ha im˝ dahay˝va tayan˝ man˝ ƒ˝tiy D˝rayavau x˝yaƒiya d˝t˝ apariy˝ya man˝ pit˝ Vit˝spa yaƒ˝˝m hac˝ma aƒahaya avaƒ˝ akunavayat˝ Vit˝spahay˝ pit˝ A˘r˝ma Ar˝mahay˝ pit˝ Ariy˝ramna DB 1.24-26 Ariy˝ramnahay˝ pit˝ Cipi ƒ˝tiy D˝rayavau x˝yaƒiya Cipai pit˝ Hax˝mani Auramazd˝maiy ima xaçam fr˝bara Auramazd˝maiy upast˝m abara y˝t˝ ima xaçam DB 1.7-8 hamad˝rayaiy ƒ˝tiy D˝rayavau x˝yaƒiya van˝ Auramazd˝ha ima xaçam d˝ray˝miy avahayar˝diy vayam Hax˝maniiy˝ ƒahay˝mahay DB 1.26-35 hac˝ paruviyata ˝m˝t˝ amahay ƒ˝tiy D˝rayavau x˝yaƒiya hac˝ paruviyata hay˝ am˝xam taum˝ x˝yaƒiy˝ ima taya man˝ kartam pas˝va yaƒ˝ x˝yaƒiya ˝ha abavam Kab¨jiya n˝ma Kurau puça am˝xam taum˝y˝ DB 1.8-11 *hauv paruvam id˝ x˝yaƒiya ˝ha ƒ˝tiy D˝rayavau x˝yaƒiya avahay˝ Kab¨jiyahay˝ br˝t˝ *Bardiya n˝ma ˝ha VIII man˝ taum˝y˝ tayaiy paruvam x˝yaƒiy˝ ham˝t˝ hamapit˝ Kab¨jiyahay˝ ˝ha pas˝va *Kab¨jiya avam Bardiyam av˝ja adam navama yaƒ˝ Kab¨jiya Bardiyam av˝ja k˝rahay˝ [naiy] IX duvit˝paranam vayam x˝yaƒiy˝ amahay azd˝ abava taya Bardiya avajata pas˝va Kab¨jiya Mudr˝yam *aiyava DB 1.11-12 yaƒ˝ Kab¨jiya Mudr˝yam aiyava pas˝va k˝ra ƒ˝tiy D˝rayavau x˝yaƒiya ar^ka abava van˝ Auramazd˝ha adam x˝yaƒiya amiy [ut˝] drauga dahayauv˝ vasiy abava ut˝ P˝rsaiy Auramazd˝ xaçam man˝ fr˝bara ut˝ M˝daiy ut˝ aniy˝uv˝ dahayuuv˝
DB 1.12-17 DB 1.35-40 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya im˝ dahay˝va tay˝ man˝ *patiy˝ia *pas˝va I martiya ˝ha Gaum˝ta n˝ma van˝ Auramazd˝ha adam˝m x˝yaƒiya ˝ham hauv udapatat˝ hac˝ *Paiiy˝uv˝d˝y˝ Arakadri P˝rsa UÚvja B˝biru Aƒur˝ Arab˝ya Mudr˝ya n˝ma kaufa hac˝ avadaa tayaiy drayahay˝ Sparda Yauna Viyaxanahay˝ m˝hay˝ XIV raucabi ƒakat˝ ˝ha M˝da Armina Katpatuka Parƒava Zraka Haraiva yadiy udapatat˝ Uv˝razm^y B˝xtri Suguda Gad˝ra Saka hauv k˝rahay˝ avaƒ˝ adurujiya ÿatagu Harauvati Maka adam Bardiya amiy haya Kurau puça fraharavam dahay˝va XXIII Kab¨jiyahay˝ br˝t˝
152 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN
DB 1.40-43 avaƒ˝ adam akunavam ˝yadan˝ tay˝ Gaum˝ta pas˝va k˝ra haruva hamiçiya abava hac˝ haya magu viyaka Kab¨jiy˝ adam niyaç˝rayam k˝rahay˝ abicar^ gaiƒ˝mc˝ abiy avam aiyava ut˝ P˝rsa ut˝ M˝da ut˝ aniy˝ m˝niyamc˝ viƒbic˝ tay˝di Gaum˝ta haya dahay˝va magu ad^n˝ xaçam hauv agarb˝yat˝ adam k˝ram g˝ƒav˝ av˝st˝yam P˝rsamc˝ Garmapadahay˝ m˝hy˝ IX raucabi ƒakat˝ ˝ha M˝damc˝ ut˝ aniy˝ dahay˝va avaƒ˝ xaçam agarb˝yat˝ yaƒ˝ paruvamciy pas˝va Kab¨jiya uv˝mariyu amariyat˝ adam taya par˝bartam patiy˝baram van˝ Auramazd˝ha ima adam akunavam DB 1.43-48 adam hamataxaiy y˝t˝ viƒam tay˝m am˝xam ƒ˝tiy D˝rayavau x˝yaƒiya g˝ƒav˝ av˝st˝yam aita xaçam taya Gaum˝ta haya magu ad^n˝ yaƒ˝ paruvamciy Kab¨jiyam avaƒ˝ adam hamataxaiy van˝ Auramazd˝ha aita xaçam hac˝ paruviyata am˝xam taum˝y˝ yaƒ˝ Gaum˝ta haya magu viƒam tay˝m ˝ha am˝xam naiy par˝bara pas˝va Gaum˝ta haya magu ad^n˝ Kab¨jiyam ut˝ P˝rsam ut˝ M˝dam ut˝ aniy˝ dahay˝va DB 1.71-73 hauv ˝yasat˝ ƒ˝tiy D˝rayavau x˝yaƒiya uv˝ipaiyam akut˝ ima taya adam akunavam pas˝va yaƒ˝ x˝yaƒiya hauv x˝yaƒiya abava abavam
DB 1.48-54 DB 1.73-77 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya naiy ˝ha martiya naiy P˝rsa naiy M˝da naiy yaƒ˝ adam Gaum˝tam tayam magum av˝janam am˝xam taum˝y˝ kaciy haya avam pas˝va I martiya AÚçina n˝ma Upadarmahay˝ Gaum˝tam tayam magum xaçam d^tam puça hauv udapatat˝ UÚvjaiy caxriy˝ k˝rahay˝ avaƒ˝ aƒaha k˝raim hac˝ daram atarsa adam UÚvjaiy x˝yaƒiya amiy k˝ram vasiy av˝janiy˝ haya paranam Bardiyam pas˝va UÚvjiy˝ hamiçiy˝ abava ad˝n˝ abiy avam AÚçinam aiyava avahayar˝diy k˝ram av˝janiy˝ hauv x˝yaƒiya abava UÚvjaiy m˝tayam˝m xn˝s˝tiy taya adam naiy Bardiya amiy haya Kurau puça DB 1.77-81 kaciy naiy adarnau ciciy ƒastanaiy pariy ut˝ I martiya B˝biruviya Naditabaira n˝ma Gaum˝tam tayam magum y˝t˝ adam arasam *Ainairahay˝ puça hauv udapatat˝ B˝birauv k˝ram avaƒ˝ adurujiya DB 1.54-61 adam Nabukudaracara amiy haya Nabunaitahay˝ pas˝va adam *Auramazd˝m patiy˝vahayaiy puça Auramazd˝maiy upast˝m abara pas˝va k˝ra haya B˝biruviya haruva abiy avam B˝gay˝dai m˝hay˝ X raucabi ƒakat˝ ˝ha Naditabairam aiyava avaƒ˝ adam had˝ kamnaibi martiyaibi avam B˝biru hamiçiya abava Gaum˝tam tayam magum av˝janam xaçam taya B˝birauv hauv agarb˝yat˝ ut˝ tayaiaiy fratam˝ martiy˝ anuiy˝ ˝hat˝ Sikayauvati n˝m˝ did˝ Nis˝ya n˝m˝ dahay˝u DB 1.81-83 M˝daiy avadaim av˝janam ƒ˝tiy D˝rayavau x˝yaƒiya xaçamim adam ad^nam pas˝va adam fr˝iayam UÚvjam van˝ Auramazd˝ha adam x˝yaƒiya abavam hauv AÚçina basta ˝nayat˝ abiy m˝m Auramazd˝ xaçam man˝ fr˝bara adamim av˝janam
DB 1.61-71 DB 1.83-86 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya xaçam taya hac˝ am˝xam taum˝y˝ par˝bartam pas˝va adam B˝birum aiyavam abiy avam ˝ha ava adam patipadam akunavam Naditabairam haya Nabukudaracara agaubat˝ adamim g˝ƒav˝ av˝st˝yam k˝ra haya Naditabairahay˝ Tigr˝m ad˝raya yaƒ˝ paruvamciy avad˝ aitat˝
153 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN ut˝ abi n˝viy˝ ˝ha Kuganak˝ n˝ma [vardanam P˝rsaiy] avad˝ ad˝raya DB 1.86-90 hauv udapatat˝ UÚvjaiy pas˝va adam k˝ram mak˝uv˝ av˝kanam k˝rahay˝ avaƒ˝ [aƒaha aniyam uab˝rim akunavam aniyahay˝ asam adam] Imani amiy UÚvjaiy x˝yaƒiya fr˝nayam Auramazd˝maiy upast˝m abara DB 2.11-13 van˝ Auramazd˝ha Tigr˝m viyataray˝ma ƒ˝tiy D˝rayavau [x˝yaƒiya] avad˝ avam k˝ram tayam Naditabairahay˝ adam adakaiy adam anaiy ˝ham abiy UÚvjam ajanam vasiy pas˝va *hac˝ma [atarsa] UÚvjiy˝ AÚçiy˝diyahaya m˝hay˝ XXVI raucabi ƒakat˝ avam Martiyam agarb˝ya haya˝m maƒita ˝ha ˝ha [utaim] av˝jana avaƒ˝ hamaranam akum˝ DB 2.13-17 DB 1.90-96 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya I martiya *Fravarti [n˝ma M˝da] hauv pas˝va adam B˝birum aiyavam udapatat˝ M˝daiy aƒaiya B˝birum [yaƒ˝ naiy] *up˝yam Z˝z˝na k˝rahay˝ avaƒ˝ aƒaha n˝ma vardanam anuv Ufr˝tuv˝ [adam Xaƒrita] amiy Uvaxatarahay˝ taum˝y˝ avad˝ [hauv] Naditabaira haya Nabukudaracara pas˝va k˝ra M˝da haya [viƒ˝patiy hauv] hac˝ma agaubat˝ ˝i had˝ k˝r˝ pati [m˝m] hamiçiya abava *hamaranam cartanaiy abiy avam Fravartim aiyava pas˝va hamaranam akum˝ hauv [x˝yaƒiya] abava M˝daiy Auramazd˝maiy upast˝m abara [van˝] Auramazd˝ha k˝ram tayam DB 2.18-30 Naditabairahay˝ adam ajanam vasiy ƒ˝tiy D˝rayavau x˝yaƒiya aniya apiy˝ *˝hayat˝ k˝ra P˝rsa ut˝ M˝da haya up˝ m˝m ˝ha hauv ˝p^im par˝bara kamnam ˝ha AÚn˝makahay˝ m˝hay˝ II raucabi ƒakat˝ ˝ha pas˝va adam k˝ram fr˝iayam avaƒ˝ hamaranam akum˝ Vidarna n˝ma P˝rsa man˝ badaka avam˝m maƒitam akunavam avaƒ˝˝m aƒaham parait˝ avam k˝ram tayam M˝dam jat˝ haya DARIUS’S INSCRIPTION AT BEHISTUN. man˝ naiy gaubataiy COLUMN 2 pas˝va hauv Vidarna had˝ k˝r˝ aiyava yaƒ˝ M˝dam par˝rasa *M˝ru n˝ma vardanam DB 2.1-5 M˝daiy avad˝ hamaranam akunau had˝ ƒ˝tiy D˝rayavau x˝yaƒiya M˝daibi pas˝va Naditabaira had˝ kamnaibi asab˝raibi haya M˝daiuv˝ maƒita ˝ha hauv adakaiy naiy amuƒa avad˝ ˝ha B˝birum aiyava Auramazd˝maiy upast˝m abara pas˝va adam B˝birum aiyavam van˝ Auramazd˝ha k˝ra [haya] man˝ avam [van˝] Auramazd˝ha ut˝ B˝birum agarb˝yam k˝ram tayam hamiçiyam aja vasiy ut˝ avam Naditabairam agarb˝yam AÚn˝makahay˝ m˝hay˝ XXVII raucabi ƒakat˝ pas˝va avam Naditabairam adam B˝birauv ˝ha avaƒ˝˝m hamaranam kartam av˝janam pas˝va hauv k˝ra haya man˝ Kapada n˝ma dahay˝u M˝daiy avad˝ m˝m am˝naiya y˝t˝ DB 2.5-8 adam arasam M˝dam [ƒ˝tiy] D˝rayavau x˝yaƒiya y˝t˝ adam B˝birauv ˝ham [im˝ dahay˝va] tay˝ DB 2.29-37 hac˝ma hamiçiy˝ abava ƒ˝tiy D˝rayavau x˝yaƒiya P˝rsa UÚvja M˝da *Aƒur˝ [Mudr˝ya] *Parƒava D˝dari n˝ma Arminiya man˝ badaka avam Margu ÿatagu Saka adam fr˝iayam Arminam avaƒ˝aiy aƒaham DB 2.8-11 paraidiy k˝ra haya hamiçiya man˝ naiy ƒ˝tiy D˝rayavau x˝yaƒiya gaubataiy avam jadiy [I] *martiya Martiya n˝ma Cicaxrai puça
154 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN pas˝va D˝dari aiyava AÚn˝makahay˝ m˝hay˝ XV raucabi ƒakat˝ ˝ha yaƒ˝ Arminam par˝rasa pas˝va hamiçiy˝ avaƒ˝˝m hamaranam kartam hagmat˝ parait˝ pati D˝darim hamaranam cartanaiy DB 2.57-64 Z¨zahaya n˝ma ˝vahanam Arminiyaiy avad˝ ƒ˝tiy D˝rayavau x˝yaƒiya hamaranam akunava patiy duvit^yam hamiçiy˝ hagmat˝ parait˝ pati Auramazd˝maiy upast˝m abara Vaumisam hamaranam cartanaiy van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram Autiy˝ra n˝m˝ dahay˝u Arminiyaiy avad˝ tayam hamiçiyam aja vasiy hamaranam akunava ÿ¨rav˝harahay˝ m˝hay˝ VIII raucabi ƒakat˝ Auramazd˝maiy upast˝m abara ˝ha van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram avaƒ˝˝m hamaranam kartam tayam hamiçiyam aja vasiy ÿ¨rav˝harahay˝ m˝hay˝ jiyamnam patiy DB 2.37-42 avaƒ˝˝m hamaranam kartam ƒ˝tiy D˝rayavau x˝yaƒiya pas˝va Vaumisa cit˝ m˝m am˝naya Arminiyaiy patiy duvit^yam hamiçiy˝ hagmat˝ parait˝ pati y˝t˝ adam arasam M˝dam D˝darim hamaranam cartanaiy Tigra n˝m˝ did˝ Arminiyaiy avad˝ hamaranam DB 2.64-70 akunava ƒ˝tiy D˝rayavau x˝yaƒiya Auramazd˝maiy upast˝m abara pas˝va adam nij˝yam hac˝ B˝birau van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram aiyavam M˝dam tayam hamiçiyam aja vasiy yaƒ˝ M˝dam par˝rasam Kuduru n˝ma ÿ¨rav˝harahay˝ m˝hay˝ XVIII raucabi ƒakat˝ vardanam M˝daiy avad˝ hauv Fravarti haya ˝ha M˝daiy x˝yaƒiya agaubat˝ ˝i had˝ k˝r˝ pati avaƒ˝˝m hamaranam kartam m˝m hamaranam cartanaiy pas˝va hamaranam akum˝ DB 2.42-49 Auramazd˝maiy upast˝m abara ƒ˝tiy D˝rayavau x˝yaƒiya van˝ Auramazd˝ha k˝ram tayam Fravartai patiy çit^yam hamiçiy˝ hagmat˝ parait˝ pati adam ajanam vasiy D˝darim hamaranam cartanaiy AÚdukanaiahay˝ m˝hay˝ XXV raucabi ƒakat˝ Uyam˝ n˝m˝ did˝ Arminiyaiy avad˝ hamaranam ˝ha akunava avaƒ˝ hamaranam akum˝ Auramazd˝maiy upast˝m abara van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram DB 2.70-78 tayam hamiçiyam aja vasiy ƒ˝tiy D˝rayavau x˝yaƒiya ÿ˝igracai m˝hay˝ IX raucabi ƒakat˝ ˝ha pas˝va hauv Fravarti had˝ kamnaibi avaƒ˝˝m hamaranam kartam asab˝raibi amuƒa pas˝va D˝dari cit˝ m˝m am˝naya Arminiyaiy Rag˝ n˝ma dahay˝u M˝daiy avapar˝ aiyava y˝t˝ adam arasam M˝dam pas˝va adam k˝ram fr˝iaya nipadiy Fravarti agarbiya ˝nayat˝ abiy m˝m DB 2.49-57 adamaiy ut˝ n˝ham ut˝ gau˝ ut˝ haz˝nam ƒ˝tiy D˝rayavau x˝yaƒiya fr˝janam ut˝aiy I cama avajam Vaumisa n˝ma P˝rsa man˝ badaka avam adam duvaray˝maiy basta ad˝riya fr˝iayam Arminam haruvaim k˝ra avaina avaƒ˝aiy aƒaham pas˝vaim Hagmat˝naiy uzmay˝patiy akunavam paraidiy k˝ra haya hamiçiya man˝ naiy ut˝ martiy˝ tayaiaiy fratam˝ anuiy˝ ˝hat˝ avaiy gaubataiy avam jadiy Hagmat˝naiy [atar] did˝m fr˝hajam pas˝va Vaumisa aiyava yaƒ˝ Arminam par˝rasa pas˝va hamiçiy˝ DB 2.78-91 hagmat˝ parait˝ pati Vaumisam hamaranam ƒ˝tiy D˝rayavau x˝yaƒiya cartanaiy I martiya Ciçataxma n˝ma Asagartiya hauvmaiy Izal˝ n˝m˝ dahay˝u Aƒur˝y˝ avad˝ hamaranam hamiçiya abava akunava k˝rahay˝ avaƒ˝ aƒaha Auramazd˝maiy upast˝m abara adam x˝yaƒiya amiy Asagartaiy Uvaxtarahay˝ van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram taum˝y˝ tayam hamiçiyam aja vasiy pas˝va adam k˝ram P˝rsam ut˝ M˝dam
155 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN
fr˝iayam avaƒ˝˝m hamaranam kartam Taxmasp˝da n˝ma M˝da man˝ badaka avam˝m maƒitam akunavam DB 3.9-10 avaƒ˝˝m aƒaham ƒ˝tiy D˝rayavau x˝yaƒiya parait˝ k˝ram hamiçiyam haya man˝ naiy pas˝va dahay˝u man˝ abava gaubataiy avam jat˝ ima taya man˝ kartam Parƒavaiy pas˝va Taxmasp˝da had˝ k˝r˝ aiyava hamaranam akunau had˝ Ciçataxm˝ DB 3.10-12 Auramazd˝maiy upast˝m abara ƒ˝tiy D˝rayavau x˝yaƒiya van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram Margu n˝m˝ dahay˝u hauvmaiy hamiçiy˝ tayam hamiçiyam aja abava ut˝ Ciçataxmam agarb˝ya ˝naya abiy m˝m I martiya Fr˝da n˝ma M˝rgava avam maƒitam pas˝vaaiy adam ut˝ n˝ham ut˝ gau˝ fr˝janam akunavat˝ ut˝aiy I cama avajam duvaray˝maiy basta ad˝riya DB 3.12-15 haruvaim k˝ra avaina pas˝va adam fr˝iayam D˝dari n˝ma P˝rsa pas˝vaim Arbair˝y˝ uzmay˝patiy akunavam man˝ badaka B˝xtr^y˝ xaçap˝v˝ abiy avam avaƒ˝aiy aƒaham DB 2.91-92 paraidiy avam k˝ram jadiy haya man˝ naiy ƒ˝tiy D˝rayavau x˝yaƒiya gaubataiy ima taya man˝ kartam M˝daiy DB 3.15-19 DB 2.92-98 pas˝va D˝dari had˝ k˝r˝ aiyava ƒ˝tiy D˝rayavau x˝yaƒiya hamaranam akunau had˝ M˝rgavaibi Parƒava ut˝ Vark˝na [hamiçiy˝] *abava *hac˝ma Auramazd˝maiy upast˝m abara *Fravartai *agaubat˝ van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram Vit˝spa man˝ pit˝ hauv [Parƒavaiy] ˝ha tayam hamiçiyam aja vasiy avam k˝ra *avaharda AÚçiy˝diyahaya m˝hay˝ XXIII raucabi ƒakat˝ [hamiçiya] abava ˝ha pas˝va Vit˝spa *aiyava [had˝] *k˝r˝ *hayaaiy avaƒ˝˝m hamaranam kartam *anuiya ˝ha *Vipauz˝ti n˝ma vardanam [Parƒavaiy] avad˝ DB 3.19-21 hamaranam akunau had˝ Parƒavaibi ƒ˝tiy D˝rayavau x˝yaƒiya Auramazd˝maiy [upast˝m abara] pas˝va dahay˝u man˝ abava van˝ Auramazd˝ha [Vit˝spa] avam k˝ram ima taya man˝ kartam B˝xtr^y˝ [tayam] hamiçiyam [aja vasiy] Viyaxanahaya m˝hay˝ [XXII raucabi] ƒakat˝ DB 3.21-25 ˝ha ƒ˝tiy D˝rayavau x˝yaƒiya avaƒ˝˝m hamaranam kartam I martiya Vahayazd˝ta T˝rav˝ n˝ma vardanam Yautiy˝ n˝m˝ dahay˝u P˝rsaiy avad˝ ad˝raya hauv duvit^yam udapatat˝ P˝rsaiy k˝rahay˝ avaƒ˝ aƒaha DARIUS’S INSCRIPTION AT BEHISTUN. adam Bardiya amiy haya Kurau puça COLUMN 3 DB 3.25-28 DB 3.1-9 pas˝va k˝ra P˝rsa haya viƒ˝patiy hac˝ Yad˝y˝ ƒ˝tiy D˝rayavau x˝yaƒiya frataram hauv hac˝ma hamiçiya abava pas˝va adam k˝ram P˝rsam fr˝iayam abiy abiy avam Vahayazd˝tam aiyava Vit˝spam hac˝ Rag˝y˝ hauv x˝yaƒiya abava P˝rsaiy yaƒ˝ hauv k˝ra par˝rasa abiy Vit˝spam pas˝va Vit˝spa ˝yasat˝ avam k˝ram aiyava Patigraban˝ n˝ma vardanam Parƒavaiy DB 3.28-33 avad˝ hamaranam akunau had˝ hamiçiyaibi ƒ˝tiy D˝rayavau x˝yaƒiya Auramazd˝maiy upast˝m abara pas˝va adam k˝ram P˝rsam ut˝ M˝dam van˝ Auramazd˝ha Vit˝spa avam k˝ram tayam fr˝iayam haya up˝ m˝m ˝ha hamiçiyam aja vasiy Artavardiya n˝ma P˝rsa man˝ badaka avam˝m Garmapadahaya m˝hay˝ I rauca ƒakatam ˝ha maƒitam akunavam
156 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN haya aniya k˝ra P˝rsa pas˝ man˝ aiyava M˝dam D˝rayavahau x˝yaƒiyahay˝ gaubataiy
DB 3.33-36 DB 3.59-64 pas˝va Artavardiya had˝ k˝r˝ aiyava P˝rsam pas˝va hauv k˝ra aiyava tayam Vahayazd˝ta yaƒ˝ P˝rsam par˝rasa Rax˝ n˝ma vardanam fr˝iaya abiy Viv˝nam hamaranam cartanaiy P˝rsaiy avad˝ hauv Vahayazd˝ta haya Bardiya K˝piak˝ni n˝m˝ did˝ avad˝ hamaranam agaubat˝ ˝i had˝ k˝r˝ pati A˘rtavardiyam akunava hamaranam cartanaiy Auramazd˝maiy upast˝m abara van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram DB 3.36-40 tayam hamiçiyam aja vasiy pas˝va hamaranam akunava AÚn˝makahaya m˝hay˝ XIII raucabi ƒakat˝ ˝ha Auramazd˝maiy upast˝m abara avaƒ˝˝m hamaranam kartam van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram tayam Vahayazd˝tahaya aja vasiy DB 3.64-69 ÿ¨rav˝harahaya m˝hay˝ XII raucabi ƒakat˝ ˝ha ƒ˝tiy D˝rayavau x˝yaƒiya avaƒ˝˝m hamaranam kartam patiy hay˝param hamiçiy˝ hagmat˝ parait˝ pati Viv˝nam hamaranam cartanaiy DB 3.40-49 Gadutava n˝m˝ dahay˝u avad˝ hamaranam ƒ˝tiy D˝rayavau x˝yaƒiya akunava pas˝va hauv Vahayazd˝ta had˝ kamnaibi Auramazd˝maiy upast˝m abara asab˝rabi amuƒa van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram aiyava Paiiy˝uv˝d˝m tayam hamiçiyam aja vasiy hac˝ avada k˝ram ˝yasat˝ Viyaxanahaya m˝hay˝ VII raucabi ƒakat˝ ˝ha hay˝param ˝i pati A˘rtavardiyam hamaranam avaƒ˝˝m hamaranam kartam cartanaiy Parga n˝ma kaufa avad˝ hamaranam akunava DB 3.69-75 Auramazd˝maiy upast˝m abara ƒ˝tiy D˝rayavau x˝yaƒiya van˝ Auramazd˝ha k˝ra haya man˝ avam k˝ram pas˝va hauv martiya haya avahay˝ k˝rahay˝ tayam Vahayazd˝tahaya aja vasiy maƒita ˝ha tayam Vahayazd˝ta fr˝iaya abiy n Garmapadahaya m˝hay˝ V raucabi ƒakat˝ ˝ha Viv˝nam hauv amu ƒa had˝ kamnaibi avaƒ˝˝m hamaranam kartam asab˝raibi ut˝ avam Vahayazd˝tam agarb˝ya ut˝ martiy˝ aiyava Ar˝d˝ n˝m˝ did˝ Harauvat^y˝ tayaaiy fratam˝ anuiy˝ ˝hata agarb˝ya avapar˝ atiy˝i pas˝va Viv˝na had˝ k˝r˝ *nipadiaiy [x x x] aiyava DB 3.49-52 avad˝im agarb˝ya ƒ˝tiy D˝rayavau x˝yaƒiya ut˝ martiy˝ tayaiaiy fratam˝ anuiy˝ ˝hat˝ av˝ja pas˝va adam avam Vahayazd˝tam ut˝ martiy˝ tayaiaiy fratam˝ anuiy˝ ˝hata Uv˝daicaya DB 3.75-76 n˝ma vardanam P˝rsaiy avadai ƒ˝tiy D˝rayavau x˝yaƒiya uzamay˝patiy akunavam pas˝va dahay˝u man˝ abava ima taya man˝ kartam Harauvat^y˝ DB 3.52-53 ƒ˝tiy D˝rayavau x˝yaƒiya DB 3.76-83 ima taya man˝ kartam P˝rsaiy ƒ˝tiy D˝rayavau x˝yaƒiya y˝t˝ adam P˝rsaiy *ut˝ M˝daiy ˝ham patiy duvit^yam B˝biruviy˝ hamiçiy˝ abava hac˝ma DB 3.53-59 I martiya Araxa n˝ma Arminiya Halditahaya ƒ˝tiy D˝rayavau x˝yaƒiya puça hauv udapatat˝ B˝birauv Dub˝la n˝m˝ hauv Vahayazd˝ta haya Bardiya agaubat˝ hauv dahay˝u hac˝ avada k˝ram fr˝iaya Harauvat^m Viv˝na n˝ma hauv k˝rahay˝ avaƒ˝ adurujiya P˝rsa man˝ badaka Harauvat^y˝ xaçap˝v˝ adam Nabukudaracara amiy haya Nabunaitahaya abiy avam puça ut˝˝m I martiyam maƒitam akunau pas˝va k˝ra B˝biruviya hac˝ma hamiçiya abava avaƒ˝˝m aƒaha abiy avam Araxam aiyava parait˝ Viv˝nam jat˝ ut˝ avam k˝ram haya B˝birum hauv agarb˝yat˝
157 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN hauv x˝yaƒiya abava B˝birauv DB 4.12-15 [I] Niditabaira n˝ma B˝biruviya DB 3.83-86 hauv adurujiya ƒ˝tiy D˝rayavau x˝yaƒiya avaƒ˝ aƒaha pas˝va adam k˝ram fr˝iayam B˝birum adam Nabukudracara [amiy] haya Vidafarn˝ n˝ma P˝rsa man˝ badaka avam˝m Nabunaitahaya puça maƒitam akunavam hauv B˝birum hamiçiyam akunau avaƒ˝˝m aƒaham parait˝ avam k˝ram B˝biruviyam jat˝ haya man˝ DB 4.15-18 naiy gaubataiy I Martiya n˝ma P˝rsa hauv adurujiya DB 3.86-92 avaƒ˝ aƒaha pas˝va Vidafarn˝ had˝ k˝r˝ aiyava B˝birum adam Imani amiy UÚvjaiy x˝yaƒiya Auramazd˝maiy upast˝m abara hauv UÚvjam hamiçiyam akunau van˝ Auramazd˝ha Vidafarn˝ B˝biruviy˝ aja ut˝ [bast˝ ˝naya] DB 4.18-20 [Varkazanahaya] m˝hay˝ XXII raucabi ƒakat˝ I Fravarti n˝ma M˝da ˝ha hauv adurujiya avaƒ˝ avam Arxam [haya] *Nabukudaracara avaƒ˝ aƒaha [duruxta]m agaubat˝ ut˝ martiy˝ tayaiaiy adam Xaƒrita amiy Uvaxtarahaya taum˝y˝ fratam˝ anuiy˝ [˝hat˝ agarb˝ya] adam x˝yaƒiya amiy M˝daiy *niyat˝yam hauv Arxa ut˝ martiy˝ tayaiaiy hauv M˝dam hamiçiyam akunau fratam˝ anuiy˝ ˝hat˝ B˝birauv uzmay˝patiy akariyat˝ DB 4.20-23 I Ciçataxma n˝ma Asagartiya hauv adurujiya DARIUS’S INSCRIPTION AT BEHISTUN. avaƒ˝ aƒaha COLUMN 4 adam x˝yaƒiya amiy Asagartaiy Uvaxtarahay˝ taum˝y˝ DB 4.1-2 hauv Asagartam hamiçiyam akunau ƒ˝tiy D˝rayavau x˝yaƒiya ima taya man˝ kartam [B˝birauv] DB 4.23-26 I Fr˝da n˝ma M˝rgava DB 4.2-7 hauv adurujiya ƒ˝tiy D˝rayavau x˝yaƒiya avaƒ˝ aƒaha ima taya adam akunavam van˝ Auramazd˝ha adam x˝yaƒiya amiy Margauv hamahay˝y˝ ƒarda pas˝va yaƒ˝ x˝yaƒiya hauv Margum hamiçiyam akunau[ abavam XIX hamaran˝ akunavam DB 4.26-28 van˝ Auramazd˝ha adami ajanam [I] *Vahayazd˝ta n˝ma P˝rsa ut˝ IX x˝yaƒiy˝ agarb˝yam hauv adurujiya avaƒ˝ aƒaha DB 4.7-10 adam Bardiya amiy haya Kurau puça I Gaum˝ta n˝ma magu hauv P˝rsam hamiçiyam akunau [hauv] adurujiya avaƒ˝ aƒaha DB 4.28-31 adam Bardiya amiy [haya] Kurau puça I Araxa n˝ma Arminiya hauv P˝rsam hamiçiyam akunau [hauv] adurujiya avaƒ˝ aƒaha DB 4.10-12 adam Nabukudracara amiy haya Nabunaitahaya I AÚçina n˝ma UvjiyaÚ puça hauv adurujiya hauv B˝birum hamiçiyam akunau avaƒ˝ [aƒaha adam] x˝yaƒiya amiy UÚvjaiy DB 4.31-32 hauv UÚvjam hamiçiyam akunau ƒ˝tiy D˝rayavau x˝yaƒiya imaiy IX x˝yaƒiy˝ tayaiy *adam agarb˝yam
158 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN
atar im˝ hamaran˝ DB 4.57-59 DB 4.33-36 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya yadiy im˝m hadug˝m apagauday˝hay naiy ƒ˝hay dahay˝va im˝ tay˝ hamiçiy˝ abava draugadi *k˝rahay˝ *hamiçiy˝ akunau Auramazd˝taiy jat˝ biy˝ ut˝taiy taum˝ m˝ biy˝ taya imaiy k˝ram adurujiyaa pas˝vadi *Auramazd˝ man˝ dastay˝ akunau DB 4.59-61 yaƒ˝ m˝m k˝ma avaƒ˝di [akunavam] ƒ˝tiy D˝rayavau x˝yaƒiya ima taya adam akunavam hamahay˝y˝ ƒarda DB 4.36-40 van˝ Auramazd˝ha akunavam ƒ˝tiy D˝rayavau x˝yaƒiya Auramazd˝maiy upast˝m abara ut˝ aniy˝ha tuvam k˝ *x˝yaƒiya haya aparam ˝hay hac˝ bag˝ha tayaiy hatiy draug˝ daram patipayauv˝ martiya [haya] *draujana ahatiy avam ufratam DB 4.61-67 pars˝ yadiy avaƒ˝ *maniy˝hay ƒ˝tiy D˝rayavau x˝yaƒiya dahay˝umaiy duruv˝ ahatiy avahayar˝dimaiy Auramazd˝ upast˝m abara ut˝ aniy˝ha bag˝ha tayaiy [hatiy] DB 4.41-43 *yaƒ˝ naiy ar^ka ˝ham naiy draujana ˝ham naiy ƒ˝tiy D˝rayavau x˝yaƒiya z¨rakara ˝ham ima taya adam akunavam van˝ Auramazd˝ha naiy adam naimaiy taum˝ hamahay˝y˝ ƒarda akunavam upariy art˝m upariy˝yam tuvam k˝ haya aparam im˝m dipim patipars˝hay naiy kauƒim naiy tunuvatam z¨ra akunavam taya man˝ kartam varnavat˝m ƒuv˝m martiya haya hamataxat˝ man˝ viƒiy˝ avam m˝taya *druxtam maniy˝hay ubartam abaram haya viyan˝ƒaya avam ufratam aparsam DB 4.43-45 ƒ˝tiy D˝rayavau x˝yaƒiya DB 4.67-69 Auramazd˝ha *ragam *vardiyaiy yaƒ˝ ima ƒ˝tiy D˝rayavau x˝yaƒiya haiyam naiy duruxtam adam *akunavam tuvam [k˝] x˝yaƒiya haya aparam ˝hay martiya *hamahay˝y˝ ƒarda haya draujana ahatiy hayav˝ z¨rakara ahatiy avaiy m˝ daut˝ biy˝ DB 4.45-50 ufrat˝di pars˝ ƒ˝tiy D˝rayavau x˝yaƒiya van˝ Auramazd˝ha ut˝maiy aniyaciy vasiy DB 4.69-72 astiy kartam ƒ˝tiy D˝rayavau x˝yaƒiya ava ahay˝y˝ dip^y˝ naiy nipitam tuvam k˝ haya aparam im˝m dipim vain˝hay avahayar˝diy naiy nipitam taya adam niyapaiƒam imaiv˝ patikar˝ m˝taya m˝taya haya aparam im˝m dipim patipars˝tiy vikanahay avahay˝ paruv ƒaday˝tiy taya man˝ kartam y˝v˝ utava ˝hay avaƒ˝di paribar˝ naiim ima varnav˝taiy duruxtam maniy˝taiy DB 4.72-76 DB 4.50-52 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya yadiy im˝m dipim vain˝hay imaiv˝ patikar˝ tayaiy paruv˝ x˝yaƒiy˝ y˝t˝ ˝ha avai˝m av˝ naiydi vikanahay naiy astiy kartam yaƒ˝ man˝ van˝ ut˝taiy y˝v˝ taum˝ [ahatiy] paribar˝hadi Auramazd˝ha hamahay˝y˝ ƒarda kartam Auramazd˝ ƒuv˝m daut˝ biy˝ ut˝taiy taum˝ vasiy biy˝ DB 4.52-57 ut˝ dargam j^v˝ ƒ˝tiy D˝rayavau x˝yaƒiya ut˝ taya kunav˝hay avataiy Auramazd˝ uc˝ram n¨ram ƒuv˝m varnavat˝m taya man˝ kartam kunautuv avaƒ˝ k˝rahay˝ *r˝diy m˝ apagaudaya yadiy im˝m hadug˝m naiy apagauday˝hay DB 4.76-80 k˝rahay˝ ƒ˝hay ƒ˝tiy D˝rayavau x˝yaƒiya Auramazd˝ ƒuv˝m daut˝ biy˝ ut˝taiy taum˝ yadiy im˝m dipim imaiv˝ patikar˝ vain˝hay vasiy biy˝ ut˝ dargam j^v˝ vikanahadi
159 July 31, 2005 APPENDIX 2. DARIUS’S INSCRIPTION AT BEHISTUN ut˝taiy y˝v˝ taum˝ ahatiy naiydi paribar˝hay tay˝m˝m maƒitam agarb˝ya ˝naya abiy m˝m Auramazd˝taiy jat˝ biy˝ ut˝im adam av˝janam ut˝taiy taum˝ [m˝ biy˝] pas˝va dahay˝u man˝ [abava] ut˝ taya kunav˝hay avataiy Auramazd˝ nikatuv DB 5.14-18 DB 4.80-86 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya avaiy UÚvjiy˝ [ar^k˝ ˝ha] ut˝˝m Auramazd˝ naiy imaiy martiy˝ tayaiy adakaiy avad˝ *˝hat˝ y˝t˝ *ayadiya adam Gaum˝tam tayam magum av˝janam Auramazd˝m ayadaiy haya Bardiya agaubat˝ van˝ Auramazd˝ha [yaƒ˝] m˝m [k˝ma] adakaiy imaiy martiy˝ hamataxat˝ anuiy˝ *avaƒ˝di akunavam man˝ Vidafarn˝ n˝ma *Vahayasparahay˝ puça P˝rsa DB 5.18-20 *Ut˝na n˝ma ÿuxrahay˝ puça P˝rsa ƒ˝tiy D˝rayavau x˝yaƒiya *Gaubaruva n˝ma Marduniyahay˝ puça P˝rsa haya Auramazd˝m yad˝taiy *y˝nam [avahay˝] Vidarna n˝ma Bag˝bignahay˝ puça P˝rsa ahatiy ut˝ j^vahay˝ ut˝ martahay˝ Bagabuxa n˝ma D˝tuvahayahay˝ puça P˝rsa *Ardumani n˝ma Vahaukahay˝ puça P˝rsa DB 5.20-30 ƒ˝tiy D˝rayavau x˝yaƒiya DB 4.86-88 pas˝va had˝ k˝r˝ adam aiyavam abiy Sak˝m ƒ˝tiy D˝rayavau x˝yaƒiya *pas˝ Sak˝ tayaiy xaud˝m tigr˝m baratiy *imaiy tuvam k˝ x˝yaƒiya haya aparam ˝hay tay˝m [pati m˝m] *˝ia imai˝m martiy˝n˝m taum˝m *ubart˝m *yadiy abiy draya *av˝rasam *draxt˝ [ava] paribar˝ *had˝ *k˝r˝ vis˝ viyatarayam [pas˝va] avaiy Sak˝ [adam] ajanam DB 4.88-92 aniyam agarb˝yam ƒ˝tiy D˝rayavau x˝yaƒiya [aniya] *basta [˝nayat˝] abiy m˝m van˝ Auramazd˝ha ima *dipiciçam taya adam *ut˝˝m [haya] *maƒita Skuxa n˝ma avam akunavam agarb˝ya patiam ariy˝ ut˝ pavast˝y˝ ut˝ carm˝ *graftam *bastam ˝naya [abiy m˝m] [˝ha] avad˝ aniyam maƒitam akunavam *yaƒ˝ m˝m *patiamciy *n˝man˝fam akunavam k˝ma *patiam *uvad˝tam [akunavam] pas˝va dahay˝u man˝ abava ut˝ *niyapaiƒiya ut˝ patiyafrasiya paiiy˝ m˝m pas˝va ima *dipiciçam fr˝st˝yam vispad˝ atar DB 5.30-33 dahay˝va ƒ˝tiy D˝rayavau x˝yaƒiya k˝ra *ham˝taxat˝ [avaiy] Sak˝ ar^k˝ ˝ha ut˝ naiy Auramazd˝[˝m] *ayadiya Auramazd˝m ayadaiy DARIUS’S INSCRIPTION AT BEHISTUN. van˝ Auramazd˝ha yaƒ˝ m˝m k˝ma avaƒ˝di COLUMN 5 akunavam
DB 5.1-4 DB 5.33 ƒ˝tiy D˝rayavau x˝yaƒiya ƒ˝tiy D˝rayavau x˝yaƒiya ima taya adam akunavam duvit^y˝mca *çit˝mca [haya] Auramazd˝m yad˝taiy [avahay˝] *y˝nam ƒardam pas˝va yaƒ˝ x˝yaƒiya [abavam] [ahatiy] ut˝ j^vahay˝ ut˝ *martahay˝
DB 5.4-14 UÚvja n˝m˝ dahay˝u hauv hamiçiy˝ abava [I martiya] Atamaita n˝ma UÚvjiya [avam] maƒitam akunavat˝ pas˝va adam k˝ram fr˝iayam [I] *martiya Gaubaruva n˝ma P˝rsa man˝ badaka avam˝m maƒitam akunavam pas˝va Gaubaruva [had˝] k˝r˝ aiyava UÚvjam [hamaranam] akunau had˝ UÚvjiyaibi pas˝va Gaubaruva UÚvjiy˝ aja ut˝ viyamarda ut˝
160 July 31, 2005
OLD PERSIAN - ENGLISH GLOSSARY abayapara (*abiyapara): subsequently Ariya-: Aryan 2 control (of: + gen.dat.) 19 Ariya-ciça-: of Aryan stock 2 azd˝ •bav-: become known (+ taya abicari-: *pasture 7 Ariy˝ramna-: Ariaramnes 2 “that”) 8 abi-j˝vaya- < •jav: to add (to: + abiy ar^ka-: disloyal 3 azd˝ •kar-: to make known (+ taya + acc.) 4 Armina-: Armenia 4 “that”) 8 Abir˝du-: place in Elam 14 Arminiya-: Armenian 1 *˝-ai-/i- < •ai: to come 6 abiy: to, over to, toward (+ acc.) 4 Ar˝da-: place name 12 *˝-bara- < •bar: bring (about), *abiyapara (abayapara): subsequently A˘r˝ma-: Arsames 2 endeavor, perform; to bring (things 19 art˝-: rectitude, righteousness 13 to) 9, 11 ˝Ÿciy: until, as long as 10 arti- fem.: spear 3 AÚçina-: proper name 6 ada-: then 15 artika- (or ˝rtika-?): spearman 2 AÚçiy˝diya-: month name (Nov.-Dec.) adakaiy: then, at that time 5 A˘rtaxaç˝- masc.: Artaxerxes 2 12 adam: I 1 A˘rtavardiya-: proper name; one of AÚdukanaia-: month name 9 afuv˝-: fear 5 Darius’s generals 6 ˝hat, ˝han he was, they were 2 agriya- (or ˝g(a)riya-): loyal 13 art˝c˝ (< art˝ hac˝): according to the ˝haya- < ahaya- *aguru- (Akk. agurru): baked brick (universal) Order 5 ˝-jamiy˝ (optative) < •gam: to come 11 art˝van-: blessed, belonging to or (to) 9 ahaya-* < •ah: to throw 13 acting according to the (universal) AÚkaufaiya-: mountain dwellers, ahmatah: from there 9 Order (after death) 5 tribal name 3 ahmiy: I am 1 aruvasta- neut.: physical ability 7 ˝m˝ta-: distinguished, noble 3 ahm˝xam: our 3 *asa-, see aspa- *˝-naya- < •nay: to bring (people to) Ahuramazd˝-: Ahuramazd˝ 1 asa-b˝ra-: a rider, on horseback 2 10 Asagarta-: Sagartia 10 ai-/i-: to go 10 AÚn˝maka-: month name (the month Ainaira-: proper name 7 Asagarta-: Sagartia 15 in which the name of God should aita: this (neut.) 4 Asagartiya-: Sagartian 10 be invoked?) 12 aitiy < ai-/i- asan-: stone 14 ˝ranjana-: decoration 11 asman-: heaven 8 aiva-: one 5 ˝vahana- neut.: settlement 14 akum˝ < •kar: we did 9 asman-: sky 10 ˝-xnau- < •xnu act./mid.: to hear akunaiy: past innitive? 19 aspa-: horse 2 10 akunau- < •kar: made 4 Aspacanah-: proper name, ˝yadana- neut.: place of worship, Aspathines 2 h h h temple 4 amata > a mata anaiy: near(?) 7 An˝hat˝-, An˝hit˝-: Anahita 19 ˝-yasa- < •yam mid.: to appropriate, ati-ay- < •ay: to *pass (near) by 12 assume command of 6 aniya-: other; aniya- ... aniya-: one ... aƒaiya (uncertain reading): *at rst another 4 ˝Ÿciy: until 10 13 baga-: god 1 antar: among, in (+ acc.) 4 aƒanga-: stone 3 Bagabuxa-: proper name; anuiya-: a loyal follower 3 aƒangaina-, fem. aƒangain^- (lesson Megabyxus, son of D˝tuvahaya- anuv: according to (+ gen.-dat.), 7): (made) of stone 3 and one of Darius’s six helpers 13 along (+ instr.-abl.) 5, 9 Aƒuriya-: Assyrian 2 Bag˝bigna-: proper name; father of ap- fem.: water 7 Aƒur˝-: Assyria 2 Vidarna-, one of Darius’s six apa-gaudaya •gaud: to hide 16 aur˝: (down) hither 15 helpers 13 apad˝na-: palace, throne hall 11 ava-jan- < •jan: to kill 6 n *apaniy˝ka- (apanay˝ka-): great- *ba daya-, pp. basta- < •band: to ava-jata- < ava-jan-: killed 8 grandfather 19 bind 10 ava-kan- < •kan: load onto 7 n aparam: henceforth, afterward 15 ba daka-: loyal subject 3 ava-rasa-: to come down to (+ abiy + apataram: further away (from), in bara-, inf. bartanaiy < •bar: to carry acc.) 8 addition to (+ hac˝) 10 4 ava-st˝ya- < •st˝: to place 7 apiy: also 15 Bardiya-: Smerdis 6 avada: thence 9 Arab˝ya-: Arabia 7 barn˝ (< *barzan-): in height, depth avad˝: there 6 Arakadri-: name of a mountain 9 4 avahaya-r˝diy: for this (the arani-: a cubit 4 basta-, pp. of *bandaya-: to bind 10 following) reason 4 Araxa-: name of an Armenian rebel, bava- < •bav: to become 5 avapar˝: thither 11 barzman-: height, the highest 5 son of Haldita- 1 h Arbair˝-: Arbela (place name) 9 avarda for ava- arda- < •hard(?): to B˝biru- : Babylon, Babylonian 2 ardast˝na-: window sill 5 leave, relinquish 15 B˝biruviya-: Babylonian 6 ardata- neut.: silver 14 avaƒ˝: thus, in that manner 6 B˝gay˝di-: month name 10 Ardumani-: proper name; av˝ < avant-: so much 14 b˝ji-: tribute; b˝jim bara-: pay tax (to: (Herodotus: Aspathines!), son of av˝karam: of such a sort 10 gen.-dat.) 10 Vahauka- and one of Darius’s six axaina-: blue-green (turquoise) 14 b˝tugara-: kind of vessel 19 helpers 13 axata-: undisturbed 15 B˝xtr˘- fem.: Bactria 7 *ayaumaini-: uncoordinated, *not in
161 July 31, 2005 OLD PERSIAN - ENGLISH GLOSSARY br˝tar-: brother 8 drayah- neut.: ocean 7 Gaum˝ta-: proper name 6 b¨m˘- fem.: earth 10 Dub˝la-: place name 22 gaua-: ear 10 caxriy˝, opt. perf. of •kar durujiya-, pp. duruxta- < •draug: to g˝ƒu-: place, throne 7 carman-: skin, hide, parchment 17 (tell a) lie, deceive 6 *grafta-, pp. of garb˝ya-: seized, cartanaiy inf. of •kar: to do 9 duruva-: healthy, whole 7 grasped caman- neut.: eye 6 duruxta-, pp. of durujiya-: false (lit. hac˝: from (prep. + inst.-abl.) 3 -c˝: and; -c˝ ... -c˝: both ... and 7 “lied up”) 2 hac˝ma: from me 6 Ciça(n)taxma-: proper name 10 duiy˝ra- neut.: bad year (famine) 9 had˝ + instr.-abl.: together with dukarta-: something badly done, evil (people) 9 Cincaxri-: proper name 7 deed 8 hadi- neut.: palace 11 Cipi-: proper name; Darius’s great- duvaita-
162 July 31, 2005 OLD PERSIAN - ENGLISH GLOSSARY hu-hamaranakara-: a good ghter 2 kaufa-: mountain 9 navama-: ninth 5 huka-: dry 15 kay˝da-: astrologer 8 nay-, see ˝-nay- hu-ƒandu-: satised, happy 13 k˝ma-: to wish, please (+ acc. of nay˝ka- (for *niy˝ka-): grandfather hu-ƒanuvaniya-: a good archer 2 subject) 6 19 K˝piak˝ni-: name of a fortress 6 n˝h-: nose 8 *huvad˝ta-: *lineage 17 k˝ra-: the people, army 2 n˝man- neut.: name 6 h uvaipaiya-: self 18 k˝saka-: glass 2 *n˝man˝fa-: *genealogy 17 h uv-artika- (uv-˝rtika-?): a good k˝sakaina-: (made) of glass 3 n˝viya-: deep (so as to require ships, spearman 2 Kunduru-: place name 9 or similar, to cross; cf. Sogdian huv-asa- = huv-aspa-: with good Kuganak˝-: place name 7 n˝yuk “deep”) 7 horses, having good horses 4 kunau-/kun-, perf. caxr-, pp. karta-, ni-ç˝raya- < •çay (< sray): put back h uv-asab˝ra: a good rider 2 inn. cartanaiy < •kar: to do 4 in place, restore 9 huv-aspa- = uv-asa- 3 Kuru-: Cyrus 6 nij-ay- < ni + •ay: to go out 9 hUvaxatara-: proper name; K¨a-: Ethiopia 9 ni-kan-: to destroy 8 Cyaxares, Median king 2 K¨iya-: Ethiopian 10 nipadiy: in pursuit (of: + acc.) 11 ni-pita-, pp. of paiƒa-: written 12 hUv˝daicaya-: place name 19 Lab(a)n˝na-: place name 11 ni-rasa- •ras: to come down 15 huv˝ipaiya-: own 6 Maciya-: Makranian 3 ni-saya- < •s˝: to bestow (upon: + h magu-: magian 6 uv˝mariyu-: self-dead, i.e., without upariy + acc.) 4 foreign intervention 9 -maiy: me (gen.-dat.) 4 Maka-: Makran 3 Nis˝ya-: place name 13 hUv˝razm˘-: Chorasmia 7 manah- neut.: mind, thought 8 ni-t˝ya- < •st˝: to lay down, order h UÚ(v)ja-: Elam, Elamite 2 manauvi-: angry, vengeful 6 (+ innitive) 12 h UÚvjiya-: an Elamite 6 man˝: me, my, mine (gen.-dat.) 4 ni-˝daya- < •had/ad: to set down h¨Ÿvnara- neut., h¨Ÿvnar˝-: talent, maniya- < •man mid.: to think 5 10 abiliy 3 Marduniya-: proper name; father of niy˝ka-: grandfather 11 i- > ay- Gaubaruva-, one of Darius’s six n¨ram: now 16 id˝: here 8 helpers 13 paiiy˝ (+ acc.): before, in the ima-: this 2 Margu-: Margiana 7 presence of 4 Imani-: proper name; an Elamite 7 mar^ka-: young man 8 Paiiy˝(h)uv˝d˝-: place name 9 isuv˝-: battle-axe 4 Martiya-: proper name 7 paiiy˝: before 17 iti- fem.: sun-dried brick 11 martiya-: man 2 paiƒa- (or pinƒa) < •paiƒ: to paint 11 iyam: this (nom. masc., fem.) 2 mariya- < •mar (cf. marta-): to die 9 para-ay-/i- < •ay: go (forth) 8 Izal˝-: place name 7 marta-: dead 5 parah: beyond (+ acc.) 4 jadiya- < •jad: ask (sb. for sth.: + mak˝-: inated cow hide (used for paradayad˝-, for *paridaid˝-?: acc. + acc.) 4 ferrying) 7 *garden, pleasure spot 19 jan-/ja- < •jan: strike, smite 4 maƒita-: greatest 4 para-drayah: beyond the sea 10 jantar-: crusher, striker (of: + gen.- mayuxa- : nail, doorknob 3 paraita- pp. < para-ay-/i-: to go off dat.) 6 m˝: let not 9 12 jiyamna-, only in: jiyamnam patiy: on M˝da-: Media, Median, Mede 2 paranam: previously 16 the last day of the month 13 m˝h˘-: month 9 parataram: farther away, beyond 15 j^va- < •j^v: live 8 m˝m: me (acc.) 4 par˝-bara- < •bar: to carry away 7 j^va-: alive 5 m˝naya-, m˝naiya- < •man: to await, par˝-gmata- < •ay/gam: gone far ka-, rel. pron., only in: tuvam k˝ wait for 8 (partic.) 15 (tuvaº ka) ... haya: you who 15 m˝niya-: household(?) 3 par˝-rasa < •ras: to arrive 6 Kamb¨jiya-: Cambyses (king of M˝rgava-: Margian 8 *par˝-y˝taya-: ? 17 Persia) 6 M˝ru-: name of town 8 Parga-: name of mountain 22 kamna-: few 8 miƒah- kunau-: to do sth. wrong to (+ pari-ay-/i- < •ay act./mid.: to behave Kampanda-: name of land 8 acc.) 4 9 kaniya-, pp. kanta-, inf. kantanaiy < Miƒra-, Mitra-: Mithra 19 pari-bara- < •bar: to reward 9 •kan: be dug 11 Mudr˝ya-: Egypt, Egyptian 6 pariy + acc.: about, concerning 4 n kantanay, inn. of kaniya- < •kan: to mu ƒa- < •mauƒ: to ee 9 par^yana-: behavior 3 dig, be dug 13 Nabukudracara-: Nebuchadrezzar 6 parsa- < •pars/fraƒ: to ask, punish 9 kapautaka-: blue 2 Nabunaita-: proper name; last (Neo- partana- neut.: ght, conict 13 Karka-: Carian 10 )Babylonian king, Nabonides, Parƒava-: Parthia, Parthian 2 Karm˝na-: Kerman, Karmania 11 Nab¨-na÷id 6 paru-: much, plur. many 2 karnuvaka-: artisan, craftsman 11 Nadintabaira-: proper name; paruva-: former 14 kara-: a measure of weight = 83.33 Babylonian rebel, Nidintu-Bˇl 6 paruvam: of old, before 5 h h g. naiba-: good, beautiful 2 paruviyata , in: hac˝ paruviyata : karta-, pp. of •kar-: done, made; naiy: not 2 from before, from old 3 work 5 napat-: grandson 5 paru-zana-: of many kinds 3 kaciy: anybody 13 *nau- (only restored): ship 13 pas˝: after (+ acc. or gen.-dat.) 4 Katpatuka-: Cappadocia 7 naucaina-: of cedar 11 pas˝va: afterward; pas˝va yaƒ˝
163 July 31, 2005 OLD PERSIAN - ENGLISH GLOSSARY
“after” (in past narrative) 6 st¨n˝-: column 7 upariy: in, on, above 4 pasti-: foot soldier 2 Sugda- = Suguda-: Sogdiana 9 upast˝-: assistance, aid; + bar-: “to patikara-: representation, statue, Suguda- = Sugda-: Sogdiana 7 bear aid” 6 picture 3 -aiy: him (gen.-dat.) 9 up˝ + acc.: under = during the reign pati-bara- < •bar: to bring back 14 -aiy: his, her, its (gen.-dat.) 5 of 4 pati-fraƒiya- = -frasiya- < •pars/fraƒ: -˝m: them, their (gen.-dat.) 5 ustaan˝-, ustacan˝-: staircase (with to be read 17 ˝y˝ta- for iy˝ti- 20 carved reliefs?) 19 Patigraban˝-: place name 19 -im: him 7 u^ (nom.-acc. dual): conscience, pati-jan- < •jan mid.: to ght 15 iyava- < •iyav: to go 6 intelligence 8 patipadam •kar: to reestablish, to put iy˝ta-: happy, blissful 3 ua-b˝ri-: camel-borne 7 back where it belongs 14 iy˝ti-: peace, happiness 10 Ut˝na-: proper name; Otanes, son of pati-paya- •p˝ mid.: to guard -i: them 6 ÿuxra-, one of Darius’s six helpers (oneself) 16 kauƒi- = skauƒi-: weak 8 13 pati-parsa- < •pars/fraƒ: to read 15 tacara-, dacara-: palace 6 ut˝: and; ut˝ ... ut˝: both ... and 2 pati-xaya- < •x˝ mid.: to rule over takabara-: petasos-bearing 10 Uyam˝-: name of a town 6 (+ gen.-dat.) 5 tan¨Ÿ- fem.: body, self 7 uzmay˝patiy kar-: to impale 7 -patiy: too 7 tarah: through, via (+ acc.) 4 vaçabara-: mace-bearer(?) 2 patiy-avahaya- mid.: to implore tarsa- < •tars: to fear (+ hac˝ + inst.- Vahauka-: proper name; (Ochus) somebody for help, to pray to (+ abl.) 7 father of Ardumani-, one of acc.) 6 tauman- neut.: power, capacity 6 Darius’s six helpers 13 patiy-ay- < •ay: to come to 7 taum˝-: family 3 Vahayazd˝ta-: proper name; rebel 6 pati-zbaya- •zb˝: to proclaim 16 tauv˘yah-: stronger, mightier 8 *Vahayaspara-: proper name; father pati: against (+ acc.) 4 *taxma-: brave 2 of Vidafarnah-, one of Darius’s six patiam: in addition 17 Taxmasp˝da-: “having a brave army” helpers 8 pavast˝-: clay tablet 17 proper name 10 vaina- < •vain: to see 4 paƒ˘- fem.: path 15 taya: that (conjunction) 8 vaja- < •vaj: to gouge out 10 paya- < p˝- tayaiy: who (nom. plur. masc.) 3 vaniya-: to be lled (poured) in 11 p˝-, pres. paya-, pp. p˝ta- < •p˝: to T˝rav˝-: place name 22 vardana- neut.: town 6 protect 8 t˝vaya- < •tav: be able, endure 4 *vardiya-, see ragam *vardiya- 5 p˝d(a)-: foot 10 tigra-: pointed 3 Vark˝na-: Hyrcania, Gurg˝n 7 P˝rsa-: Persia, Persian 2 tigra-xauda-: wearing pointed hats 3 varnava- < •var mid.: to choose 15; Pir˝va-: the Nile 6 Tigr˝-: Tigris 6 + pers. pron. acc.: to believe (see piru-: ivory 14 tunuvant-: mighty 8 grammar) 6 vasiy: greatly, mightily (only form of pitar-: father 5 ƒadaya- < •ƒand: to seem (to: + gen.- this word) 2 pita-, pp. of paiƒa- (pinƒa-): to paint dat.) 15 van˝ (instr.-abl. of *vazar): by the 14 ƒaha-, inn. ƒastanaiy < •ƒah: say, greatness of (often translated as: by puça-: son 1 speak, announce (to: gen.-dat.) 6 the favor of, by the grace of; see Put˝ya-: Libyan 10 ƒahaya- < •ƒah: be said (by: + hac˝, lesson 9) 1 ragam *vardiya- mid.: to swear 5 to: gen.-dat.), be announced (as), Vaumisa-: proper name; a Persian Rag˝-: Rhaga, Ray 11 be called (+ nom.) 3, 4 vayam: we 3 rasa- < •ras: to arrive 8 ƒakata-: passed 4 vazarka-: great 1 raucah-: day 9 ƒanuvaniya-: a marksman (lit. bow- v˝: or; v˝ ... v˝ “either ... or” 13 rautah (nom.-acc. sing. of rautah- man) 2 Vidarna-: proper name; Hydarnes, neut.): river 6 ƒard- (or ƒarad-) fem.: year 5 son of *Vahayaspara-, one of Rax˝-: name of a town in Persia 6 ƒarmi-: timber 11 Darius’s six helpers 8 raxƒa-: ? 8 ÿatagu-: Sattagydia 7 vi-kan- < •kan: to destroy 4 r˝diy: from, by, on account of 15 ƒava- < •ƒav: to burn (intr.) 11 vi-marda- (-marda-) < •mard: to wipe r˝sta-: right 2 ƒ˝- < ƒaha-- out, destroy 11 saiyma-: silver 19 ÿ˝igraci-: month name 12 vi-nasta-: offense 9 Saka-: Scythian, Scythia 3 ƒ˝tiy < *ƒahatiy < •ƒah: he says 4 vi-n˝ƒaya- < •naƒ: to do harm, do Sikayauvat˘-: name of a fortress in ƒik˝-: gravel 11 wrong 9 Media ÿuxra-: proper name; a Persian , Vindafarnah-: proper name; sinkabru-: carnelian 14 father of Ut˝na 13 Intaphernes, one of Darius’s six skauƒi- = kauƒi-: weak, poor 16 ÿ¨rav˝hara-: month name 14 helpers 13 Skudra-: a people north of Greece ub˝: both 10 visa-: all 9 (Thrace, Thracian?) 10 ud-pata- < •pat: to rise up (in n rebellion) 6 visa-dahayu-: of all nations 4 Sku xa-: proper name; Scythian rebel vispad˝: everywhere 14 Sparda-: Sardis 7 Ufr˝t¨-: Euphrates 9 upa-ay- < •ay: to come close to 13 vispa-zana-: of all kinds 5 Spardiya-: Sardian 11 Vipauz˝ti-: name of town 21 sp˝yantiya-: *army camp 15 Upadarma-: proper name 6 upariy-ay- < •ay: to abide (by: + Vit˝spa-: proper name; Hystaspes, stambava- < •stamb: to rebel 11 inst.-abl.) 14 Darius’s father 1 st˝na-: *niche 13 vi-taraya- < •tar: to convey across 6
164 July 31, 2005 OLD PERSIAN - ENGLISH GLOSSARY viƒ- fem.: house 5 Viv˝na-: proper name; Persian satrap of Arachosia 12 Viyaxana-: month name 9 xaud˝-: hat 3 xraƒu-, xratu-: mind, understanding 12, 13 xaça-:neut.: power, (royal) command, empire 3 xaçap˝van(t)-: satrap 8 xap- fem.: night 16 xaya- < •x˝ mid.: to rule, control (+ gen.-dat.) 13 Xayaar˝- (Xay˝r˝-) masc.: proper name; Xerxes, son of Darius 2 xayamna- < •x˝ (see lesson 13): being in control 9 Xaƒrita-: proper name; a Mede 7 x˝yaƒiya-: king 1 xn˝sa- < •xn˝: to know 15 xnuta-: pleased 9 yaciy: whatever 10 yada- < •yad mid.: to worship 5 Yad˝-: Anshan 9 yad˝y˝: where(ever) 16 yadiv˝: or 9 yadiy: if, when 5 yak˝-: yak tree, sissoo 11 yaniy: where, in which 14 yaƒ˝: as, than, when 5 yaƒ˝: so that 6 yaudant^- (fem.): (being) in turmoil 10 Yauna-: Ionian, Greek 7 Yautiy˝-: place name 22 yauviy˝-: canal 13 y˝na- neut.: boon, favor, gift (d˝- “grant,” jadiya- “ask”) 4 y˝tu-: sorcerer 8 y˝t˝: until 6 y˝t˝ ˝: up to, until (+ instr.-abl.; local) 9 y˝umani-: coordinated, being in control 10 y˝v˝: as long as 14 Z˝z˝na-: place name 13 Zranka-: Drangiana 7 z¨rah- neut.: crooked deed, wrong(doing) 8 z¨rahkara-: doer of crooked deeds, crook, wrong-doer 13 Z¨za-: place name 21
165 July 31, 2005