TRUBNER*S 
 COLLECTION OF SIMPLIFIED GRAMMARS 
 
 LANGUAGES OF THE 
 CUNEIFORM INSCRIPTIONS 
 
 GEORGE BERTIN, M.R.A.S. 
 

 THE EISENLOHR COLLECTION IN 
 EGYPTOLOGY AND ASSYRIOLOGY 
 
 PRESENTED TO CORNEI,!, UNIVERSITY BY 
 
 X902 
 
 AjjyMi. 0/../fAz. 
 
 3947 
 
DATE DUE 
 
 
 
 
 
 — -f 
 
 ^irnir-*" 
 
 i' 
 
 
 
 
 
 AfeM^ 
 
 WK^ 
 
 19 ^ 
 
 
 App /-> 1 
 
 iofti r 
 
 
 
 oLr^^ 
 
 l3ol-il_- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 PJ 4013.B5T" ""'""'"^ '■''"''^ 
 *'''iIi3nl,9M!!!.&,2f...!!!e..Janguages of th 
 
 3 1924 026 822 431 
 
Cornell University 
 Library 
 
 The original of tliis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924026822431 
 
TKUBNER'S COLLECTION 
 
 OP 
 
 SIMPLIFIED GEAMMAES 
 
 OF THE PKINCIPAL 
 
 ASIATIC AND EUROPEAN LANGUAGES. 
 
 EDITED BY 
 
 EBINHOLD EOST, LL.D., Ph.D. 
 
 XVI L 
 
 THE 
 
 LANGUAGES OF THE CUNEIFOEM INSCEIPTIONS 
 
 By GEOEGE BERTIN, M.E.A.S. 
 
TEtJBNEE'S 
 COLLECTION OF SIMPLIFIED GRAMMARS 
 
 OF THE 
 
 PRINCIPAL ASIATIC AND EUROPEAN LANGUAGES. 
 
 Edited bv KEINHOLD EOST, LL.D., Ph.D. 
 
 I. HINDUSTANI, PERSIAN AND ARABIC. By 
 
 the late E. H. Palmek, M.A. Second Edition. Price 5s. 
 
 It. HUNGARIAN. By I. Sihqer. Price is. M. 
 
 III. BASQUE. By W. Van Eys. Price 3s. M. 
 
 IV. MALAGASY. By G. W. Pakkek. Price 5s. 
 
 V. MODERN GREEK. By E. M. Geldakt, M.A. Price^s. 6d- 
 VI. ROUMANIAN. By R. Torceahu. Price 5s. 
 VII. TIBETAN. By H. A. Jaschke. Price 5s. 
 VIII. DANISH. By E. C. Oixfi. Price 2s. Gd. 
 IX. OTTOMAN TURKISH. By J. W. Eedhouse. Price 
 
 10s. ed. 
 
 X. SWEDISH. By E. G. OttiS. Price is. Qd. 
 XI. POLISH. By W. E. MoEPiLL, M.A. Price 3s. M. 
 XII. PALI. By Edwakd Mullee, LL.D. Price 7s. M.. 
 
 XIII. SANSKRIT. By Hjalmak Edqken, Ph.D. Price 10s. 6d. 
 
 XIV. ALBANIAN. By P. W. Pnce Is. Gd. 
 
 XV. JAPANESE. By B. H. Chamberlain. Price 5s. 
 
 XVI. SERBIAN, By W. E. Morfill, M.A. Price is. Gd. 
 
 XVII. LANGUAGES OF THE CUNEIFORM IN- 
 
 SCRIPTIONS. By Geoeqe Bektin, M.E.A.S. Price 5s. 
 
 Grammars of the following are m preparation : — 
 
 Anglo-Saxon, Assyrian, Bohemian, Bulgarian, Burmese, Chinese, 
 
 Cymric and Gaelic, Dutch, Egyptian, Finnish, Hebrew, Khassi, 
 
 Kurdish, Malay, B,us3ian, Siamese, Singhalese, &o. &c. 
 
 London: TRUBNER & CO., Ludgate Hill. 
 
LANGUAGES OF THE CUNEIFOEM 
 INSCRIPTIONS. 
 
ABRIDGED GRAMMARS 
 
 OF THE 
 
 LANGUAGES OF THE CmEIEORM 
 INSCRIPTIONS 
 
 CONTAINING 
 
 I.— A SUMERO-AKKADIAN GRAMMAE, 
 II.— AN ASSTKO-BABYLOMAN GRAMMAR 
 III.— A TANNIC GRAMMAR 
 IT.— A MEDIC GRAMMAR 
 
 T.— AN OLD PERSIAN GRAMMAR 
 
 GEOEGE BERTIN, M.E.A.S. 
 
 LONDON 
 
 TEUBNEE & CO., LUDGATE HILL 
 
 1888 
 
 [All rights reserved] 
 
35^ 
 
 ^.\W^'i ^^% 
 
 BALLANTYNK, HANSON AND CO. 
 fiDINBUKGH AND LONDON 
 
 (i... 
 
 c^ 
 
 '^. 
 
TO 
 'a. H. SAYCE, M.A., 
 
 DEPUTY PROFESSOR OF COMPARATIVE PHILOLOGY IN THE 
 UNIVERSITY OF OXFORD, 
 
 WHO HAS, BY HIS EESBAEOHES, HIS TBAOHINO, AND HIS DISC0VEKIE8, 
 
 DONE SO MUCH TO ADVANCE CUE KNOWLEDGE OF THE 
 
 LANGUAGES OF THE CUNBIEOBM INSOEIPTIONS, 
 
 THIS WOEK IS 
 
 2Del]icateti 
 
 AS A SMALL TEIBUTE 
 
 THE AUTHOR. 
 
PREFACE. 
 
 In the following grammars I have endeavoured to follow 
 the plan laid down by the late Professor Palmer in the first 
 volume of the series, which is a model of concision and 
 lucidity. The great difficulty I had to contend with is 
 the fact that these languages have been dead for many cen- 
 turies, and the texts — except for Assyro-Babylonian — are still 
 very limited. The task would have been impossible without 
 the efforts and labours of the first pioneers in the field. Sir 
 H. 0. Rawlinsqn, Dr. Hincks, Norris, Dr. J. Oppert, Professor 
 Sayce, Professor Schrader, &c., to all of whom I am largely 
 indebted. 
 
 For the transcription I have adopted the principle that 
 the inscriptions themselves must be our sole guide, and that, 
 rejecting all theories, all words ought to be transcribed as 
 they are actually written, whatever might have been the pro- 
 nunciation, as this cannot be stated with, certainty. I give 
 these languages as they are found on the inscriptions. Theori- 
 cians may afterwards, if they like, discuss the question of 
 what might have been the' pronunciation of the people. All 
 the forms — except stem- words, and, for Babylonian, the infini- 
 tives of hal — given as examples in the following pages, are 
 actually found in the inscriptions. 
 
 It must not be forgotten that the object in view in these 
 grammars — as with all those of dead languages — is to facili- 
 tate to the student the reading of the texts, and not to teach 
 
viii PREFACE. 
 
 him how to speak. Particular attention has, however, been 
 given to the Syntax, so often neglected in works of this kind. 
 
 The Sumero-Akkadian grammar is practically the first 
 written for this language, as the attempt of F. Lenormant 
 was made at a time when a correct analysis of it was im- 
 possible. 
 
 In the Assy ro- Babylonian grammar I have adopted an 
 entirely new point of view, which has the advantage of making 
 clear what could not be rationally explained before. 
 
 The Vannic grammar is based on the works of Professor 
 A. H. Sayce, though I believe I have improved certain parts. ^ 
 
 For the Medic I have followed my first master. Dr. Oppert, 
 and I am also much indebted to Professor Sayce. 
 
 In the Old Persian studies all is due to Sir H. Rawlinson, 
 Dr. Oppert, and Dr. Spiegel, and I follow them on the main 
 points. 
 
 I must also acknowledge the help I have received from 
 my friend Mr. Pinches, of the British Museum, who pointed 
 out to me many things which might have escaped my atten- 
 tion, and who placed at my disposal his vast acquaintance 
 with the texts. 
 
 If the student wish for a syllabary, I cannot but recommend 
 that of Professor Sayce, still the best ; but Mr. Pinches is 
 preparing a very complete one, in which will be given the 
 characters of various epochs and of all the different languages, 
 which will be published shortly by Messrs. Triibner <fe Co. 
 
 G. BERTIN. 
 
 London, Sept. 1887. 
 
 ■^ Professor Sayce having kindly communicated to me advance proofs of 
 his supplementary paper on the inscriptions of Van, I have been able to 
 incorporate the new discoveries and post my grammar up to date (January 
 1888). 
 
SUMERO-AKKA.DIAN GRAMMAR. 
 
 PHONETIC. 
 
 Akkadian is written almost exclusively by means of ideo- 
 grams, with, or without phonetic complements ; before and after 
 certain words use is also made of determinative affixes. The 
 phonetic pronunciation of the words, though still uncertain for 
 some, is pretty well ascertained from the glosses given in the 
 syllabaries and word-lists, and from comparison and- analogy. 
 
 The alphabet is composed of four vowels and eighteen con- 
 sonants. 
 
 The vowels are : a, e, i, and u. 
 
 The consonants may be grouped thus — 
 
 Two aspirates : ', g (k). 
 Three gutturals : g, Jc, k. 
 Three labials : b, p, m (w). 
 Three liquids : n, I, r. 
 Four sibilants : s, ^, z, s. 
 Three dentals : d, t, t. 
 
 Either from the very nature of the sounds of the language, or 
 because the characters representing these sounds were borrowed 
 from a foreign language, the letters of each group are easily 
 confounded : g, k and k, d, t and t, &c. There never was any 
 difference made between b and p, these two letters being indiffer- 
 ently used for one another ; m was never distinguished from w, 
 and both letters always were written with the same characters. 
 
 At the end of a closed syllable the consonants are obscured 
 
 A 
 
2 SUMEEO-AKKADIAN GRAMMAR. 
 
 and take an indistinct sound, which can be transcribed by either 
 letter of each group. The syllables ad, at, and at give a final 
 dental having an intermediary and uncertain sound between the 
 three normal dentals ; in the same way ag, ah, and ak give an 
 intermediary guttural, &o. The final consonant of a word is 
 ascertained only when followed by a vowel. 
 
 It must be noticed that ^ is, however, generally distinguished 
 from the other sibilants. 
 
 Vocalic harmony is observed in a certain measure, but is often 
 difficult to detect on account of the ideographic system of writing ; 
 the vowels of the suffixes or prefixes are generally assimilated 
 to those of the noun or verb to which they are j^oined. 
 
 Sumerian differs from Akkadian only in its phonetic peculi- 
 arities. The letter-changes between the two dialects appear at 
 first sight to be numerous and irregular, but are easily explained 
 by supposing that, when the Akkadians borrowed their system 
 of writing from the Semites, they were obliged to represent the 
 sounds of their own language approximately by the characters 
 giving the normal sounds of the Semitic tongues, and that 
 the Semitic consonants represent really, each several letters in 
 Akkadian. In his transcription the scholar is obliged to follow 
 the apparent value of the letters as given by the phonetic read- 
 ings of the inscriptions, but in the following table of letter- 
 changes each Akkadian letter is followed by the supposed real 
 or primitive pronunciation. 
 
 ' the spiritus lenis^ remains unchanged or disappears en- 
 tirely, as also sometimes in Akkadian. 
 g (h) hard aspirate remains unchanged, or is hardened into 
 
 g or Tc. 
 g normal remains unchanged or is replaced by hard Ic. 
 slightly aspirated is replaced by ?. 
 nasal (m), like ng in " sing," is replaced by m (w). 
 dental (g) or English j is replaced by d. 
 labial (z) or French/ is replaced by b. 
 
 1 The spiritus lenis and forte appear to have been confounded or 
 expressed by the same character. 
 
PHONETIC. 3 
 
 h normal remains unchanged or is replaced by hard g. 
 
 aspirated is replaced by t. 
 k normal remains unchanged or is replaced by hard g or Jc. 
 
 clapping is replaced by d ov t. 
 b and p, these two letters, which appear to be confounded in 
 Akkadian, are more clearly distinguished in Sumerian, 
 but the b predominates. 
 m (w) remains unchanged. 
 n normal remains unchanged. 
 
 palatal is replaced by Z. 
 
 aspirated is replaced by S. 
 
 soft is replaced by m. 
 s remains unchanged or is replaced by ? or s. 
 .? normal remains unchanged. 
 
 strong is replaced by r or Z. 
 z remains unchanged or is replaced by s. 
 s remains unchanged. 
 I remains unchanged or is replaced by r. 
 r remains unchanged or is replaced by n. 
 d normal remains unchanged or is replaced by t. 
 
 aspirated (soft th) is replaced by s or z. 
 f remains unchanged. 
 
 strong is replaced by n. 
 
 In the following pages the Akkadian forms will be always 
 given ; in a few cases, when the Sumerian forms are giveUj they 
 will be preceded by S. 
 
 Vocalic and Consonantio Harmony. 
 
 As will be seen in the course of this grammar, vocalic 
 harmony played probably a great part iii Akkadian pronun- 
 ciation, but it is very difficult for us to detect the changes, as 
 the words are mostly written ideographically. The consonants 
 exercised also an influence on one another ; this is shown by the 
 poetic texts, in which alliteration was largely resorted to ; but 
 as for the vocalic harmony, it is impossible to draw any rule on 
 account of the ideographic system of writing. 
 
SUMEEO-AKKADIAN GRAMMAR. 
 
 NOUNS. 
 
 Nouns have not any particular forms or suffixes, the meaning 
 and the context only show if a word is to be regarded as a 
 noun: ad, "father;" am, "mother." 
 
 All nouns are susceptible of taking the lengthening a^ga, 
 S. a-ka, often reduced to ag and ak, and also weakened into 
 ag. 
 
 R. As the Sumero-Akkadian is written mostly by means of 
 ideograms, the lengthening is often added with the phonetic 
 complement, so that the final consonant appears to be doubled : 
 
 kalam, " land ; " state of prolongation, kalam-ma-ga. 
 
 R. The vowel of union is affected by that of the word and 
 harmonises with it : 
 
 lid, " sun ; " state of prolongation, ud-dug. 
 
 The most common way to lengthen the nouns is to add 
 simply a vowel ; it may be the same suffix as ag with the loss 
 of the g : 
 
 ad, " father ; " lengthening, ad-da. 
 
 The state of prolongation and the lengthening express em- 
 phasis, and answer in a certain measure to our article. 
 
 For the same reason, when the noun is used in the vocative, 
 it is to be read with the lengthening, though it is not always 
 written : am for ama, " mother ! " 
 
 The genitive is generally expressed by simple apposition : 
 6-ad-da, " the house of the father.'' 
 
 The dative is also often expressed by the position of the 
 word in the sentence (see Syntax), but more often by a 
 postposition. 
 
 In the accusative the nouns appear with or without the 
 lengthening; in some cases the accusative is indicated by a 
 postposition. 
 
NOUNS. 
 
 Postpositions. 
 
 Postpositions are added to the nouns to express the relations 
 expressed in other languages by cases or by prepositions. The 
 meanings of these postpositions are very difficult to render, as 
 they depend in a great measure on the conception of the people, 
 and cannot be always translated by the same prepositions of 
 our idiom. 
 
 R. The postpositions are always added to the noun with the 
 lengthening. 
 
 R. The vowel of the postposition is often affected by that of 
 the noun, or dropped altogether. 
 
 The postpositions are nine in number — 
 
 -na, " in, at, to, as," primitively en-na. 
 
 -da, "at, in, on, with,'' generally without movement, 
 
 -ta, " in, from, by, for," generally with movement. 
 
 -a, " in, at," sometimes simply marks the accusative. 
 
 -ra, " to, towards, far from." 
 
 -^u (written 7m, the pronunciation ^u is^the Sumerian 
 
 which passed into Akkadian at an early date), " on, at, 
 
 over," generally without movement. 
 -7ca, "in, of" (may be another form of the preceding or 
 
 the following particle). 
 gi, S. ki, " of." 
 gime or Mine, S. dime (more often written gim, s. dim), 
 
 " as, like." 
 
 Akkadian has no adjectives properly so called; all n"olinS are 
 substantives ; when they are used to express what would be an 
 adjective in our language, they are merely put in apposition. 
 
 Gender. 
 
 The genders are not expressed by any particular forms in 
 the nouns ; there is, therefore, no gender from the grammatical 
 point of view. Words like ada, " father," and ama, " mother," 
 have of course a gender of meaning, but that is all ; words like 
 
6 SUMEEO-AKKADIAN GRAMMAR. 
 
 gihil, S. Jcibir, " fire " or " burning," saHm, " peace," gissigar, 
 " barrier," have really no gender. 
 
 Sometimes, but rarely, the gender is indicated by another 
 word added to the noun: tur-nita, "a little one, male " = boy; 
 tur-sal, " a little one, female " = girl. 
 
 Number. 
 
 There are only two numbers, the singular and the plural 
 The plural is, however, often not expressed, and must be 
 detected from the context. When the plural is expressed, 
 it is generally by simple reduplication : Icur, " mountain ; " 
 kur-kur, " mountains.'' 
 
 R. The lengthening is placed only once after the group : 
 
 kur-kur-ra, "the mountains." 
 
 We find in the texts two other contrivances to express the 
 plural clearly — 
 
 1°. By placing after the noun the plural of the demonstrative 
 or third person pronoun : dingir-ri-e-ne, or dingir-ri-ne-ne, " the 
 gods," lit. " the god, they." 
 
 2°. By placing after the noun the plural of the verb " to be: " 
 dum^ma-mes, "the sons," lit. "the son, they were." 
 
 NUMEEALS. 
 
 
 
 r ge^, S. det 
 
 6, 
 
 7, 
 8, 
 9. 
 
 • 
 
 Imin. 
 
 v£. 
 
 Uim. 
 
 The numerals are- 
 
 1, ge or a^ 
 
 2, min. 
 
 3, el 
 
 4, Urn. 
 
 5, id. 
 
 R. They can take the lengthening ; often the final consonant 
 is doubled : mina, limma or limmu, Imina, llimma ; eS and vM 
 make eUe, uSla. 
 
NUMERALS. ] 7 
 
 B. It is to be observed that geS became a-eS and then a^, or 
 that the final S of ge^ was lost ; so we have the double form ge 
 and aS. 
 
 E. The four last units are formed with the four first added 
 to id, " five," aS for id-aS, Imin for ia-min, &c. 
 
 The tens are — 
 
 10, gu, ga, u or a. 40, nimin. 
 
 20, nin, nis, han. 50, ninnu. 
 
 30, v^u, e^, la. 60, ^u^, S. mut 
 
 R. The forms for 10 are the produce of successive weakening, 
 
 R. The words for 30 and 40 are perhaps formed with 10 
 or 20. and the units u^u for it-e^^e and e& for a-el¥e (10 x 3); 
 nimin for nin-min (20 x 2), ,or for nin-nin (20 + 20), ^zj^ for u^-u^ 
 (30 + 30). The S. mu^ is probably from another root. 
 
 R. The forms ni^ and &aw for 20 and ha for 30 are probably 
 dialectical. 
 
 R. The numeral for 60 being taken as unit, " the soss," the 
 words for 70, 80, and 90 were not used. 
 
 For 100 we have me, probably borrowed from Babylonian. 
 
 For 600, a multiple of 60, we have ner, also probably borrowed. 
 
 For 3600 (600 X 6), we have sar, lit. " multitude." 
 
 Fractions. 
 
 The word for " half " is Curiam. 
 
 The other fractions are (expressed by U placed before the 
 numerals, generally written in ciphers — 
 
 ^ U-III or U-et. ^ U- VI or ^i-aM. 
 
 i M-IV or %i-lim. yg- H-X or ^i-gu. 
 
 ^ U- V or ^i-id. 
 
 R. The soss being taken as unit, the fractions may be ex- 
 pressed from that standpoint, in which case special words are 
 
 used — 
 
 ^lu-^a-na, . . f § or ^, also used for 20. 
 
 Sor-na-bi, . . f § or f, also used for 40. 
 
 kin-gu-sil-la, . |-g or |-, also used for 50. 
 
8 SUMERO-AKKADIAN GRAMMAR 
 
 Oedinal Numbers. 
 
 The numeral adjectives, or ordinal numbers, are formed by 
 adding gan, s. kam, to the numerals, generally written in cipher — 
 
 Second, Il-gan or mi-na-gan. 
 Third, Ill-gan or eS-h-gan, &c. 
 
 DEMONSTKATIVE ADJECTIVES. 
 There are five demonstrative pronouns or adjectives — 
 1°. Am or ma, primitively ama, " this very " (ipse), very 
 emphatic pronoun. 
 
 2°. An or na, primitively ana, " this," pronoun of proximity 
 (hie), is rarely used as a demonstrative with the nouns, no 
 doubt because easily confounded with the postposition -na. 
 
 3°. ^6 or ba, primitively abi, first pronoun of remoteness, 
 " that " (i7Ze). 
 
 4°. A, primitively 'e (the aspirate is never written), second 
 pronoun of remoteness {isie) ; it may be the adjunction of 
 this pronoun to the nouns which in some cases appears as 
 the lengthened forms. 
 
 5". The compound an-ga, emphatic pronoun "this very" 
 (is ipse). 
 
 It. The vowels of these pronouns are affected by vocalic 
 harmony ; so, for instance, the second may become an, un, in, 
 en, or na, nu, ni, ne, according to the cases. 
 
 S. These pronouns never appear isolated, but always either 
 suffixed to the noun or united to the verb. 
 
 PEESONAL PRONOUNS. 
 
 The personal pronouns appear to have been mostly derived 
 from the demonstratives. The isolated pronouns are, except 
 
PRONOUNS. 9 
 
 for the first and second persons of the singular, seldom used in 
 the texts. They are — 
 
 ist pers. ga-e, S. ma-e, plur. ge, S. me, 
 
 gu-un, S. mu-un. 
 2nd pers. za-e, S. sa-e, plur. e-ne. 
 
 ib-e. 
 3rd pers. e-ne, plur. e-ne-ne-ne, e-ne-ne, or e-ne, 
 
 ni or ni-na, an-ne-ne or an-ne. 
 
 These pronouns can take the postpositions, but only with 
 certain forms. 
 
 In the singular, with the first person : 
 
 ga-da, S. ma-da, " with me ; " ga-ra, S. ma-ra, " towards 
 me," &c. 
 
 With the second person : 
 
 za-a-gi, S. sa-a-ki, "of thee = thine." 
 e-da, "with thee," e-ta, "from thee," &c. 
 
 With the third person : 
 
 ni-na-a, " to him,'' abbreviated into na. 
 U-a, " to him," abbreviated into ha. 
 U (for e-^u), " to him." 
 
 In the plural, with the first person : 
 
 ge-^u, S. me-^u, " to us," ge-a, S. me-a, " to us " or " us," accu- 
 sative ; abbreviated sometimes into ga, S. md. 
 
 With the second person : 
 
 e-nt-a, " you," accusative. 
 
 With the third person : 
 
 an-ne-da, "with them; " an-ne-a, "to them" or "them." 
 
 R. As with the nouns, the postpositions are added after the 
 lengthening; this explains the forms gd-ra for ge-a-ra, zaa-gi 
 for zors-a-gi, &c. 
 
lO 
 
 SUMERO-AKKADIAN GRAMMAR. 
 
 Ifcoeporating Personal PEONorNS. 
 
 Certain compound forms of the personal pronouns incorporate 
 the whole sentence. We know these forms only for the plural ; 
 they always imply the verb " to be." 
 
 ist pers. ge-en-ne 
 
 an-ne . 
 2nd pers. ge-en-zi 
 
 an-zi. . 
 
 ab-zi. . 
 3rd pers. an-ne . 
 
 an-ne-ne 
 
 en, "we are. 
 
 en. 
 
 en, or ge-zi . . . 
 
 en. 
 
 en. 
 
 en, "they are." 
 
 en. 
 
 en, '• you are." 
 
 For the singular the emphatic personal pronouns are used 
 with the verb " to be." , 
 
 ga-e 
 
 . me-en, " I am ; " za-e 
 
 me-en, " thou art." 
 
 Peonominal Suffixes. 
 
 The possessive is expressed by the emphatic pronoun under a 
 certain form put in apposition after the noun, that is, suffixed 
 to the lengthened form : 
 
 dih-ha-gu, S. gub-la-mu, " the doing of me" = "my doing." 
 e-a-zu, " the house of thee " = " thy house." 
 Svra-ni, " the hand of him " = " his hand." 
 
 For the third person the pronoun hi is also used, but is more 
 often to be translated by the demonstrative : 
 
 lu-bi, "his man" or "that man." 
 
 As we say in English " my man " for "this man I spoke of." 
 
 The plural is not generally expressed in the possessive pro- 
 nouns ; when this is done, it is like in the noun by means of the 
 plural of demonstrative pronoun : 
 
 e-a-zu-ne-ne, lit. " the house of thee these " = " your house." 
 
 We have therefore for the possessive sufi&xes the following 
 table : 
 
PRONOUNS. II 
 
 1st pers. -gu, S. -mu,, "my " or "our ; " emphatic plur. -gu-ne-ne, "our." 
 2nd pers. -zu, S. -su, "thy" or "your," „ -zu-ne-ne, "your." 
 
 3rd pera. -m, " his " or " their," ,, -m-me-mc, "their." 
 
 ■e-ne.-ne, 
 •e-ne. 
 
 ^ JR. On account of tho vocalic harmony, the pronominal 
 forms, such as an-ne, an-ne-en, ab-zi-en, &c., can become en-ne, 
 in-ne, un-ne, erir-ne-en, in-ne-en, &c. 
 
 Inteeeogativb Pronouns. 
 
 A-ha, "who;" a-na, "which, what;" e-h or iS-en, "who, 
 which." 
 
 Indefinite Peonouns. 
 
 Na-mi, " some one ; " nig, " something," lit. " a thing ; " abbre- 
 viated also in ig, S. ag or l; nu-vM and nu^b-da, "no one." 
 
 Eelative Peon guns. 
 
 Nig, "which," lit. "a thing." 
 
 Por " beings " the demonstrative emphatic pronouns are used. 
 
 Htpeebolio Peonouns. 
 
 Sometimes nouns are used instead of personal pronouns, as 
 lu, " man," for " I ; " ka, "mouth," for " thou," &c. As in the 
 Semitic tongues, "head" is used with the possessive suffixes for 
 the emphatic pronouns. 
 
 E. With these words there is no localisation more than 
 with the demonstratives ; lu is used for " he " as well as for 
 " I," and may be also used for " thou." 
 
 Nouns with the possessive adjectives are also used : im-zu, 
 "thy breath"=" thou." 
 
 B. The use of those hyperbolic pronouns is not restricted 
 to personal pronouns, but they may be used for indefinite and 
 other pronouns : lii-na, " the or this man," for " somebody," and 
 also for "whoT' 
 
SUMERO-AKKADIAN GRAMMAR. 
 
 VERB. 
 
 All verbs may have the active or passive, transitive or 
 intransitive meaning, as some of our verbs : "to look," actively 
 " to fix the eye on something," passively " to appear ; " so gin, 
 act. "to fix," pass, "to be fixed," i.e., "to stand." 
 
 R. Properly speaking, the verb has always a nominal mean- 
 ing; it is only its position in the sentence, the context, or the 
 addition of pronominal suffixes and particles, which show how 
 it is to be translated. 
 
 As the noun, the verbal theme or stem appears under two 
 forms — the bare form and the lengthened form. 
 
 The lengthening is added exactly as with the noun : 
 
 dun, " to go ; " lengthened form, dun-na. 
 
 The lengthening was primitively a demonstrative pronoun 
 suffixed to the noun, and for this reason it represents with 
 a word used as verb the action as present : 
 
 dun, " a going," that is, the action of going being completed, 
 
 which consequently expresses the past, " a having 
 
 gone " or " went." 
 dun-na, " the or this going," that is, the action in the act 
 
 of being performed, which consequently expresses the 
 
 present, "a being going." 
 
 R. The verbal stem, bare or lengthened, may be used with- 
 out any pronoun, the bare form having the force of past and 
 the lengthened of present or future. 
 
 The plural is often not expressed, being implied by the sub- 
 ject given in the sentence ; but when the writers wished to 
 express it emphatically, it was done by means of eh, a disused 
 word meaning " many," abridged into es, and added to the 
 stem with the lengthening : 
 
 gar, " to do," " doing," or " he did ; " plur. gar-ri-es, " to 
 do (many)," " doings," or " they did." 
 
VERB. 13 
 
 R. The vowel of eS is often assimilated : 
 
 ^um-mu-u^, " they gave." 
 
 When the lengthened form is to be used, the plural, if 
 expressed, is formed by adding the plural of the second de- 
 monstrative pronoun : 
 
 gar-ra, "the doing," or "he does;" plur. gar-ri-e-ne-ne, 
 " these doings," or " they do." 
 
 Eeduplication. 
 
 Reduplication gives to the stem a factitive or causative 
 meaning : 
 
 gi, " to turn;'' gt-gi, "to cause to turn " or " to send away." 
 
 In many cases, however, it does not alter the primitive mean- 
 ing, and only expresses emphasis : 
 
 gi-gi-e-ne, " they are turning about." 
 
 Pebcative. 
 
 The precative is expressed by the prefix ga : 
 
 ga-gar, " may he do," lit. " be a doing." 
 
 It. The vowel of this prefix maybe changed under the influ- 
 ence of vocalic harmony, and the prefix becomes ge-, gi-, gu-. 
 In some cases the consonant is weakened into g : ga-, ge-, gi-, 
 gu; and is even dropped, or the prefix is reduced to a- : 
 
 ga-gar, " may he do." 
 
 a-ba-ni-in-gin, " may he return to it." 
 
 Negative. 
 
 The negation with the verb is expressed by the prefixes nu- 
 or la- (this last is very rare, and may only be but a dialectical 
 variation) : 
 
 nu-zu, "not to know." 
 
 la-gin, " not to fix." 
 
14 SUMERO- AKKADIAN GRAMMAR 
 
 R. Through, the vocalic harmony the prefix can become na-, 
 rarely ne or ni; before the labials h, p, and m it is sometimes 
 nam-. 
 
 iNCOfePOEATION. 
 
 The verbal stem can incorporate not only pronouns repre- 
 senting all the various elements of the sentence, but also the 
 particles. 
 
 The incorporated pronouns are forms of the five demonstra- 
 tives modified through vocalic harmony. 
 
 The first pronoun appears under the forms ma, mu, mi, me, 
 or am, um, im ; it is always placed first before all the other 
 incorporated pronouns or particles, never expresses the sub- 
 ject, and often represents an adverb, an indirect object, or an 
 expression modifying or explaining the meaning of the verbal 
 stem. 
 
 The second or pronoun of proximity appears under the forms 
 na, nu, ni, ne, or an, un, in, en, and after m, the n being 
 assimilated, -ma, -mu, -mi, -me. It may represent the subject 
 or object, direct or indirect, or even as the first pronoun, an 
 adverb, &c. 
 
 The third pronoun or first of remoteness appears under the 
 forms ia, hu, hi, be, or ab, ub, ib, eb ; the use of it is the same 
 as for the second. 
 
 The fourth pronoun or second of remoteness appears under 
 the form a, u, i, e ; it is easily absorbed by the vowel of the 
 next element in the incorporation, and even disappears entirely ; 
 the presence of a postpositive particle without a pronoun or 
 the vocalic alteration of the other pronoun shows that it once 
 existed in the group. 
 
 The fifth or compound emphatic pronoun appears under the 
 forms an-ga, un-ga, in-ga, en-ga ; sometimes the n is dropped, 
 and we have a^ga, &c. Like the first pronoun, it never repre- 
 sents the subject, but the object or an adverb, &c., on which the 
 writer wants to lay a special stress. 
 
 R. As a rule, these pronouns represent in the incorporation 
 either a singular or a plural ; sometimes, but rarely, the plural 
 
INCORPORATION. 15 
 
 is represented by the syllable -we suffixed to tbe incorporated 
 pronoun. 
 
 The incorporated particles are the same as the postpositions 
 added to the nouns, but only the six first are used in the incor- 
 poration : 
 
 (i.) The first appears under the forms na, nu, ni, ne, or an, 
 un, in, en, and by assimilation ma, mu, mi, me. 
 (2.) The second under the forms da and di. 
 (3.) The third under the forms ta and te. 
 (4.) The fourth under the forms ra and ri. 
 (5.) The fifth under the forms U and h. 
 (6.) The sixth under the form a, but often disappears alto- 
 gether. 
 
 R. The particles are sufiixed to the incorporated pronouns 
 which represents in the verbal form- the noun with the post- 
 position of the sentence. 
 
 R. They have the same force and meaning as the postpositions 
 added to the noun, but do not always in the incorporation 
 correspond to the postpositions used in the sentences. The 
 incorporated na may correspond to the postposition da, the 
 incorporated ta to gime, &c. 
 
 The incorporated pronouns represent the various elements of 
 the sentence irrespectively of their gender, and often number, 
 and of their person ; they only represent the words in their rela- 
 tion to each other in space or time. The subject will be treated 
 in the syntax. "We have here to consider these pronouns only 
 under their grammatical aspect ; to do this it is more conve- 
 nient to use the following abbreviations : — 
 
 s = subject; it is placed always at the end of the incorporated 
 group, that is, prefixed direct to the verb. 
 
 o = object, that is, the direct object of the verb or accusative ; 
 it comes next to the subject when there is no indirect object. 
 
 i = indirect object, which always has an incorporated particle ; 
 it is placed between the direct object and the subject. 
 
 p = particle, that is, the incorporated particle which governs 
 the indirect object ; it naturally always follows the incorporated 
 pronoun representing the indirect object. 
 
i6 
 
 SUMERO-AKKADIAN GRAMMAR. 
 
 c = complement explicative, |;hat is, the incorporated pro- 
 noun representing in the verbal form the adverb, locution, or 
 expression explaining, limiting, or specialising the meaning of 
 the stem. 
 
 V = verbal stem, that is, the verb or noun used as such. 
 
 If I wish to say, "I gave with pleasure the book to my 
 brother," the verbal form may contain incorporated pronouns 
 representing every element of the sentence :— I = s, with pleasure 
 = c, the book = o, to = j3, my brother = o. "We wiU have as verbal 
 form c-o-i-p-s-v, sic-illud-hunc-ad-ego-dedi. 
 
 Except the first pronoun (ma, &c.), and the fifth or emphatic 
 compound pronoun {an-ga, &o.), any of the pronouns may 
 represent any of these grammatical relations. 
 
 The incorporation is not always complete ; only part of the 
 sentence may be represented; often the subject is omitted, or 
 the object, direct or indirect, and also the complement ; but if 
 two or more of the sentence-elements are represented, the incor- 
 porated pronouns keep the Telative position of the grammatical 
 relation they represent. 
 
 There are therefore fifteen possible forms : — 
 
 Complete form 
 
 Incomplete (ist degree) 
 
 c-o-i-p-s-v, all the elements being 
 represented. 
 
 three elements only 
 being represented. 
 
 (2nd degree) 
 
 (3rd degree) 
 
 two elements only 
 being represented. 
 
 one element only 
 being represented. 
 
VERB. 17 
 
 R. All tliese combinations may take place -with, the stem 
 either bare or lengthened, and also with their plural forms 
 when the subject of the verb is a plural ; but often the verbal 
 stem is not in the plural with a plural subject. 
 
 R. When the prefix of the precative or the negative is used 
 with a verbal stem with an incorporated group, it is placed 
 first ; so the incorporated group is placed between this prefix 
 or negative and the verbal stem. 
 
 Emphatic Imperative. 
 
 A kind of emphatic imperative is formed by prefixing to the 
 verbal form the copula u or u, which in this case may be trans- 
 lated by "then,'' "indeed," or "ha!" and the complement is 
 always expressed by the first incorporated pronoun : tt-me-ni- 
 ^ar, "thou repeat then so." 
 
 Peonouns Suffixed to the Stem. 
 
 Part or the whole of the incorporation may be placed after 
 the verbal stem. When only one pronoun, as am, en, ib, &c,, 
 is suffixed, it may be considered as a demonstrative determining 
 the stem itself, and for this reason represents, in a certain 
 measure, the subject. When the whole group of the incorpora- 
 tion is suffixed, its elements keep the same position, v-c-o-i-p-s ; 
 therefore the complement and the objects, direct and indirect, 
 with the particle are practically incorporated between the stem 
 and the pronoun-subject to be considered in this case as a 
 demonstrative. 
 
 R. The suffixed incorporated group may offer all the same 
 combinations as when prefixed. 
 
 R. When more than one element of the incorporation is 
 suffixed, none can be prefixed. 
 
 Compound Verbal Stems. 
 
 The compound verbal stems, that is, those formed by a word 
 used as a verb and a noun simple or compound, incorporate the 
 pronouns and particles between their two elements : 
 
 B 
 
1 8 SUMERO- AKKADIAN GRAMMAR. 
 
 igi4um, "to look," lit. "to give eye,'' 
 gah-ri, "oppose," lit. "to turn the breast," 
 
 make 
 
 1234s 4SI 
 
 igi-im-ma-an-tum, "he took to him," lit. "he gave the eye 
 
 to him." 
 gdh-im-moran-ri, "he opposed him," lit. "he turned the 
 breast to him." 
 
 R. When the verbal stem is composed of more than two 
 elements, only one, the last, is the verb ; the others form one 
 word, which is placed before the incorporation : hi-gar — gi, 
 " to revenge," composed of ^u-gar and gi ; the incorporation is 
 placed between gar and gi. 
 
 R. The first element may be considered as an explicaiiive 
 complement, and as such placed at the head of the incorpora- 
 tion ; but when the pronouns and particles are suffixed, it 
 keeps its place before the verb. 
 
 ADVEEBS. 
 
 Properly speaking, there are no adverbs ; these are expressed 
 by means of nouns with a demonstrative pronoun or a post- 
 position : 
 
 iur-li or tur-rornd, " strongly ; " agi-gim, " like dark water. " 
 M-ta, "on the earth " with movement; u-kur4u, " in another 
 day = at a future time ; " tirbi-a, " then," &c. 
 
 CONJUNCTION. 
 
 The only conjunction is the copula u * " and," rarely " or.'' 
 Sometimes postpositions are used for the copula, ana-ki-ta, 
 " heaven and earth," lit. " heaven earth — with." 
 
 * It is read Sa by some. 
 
SYNTAX OF WORDS. 19 
 
 SYNTAX. 
 I. Words. 
 
 All nouns in their bare or lengthened forms may be used for 
 the masculine and feminine singular or plural. 
 
 When the genitive is expressed by apposition, the regent is 
 placed first, and the governed word only takes the lengthening : 
 e-ad-da, " the house of the father." 
 
 R. In a few expressions, probably archaic, the order is inter- 
 verted : ana-Saga, " the sky's middle." 
 
 The noun taken as adjective follows the word it qualifies : 
 kvrazag-ga, " the metal shining " or " bright metal " ( = silver). 
 
 When the noun expresses the person for whom the action is 
 done, it may take no postposition, but simply be placed at the 
 head of the sentence (see further Syntax op Sbntbncb). 
 
 The postpositions are added to the lengthened forms of the 
 noun or compound expression ; though not always written in 
 the texts, it must be supplied in reading : sil-na, for sila-na, 
 " in the street ; " an-Sag-Su, for anaSaga-Su, " in the middle of 
 the sky." 
 
 The noun in the accusative or as direct object may be in 
 the bare or lengthened forms, but sometimes the accusative 
 relation is indicated by a postposition : -na, -da, or -a. 
 
 The demonstrative adjectives suffixed to the nouns do not 
 always indicate relation of distance ; they also serve to denote 
 the instrument or the manner : Su-M, " with the hand ; '' sil-la- 
 am, "in the street;" ner-ri-ni, "with the foot." 
 
 The emphatic personal pronouns, as subjects, are generally 
 ' placed at the beginning of the sentence (see Syntax of Sen- 
 tence). 
 
 They may be used as objects. 
 
 For the use of the incorporating pronouns we have the 
 following example : me-en-ne uru-iu ga-ni-dudu-en, precative 
 form, "may we be going to the city." 
 
 The interrogative aha, "who," ana, "what," "which," are 
 always placed at the beginning of the sentence. 
 
20 SUMERO-AKKADIAN GRAMMAR. 
 
 The verbal stem is always placed at the end of the sentence. 
 
 When it is placed in the middle, it is to be considered as an 
 adjective or participle. 
 
 The incorporated pronouns never distinguish the person ; 
 the same form is used to represent either the first, second, or 
 third person. 
 
 In some texts, however, there is a tendency to use the 
 pronouns of remoteness for the second person (comp. the use 
 of iste in Latin). 
 
 The relations expressed by the pronoun of proximity and 
 the two pronouns of remoteness are not always relations of 
 distance ; but they indicate also (i) the comparative relations 
 of the various elements of the sentence as to time (if the' 
 action was performed at some period previous to the narration, 
 or is to be performed later on, the remote pronouns will 
 be used) ; (2) the comparative importance of the various 
 elements represented in the incorporation as to the performance 
 of the action. 
 
 When translating a text, it is, therefore, to be considered if 
 the author had in view to indicate, with the three different 
 pronouns, a relation of distance, of time, or of importance. In 
 the first case, the pronouns may in a certain measure represent 
 a relation of persons, like in Latin, na (hie) = " I or me," ba 
 (ille) = "he or him'.' and e (iste) = "thou or thee," but not 
 necessarily, as the relation of distance may be considered from 
 the point of view of the performance of the action ; so, if the 
 subject is a first or second person, the near pronoun would repre- 
 sent the first or second, and the two others may represent the 
 person of the object, direct or indirect. 
 
 The first and fifth pronouns never express a relationship, 
 but only represent the element to which the speaker wishes to 
 draw special attention ; they never represent the subject. 
 
 The various elements of the sentence are sometimes re- 
 presented by the same incorporated pronoun, which is, in that 
 case, generally na more often than ba. 
 
 As a rule, the subject and object are represented by diffe- 
 rent pronouns. 
 
SYNTAX OF SENTENCE. 21 
 
 With, the particles ta and ra, the subject and indirect object 
 are never represented by the same pronoun. 
 
 The desire for clearness or emphasis seems to have guided 
 the authors in the use of a complete or incomplete incorporation. 
 
 The presence of the incorporated particles are, with some 
 verbal stems, indispensable to determine the meaning : 
 
 i is " to go," with the incorporated ta, " to send away." 
 ib-ia-an-S, " he send (him) away from it.'' 
 lal is " to weigh," with na, " to weigh for " or " to pay." 
 in-na-an-lal, " he weighed to him" or "he paid to him." 
 
 When one of the incorporated pronouns, or the whole of 
 the group, is placed after the verbal stem, it gives to it a 
 participial force. 
 
 All verbal forms, with complete or incomplete incorporation, 
 may be used as imperatives. All the forms, as well as 
 the precative and the imperative proper, seem to be used 
 indifferently ; it may, however, be stated, that the precative 
 expresses a wish, the simple imperative an order, and the 
 emphatic imperative a prayer or entreaty. 
 
 The copula u is placed either between the two nouns to 
 be united, or after the second, like que in Latin. 
 
 II. Syntax op Sentence. 
 
 The order in which the words are placed together to form 
 a sentence is not the same as that of the incorporated pronouns 
 which represent them in the verbal form, and not so regular. 
 This irregularity, when it is not through poetical license, is, 
 however, only apparent, as the position of the words in the 
 sentence depends on the point of view from which they are 
 grammatically considered. 
 
 To understand the Sumero-Akkadian ideology, it is indis- 
 pensable to analyse the elements or members of the sentence 
 as they are conceived in this language. These elements can be 
 classed under seven different headings : 
 
22 SUMERO-AKKADIAN GRAMMAR. 
 
 1. Tte subject (s), or nominative. 
 
 2. The object (o), or accusative. 
 
 3. The indirect object («'). 
 
 4. The motive or reason (r), for which the action is per- 
 formed. 
 
 5. The complement (c), that is, the manner in which the 
 action is performed. 
 
 6. The determinative (d), generally expressed by what is 
 called by grammarians the adverb or adverbial locution, and 
 marking either the time (dt), place (dp), or state {ds), in which 
 the action is performed. 
 
 7. The verb (v), or the verbal stem, with or without the 
 pronouns and particles. 
 
 To these we may add the qualificative (q), which is a secon- 
 dary element, explaining or specifying primitive elements, 
 subject, object, &c., and which may be either a noun-adjec- 
 tive, a noun in the genitive, or with postposition, or even a 
 whole sentence, and always follows immediately the word it 
 qualifies. 
 
 To give an example in our language, if we say, " Yesterday 
 the king, to recompense him, gave magnanimously to the soldier 
 a sword of honour," it is analysed thus : yesterday = d, the 
 king = s, to recompense him = r, gave = «;, magnanimously = c, 
 a sword of honour = 0, to the soldier = »y giving the formula 
 d-s-r^v-c-o-i. In this sentence we have besides " of honour " = q, 
 qualifying "a sword." 
 
 In Sumero- Akkadian the formula would be d-r-s-o-i-c-v ; 
 that is, ist, the determinative or locution marking time, place, 
 or state ; and, the expression, stating the reason or motive 
 of the action ; 3rd, the subject ; 4th, the object ; 5th, the 
 indirect object ; 6th, the complement or expression explaining 
 how the action is done ; and 7th, the verb. 
 
 M. Sentences containing all the elements are never found, 
 but the elements expressed keep the same relative position. 
 In the case of a narration or speech, some are naturally often 
 understood ; as, for instance, the subject, which is not repeated 
 
SYNTAX OF SENTENCE. 
 
 23 
 
 in every sentence. The same happens also for the other ele- 
 ments. 
 
 B. It is evident that in every sentence the different words 
 or expressions might be viewed in different ways, and their 
 position altered according to the view taken by the writer. 
 Por instance, "magnanimously" might be considered as a 
 qualificative either " to recompense," the reason of the action, 
 or to "king," the subject, and "recompense" might be considered 
 as a qualificative to " sword of honour " and expressed " as 
 recompense," &c. It is these different possible conceptions 
 which give flexibility to the language. 
 
 B. The complement, in some cases, is really qualificative to 
 the verb, and the word which plays this part may be considered 
 in a certain measure as forming a compound stem with the 
 verb. 
 
 R. The three kinds of determinatives (adverbial expressions) 
 (d), marking the time (dt), the place (dp), and the state (ds), 
 might be found in the same sentence. 
 
 E. The fixed place assigned to each member of the sentence 
 explains how the postpositions are often neglected, as the 
 position of the members of the sentence is sufficient to mark 
 their relation and the part they play. 
 
 A complete sentence would contain nine members, disposed 
 according to the following formula : dt-dp-ds-r-s-o-i-c-v. "We 
 therefore can obtain 511 combinations, thus : 
 
 Complete sentence, all the members being expressed. 
 
 Incomplete (ist degree), one member missing, 9 combinations. 
 
 (2nd „ 
 
 ), two members 
 
 „ 36 
 
 (3rd „ 
 
 ), three ,, 
 
 „ 84 
 
 (4th „ 
 
 ), four ,, 
 
 „ 126 ,, 
 
 (Sth „ 
 
 ), five 
 
 I) 126 „ 
 
 (6th „ 
 
 ), six „ 
 
 „ 84 
 
 (7th „ 
 
 ), seven „ 
 
 „ 36 
 
 (8th „ 
 
 ), eight „ 
 
 9 
 
 Any sentence which 
 
 is not complete, 
 
 that is, which does not 
 
 contain the nine different members, must belong to one of these 
 
 eight classes. 
 
 
 
24 SUMERO-AKKADIAN GRAMMAR. 
 
 B. The qualificative is not by itself a member of the 
 sentence, but only forms part of a member with the word or 
 expression it quahfies. 
 
 B. The particles incorporated, the prefixes of the negative 
 and precative are to be considered as qualifications to the 
 verb. 
 
 B. Wlien the verb has a neuter meaning, there is naturally 
 no direct object, and the sentence therefore cannot be com- 
 plete. 
 
 B. In the texts it would seem that there is sometimes a 
 transposition of the three first terms (dt, dp, and ds) ; this 
 comes from a confusion in the mind of the authors. The 
 determinative of state may indeed in many cases be considered 
 as a determinative of time or of place. 
 
 B. In Sumero-Akkadian, as in the other languages, the com- 
 plete sentences are avoided, as they would render the periods 
 heavy and intricate. For quickness and clearness, authors 
 prefer to break up the period in small sentences more or less 
 connected ; for this reason most of the sentences found in the 
 texts belong to the last classes. 
 
 B. Sentences of the last class (incomplete 8th degree) are 
 only imperatives or vocatives used alone. 
 
 dingir-ra-ni-ir lugal-la e rmi-ne-en-du, "for his god the king 
 built a house," incomplete 5 th degree, formula r-s-o-v. 
 
 ib-ba-hi-ta dingir Jci-gi Ma ba-an-dvr-ne-es, " in this (for their) 
 anger the gods of the earth go on the earth," incomplete 5th 
 degree, formula ds-s-i-v : ki-gi, "of the earth,'' is qualificative 
 to the subject dingir. 
 
 an-bara nig-si-sa gu-bi ma-ra-an-ri, " the Sun-god sets the right 
 thing by (a sign of) his head," incomplete 5th degree, formula 
 s-o-c-v. 
 
 as gul galla-gim lii-ra ba-ni-inrgin, " the demon-like curse 
 is fixed on the man," incomplete 6th degree, formula s-i-v : 
 galla-gim is qualificative to the subject. 
 
 u-ebur-ra-7ca oMaga gisab-ur-ra, " at the time of work he 
 ploughs with wood (i.e., rakes) the field," incomplete 6th degree, 
 formula dt-o-v : gis forms with the verb a compound stem. 
 
COMPLEX PERIODS. 25 
 
 Similarly in the following example : 
 
 aiaga a-ih-ta-an-de, " he gives water (i.e., he waters) the 
 field," incomplete 7th degree, formula o-v : a forms with de a 
 compound stem. 
 
 Complex Periods. 
 
 Sometimes some members of the sentence are composed 
 of another sentence, which is then secondary or incidental, 
 and the qualificative may also be expressed by a secondary 
 or incidental sentence. 
 
 In these two cases the secondary sentence keeps the place 
 of the member it represents or follows the member it qualifies. 
 
 The secondary sentence is either placed in apposition, its 
 relation to the principal sentence is then shown only by its 
 position, or takes a postfix which governs the whole of it, or 
 its verb takes the participial form. 
 
 JR. Several secondary sentences may follow one another 
 united or not by the copula, but as a rule in the second case 
 they represent different numbers of the sentence. 
 
 Secondary or incidental sentence in apposition : 
 
 azag-tag-a-ni in-na-an-^e | ur-ra-na-nam ne-en-gir | e-ta ib-ta- 
 an-e, " he gave to him her dowry ; he bound it in his girdle ; 
 he sent him away from the house." The principal sentence is 
 the last ; the two first play the part of determinatives ; the first 
 being considered as marking the time {dt) " when he had given to 
 him," &c., and the second the state (ds), "having bound it," &c. 
 
 Participial secondary or incidental sentence : 
 
 salmi-ni-in-dug-ga \ te-hi nu-uh-ra- ag mi-ni-in-diin, "a woman 
 having spoken to him, he withheld himself from her.'' The 
 first sentence is secondary, and plays the part of the deter- 
 minative, marking the time or the state (dt or ds). 
 
 Adverbial secondary or incidental sentence : 
 
 a-iaga al-sal-ta ha-ah-ag-ta \ gia-gan-uru mu-^ar-ata ba-ab-icr- 
 ra, " after he had possession of the field, he drives palings in 
 the furrow." The first sentence, which plays the part of the 
 determinative of time' (dt), has the postposition ta governing- 
 
26 SUMERO-AKKADIAN GRAMMAR. 
 
 the whole of it. This postposition is the one mostly used 
 to form determinative or adverbial secondary sentences. 
 Qualificative secondary or incidental sentence : 
 u-hi-a el-Ul-la \ dim-hi gis-ne-in-du-a \ enim ^aga-M-^u ha-an- 
 hu-i, " then Bel, who heard this news, took the matter at heart." 
 The secondary sentence plays the part of an adjective. 
 
ASSYRO-BABYLONIAN GRAMMAR. 
 
 PHONETIC. 
 
 The alphabet is the same as for Sumero-Akkadian, but the 
 sound of each letter is more fixed. 
 
 -The vowels may be long or short ; a, ?, and u are weak 
 vowels, and l is the weakest, often taking the sound of the 
 Prench mute e; the other three weak vowels have also a 
 tendency to be obscured to the same extent, and often they are 
 written only, on account of the syllabic system of writing, to 
 give a consonant which could not be written alone, just as 
 in Ethiopian. The vowel e is never weak ; when weakened it 
 becomes ^; e and I are confounded. Long vowels are some- 
 times, but not always, represented by repeating the vowel, 
 as lu-u for lu ; i-e stands often for I or e. Long vowels are 
 found even in closed syllables, where there is nothing to show 
 us that they are such : lut for lut, tir for fir, &o. When two 
 vowels meet they are often contracted : ai, ae and ea become e, 
 ia becomes I, ie and ei become e. The vowel u absorbs all the 
 others and becomes long, so that au, ita, iu, ui, ue, eu become u. 
 In some cases, however, ue or eu are represented by e ; hence ui 
 may become ^ for e. 
 
 A weak vowel not accentuated, following an open syllable 
 and followed by a single consonant and vowel, is generally 
 dropped : belit and beltu, "lady." 
 
 The vowel e is generally employed to lengthen a syllable : 
 Sarrie for ^arrl ; so %arru has passed through iarrue. 
 
28 ASSYRO-BABYLONIAN GEAMMAR. 
 
 The length, of a long vowel is sometimes represented by the 
 doubling of the following consonant : amma for dma, atta for 
 dta, illi for Hi, &c. Sometimes the addition of w or m is used 
 for the same purpose ; an or am may therefore stand for a. 
 It is probable that in those cases it represented a nasalisation. 
 The n is used in preference before a dental, and m before a 
 labial, but both are found before other consonants. 
 
 The doubling of consonants was by analogy often represented 
 by lengthening the previous vowel or by placing n ox m before 
 the consonant to be doubled. 
 
 The aspirate ' indicates sometimes that the preceding vowel 
 is long, and also represents a primitive nasalisation. 
 
 Very often the aspirate ' is not represented at all by the 
 Avriting; so when two vowels follow one another, we may 
 suppose an aspirate : ta-d-ru for tddru, tairu for tdiru. 
 
 Sometimes h and g are confounded : hatu and gatu, " hand ; " 
 also A: and A : hibii and Mhit, "command," k becomes usually 
 k before i and e. 
 
 In some cases m (w) seems to have grown out of & u : ahumi 
 for dbui, " my father ; " in others, m is weakened into u. 
 
 Consonants of the same organ are sometimes assimilated : 
 erum-ma, " he went," for eruh-ma. 
 
 Before a guttural and a sibilant m often becomes n : dunku 
 for dumku, "lucky;" han^a for hamia, "five." 
 
 After a guttural t becomes t or d: aktirib for akttrib, "1 
 approached." 
 
 In the same way after m (and also, but rarely, after b and p) 
 t becomes d : amdahis for amtahis, " I fought." 
 
 When n comes into contact with another consonant it is 
 generally assimilated : Mttu for iantu, " year ; " ibhu for inhu, 
 " he proclaimed." 
 
 Sometimes I is assimilated when it comes into contact with 
 a sibilant or an r. 
 
 Sometimes r is assimilated : annu for arnu, " sin ; '' kussu for 
 kursu, " throne." 
 
 The two sibilants s and 2 are sometimes interchanged : irzitu 
 and irsitu, " earth." 
 
PHONETIC. 29 
 
 When two sibilants meet, tliey have a tendency to be 
 assimilated or to be changed in ss. 
 
 When ^ meets a i( it is changed into I: hamiltu for hamiku, 
 " five ; " the other sibilants assimilate the t. 
 
 When t is followed by a sibilant the two consonants become 
 i's : mussu for mwt-^u, "her husband." 
 
 Two ss become s simple : halasun for halassun (which is 
 itself for halat-^un), "their life.'' 
 
 These letter-changes and phonetic peculiarities are most im- 
 portant, and must always be borne in mind when decipherin"- 
 a text ; they will explain the apparent irregularities. 
 
 Often, however, the scribes appear to have retained the ety- 
 mological orthography, though no doubt it was read according to 
 the phonetic laws of the language. We find kat-ki, balaf-Su, &c., 
 which was probably read kasu, halasu, &c., according to the rule 
 given above. 
 
 Vocalic and Consonantio HARMONr. 
 
 There appears to have been a tendency to harmonise all the 
 vowels of each word and to bring them to one type. The stem 
 ' EPS makes epesu (instead of eapdsu) ; ^olaMu makes hlalti 
 (probably read hlelti). This harmonic tendency was carried 
 even to the consonants, and often a consonant is altered by the 
 presence of another in the same word, though separated by other 
 letters : iddidin for intadin. The Babylonians had a liking 
 for accumulation of the same consonants, and as result of assimi- 
 lation we get themes like zzz, bbb, &c. 
 
 ACCENTUATIOX. 
 
 The accent is almost invariably on the penultimate, and 
 exactly as in Italian, when the penultimate has a weak vowel 
 the accent is thrown on the ante-penultimate. The accent is 
 often represented by doubling the initial consonant of the next 
 syllable, and for this reason accent is often confounded with 
 length of the vowel. 
 
 The suffixes and the enclitics generally throw the accent on 
 
30 ASSYRO-BABYLONIAN GRAMMAK. 
 
 the preceding syllable. The word which receives the accent 
 from the enclitic seldom loses its own, but sometimes it is 
 thrown farther back ; the word has therefore two accents. 
 
 Teiliterism. 
 
 Though Assyro-Babylonian was written with syllabic signs, it 
 offers to a great extent the sanie phenomena exhibited by the 
 other Semitic tongues and produced by the triliterism. As 
 explained by the late Professor Palmer,^ the Semites treated the 
 fundamental letters of every root as algebraic elements, assuming 
 that every root was composed of three letters : xyz. When 
 one of the radical letters was a vowel, the root was called weak 
 of the first, second, or third radical, as the case might be. 
 When the three radicals were consonantic, the root or stem was 
 called strong. 
 
 To preserve the triliteral harmony of the words when they 
 have only two consonants, either one of them is doubled, or 
 one of the vowels is lengthened to represent the supposed weak 
 radical. The Babylonian and Ninevite scribes have often 
 shown a certain hesitation between the two alternatives; so 
 we have rahbu (supposed root ebb) and raMi (supposed root 
 EBu). There was also some hesitation as to the place of the 
 weak radical : ataru (ate) and tdru (tae). Sometimes the 
 weak radical is supposed to have been primitively an aspirate, 
 and the scribe re-establishes it : tdaru (t"e). 
 
 For the same reason the derivatives of weak stems are often 
 treated as primitive, and such is the origin of many of the 
 so-called triliteral strong verbs : tahanu, " to place," is a Shaphel 
 of Imnu, "to stand.'' 
 
 FOEMS OP THE WoEDS. 
 
 The parts of the speech may be divided into three : nouns 
 (and adjectives), verbs, and particles. 
 
 There are only two genders, masculine and feminine, and 
 three numbers, singular, dual, and plural. 
 
 ' See his abridged Arabic Grrammar, p. 59. 
 
NOUNS AND ADJECTIVES. 31 
 
 NOUNS AND ADJECTIVES. 
 
 As nouns and adjectives follow in most cases tlie same rules 
 of formation, it is as well to treat them together. 
 
 All nouns and adjectives have two forms, one short, which is 
 considered as primitive and called construct state, because used 
 when the word is constructed with another forming with it one 
 expression ; the other longer, which we call the cased form, has 
 three different endings, answering to the three cases of Arabic. 
 
 In the construct state the word has no foreign element 
 attached to it : ndkir, "enemy;" zumur, "body." 
 
 When the noun ends in a vowel this is often dropped : ah, 
 " father." 
 
 When the noun ends in a double consonant (as in the words 
 having the same consonant as second and third radical or those 
 formed by the doubling of the last radical), this double letter 
 is expressed by a single one: ^r, "king" (for Sarr); nvdun, 
 "dowry" (for nudunn). 
 
 The casal form is terminated in u for the subjective, in i for 
 the dependent, and in a for the objective case. 
 
 Sometimes the casal form is simply the noun in the construct 
 state with these vowels added to it : nakir, subjective case 
 nahiru. 
 
 When the last vowel of the word in the construct state is 
 weak and unaccented, it is generally dropped ; zumur, depen- 
 dent case zumri. 
 
 When the word is a noun derived from a root weak of the 
 third radical, the vowel of the casal form is long : ah, subj. 
 abu ; dependent abl ; obj. ahoL 
 
 The words which drop one of their double consonant in the 
 construct state restore it in the casal form : $ar, subj. case 
 iarru ; nudun, dependent case nudunni. 
 
 The feminine in the construct state is formed from the 
 masculine by the suffix at or it substituted to the vowel of the 
 casal form. Vowel harmony seems to rule the choice of one or 
 
32 ASSYEO-BABYLONIAN GKAMMAE. 
 
 the other suffix: Icalhu, "dog," fem. Imlbat ; helu, "lord," fem. 
 helit. 
 
 In the casal form the vowel preceding the t of the suffix is 
 sometimes preserved, but more often is dropped ; in this case 
 the vowel preceding the last consonant of the word must be 
 restored if it be one containing three consonants, and in words 
 having a double consonant like iarrat the a of the suffix is 
 necessarily maintained ; kalbai makes subj. case Jcalbatu ; tarrat, 
 subj. case tarratu; aAa^, sister, subj case ahatu ; but 6eKt makes 
 subj. case heltu. 
 
 Dual. 
 
 The suffix of the dual was primitively an, but it became a in 
 the course of time, though the two endings are found : 
 
 saptu, " the lip ; " iwptan and ^aptd, " the two lips." 
 uznu, "the ear;" uznd, "the two ears." 
 
 Plukal. 
 
 A primitive way to express the plural, imitated perhaps from 
 Akkadian, is by reduplication. It is, however, of a very limited 
 use. When the word was polysyllabic, only one syllable was 
 repeated : 
 
 mil, " water ; " plur. mami. 
 
 iamu, "sky;" plur. ^amame. 
 
 The plural was more often formed by lengthening the vowel 
 of the casal form, the lengthening being made by adding the 
 vowel e. We ought to have : subj. we, dep. ie, and obj. ae ; but 
 the vowels are often assimilated, and we find subj. u, dep. I, 
 obj. a. In general it is the vowel e which has absorbed the 
 other, and we have e for all cases of the plural. Sometimes 
 the same word is found with the two forms umu and ume, 
 " days." 
 
 Another way of forming the plural is by the suffix nu, 
 generally weakened into ni, added to the casal form of the 
 
NOUNS AND ADJECTIVES. 33 
 
 singular. We might expect forms in uni, ini, ani, or unu, 
 inu, anu, but tlie form in ani is almost exclusively used, and 
 is found side by side with the plural in ie .- ^arrani and Mrrie, 
 "kings." 
 
 The plural of the feminine is formed by changing at into ate, 
 sometimes dti, very rarely dtu. 
 
 This formation probably originated from adding the feminine 
 suffix to the plural of the masculine ; cite would be for OrOte. 
 We might then expect the forms utu for u-atit, dtu for d-atu, 
 Itu for l-atu, but the forms in ate and dti only are preserved ; 
 the final e or i is, no doubt, by analogy to the masculine plural 
 in e. 
 
 The words taking itu in the singular make Iti or eti in the 
 plural: Narrate, " queens;" bSlieti, "ladies." 
 
 The forms in Utu or Uti are applied to form the plural of 
 adjectives in the masculine, as Hani rahuti, "the great gods." 
 
 The form in utu is used to form abstracts, which are used 
 as singular nouns, and have no plural : Sarrutu, " kingship " 
 or "royalty.'' 
 
 The construct state is generally the same as the casal form. 
 In a few rare cases, however, it is formed by dropping the 
 final consonant. 
 
 MiMATION AND EMPHATIC StATE. 
 
 Primitively the casal form was terminated by an m, and was 
 subj. um, dep. im, obj. am. Subj. Uprum, "the message;" 
 dep. irsitim, "the earth;" obj. hirram, "the free-born child." 
 
 These mimated forms were soon more weakened still, and 
 the m dropped was replaced by a simple aspirate ; it is what 
 we call the emphatic state (see Syntax), to distinguish it from 
 the mimation : obj. koMpd, " the money." 
 
34 
 
 ASSYRO-BABYLONIAN GKAMMAR. 
 
 IJUMEEALS. 
 
 
 The numerals have two genders ; they are from one to ten. 
 
 Masculine. 
 
 
 Feminine. 
 
 I. eUen, c.f. eManu. 
 
 edit, 
 
 c.f. editu. 
 
 2. ^ina, „ Unu. 
 
 Unat, 
 
 „ Unatu. 
 
 3. Mai, „ Miu. 
 
 Miat, 
 
 „ ialaitu or ialaltu 
 
 4. arba, „ arbu. 
 
 arbat, 
 
 „ arbatu. 
 
 or arba, „ arbdu or arbau. 
 
 or «>6ji, 
 
 ,, irbittu. 
 
 5. hamii, „ JiamMu 01 haniu. 
 
 hamMat, 
 
 „ hamiltu. 
 
 6. M, „ kiiu. 
 
 mMt, 
 
 „ Uiiatu. 
 
 7. siba, „ sibau or sibiZ. 
 
 sibit, 
 
 „ sibitu or sibittu. 
 
 8. iamna, „ iamnu. 
 
 iamnat, 
 
 „ iavmatu. 
 
 9. tiioitiia,,, tisu. 
 
 tisit, 
 
 „ tiUtu or tiiittu. 
 
 10. eJ'er, „ eJ'rw. 
 
 e%rit. 
 
 „ eiritu. 
 
 or eterit, „ e^eritu. 
 
 The numeral for "twenty" is the dual of "ten;" the other 
 tens are formed from the units by lengthening the vowel of the 
 casal form, but they are always vocalised in a : 
 
 20. eird (dual of 10). 60. Bid. 
 
 30. keloid. 70. sibd. 
 
 40. irbd or irbdia. 80. iamnd. 
 
 50. hansd. 90. tiid. 
 
 The word for 100 is me invariable ; Sina me, 200. 
 
 From II to 19 the units are placed, as in Latin, before ten : 
 eiten-eirit, "eleven;" and the compounds are considered as 
 forming only one word : hamiihrit, " fifteen." 
 
 Numeral Adjectives. 
 
 The ordinal or numeral adjectives are formed by lengthening 
 the final vowel of the casal form, so that they appear to have a 
 plural form; for "first," the word is not derived from the 
 numeral " one," but we have riUdnu, lit. " the head one,'' for 
 
PRONOUNS. 35 
 
 both, genders, and mahru, lit. "the foremost" or "front one," 
 fern, mahritu. 
 
 The others are : 
 
 2iid. ^anu, fern, ^andtu. 6th. Uitu or Mm, f. U^dtu. 
 
 ^rd. taMii, „ %aMdtu. "jth. sibu, „ sibdtu. 
 
 4th. ribu, „ ribdtu. 8th. Samnu, „ Samndtu. 
 
 ^th. hansu, „ han^dtu. gth. ti^ii, „ tiidtu. 
 
 or lia^^, „ IwMdtu. 10th. esru, ,, uhdtu, 
 
 or eh'dtu. 
 For the tens we have a similar formation : the 30th, MaMu. 
 
 Feaotions. 
 
 The fractions are expressed by the cardinal numerals followed 
 by the feminine of the ordinal : 
 
 ^alsi ribati, "three-fourths." 
 
 At a later period the masculine is found, iii ribi, "three- 
 fourths.'' Half is meSlanu, or menu. 
 
 The number 60 or the soss being taken as unit, the tens are 
 used as the fractions of 60, and words used for these fractional 
 numbers also used for the tens, as in Akkadian. 
 
 Unlbu, f. Unipatu, &c. 
 
 PEESONAL PKONOUNS. ' 
 
 The pronouns are the only part of the speech which has 
 preserved flexions indicating the eases, like in the Aryan 
 tongues. 
 
 Singular. 
 
 1st Pers. 
 
 2nd Pers. 
 
 3rd 
 
 Pers. Masc. 
 
 Pern. 
 
 Nom. . . andku, 
 
 attahau, 
 
 
 Sm, 
 
 U. 
 
 Gen. . . ia. 
 
 Tta, 
 
 
 &, 
 
 U. 
 
 Dat. and Abl. iaSi, 
 
 MU, 
 
 
 SuaM, 
 
 %ias; 
 
 Ace. . . iati, 
 
 kdti, 
 
 
 ivati, 
 
 iiat 
 
36 ASSYEO-BABYLONIAN GRAMMAR. 
 
 Plural. 
 
 1st Pers. 2nd Pers. 3rd Pers. Masc. rem. 
 
 Nom. . . ninu or aninu, attunu, tunu, Una. 
 
 Gen. . . ni, Itunu, ^unu, Sina. 
 
 Dat. and Abl. nia^i or ndU, hunuU, ^unv^i, &ina^i. 
 
 Ace. » . niati, kunuti, ^unuti, Unati. 
 
 R. The form attakau was soon disused, and replaced by atta 
 for the masculine, and atti for the feminine. In the genitive 
 of the same person the genders are also distinguished : masc. 
 ka, fem. ki. 
 
 The pronouns in the genitive being placed after the nouns 
 in the construct state, are considered as possessive suffixes. 
 
 For the first person singular ia often becomes a, a, or i, or is 
 absorbed by the final vowel of the noun, ahi, " my father." 
 
 For the second person singular, besides ka, we find hu, ku, and 
 ka. These forms are used for the masculine. For the feminine 
 ki is used. 
 
 For the third person feminine la is also used, and at all the 
 cases %i may become d. 
 
 The genitive forms used as suffixes may take a vowel of 
 union, which is only a prolongation of the vowel of the noun, 
 and the length may of course be represented by the doubling of 
 the following consonant: ahua, "my father;" kibitukka, "thy 
 will." 
 
 The final vowel of the genitive may be dropped : ahlu^, " his 
 son ; " ^arrikun, " your king." 
 
 When the dative or ablative and accusative forms follow 
 immediately the verb of which they are the indirect or direct 
 objects, they may be considered as suffixes, and the initial con- 
 sonant may be doubled. 
 
 The flexions of the pronouns are retained only in the earliest 
 inscriptions ; at the later period it is maintained only, and not 
 always, for the third person plural. 
 
 As a rule, when the pronouns are direct or indirect objects 
 of the verb, they are suffixed to the verb under a short form : 
 
PRONOUNS. 37 
 
 ist pers. -ni. plur. -nini or -nu. 
 
 2nd pers. -ka, fem. -^«. , plur. -kunu, fern, -fe'na, 
 
 3rd pers. -Su, fem. -&'. plur. -Sunu, fem. 
 
 The final vowel may be dropped, as for the possessive suffixes. 
 
 The Babylonians having forgotten the origin and notion of 
 these flexions, assimilated the final vowel to the case-endings of 
 the noun ; so we find Tca^u, TcaMi, and IcaMa ; ^unutu, ^unuti, and 
 iunuta, &c. 
 
 The plural formative in the pronoun, -nu, is sometimes added 
 to the flexion : tutunu, titina, Jca^unu, &c., for ^unutu, Enati, 
 JcunuSi. 
 
 The mimation is also sometimes added to all the flexions : 
 ia%im, IcaUm, haSam, Tcatim, &c. 
 
 From the genitive ia, Jcu, &c., are formed possessive adjec- 
 tives by adding u : iau, mine ; kau, thine. 
 
 An independent possessive adjective is formed by adding the 
 sufiixes to atu or attu: ahu, atua, or attua, "my father;" but 
 this is rare before the Persian period. 
 
 Demonstratives. 
 
 The personal pronoun of the third person is used as demon- 
 strative adjective : %u, "this," fem. H. 
 
 The accusative tuati, fem. Uati, was used for the nominative, 
 and the final vowel assimilated to the case-endings ; but the 
 forms in ti were always used in preference even for the nomi- 
 native. 
 
 By a similar assimilisation the dative form is used as a 
 demonstrative : Suasu, %uaMi, ^ua^a, fem. iia^u, &c. 
 
 We have also the relative combined with the possessive suf- 
 fixes : ^aMu, ^aMi, %dsunu, ^aMina. 
 
 The other demonstratives are : 
 
 ammu, fem. ammatu, plur. ammutu, fem. ammdtu. 
 annu, „ annitu, „ annutu, „ anndtu. 
 ullu, „ ullatu, „ ullutu, „ ulldtu. 
 
38 ASSYRO-BABYLONIAN GKAMMAR. 
 
 These three pronouns, instead of the plural in utu, can also 
 take another plural in e : amme, anrie, uUe. 
 
 There is another demonstrative of doubtful origin : 
 
 aga, agd, or agd, fern, dgata, plur. aganutu, fern, agandtu. 
 
 It may be combined with the personal pronoun, ogaMu or 
 Suaga, &c., also with the demonstrative annu. 
 
 Eelative. 
 
 There is only one relative, which is invariable : ^a, " who, 
 whom, which," &c. 
 
 Interrogative. 
 
 The interrogatives are mi, " who, why ; " man, manu, " who ; " 
 manume, "who." 
 
 The Indefinite Pronouns are : 
 
 mamma, manma, manamma, \ "anything," 
 
 mimmu, plur. mimmemi or mimemi, j " everything." 
 iaumma, "whatever." 
 Manamma, " another." 
 mala, "everything." 
 
 For the negative, la is simply put before. 
 For " one another," ahames is used, but for " some . . . 
 others," we find also anute . . . anute, ahadi . . . ahadi, &c. 
 
 VEEB. 
 
 The verb has four principal voices, having each a secondary 
 and a tertiary form or voice. 
 
 The first voice, or Kal, is the stem simple without addition. 
 
 The second, or Niphal, is formed by prefixing na to the stem. 
 
 The third, or Shaphel, is formed by prefixing la to the stem. 
 
 The fourth, or Pael, is formed by doubling the middle radical 
 of the stem. 
 
VERB. 39 
 
 The secondary voices are formed by the insertion of ta after 
 the first syllahle of the principal or primary voices. 
 
 The tertiary voices are formed by inserting tan or tana in 
 the same way. 
 
 I Moods and Tenses. 
 
 The Assyro-Babylonian verb is rich in forms. There are 
 nine moods, including the infinitive, which is, however, not a 
 mood of the verb, but which, properly speaking, is composed of 
 nominal and adjectival forms, as will be seen farther on. 
 
 The moods and tenses may be classified thus : 
 
 I. Infinitive: simple, participle present, participle continu- 
 ous, participle past-active, and participle past-passive. 
 2 Indicative : aorist, mutative, precative. 
 
 3. Imperative : aorist, mutative. 
 
 4. Permensive: simple, precative. 
 
 5. Subjective: aorist, mutative, precative. 
 6: Dependent; „ „ ,, 
 
 7. Objective : „ „ „ 
 
 8. Energetic: „ „ ,, 
 
 9. Paragogic: „ „ 
 
 JOEMATION OF TeNSES. 
 INFINITIVE. 
 
 In the Assyro-Babylonian language, as it is known to us, all 
 the words were considered, by those who spoke and wrote it, as 
 derived from a theme or root formed of three letters, generally 
 consonants, and called radicals : XTZ. These themes repre- 
 sented to the mind an abstract notion or idea, and to give to 
 them life, so to say, vowels were to be added. 
 
 The words so formed from the abstract themes do not, pro- 
 perly speaking, form part of the verb, and they are generally 
 considered as nouns or adjectives ; but as it is from them 
 that the various tenses are formed, they must be mentioned. 
 
40 ASSYRO-BABYLONIAN GEAMMAR. 
 
 All the verbal forms may be brought back to five nominal 
 formations, obtained from the theme by different vocalisa- 
 tions. 
 
 (i.) Infinitive simple. — XaYdZ gives the abstract notion of 
 the theme, and answers exactly to our infinitive, no account 
 being taken of time or duration: SaJcdnu, "to place;" rahdsu, 
 " to inundate ; " zakdru, " to record ; " hahdtu, " to plunder ; " 
 rahdsu, "to bind;" makdsu, "to impose (a tax);" rakabu, "to 
 ride;" haldtu, "to live;" ^atdru, "to write," &c. 
 
 All these words are really abstract nouns of action or state, 
 and might therefore be translated : a placing, inundation, a 
 recording, &c. But the close connection of these words with 
 the real verbs was always preserved in the mind of the Baby- 
 lonians, and they were used exactly like our infinitives. 
 
 (2.) Participle present. — XaYiZ forms nouns of agents, and 
 answers exactly to our present participle. It gives the notion 
 of the action being done for the time being : iaidn, " placing ; " 
 rahis, " inundating, a shower ; " zaltir, " recording, recorder ; " 
 rahii, " binding ; " rakihu, " riding, a charioteer ; '' maJcis, " tax- 
 ing, a taxer," &c. 
 
 (3.) Participle continuous. — XaFa.^ expresses the action being 
 done and lasting, and answers more exactly to our names of ' 
 agents. The examples of this form, as of the following, are less 
 numerous, as their use was subjected in a great extent to the 
 caprice of the nominal formation and of the custom of the 
 speakers : ^akan, subj. ^aknu, " the placer or doer, i.e., governor ; " 
 zakar, subj. zakru, " the one who records," taken as adjective, 
 " mentioned," because the thing mentioned, which has a name, 
 is a perpetual recorder ; rakah, subj. rakahu or rakbu, " ambas- 
 sador,'' because he is constantly riding or travelling ; Sataru, 
 "inscription, writing," because the inscription keeps perma- 
 nently the writing, is being inscribed; also used as adjective, 
 " inscribed," &c. 
 
 (4.) Participle past-active. — XiYiZ indicates the result of 
 the action completed ; it answers to a great extent to our nouns, 
 like " rain," the result of " raining ; " " deed," result of " doing ; " 
 "a run," the space run through, &c. : Min, subj. Uknu, "the 
 
VERB. 41 
 
 space," from the meaning " to do ; " rihis, subj. rihsu, with the 
 fern, form rihistu, " inundation," or " a shower " after it is over ; 
 zildr, " name," result of recording ; ^itir, subj. Etiru, " a writ- 
 ing," the body of the contract, result of "writings" ^ihib, 
 subj. Ebbu, " crown,'' result of the action of " bending round " 
 i^ababu), &c. 
 
 (5.) Participle past-passive.— XuTiiZ expresses the result of 
 the action as suffered by the object, and answers to our past- 
 participle : %ukun, in the fern, ^ulcuttu (for ^ukuntu), " tool," 
 what has been done ; hubut, subj. hubtu, " booty," what has 
 been plundered ; bulut, subj. bultu, " alive," as a noun " the 
 living,'' what has received life ; rukub, subj. rukubu, " a chariot,'' 
 what is ridden, also " the course," the ground which is ridden 
 on, &a 
 
 The two last formations have been often confused, and the 
 same theme appears in the two forms with the same meaning, 
 as ^ikittu (for Ukintu) and ^ukuttu (for %ukuntu), " tool ; " or the 
 same form has the meanings of the active and passive, as from 
 rakdsu, " to bind ; '' rikis, subj. riksu, " a contract," the result 
 of " binding," and " the waist," the part which is bound (cf. 
 French ceinture). 
 
 All these forms, being in reality nouns or adjectives, follow 
 the same rules as to the formation of the feminine and plural, 
 and as to the case-endings. 
 
 When the theme has two radicals the same letter, one of 
 them is often doubled : iulullu by the side of iullu (subj. of 
 Sulul), participle past-passive of ialdlu, " to carry away." 
 
 Infinitives of the other voices are very rarely found in the 
 texts, and appear only as nouns or adjectives. 
 
 The infinitives, as given by modern Assyriologists, are in all 
 the voices formed from the participle past-passive, to which is 
 added the formative prefixes of the various voices. For the 
 theme ikn the twelve infinitives restored by analogy are : ^ 
 
 1 In the excellent paper of Mr. Pincher on the Permensive, the nomen- 
 clature of the Hebrew grammar is preserved, Kal, Niphal, &o., because it 
 has the advantage to be understood. 
 
42 ASSTEO-BABYLONIAN GEAMMAR. 
 
 Kal, iahdnu. K^, iitdkunu. K^, Utanhunu. 
 
 Mphal, naMkunu. W, itoMkunu. W, itanaHcunu. 
 
 Sliaph.el, ^iiMhunu. S^, ^uta^kunu. S', tuianaskunu. 
 
 Pael, tukkunu. P^, Utdkkunu. P^, Mtanakkunu. 
 
 These infinitives are real nouns, and. follow tlie same rules as 
 the nouns for the formation of the feminine and plural, &c. 
 
 The secondary and tertiary voices of the Niphal shovf the 
 weakening and the total loss of the characteristic n / itasJmnu 
 and itanaSkunu is for nitaskunu and nitanaMhunu. 
 
 It is to be noticed that all these forms are passive, and really 
 participles past-passive of the same type as in the Kal XuYuZ. 
 It is, therefore, not correct to call them infinitives simple. 
 
 A few forms accidentally found show besides that real infini- 
 tives simple, corresponding to the Kal form XaYdZu, existed in 
 the language, but those found are considered as nouns or adjec- 
 tives, like the Niphal naptdru, " to defend," and the Shaphel 
 iap^dhu, "steep." 
 
 These forms were considered as the real infinitives or nomina 
 verbi by the Babylonians, and are those given in the sylla- 
 baries. 
 
 The infinitives of the secondary voices have always i before 
 the inserted t, as in gitmalu, " benefactor," from gamalu, " to 
 spare." 
 
 The infinitives of the tertiary voices have not been found. 
 
 The forms of the infinitive simple may be restored, for the 
 stem BLT (life, to give life, to live, &c.), thus : 
 
 Kal, balatu. K^, hitldtu. 'K?, hitanlatu. 
 
 Niphal, nahldtu. N^, {n)itdblatu. 'S^, (n)itanahlatu. 
 
 Shaphel, $ahldtu. S^, Utdbldtu. S', ^itanablatu. 
 
 Pael, balldtu. P^, hitalldtu. P^, hatanallatu. 
 
 Examples of the participles present are found only in nouns 
 and adjectives like nahliptu, feminine form of the participle 
 present Niphal of haldpu, "to cover." 
 
 The examples of the participle continuous are not so scarce. 
 For the Niphal we have namrasu, " impassable," from marasu, 
 
VEEB. 43 
 
 " to be unwell ; " for the Pael, raklcahu, " a chariot ; " but these 
 forms must have been easily confounded with those of the true 
 infinitive simple. 
 
 The participle past-active was naturally very rare in those 
 voices having a tendency to be considered as passive. Only 
 the Pael, Jdssillu, " royal," from Icasdlu, may be quoted, and has 
 the last radical doubled as well as the second. 
 
 The past-participle passive has been more preserved, and 
 forms regularly the permensive of all these voices ; the restored 
 forms have been given above, because it is the ones generally 
 given as the infinitive by Assyriologists. 
 
 The words formed with these participles were no doubt 
 avoided on account of their length, and examples are not so 
 rare with the weak stems. 
 
 All these voices have also a compound participle formed by 
 the prefix mu added to the other participle. It is found with 
 the Kal forms, but in the other voices it is more regularly used : 
 
 muntahsu (constr. muntahis), " soldier," from the Niphal 
 
 of iahasu, "to fight." 
 munndltalu, " concealed," from the Niphal of naltdlu, " to 
 
 build." 
 
 In the Shaphel the vowel following the ^ is either a or e .• 
 
 mu^dlmi^ and muhhnii, from hana^u, " to subject." 
 
 In the Pael the double letter of the middle radical is often 
 written simple : 
 
 munakkir, "destroyer," from nakdru, mumahir, "director," 
 from mahdru. 
 
 The secondary and tertiary voices are regularly formed from 
 the participles of the Kal : 
 
 muhtahil, subj. muMablu, " fighter, one who places himself 
 in the middle (of the battle)," from kahdlu, " to be in 
 the middle." 
 
 These compound participles may be formed from all the parti- 
 
44 AS3YR0-BABTL0NIAN GRAMMAK. 
 
 ciples of Kal, but are generally considered as nominal or 
 adjectival formations : 
 
 munirritu, " struggling," from the Pael of naratu. 
 
 Indicative or Construct Mood. 
 
 AOEIST. 
 
 The aorist is formed by means of preiixes added to one of 
 the participles ; the genders and numbers are distinguished by 
 the vowel suffixed to the form. 
 
 Sing. 1st pers. com. a — 2nd pers. masc. ta — 3rd pers. masc. i — 
 
 fem. to — i fern, to— 
 
 Plur. „ „ ni — „ masc. to — u „ masc. i — u 
 
 fem. ta — a fem. i — a 
 
 When these prefixes are added to the participles, the first 
 vowel of this is almost invariably dropped ; we have, therefore : 
 
 From the participle continuous : ist pers. aXTaZ, 2nd pers. taXYaZ, &c. 
 From the participle past-active : „ aXYiZ, „ taXYiZ, &o. 
 From the participle past-passive : „ aX TuZ, „ taX YuZ, &c. 
 
 When the aorist is derived from the participle present, what 
 is exceedingly rare, the vowel of the first radical is preserved, 
 unless it would be confounded with a form derived from the 
 participle past-active, and it is ist pers. aXaYiZ, 2nd pers. 
 taXaYiZ, &c. 
 
 The prefixes given above, which form the various persons, 
 are the remnants of auxiliary verbs ; if their auxiliary character 
 is not lost sight of, the formation of the aorist appears quite 
 consistent and rational. For instance, a&kun, " I placed," from 
 Sakdnu, " to place," is a form exactly parallel to " I have 
 placed," formed from the auxiliary a and the past-participle 
 passive, ^ukun. In the same way : 
 
 amiuh= " I have measured," from maMdhu, " to measure." 
 ta^kun = " thou hast placed," from ^akdnu, " to place." 
 mStur=" we have written," from ^atdru, " to write." 
 
VERB. 45 
 
 And these forms may be explained more correctly thus : 
 
 aMkim, " I have (the thing) placed ; '' amiuh, " I have (the 
 thing) measured," &c. 
 
 "With the participle continuous the form may be analysed in 
 the same way : 
 
 isbat - " he has a taking," from sdbatu, " to take." 
 
 But there is a difference in meaning corresponding to that 
 attached to the participle used as formative : 
 
 ishat is " he has a taking of a thing, and retains possession 
 of it." 
 
 With the participle past-active : 
 
 dbUl=" I have a cooking (a stew)," from ha^dlu, " to cook," 
 ■which really is " I have the thing which was once 
 cooked (as a stew, a piece of roast-beef, &c.)." 
 
 Originally no doubt every verb could form its aorist from all 
 the participles indifferently, and had in reality four aorists ; but 
 in practice only one was retained, and it is only accidentally 
 that the same stem is found to have several forms for the 
 aorist, as : 
 
 ishat and isbut, "he took." 
 
 The notion of the auxiliary character of the prefixes was lost, 
 and at the same time also the origin of the formation forgotten. 
 The verbal forms were then considered as uncompound, and to 
 distinguish the genders and numbers a vowel was suffixed — i 
 for the feminine singular, and in the plural u for the masculine, 
 a for the feminine, and for the dual a. 
 
 The greatest number of the verbs retained exclusively the 
 aorist (in u) formed froni the participle past-passive, as the 
 following : 
 
 hahdiu, "to plunder;" hajndtu, "to hasten;" hasdhu, "to 
 desire;" gamdru, "to be complete;" kahaSu, "to trample upon;" 
 kandku, " to seal ; " kardbu, " to be propitious ; " kqpddu, " to 
 
46 ASSYRO-BABYLONIAN GRAMMAR. 
 
 level;" kasdru, " to -niu awaj ;" babdlu, "to pull down;" hatdku, 
 " to detacli ; " pataru, " to free ; " paharu, " to collect ; " parasu, 
 " to destroy; '' magdru, " to obey ; " makdtu, " to fall;" matdhu, 
 " to measure ; " iadddu, " to drag ; " ragdmu, " to ask ; " raJcdsu, 
 " to bind;" ramdku, " to pour out;" radddu, " to pursue;" ^aldlu, 
 " to carry away;" tardpu, " to burn;" iatdru, " to write;" %ahdpu, 
 "to overwlielm;" zakdru, " to record;" iahdpu, " to sweep away;" 
 sardpu, " to melt ; " dabdJcu, " to spread ; " dardhu, " to be cast 
 down;" tarddu, "to drive away;" tdkdpu, "to take bold;" 
 tdbdku, "to pour out;" tamahu, "to grasp;" and many others. 
 
 A small number has retained the exclusive use of the aorist 
 (in i) formed from the participle past-active, as the following : 
 
 hapdru, " to bring out ; " haldku, "to injure;" gaMdru, "to 
 strengthen ; " baMdlu, " to cook ; " haidmu, " to dispose ; " patdhu, 
 " to build; " pardru, " to disperse;" raidpu, " to build;'' rasdpu, 
 " to join;" labdnu, " to cast down;" ^aldmu, " to pacify;" ^azdnu, 
 " to be angry; " sakdbu, " to overthrow;" dagdlu, " to see ; " and 
 a few others. 
 
 A few verbs have exclusively the aorist (in a) formed with 
 the participle continuous, as the following : 
 
 pardku, "to divide;" patdhu, "to fear;" mahdsu, "to fight;" 
 iandnu, " to be equal ; " takdlu, " to trust," &c. 
 
 As said above, in a few cases verbs have preserved the use 
 of two aorists, though the same meaning is attached to both : 
 Jeardbu, " to collect," has the two forms ikrub and ikrib; sabdtu, 
 " to take," isbat and isbut, mahdru, " to receive," imhur and 
 imhar, &c. 
 
 As for the aorist derived from the participle present, it must 
 be acknowledged that its very existence is doubtful, for such 
 forms as ipa^itu, "they destroyed," from pa^dtu, "to destroy," 
 may be mutatives, in which the second or middle radical has 
 not been doubled, as it often happens (see farther on), and used 
 for aorists. 
 
 In the other voices the aorist is derived from the participle 
 present, rarely from the participle past-active. 
 
 The prefixes indicating the persons remain the same in the 
 secondary and tertiary voice of Kal and in Niphal and its two 
 
VEEB. 47 
 
 voices, but for the other voices the vowel of these prefixes is u 
 in all the persons, thus : 
 
 Singular. 
 
 ist Pers. 2nd Pers. Masc. rem. 3rd Pers. Masc. Fem. 
 U — '■ tu — tu — i U — tu — 
 
 Plural. 
 
 St Pers. 
 
 2nd Pers. Masc. 
 
 Pem. 
 
 3rd Pers. Masc. 
 
 Fem. 
 
 nu — 
 
 tu — U 
 
 tu — a 
 
 u — u 
 
 u — a 
 
 In the Niphal the characteristic n is generally assimilated : 
 ikldhus (for inlcibus), "he was conquered," from Icabdsu. 
 
 In the Shaphel the vowel following the I' characteristic may- 
 he a, e, or i : 
 
 u^akniS, u^ekn&, and ti^ikni^, from hana^u, "to submit." 
 uMhlika, from haldku, " to destroy." 
 u^iklil, from halalu, " to end.'' 
 
 In the Pael the doubling of the middle radical is often 
 neglected : 
 
 uhallik, "I divided," from halaku. 
 
 tukattiri, " thou hast surrounded," from katdru. 
 
 uzakkir and uzdkir, " I completed," from zakaru. 
 
 In the secondary voices the vowel of the characteristic letter 
 t may be a, e, or i : 
 
 iktalik, from haldku, " to destroy." 
 ihtesin, from hasanu. 
 uUahsir, from kasdru. 
 uptaMit, iiompaMtu. 
 aktirib, from kardhu. 
 
 For the tertiary voices the aorist is regularly formed from the 
 participle past active or passive. 
 
 ikdanaludu 'K? from kalddu. 
 ittandbrik N^ from hardku. , 
 
48 ASSYRO-BABTLONIAN GRAMMAR. 
 
 Mutative. 
 
 The mutative is formed with the same prefixes as the aorist 
 added to the participle present or the participle continuous, 
 but the middle radical of these participles is doubled in order 
 to either strengthen the idea conveyed to the mind by the verbal 
 noun, or, most probably, because the action is actually performed 
 by the subject. The origin of this tense is the same, in fact, 
 as that of the Pael, having the force of a passive, or rather 
 middle voice, as it is called in Greek. This explanation is sup- 
 ported by the fact that in great many cases the mutative of 
 the Kal is replaced by the mutative of the Pael (see Syntax). 
 The mutative may therefore be explained in the same way as the 
 aorist, but the auxiliary is to be translated by the verb " to be " : 
 
 araJcMs = "l am binding," from rakdsu, "to bind." 
 inadin = " he is giving," from naddnu, "to give.'' 
 taialildn = " thou art placing," from iakdnu, " to place." 
 
 When derived from the participle continuous, the mutative 
 had naturally a slight difference in meaning : 
 
 oMalclcan = " I am placer," from ^akdnu, " to place." 
 inaddan = "\\e, is giver," from nadanu, "to give." 
 ihassas= "he is sinner," from hasdsu, "to sin." 
 tasabbat="th.o\ia,xt taker," from sabdtu, "to take." 
 
 From the origin of these two forms it appears quite rational 
 that they should have been used to express the present or the 
 future. The first is more properly the present, for it indicates 
 the subject of the verb as being in the act of performing the 
 action, and the second is a continuous present or future. In 
 practice, however, the two forms were confounded, and most 
 of the verbs have retained only one. 
 
 As noticed in the phonology, a double consonant was often 
 written simple ; the mutative in this case is distinguished from 
 the aorist only by its vocalisation and the presence of a vowel 
 after the first radical. We have : 
 
VERB. 49 
 
 i'^kun, "he placed;'' iUlcan, "he places" or "shall place." 
 izlcur, "he recorded;" izakar, "he records" or "shall 
 
 record." 
 indin, "he gave;" inadin, "he gives" or "shall give." 
 imhar, "he received;" imahar, "he receives" or "shall 
 
 receive." 
 
 Sometimes we find abnormal vocalisations, as iraggum, " he 
 disputes," from ragdmu, but it is no doubt due to the fact 
 that in these cases the mutative has been derived from a nominal 
 form not noticed above. 
 
 , For all the other voices, the mutative is formed from the 
 participle continuous with the same prefixes as in the aorist of 
 the same voice : 
 
 makan (for in^akan), N., from iakdnu. 
 
 attakil and atakil (for antakil), N., from tahdlu. 
 
 u^ahnap, S., from handpu. 
 
 usahhar, P., from sahdru. 
 
 upahar, P., hompahdru. 
 
 aUakan, K^, from ^akdnu. 
 
 ittaihan, N^, from Sakdnu. 
 
 uMaklal, S^, from kaldlu: 
 
 ittanagrar, N^, from gardru. 
 
 Peecativb. 
 
 The precative is formed for all the voices by the particle hi, 
 "be it," prefixed to the verbal forms of the aorist or mutative. 
 There are therefore two precatives, but in practice only one of 
 them has been, as a rule, preserved with the same meaning of 
 simple precative, though primitively there must have been a 
 slight shade of meaning between them. 
 
 When the verbal form begins with a, this letter is dropped 
 or assimilated to the u of the prefix : lublut, "may I live," from 
 baldtu. 
 
 When the verbal form begins with i, the vowel u of the 
 prefix is dropped or assimilated : lihlik and lihcdlik, frorn, 
 haidku ; Mkun, from Sakanu. 
 
 D 
 
50 ASSYEO-BABYLONIAN GRAMMAE. 
 
 Impbeativb. 
 
 AoEiST. — Tlie aorist has only one person, the second of the 
 singular and plural. It is formed from one of the participles ; 
 the great majority of the verbs form it from the past-participle 
 passive {XuTuZ), a small number from the past-participle 
 active {XiYiZ), and a few from the participle continuous 
 {XaYaZ), and fewer still from the participle present {XaYiZ). 
 The same verb might take two different forms. Probably at 
 the origin the aorist of the imperative was formed indifferently 
 from the various participles, according to the idea which the 
 speaker wanted to convey, the imperative being, in fact, nothing 
 else than the stem. When it took the vocalisation of one of the 
 participles, it also took the idea attached to it. In practice, 
 however, one form only survived for each verb, and it follows 
 that adopted for the aorist of the indicative : 
 
 iikun, "he placed;" ^ukun, "place thou." 
 iprus, "he freed ;" ^mj'ms, "free thou." 
 
 The feminine is distinguished from the masculine in the 
 singular by an addition of the vowel i, and the plural takes 
 u for the masculine and a for the feminine ; but in these cases 
 the second vowel (that placed after the second radical) is generally 
 dropped: iulma, "place ye." 
 
 Mutative. — The mutative of the imperative is formed in the 
 same way as the aorist, and as for the mutative of the indica- 
 tive by the doubling of the second radical, from the participle 
 present or the participle continuous. It has only one person 
 (the second), the genders and numbers being distinguished in 
 the same way as for the aorist. The two forms {XaYTiZ and 
 XaYYaZ) might be found with the same stem, but the form 
 preserved generally follows the same formation as the mutative 
 of the indicative. The use of this tense being very rare, the 
 examples are very scarce. 
 
 In the other voices the imperative follows also the indicative 
 in its formation in some cases, but, as a rule, it is derived 
 from a special form, XuYiZ. The vowel following the n of the 
 
VEEB. 51 
 
 Niphal is a, but that following the ^ of the Shaphel is u. In 
 the secondary voices the vowel preceding the t characteristic 
 is always i. 
 
 Pbkmensivb. 
 
 The permensive is formed from the participle present for the 
 Kal, with a few exceptions, and from the participle past-passive 
 for all the other voices. This tense is, properly speaking, nothing 
 else than a participle ; in the third person the subject being 
 generally expressed by a noun or an emphatic pronoun, the 
 participle is used without addition, but the other persons are 
 distinguished by suffixes added to the participle, in the sub- 
 jective or objective case, its final vowel being elided, it is 
 difficult to say why. The suffixes are : 
 
 Sing, ist pers. -aim or -alt; 2nd, -dta. 
 Plur. „ -dni 01 -dim ; ,, 'diunu. 
 
 The genders are not distinguished in the i stand 2nd persons. 
 
 Sing, ist pers. rahsaJc, "I am trusting, or one who trusts," 
 from rahdsu ; 
 taklaJc, "I am protecting, or one who pro- 
 tects," from takdla. 
 2nd masc. ^aknata, " thou art establishing," from iakdnu. 
 3rd masc. ^akiti, " he is establishing or making," from 
 ^ahdnu. 
 Plur. ist pers. haltdnu, "we are living," from baldtu. 
 
 3rd masc. sabtu or sabtuni, " they are taking," from 
 sabdtu. 
 „ fem. sahrd, " they are surrounding," from saJidra. 
 
 In a very few verbs the permensive Kal seems to be formed 
 from the participle past active or passive, more often from the 
 latter, and as a rule this is found with verbs denoting a state. 
 
 3rd pers. lumun or Umun, "he is evil," fem. limnit ; 
 2nd pers. plural, Umniiunu, " you are evil," 
 from lamdnu, " to be evil." 
 
52 ASSYEO-BABYLONIAN GRAMMAR. 
 
 It is seen by the last example that the connecting vowel is 
 harmonised : 
 
 maldlu, " to be like," makes maMil and mu%ul. 
 
 mardsu, "to be ill," makes marus / maris is also found. 
 
 The permensive appears to have been formed originally with 
 all the participles, and could be formed even with nouns and 
 adjectives, like, for instance, Sarraku, " I am king." 
 
 For the other voices : 
 
 JS". 3rd pers. naMmur, "he holds or watches," from Sanidru. 
 S. „ suldul, " he has caused to complete," from haldlu. 
 
 P. ,, dulluhu, "he was troubled," from dalahu. 
 
 K^. ,, (fern.) ^itkunat, "she was situated," from Salcdnu. 
 N^- , , itamgur (for nitamgur), "he obeys," from magdru. 
 S^. ,, Suiashur, "he cause to enclose," from saharu. 
 P^. ,, not found. 
 
 The tertiary voices are not found. 
 
 A precative could be formed from the permensive by prefixing 
 the affix lu to all the forms. 
 
 Subjective Mood. 
 
 All the tenses of the subjective mood are formed from the 
 corresponding ones of the indicative by adding u to the forms 
 of the indicative in the singular : ind. aorist, iSkun; subj. aorist, 
 i^kunu ; and nu or ni in the plural : 
 
 ikpuduni, from kapddu, " to cover." 
 lihrubuni, from karabu, " to collect." 
 
 Dependent Mood. 
 
 All the tenses of the dependent mood are formed in the same 
 way by adding i for the singular, and uni for the plural : 
 
 Indicative, Hkim. Dependent, iSkuni. 
 „ ishatu „ isbatuni. 
 
VERB. 
 
 Objective Mood. 
 
 53 
 
 All the tenses of the objective are formed in the same way 
 by adding a for the singular : 
 
 Indicative, ithun. Objective, Wcuna. 
 „ tasbat. „ tasbafa. 
 
 Enbegbtio Mood. 
 
 This mood includes all the verbal forms having the mimation ; 
 but this mood is rare, it being confounded with the following. 
 From this fact it may be inferred that the doubling of the m 
 only indicates the length of the preceding vowel, or is written 
 only to prevent the weakening and dropping of the character- 
 istic m. 
 
 Paragagio Mood. 
 
 Under this name we class all the forms having the mimation 
 followed by the enclitic ma : 
 
 akhaifumma, from Jcabasu. 
 ittdhlibamma, from N^ of halabu. 
 vpUlunimma, from patdlu. 
 
 WEAK VERBS. 
 
 All the apparent irregularities of the weak verbs, that is, the 
 verbs having an n, the aspirate ', or a vowel as one of their radicals, 
 are easily explained by the phonetic laws of the language. 
 
 For instance, a verb having n for first radical, like naMdku, is 
 to a certain extent treated as Niphal of a supposed ^ahu, and 
 makes its infinitive of the secondary Kal voice in ita^ulcu. When 
 the n comes into contact with the second radical, it is assimilated, 
 and then considered as only indicating the existence of a long 
 vowel ; hence the verb falls into the category of those having 
 a vowel as first radical. 
 
54 ASSYEO-BABYLONIAN GRAMMAE. 
 
 The length of a syllahle being often represented hy the 
 doiihling of the next consonant, we have therefore illik, " he 
 went,'' for llik from dldku. 
 
 In the Niphal the length of the vowel of these verbs is often 
 represented by the n following it, and we have for the infinitive, 
 nanluhu, from dldJcu, nenpusu, from epehi, "to do," &o., and in 
 the aorist 3rd pers. sing, innalik, innepH, &c. 
 
 The verbs having ' or u for first radical are treated as those 
 having a, those having 1 as those having e. The personal pre- 
 fixes remain the same in the first, case, the vowel being only 
 lengthened. In the second case the vocalisation is e or * for all 
 persons. Some verbs having primitively u as first radical pre- 
 serve it, and this u absorbs the vowel of the prefixes ; so the pre- 
 fixes of the persons are the same as for the Pael : ulid, " I bore 
 (a son)," for (aulid), from {u)alddu. 
 
 For the verbs weak of the second radical the vowels of the 
 first and second radicals are generally contracted into one, but 
 sometimes there are attempts to represent them ; in this case 
 we may suppose the presence of an aspirate unwritten : Icain 
 (for Ita'in) by the side of kin, third person permensive of 
 kdnu. 
 
 It is to be noticed that the dififerent kinds of weak verbs are 
 often confounded ; we find attur by the side of atur (for atur), 
 ist. pers. aorist of tdru. 
 
 The verbs weak of the last radical preserve their final vowel 
 in all the tenses, but it is generally the same for the aorist or 
 the mutative : kdbu, " to speak," makes aorist akli, mutative 
 akabhi. 
 
 It is to be noticed that there was always a tendency to 
 strengthen the weak verbs, and the secondary and tertiary voices 
 are, for this reason, no doubt, of a more frequent use with the 
 weak verbs. 
 
 Q0ADRILITARY AND QUINQUILITART StbMS. 
 
 The stems having four or five consonants as radicals are 
 treated as derivative voices, even if the radicals which form 
 the first part of them are not those used to form derivatives. 
 
PARTICLES. 55 
 
 PEEPOSITIONS. 
 
 There are only a few real prepositions : ana, " to, in ; " ina, 
 "in, from, with;" uUu ot: iMu, "from;" adi, "till, to;" ki or 
 Icima, "as." 
 
 The other prepositions are really nouns, and may take the 
 possessive suffixes: itti, "with;" from ittu, "place;" itti-Tta, 
 "with thee" or "thy place;" eli, "over;" arid, "after, he- 
 hind;" pani, "before, in front of;" mahri, "in front of," &c. 
 
 The feminine form of the same nouns which form these 
 prepositions can he used in the same way : arkat, mahrit, &c. 
 
 It is generally the construct case which is used. 
 
 ADVEEBS. 
 
 The adverbs of manner can be derived from all nouns by 
 adding % to the- dependent case: rati, "great," rabi^ ; Itbbi, 
 "heart," libhi^, "cordially;" edi, "one," edis, "alone;" eli, 
 "high," eliS, "above;" hahkabi, "star," kakkabi^, "as a star" 
 or " starlike.'' 
 
 Nouns in any form can be used adverbially (see Syntax). 
 
 Adverbial expressions are formed by prepositions : ina umu 
 annute, " in these days ; " uUu ulla, " from old." 
 
 Adverbs are formed by the enclitic ma : umma, " thus ; " 
 matema, "in times past;" kalama,"oi all sorts ;" ^a«a?n as, 
 "formerly;" kiaam, "thus," &c. 
 
 The negatives are : la, ul (before a verb), a (before a verb or 
 in composition), aumma, "never." 
 
56 ASSYRO-BABYLONIAN GRAMMAR. 
 
 COifJUNCTIONS. 
 
 The principal conjunctions are : m or m, " and," sometimes 
 "or;" Zm, "or;" ma (enclitic), "and, so that, in order that;" 
 ki, "if, -when, as, while;" aMu or ciMSum, "when, as for;" 
 tumma, "if, when ;" ni (enclitic), "so." 
 
 The relative ^a is also used as a conjunction, " that, because," 
 &c., and forms compound conjunctions with the prepositions : 
 adi &, " in so far as ; " adi eli Sa, " till ; " a7-ki Sa, " after that," 
 exactly like " that" in English and que in French. 
 
 Some conjunctions are also formed with nouns followed by 
 ^a: libbu &, "just as." 
 
 SYNTAX. 
 
 NODNS. 
 
 In the earliest inscriptions and documents (age of Hammurabi) 
 the three cases terminate with the mimation ; at the other 
 periods it is only preserved, and not always with feminine or 
 abstract nouns in t and a few others. In the syllabaries and 
 word-lists the mimation is generally given as the regular and 
 proper ending, but in poetry the mimation is also often 
 neglected. 
 
 At a later period the mimation or the emphatic state are used, 
 very much as our article : kaspam or kaspa', " the money (which 
 has been mentioned before)." 
 
 When a noun is followed by another which depends upon it, 
 or by the possessive suffixes, it cannot take the mimation or the 
 emphatic state. 
 
 The subjective case is used when the noun is the subject. 
 The dependent case is used when the word is governed by 
 another noun, i.e., in the genitive or after a preposition. The 
 
SYNTAX OF WORDS. 57 
 
 objective case is used when the word is the object of the verb, 
 i.e., in the accusative. The construct state is used when the 
 word governs another noun, which is then in the dependent 
 case. The possessive suffixes are sometimes added to the con- 
 struct state. 
 
 We find compound words formed with two nouns, one in 
 the construct state and the other in the dependent case : bel 
 eJcU, "landlord," lit. "lord of the field." 
 
 When the genitive is expressed emphatically by means of the 
 relative ?a, the regent takes the case-ending : belu & ekli, " the 
 lord of the field," lit. " the lord who (is) of the field." 
 
 The nouns with possessive suffixes are sometimes put in the 
 construct state even when not followed by a genitive. 
 
 The plural in u appears more often under the form e — He for 
 ilu, "the gods," — or even in i, Hi. 
 
 The construct state of the plural in anu is hardly ever found ; 
 the form in ani is used for all cases. 
 
 The abstract nouns in utu or uttu may be used as singular 
 nouns: ameluUu, "the slave " or "the slaves." Other nouns 
 are taken collectively : ni^u, " men " or " people." 
 
 Certain words are used more often in the plural : me for mue, 
 *' water," by the side of mu. 
 
 When two nouns are placed in apposition, the second is 
 treated as an adjective, and agrees with the first in number, 
 gender, and case. 
 
 With numbers, names of measures, weights, &c., are put in 
 the dependent case of the singular : xv gum, " fifteen tons," 
 lit. " fifteen of ton." 
 
 When a noun is used as explicative complement, explaining 
 or emphasising the meaning of the verb, it may be in the 
 subjective, objective, or dependent case, or in the construct . 
 state. 
 
 Except in a very few cases, and especially in poetry, the 
 adjective follows the noun it qualifies and agrees with it in 
 number, gender, and case. 
 
 When the noun is in the construct state (or in the dependent 
 case), with a possessive suffix, the following adjective is put 
 
58 ASSYRO-BABYLONIAN GRAMMAR. 
 
 in the normal case which the noun would take if it had no 
 suffix. 
 
 The participles may govern the dependent case, but the noun 
 governed by it may also be put in the objective case, and then 
 may precede it. 
 
 In a few cases adjectives are treated in the same way. 
 
 The comparative and superlative of comparation are expressed 
 by the adjective followed by a preposition : rahu ina Hani, 
 "the greatest of the gods," lit. " the great among the gods." 
 
 The superlative of abstraction is sometimes expressed by a 
 reduplicate form of the adjective : dandannu, " very strong." 
 
 The comparative and superlative are also expressed by the 
 adjective followed by the relative : rabu &i Hani, lit. "the great 
 of the gods." 
 
 KUMEEALS. 
 
 The cardinal numerals are treated as abstracts, and there- 
 fore placed before the nouns : eUen ameluttu, " one slave " or 
 " one of a slave." 
 
 In practice the masculine forms of the cardinals from 3 to 
 10 were used with feminine nouns, and the feminine forms 
 with the masculine nouns. Sometimes the numeral agrees with 
 the noun and follows, it being treated as an adjective. These 
 rules are not, however, always adhered to. 
 
 The cardinals treated as abstracts may be followed by the 
 noun in the dependent case of the singular. 
 
 The ordinals are real adjectives and follow the same rules. 
 
 In the Babylonian contracts the month is given first, and 
 is followed by the day, then the year and the name of the 
 king, thus : arhu adaru, umu iaUU, ^attu eiru, Nabu-nd id tar 
 Babili, " month Adar, day third, year tenth, Nabonidus, king 
 of Babylon." 
 
 Sometimes the cardinals are, as in French, used for the ordi- 
 nals, and in this case generally placed before the nouns. 
 
SYNTAX OF WOEDS. 59 
 
 PEOJS-OUXS. 
 
 The personal pronouns in the nominative, when used empha- 
 tically, are placed before the verb, and generally at the beginning 
 'Of the sentence. 
 
 When the verb substantive is understood, the pronoun is 
 generally thrown to the end of the sentence, taking thus the 
 place of the verb. 
 
 The use of the two oblique flexion al cases are regularly 
 observed only in the documents of the time of Hammurabi 
 and in a few old poetic texts. In the others, the pronominal 
 suffixes are used in the same form with the verb to express 
 the dative and accusative: iddin-hmuti, "he gave them or to 
 them." 
 
 At all periods the flexional dative in dU (dSu, aMa) is used 
 also to express the ablative : Itdii, " to, by, or from thee ; " and 
 after a preposition: ana hdU, "to thee." 
 
 After a preposition the accusative is also used : ana idti, 
 "to me.'' 
 
 The dative is sometimes used to express the genitive case. 
 
 The plural of the second person is used in the nominative, 
 attunu, as dative and accusative, and even as a suffix attached 
 to the verb. 
 
 The accusative is used for the vocative case. 
 
 The pronouns with the suffix of the accusative {iati, Icdti, &c.) 
 are used as subject of a secondary clause of a complex sentence 
 (see Syntax of Sehtexces). 
 
 The pronominal possessive adjectives may be used indepen- 
 dently or as the possessive suffixes after a noun. In the first 
 case they precede the noun, which is considered as being in 
 apposition : iau abil liUbia, " the son oi my heart," lit. " mine, 
 the son of my heart." 
 
 Sometimes these possessive adjectives imply the verb sub- 
 stantive ; they are then thrown to the end of the sentence, and 
 may take the enclitic ma or a lengthening m vocalised by har- 
 mony : kummu, "it is thine " or "they are thine." 
 
 In the same way the personal pronouns can take the enclitic 
 
6o ASSYRO-BABYLONIAN GRAMMAE. 
 
 ma with the same force : attama, " thou art " or " it is thou ; " 
 iatima, " it is I." 
 
 The pronominal suffixes are used to denote the direct or 
 indirect object: iddin-^u is "he gave him" or "he gave to 
 him." When the two objects are expressed, or when the writer 
 wishes to avoid ambiguity, the pronominal suffix representing 
 the indirect object is preceded by a preposition or another turn 
 is used : ina hati-ht, iddin, "he gave into his hand." 
 
 The suffixes are added sometimes emphatically to the verbs 
 when the words they represent are already expressed, even if the 
 object is an emphatic pronoun with the flexion : kd^a lukbi-lca, 
 " may I relate to thee." 
 
 The possessive suffix of the third person masculine, Su, is used 
 fpr the feminine in the dialect of Babylon. 
 
 The third person pronoun is used with an indeterminate 
 meaning representing "some one" or "some of them" under- 
 stood. 
 
 Eblativb Pronoun. 
 
 There is only one relative pronoun, lo, for all genders, 
 numbers, and cases. The oblique cases and other relations are 
 expressed by means of the possessive suffix added to the 
 regent : ^a dbi4u, " whose father," lit. " who his father ; " ki 
 ina eli-^u, "over whom," lit. "who over him." In the same 
 way abU4u ^a, " son of," lit. " son of he who (is)." 
 
 Sometimes the relative is followed by the noun or expression 
 it represents placed in apposition : & Gambuzla ahu-^u, " whose 
 (Cambyses') brother," lit. "who, Cambyses his brother." 
 
 Demonstratives. 
 
 The pronouns of the third person are used as demonstratives : 
 amelu ?m, " this man ; " a^^atu .?i, " this woman ; " they always 
 follow the noun, but are not to be considered as suffixes : ^elalti 
 SuTM, " these three." 
 
 The accusative form has come to be used for the nominative : 
 bitu Suati, "this house.'' 
 
 The same anomalies are noticed for the other demonstratives. 
 
SYNTAX OF WORDS. 6i 
 
 VEEBS. 
 
 The third person masculine ia the singular is generally used 
 for the feminine, and the feminine for the masculine in the 
 plural. In the other cases the verb agrees with its subject in 
 number and gender. 
 
 When two subjects are united by the copula u, the verb 
 agrees with the first. 
 
 The verb is put in the plural with a singular subject when it 
 is considered as a collective noun. 
 
 Verbs can have only one direct object ; when there are two, 
 the second must be considered either as a noun in apposition or 
 as the name of manner explaining the action of the verb ; in the 
 latter case it is really a noun or expression taken adverbially. 
 Sometimes this explanatory noun is preceded by ana or Icima, 
 or takes the adverbial form in it When there is no preposi- 
 tion before it, it may be in the construct state or have the 
 casal form, according to the view taken by the writer. When 
 in the construct state it is considered as forming with the 
 verb a kind of compound ; and when in the casal form, it is 
 either a noun of manner, an adverbial expression, or a noun in 
 apposition. 
 
 Neuter verbs take sometimes the indirect object as direct 
 object without preposition. 
 
 The third person plural is used as indeterminate as in 
 English : ikhu, " they say,'' " it is said." 
 
 The various moods are irregularly used, because the notion of 
 their value seems to have been lost at an early date, as that of 
 the case-endings in the noun ; but the following principles may 
 be given, though in practice they were often ignored. 
 
 The infinitive is not really a mood; its forms, infinitive 
 simple and the four participles, are treated as nouns. 
 
 The permensive carries an idea of continuity ; it expresses an 
 action the effect of which is still lasting when the narration 
 takes place; for this reason it has neither past, present, nor 
 future. 
 
62 ASSYRO-BABYLONIAN GRAMMAR 
 
 The indicative is the narrative mood far excellence. It is 
 the one by which any statement is made. The shortness of its 
 forms give a great vivacity to the narration. 
 
 The subjective is more solemn, and might be called the oratory 
 mood ; it ought to be used for principal sentence only. 
 
 The dependent mood is used in secondary relative sentences, 
 or in secondary sentences beginning by a conjunction like 7a, 
 hjmvma, &c. 
 
 The objective mood is used generally in secondary sentences 
 brought on as a consequence of the principal ; but is often con- 
 founded with the previous mood ; from its consequential value 
 it sometimes implies the idea of futurity. 
 
 The energetic mood is composed of archaic forms. It gives 
 strength to the meaning of the verb, but does not alter it ; it 
 has forms corresponding to the three previous moods, um corre- 
 sponding to u, im to i, and am to a, and ought to be used accord- 
 ingly, but these energetic forms are rare in the modern texts, and 
 their use not always consistent. 
 
 The paragogic is merely the previous mood with the addition 
 of the enclitic ma, which generally implies a consequence, and 
 ought to be followed by the objective mood ; in many cases the 
 enclitic, however, simply represents the copula, and the double 
 m is the consequence of the accent being thrown back on the 
 previous vowel. 
 
 The aorist expresses the past ; it is used sometimes, but rarely, 
 for the permensive with an idea of continuity. 
 
 The mutative is used for the present and future ; and also 
 in a very few cases, when the narration represents the action as 
 actually taking place, for the past ; but in that case it is only a 
 figure of speech. 
 
 The imperative is used for command, but the precative ex- 
 presses only a wish. 
 
 The infinitive simple can be used for the future : la epHu, 
 " he will not do.'' 
 
 The Kal gives the primary meaning of the stem or theme. 
 
 The Niphal has generally the passive meanings of the Kal ; 
 in other cases a developed meaning. 
 
SYNTAX OF WORDS. 63 
 
 ihhibvii {ioT inldbus), "he was conquered," from kabasu, "to 
 trample upon;" i^^aknu, "they were made or placed," from 
 ialcdnu, naptdru, "to defend," ivom'patdru, "to free." 
 
 . The Shaphel is mostly used as a factitive or causative : 
 v^e^Mn, " he caused to be placed or he established, and he 
 caused to be made or he achieved," from takdiiu ; u^aklil, " he 
 caused to be completed, i.e., he finished," from kaldlu, "to be 
 complete." 
 
 The Pael gives to the stem an intensitive meaning, or simply 
 emphasises the primary meaning, and is rarely used as a causative. 
 On account of its intensitive character it has often the same 
 meaning as the Kal, and its tenses are used for those of the 
 Kal, especially with the weak verbs : u^ahkan, " I placed," from 
 Sakdnu ; luhallik, " may he destroy," from haldku, " to injure ; " 
 uparrir, " I dispersed," from pardru, " to hinder ; " uhubhu, " he 
 says," from kabu. 
 
 The secondary voices have often a reflective force, and indi- 
 cate that the action is done by the subject for himself ; it may 
 also be used as passive. Sometimes their force is simply in- 
 tensitive : iktanak, " he places his own seal," " he seals for 
 himself," from ^akdnu, " to seal ; " iktdbi, " he declared or 
 spoke for himself," from kabu, " to speak ; " uUaksiru, " they 
 assembled," lit. "they caused to come together for themselves," 
 from the Shaphel of kasdru, "to collect;" uptehir, "he 
 gathered," from paharu, " to collect ; " iptalhu, " they shall 
 fear," from paldhu, " to fear ; " iktcdik, " he was injured," from 
 haldku, " to injure." 
 
 The tertiary voices have a frequentative meaning or express 
 energy : iktanahhatu, " they go about plundering," from habatu ; 
 ittanallaka, " he goes repeatedly," N^ from aldku, " to go." 
 
 The negative la is put before nouns, adjectives, and verbs. 
 In composition it corresponds to the English "un : " la-magiri, 
 " un-obedient ; " and forms negative nouns, la mami, the "no- 
 waters," that is " the want of water." 
 
 The negative ul is placed exclusively before verbs, and the 
 negative aa forms prohibitives with the verbal forms. 
 
 Lu prefixed to the verb gives it a past meaning. This mode 
 
64 ASSYEO-BABYLONIAN GKAMMAR. 
 
 of expressing the past is only used in a few inscriptions of the 
 time of Hammurabi and of the second Ninevite Empire. 
 
 The postfix -ni is placed after the verb (and after the pronomi- 
 nal suffix when there is one) to give to the sentence a dependent 
 character. It is even to be found at the end of a sentence 
 beginning with Id, " if." 
 
 The copula u is used to join two nouns or two sentences 
 when they are treated as being in apposition. 
 
 The enclitic ma is only placed after the verb, and implies a 
 consequence ; sometimes, however, it has only the force of the 
 simple copula. 
 
 The conjunction Zm is distributive : lu , . lu . . lu . . "either 
 . . . or . . . or." It has also the force of "whether" 
 repeated. 
 
 SYNTAX OF SENTENCES, 
 
 The disposition of the words in the sentence is freer that in 
 Akkadian ; the formula would therefore vary. In the simple 
 period, or in the principal sentence of a complex period, the 
 same order as in Akkadian is often observed, that is, the formula 
 dt-dp-da-r-s-o-i-c-v. This order is generally followed in the 
 bilingual texts, the object of the scribe having apparently been 
 to give in most cases the nearest translation possible. 
 
 The direct object is often placed before the subject. This 
 happens especially in private contracts, and when greater im- 
 portance is given to the object. 
 
 The qualificative, either an adjective, a noun in apposition, or 
 a noun with a preposition, follows the word it qualifies. The 
 exceptions to this rule are exceedingly rare. 
 
 The indirect object very seldom precedes the subject, and 
 follows sometimes the direct object. It is placed generally 
 before the verb, but also may foUow it. 
 
COMPLEX PERIOD. 65 
 
 The complement explaining the action of the verb is always 
 placed near it. 
 
 When the direct object or the indirect object is represented 
 by a pronominal suffix, it naturally always follows the verb. 
 When it is a flexional pronoun (m&', idti, &c.), it often also 
 follows the verb. 
 
 The reason of the action is generally placed before the subject 
 and the objects, rarely between the subject and the objects, 
 but often also after the verb. 
 
 The determinative of time, place, and state is placed at the 
 beginning of the sentence. 
 
 In a secondary sentence, especially when introduced by the 
 enclitic ma suffixed to the verb of the principal sentence, the 
 order of the words is reversed : the verb stands first and is 
 followed by the objects ; sometimes even the subject is placed 
 after the verb ; the most common order is v-s-o-i. 
 
 Complex Peeiod. 
 
 It has been observed that the moods correspond to the diffe- 
 rent forms of the nouns ; these nominal forms were no doubt 
 the origin of the characteristics of the moods, and analogy 
 guided the writer in the use of them ; but as the casal endings 
 of the noun were so weakened as to acquire the obscure sound 
 of a mute e and were then confounded, so were the character- 
 istics of the moods. The moods having become in this way 
 undistinguishable to the ear, it threw great confusion in their 
 use, and that is the reason why the forms of the subjective are 
 found when we ought to expect the dependent or the objective, 
 those of the indicative irregularly employed for the others, &c. 
 
 Eegularly the indicative answers to the construct state and 
 ought to be used : only when the principal sentence in which it 
 is employed requires a secondary consequential sentence, having 
 its verb in the dependent mood. 
 
 The subjective was at first reserved for the principal sentence 
 when it was complete in itself and did not necessarily require 
 other complementary sentences. In the usage, however, it 
 
 E 
 
66 ASSYRO-BABYLONIAN GRAMMAK. 
 
 acquired a mere idea of importance, and secondary sentences 
 were put in this mood to give them preeminence in the 
 narration. 
 
 When a secondary sentence is introduced by a conjunction, 
 it ought regularly to have its verb in the dependent mood ; it 
 is also the mood to use in a secondary sentence in apposition, 
 if this sentence is considered as the immediate consequence 
 of the principal sentence. A dependent sentence (that is, one 
 having the verb in the dependent mood) precedes the principal. 
 
 The objective sentence generally follows the principal; it 
 may be put in simple apposition or connected by the enclitic ma 
 added to the verb of the preceding principal 
 
 "When an objective sentence is introduced by a flexional 
 pronoun in the accusative, but used as subject {idti, kdti, &c.), it 
 precedes the principal or governing sentence. 
 
 When the principal sentence follows a secondary sentence 
 introduced by a conjunction, its verb is often put in the 
 objective mood, because the action is considered as the natural 
 sequence of the statement made in the conjunctive sentence. 
 
 The conjunctive sentences often represent the determinatives 
 of time, place or state, and must therefore precede the principal. 
 
 When two sentences are placed in apposition, the verb of 
 the first one is often put in the indicative (mood of the construct), 
 being to a certain extent considered as the complement ex- 
 plaining the second, just as the complement explaining the 
 action of the verb is put in the construct state. In that case 
 the second sentence ought to have its verb in the dependent 
 mood. 
 
 When a sentence is followed by the enclitic -ni suffixed to 
 its verb, it indicates generally a statement, the consequence of 
 which is expressed by the principal sentence. 
 
 (t) Sumd abliir-^u^a Sumii^iddina abil Gahal, (2) ultu Elamti 
 hi ihlihu, (3) adi Tahha' ittalJca, (4) TaJiha' hata-^u Ttl asbata, 
 (4) uUeUra^^u ; (6) marus ; (7) adi zime-Su mala isabhatu, 
 (8) ana Sarri enia a^apparoM^u. 
 
 "(As to) Suma son of Sumuiddina son of Gahal, he had fled 
 then from Elam, he went to the Tahha, I took him {lit. his 
 
COMPLEX PERIOD. 67 
 
 hands) from the Tahha, I brought him back ; he is ill, when he 
 will have resumed his complete health, I will send him to the 
 king, my lord." 
 
 This example taken from a letter-tablet is very important. 
 There are two periods : in the first one extending to (6) all 
 the verbs are in the aorist, the first sentence (i) having no verb 
 plays the part of a determinative, the second is the principal, 
 and its verb is in the subjective mood, the three others (3, 4 
 and s) are considered as sequence, and their verbs are therefore in 
 the objective mood. The second period begins by a permensive 
 (6) playing the part of a determinative, in the two other 
 sentences the verbs are in the future, the first (7) is the prin- 
 cipal and therefore in the subjective mood, and the second, 
 (8) the sequence, in the objective mood. 
 
 (i) Udte' manjMtu imhar-^u-ma edi^u innabit ana Nabati 
 (2) TJdte ahlu Hazailu ahi ahi & Udte' dblu Pir-Rammanu 
 raman-iu i^kunu ana tarruti Arabi ; (3) AMur temu uSanni-ma 
 (4) ilUka adi mdhna. 
 
 "Misfortune happened to him, Uate, and he fled to Nabata, 
 Uate, son of Hazailu brother of the father of Uate son of 
 Pirrammanu, made himself king of Arabia, Assur gave an order 
 so that he came to my presence." 
 
 The first part of the period is the determinative of time and 
 state and is in the indicative, or construct mood, the second is 
 the statement and is in the subjective mood, the third is the 
 consequence and is in the dependent mood, and the fourth, 
 the final result, is in the objective mood. 
 
 Ilani rahuti idti AMurhanapla palah-^un ana sapah Elamti 
 uma'iru. "The great gods deputed me their worshipper to 
 destroy Elam." 
 
 The following are two examples of very quick narrations; 
 the subject is often thrown after the verb, and the verbs are 
 most of them in the indicative or construct mood : 
 
 (i) Innindu-ma Warrant Idlallan ippu^u tahaza (2) ina hiri-^unu 
 innapih i^atu, (3) ina turhuti-^unu ndduru pan SamSi, (4) 
 oMamlatu issanunda (5) isar mehu (6) ina mihie tahazi -^unu iila 
 bel narlcdbti ippallasa Mnd Sa itti-^u. (i) "The kings of the 
 
68 ASSYRO-BABYLONIAN GRAMMAR. 
 
 two regions gathered so that they made battle, (2) among them 
 fire blazed, (3) the face of the sun was darkened (subj. mood) 
 ■with their clouds of dust, (4) a hurricane had gathered (ob- 
 jective mood, may be translated : having gathered), (5) the 
 storm broke (construct mood, answers to the determinative of 
 state), (6) in the storni of their battle the warrior charioteer 
 does not recognise (subj. mood) his companion (lit. the other 
 who is with him). 
 
 (i) Itm pi IMar u Rammani, Hani heE tahazi limutte, (2) 
 lu ana ^arri Elamti (3) iteziz mati-^u, (4) u ^arru Nabuku- 
 durrizur itta^iz (5) ina liti isabat Elamti iUalal ^aga%a. (i) 
 " By the command of Istar and Eimmon, the gods lords of battle 
 evil rose (2) be it with the King of Elam ! (3) his country 
 was taken (4) and Nabuchadnezzar joined (him) (5) with 
 might he took Elam (and) confiscated his goods." 
 
 (i) lati Sin-ahe-irib ^ar AMuri epi^ Upri ^uatu (2) ki tern 
 Hani ina uznvx ipU-{^)ma kdbatti uilam-(4)ma teniUt Kaldi 
 & ana niria la kitn-uMu assu]iam-{^)ma umUkhi u^aMU4unuti- 
 (6)ma ilbinu Ubitti (7) apikupe &s kirib Kaldi aMid-{8)ma 
 appari-iun Samhufi ina bahulati nakirl ki^itti katia vMaldida 
 ana epiS Epri-Sa. 
 
 (2) "When the command of the gods came to my ears (3) so 
 that it was important (i) that myself, Sennacherib, king of 
 Assyria, should do this work, (4) therefore I removed the people 
 of Chaldea who (were) unsubmissive to my yoke, (5) so I 
 cause them to carry burdens (6) so that they should make 
 bricks ; (7) I took baskets from the middle of Chaldea (8) so 
 I caused the best wicker to be brought by the rebel people 
 captured by my hands to do its work." 
 
 In this example each sentence is introduced by ma as being 
 a consequence of the preceding one. In sentence (7) the verb 
 is in the indicative or construct mood, because it is a kind of 
 determinative, " I took or having taken," and the verb of the 
 next sentence is in the objective mood as being the result of 
 the preceding one. 
 
 The following are two examples from a private contract, a 
 statement before court : 
 
COMPLEX PERIOD. 69 
 
 Binaddunatan ana a^^utu ir^an-ni-ma Sal^u mana Icatpi nu- 
 dunnd ilkie-ma edit martu ulid-iu. " Binaddunatan had me 
 to wifehood, so he took three mana of silver, my dowry, and so 
 I bore him one daughter." 
 
 iatu u mutia nadanu u mdhari ina eli rmdunniea neput-ma 
 Samna Jcane hitu eMu irsitim ahuld galld kirib Barsip ana 
 UUt ivM^anu mana ka^pi, adi iane ma^u mana hoMpi & ultu 
 Iddin-Marduk ana ni^hu nit^am-mM ina Mm biti ^uatu niddidin- 
 ma itti ahame^ nimhur ina %atti arbati Nabundid ^ar Babili. 
 
 " Myself and my husband we made selling and buying with 
 the money of my dowry, namely : eight canes (ground and) 
 a ruined house, territory of a large estate within Borsippa, for 
 nine and two-thirds mana of silver, we made up the amount 
 with two (and) a half mana of silver, which (was borrowed) 
 from Iddinmarduk, (added) to the former, so we gave (it) as 
 price of that house, in this way we traded together in the 
 fourth year of Nabonidus, king of Babylon." 
 
 The last sentence from itti ahame^ really governs all the 
 preceding ones ; it might be translated, " we traded in this 
 way : that I and my husband," &c. 
 
VANNIC GRAMMAR. 
 
 PHOISrETIC. 
 
 The Vannic alphabet is the same as in Akkadian and Assyrian; 
 this is naturally the result of its syllabary being borrowed from 
 Nineveh, and may only be but apparent, as the value of the 
 characters have been found by comparing the signs given in the 
 inscriptions with those of the Assyrian syllabary. 
 
 In deciphering many things are to be taken into account. 
 
 A great use is made of determinative perfixes, or, as better 
 termed, of ideograms followed either by the phonetic comple- 
 ment or by the phonetic determinative {i.e., the whole word 
 written phonetically). In some cases, however, the ideograms 
 are used, as in Akkadian and Assyrian, .without any comple- 
 ment or determinative. There was a tendency to the alpha- 
 betism,- so that the vowel is often repeated after a syllabic 
 sign ending in a vowel, even when it is not long ; and the 
 final vowel of a syllable is not to be read in some cases, 
 being given merely, because the characters being syllabic, the 
 consonant could not be written alone. All the inscriptions 
 do not belong to the same region, and some show dialectical 
 variations. One inscription gives hi for ni, in another the 
 guttural h is dropped : aldi^ for haldi^. The documents are still 
 too scanty to enable us to give the rules of such changes. 
 
 The vowels are often assimilated. As in Assyrian, when two 
 vowels are not assimilated it may be due to the presence of an 
 aspirate, which is not written. The two letters I and e are 
 confounded. The consonants b and p are often confounded, bo 
 
NOUNS AND ADJECTIVES. 71 
 
 are also s and 2, d and t. A dental between two vowels is 
 easily dropped. 
 
 The vowels are lengthened generally by the addition of the 
 letter e, ae = a, ie = l, &c., but often aa, ai are found for ae 
 (a), ii for ie (i), &c. The vowel u seems to have had sometimes 
 the value of the semi-vowel v (or w) : ua, ue, ui, uu, might 
 represent va, ve, vi, vu. 
 
 FOKMS OF THE WOEDS. 
 Nouns and Adjectives. 
 
 In Vannic there are two genders and two numbers, but 
 neither gender nor number are distinguished clearly, from the 
 grammatical point of view. 
 
 All nouns are appellatives, and may be either substantives or 
 adjectives. The relations of the nouns to the other words and 
 the special extension of the meaning attached to them are 
 indicated by means of suffixes. 
 
 The nominal stems are always terminated by a vowel, a, i, or 
 u, but are never found without suffix. The simplest form is 
 that of the noun in the dependent case answering, in a certain 
 measure, to the construct state of the Semitic tongue. This 
 case is formed by lengthening the final vowel of the stem : esi, 
 "law," dependent case, esi, giMiiri, "mighty," d.c, gvM^rie. 
 
 The suffixes may be followed by the enclitic demonstrative 
 pronoun ni. The most important are the following : 
 
 (i.) ?, which is added to the vowel of the noun : its fuU form 
 is U, and formed primitively names of state, but was to a great 
 extent used to denote the nominative : Haldi-S, Menuort 
 
 (2.) di, forms names of agents, and is used also to form a kind 
 of locative : huta-di, "on leaving." 
 
 (3.) Z» (formerly read <ia), forms local names : from^i, "name," 
 pi-li, "a memorial," lit. " the place of the name." 
 
 (4.) ki, forms instrumentals, and for this reason is often to 
 be translated by an adverb : ali-M, " partly ; " kau-ki, " with 
 
72 VANNIC GRAMMAR. 
 
 ■weapons." In some cases it is written kid, perhaps for Td-di, 
 the second suffix being superposed. 
 
 (5.) U, indicates that the action is made after the completion 
 of the object taking this suffix, and has been called the per- 
 fective case : meivrli, " after the libations ; " zatua-U, " after the 
 gate (was built)." 
 
 (6.) hi, forms patronymics, and also names of people, belonging 
 to or followers of : I^puni-hi-ni, " the son of Ispunis ; " Eriorhi, 
 "son of Erias;" tu-hi, "captive." 
 
 (7.) na, indicates the country: Hcddi^-na, "of the land of 
 Haldis;" Hati-na, "of the land of the Hati." 
 
 (8.) a, forms names of population or collectives : Haldi-a, 
 " the Haldians ; " haradi-a, " armies." 
 
 (9.) ka, forms gentilic names, but having a more limited 
 extension, and forms also serial names : AgiBi-ka, " the family 
 of Agistis." 
 
 ■ (10.) ue (to be read perhaps ve), marks the possession: 
 hulorite, "having kings." 
 
 (11.) si, marks relation: asi-si, "that which belongs to the 
 cavalry." 
 
 There are few other suffixes, very rare in the inscriptions : 
 zi, in armu-zi, " altar," perhaps another form of S or si ; ri, an 
 individualising suffix, more used in the cognate tongues ; hi, 
 found in ati-hi, "thousands," and niri-hi, "dead," &c. 
 
 The suffixes can be superposed, that is, a derived word 
 formed by means of a suffix may take the suffixes as a primi- 
 tive word, so we have for instance : huradi, " soldier " (this 
 word being no doubt itself a derivation of hura) ; hura-di-a, 
 " soldiery ; " hura-di-ni-li, " camp ; " hura-di-ni-li-tie, " belonging 
 to the camp," &c. 
 
PRONOUNS. 73 
 
 NUMERALS. 
 
 The numerals are always •written ideographically. The only 
 one found written phonetically is atihi, " thousands or myriads." 
 
 It is only to be noticed that the word for " thousand " seems 
 to have been repeated after each number. For instance, 23,000 
 was written xxmiiim, which implies that it was literally " twenty 
 thousands (and) three thousands." 
 
 Of the ordinals we know %uUni, " first," tarani, " second," and 
 Mtini, " third," evidently formed with the enclitic ni. 
 
 DEMONSTRATIVES. 
 
 The demonstrative mostly used is i, followed by the enclitic 
 ni : ini, " this ; " ea may only be another form of the same. 
 
 iM, "this" and uU, "that," are not so much used. Both 
 can take the enclitic ni and all the suffixes, like the nouns, and 
 be used as abstracts : iki-ni-ue-di-a, " the people belonging to 
 them." Properly uli means "another.'' 
 
 IwAe, "these," which appears to be invariable, seems to be 
 formed from %a, the correlative of ini ; %a receives itself the 
 suffixes and the enclitic: ta^ue, "belonging there." 
 
 The distributive is manu : asut manu^, " every mouth ; " ali 
 nianu, "some and each." With the enclitic u is assimilated 
 mani-ni. 
 
 For "some , . . some," ali-ki . . . ali-ki, primitively 
 " partly . . . partly," is used. 
 
 PERSONAL PRONOUNS. 
 
 Of the personal pronouns only the first and third persons 
 have been found ; they may take the suffixes as a noun. 
 Nominative me-l, " the his " or " he " (himself). 
 
74 VANNIC GRAMMAB. 
 
 Genitive or dependent ine-i, "of him." 
 
 With the local suffix, me-li or mei-ali, " the place of it." 
 
 All the forms with suffixes may also take the enclitic ni: 
 mei-^i-ni, " his, its," " what is his or its." 
 
 The possessive is found under the three forms, mu, ma, and 
 me ; the first is the one mostly used. 
 
 Of the first person only the nominative is found : ie%, "I." 
 
 YEEB. 
 
 The Vannic verb has only one tense, the aorist, and only the 
 first and third persons of it are known. The same form stands 
 for the singular and plural of the third person, and the genders 
 are not distinguished. 
 
 The infinitive, or nomen verhi, always ends in u : 
 
 kugu, " to write," or " the action of writing." 
 
 From it is formed the aorist with the suffix hi for the first 
 person, and ni for the third sing, or plural : 
 
 ktigu-hi, " I wrote." 
 
 kugu-ni, " he, she," or " they wrote." 
 
 A kind of gerund is formed by the suffix U or lie, which 
 takes sometimes a as vowel of union, and is used for present or 
 future : 
 
 du, "to destroy;" du-U or du-lie, "he destroys." 
 
 This form was no doubt used for all persons. 
 The present of the participle is formed from the infinitive by 
 the suffix e^ in the nominative : 
 
 Siu, "to remove," pres. part. H-e^{for Hu-ef), "removing." 
 
VERB AND ADVERB. 75 
 
 Tho past participle is formed by the suffix al in the nomin- 
 ative : 
 
 zadua%, " being built." 
 
 This, as the infinitive and the present participle, can take 
 the suffixes : 
 
 zadua-li, " after being built." 
 
 A few examples are found of a form composed with the suffix 
 me ; these are imperatives or precatives : 
 
 ashu-me, " may they occupy ; " 
 
 A derivative causative voice is formed with the suffix %u 
 or ia, which is then treated as an ordinary verb : 
 
 ha-^ti-hi, " I cause to possess.'' 
 harliar4arbi, "I dause to be excavated." 
 
 The form aptini, " it is called," or " they call it," by the side 
 of tini, " he called," has suggested the belief in a passive voice 
 which would be formed by a prefix op; but a formation by 
 prefix is too much opposed to the genius of the language, it is 
 preferable to see another stem, perhaps a compound one, in 
 aptini. 
 
 ADVERBS. 
 
 As in Akkadian the adverbs are formed by means of the 
 postpositions or suffixes added to nouns, demonstratives or pro- 
 nouns, in fact many of the words having the suffixes may be 
 considered as adverbs and translated by our adverbs of time, 
 place or state : 
 
 ali-ki, " partly ; " Si-li, " after dawn ; " kuta-di, " at ihe 
 departure ; " ini-li, " here ; '' manu-li, "in each place ; " 
 uUordi, " in approaching ; " 'd-li, " for sacrifice." 
 
76 VANNIC GRAMMAR. 
 
 As in Assyro-Babylonian nouns and pronouns in the depen- 
 dent and nominative cases, and even witli the enclitic pronoun 
 ni, are used as adverbs : 
 
 gunuM; "as a spoil;" ainei, "with dust;" iu, "thus;" 
 aie, " in the land ; " &W-I, " anew ; " ardi-ni, " during 
 the day = publicly ; " nara-ni, " with fire." 
 
 pari or parie requires a special mention ; primitively an adjec- 
 tive, it has come to mean " out of," and to be used apparently 
 as a preposition. 
 
 The negative has not been found yet. 
 
 CONJUNCTIONS. 
 There are two words to express the copula, ui and cdi. 
 
 SYNTAX. 
 
 As already stated, the gender was. not distinguished by any 
 particular forms in the nouns or adjectives; in the cases of proper 
 names, as in Assyrian, those of men were preceded by the 
 straight wedge, and those of women by the sign of female. The 
 plural was not distinguished either, but merely indicated by the 
 ideographic sign of " many " (which was probably not pro- 
 nounced) placed after the word. 
 
 The demonstrative enclitic ni could be added to any nominal 
 form, and has in a certain degree the same force as our article 
 "the:" ina4, "a city" (with ?, the mark of the nominative), 
 inai-ni, "the city." 
 
 Sometimes this enclitic seems to be used to indicate the 
 accusative. 
 
 When a noun is used in the plural, an attempt has been made 
 
SFNTAX. 77 
 
 in some cases to distinguish it from the singular by repeating 
 the enclitic twice, as in Akkadian with the demonstrative we ; 
 guiu-ni-ni, " the slaves." 
 
 "When a noun is used in the dependent case, it is considered 
 as forming a kind of compound word with the word from which 
 it depends and precedes it : Menuai pili, " the memorial of 
 Menuas." 
 
 The suffix S, when it marks the nominative, is only added to 
 the word in the singular. Por the plural the word is placed in 
 the dependent case, or takes the enclitic simple or repeated. 
 
 The words used as adjectives are either considered as form- 
 ing, with the noun they qualify, a kind of compound, or as 
 being in apposition. In the first case they are placed in the 
 dependent case and precede the qualified noun : gi^Surie kuru- 
 ni, " the mighty giver." In the other case they follow the 
 qualified noun, and take the same suffix: asuS manu-S, "each 
 mouth;" Haldie eurie, "Haldis, the lord." 
 
 B. The word with a suffix is considered often as a derivative 
 word, and being then treated as a new stem, can take the ^ of 
 the nominative and the other suffixes : I^pUini-S Saridurie-hi- 
 ni4, " Ispunis Sariduris's son;" Haldini-ni (dsuiii-wi, "the 
 numerous Haldians." 
 
 Sometimes the word used adjectively is considered as giving 
 an explication in parenthesis, so to say; then it takes the 
 enclitic ni, and may or may not have the same suffixes as the 
 noun explained by it : M.enua4, Rpuni-hi-ni, " Menuas, the 
 Ispunian (son of Ispunis)." Hdldi-a i^ti-ni, "the people of 
 Haldi, these." 
 
 The regimen of an adjective generally follows it : Karuni 
 ebanie, " who has given the countries." 
 
 The same rule applies to participles and all words used 
 adjectively. 
 
 When the participle has a suffix, its regimen takes the same 
 suffix : ^ia-di Eriahinie ebanie-di, " in despoiling the Erianians' 
 land ; " u^ta-di Urme-di, " in approaching Urme." 
 
 The numerals stand before the nouns : ill a^ida, " 3 
 palaces;" TjX tumeni, "60 villages." 
 
78 VANNIC GRAMMAR. 
 
 The demonstratives are considered as adjectives, and follow 
 the same rules, but they may be used abstractly. 
 
 The personal pronoun is really an abstract noun, and follows 
 the same rules. 
 
 The possessive pronoun follows the noun possessed : ikukani 
 mu, "his goods." 
 
 The relative pronoun is always understood ; the incidental 
 sentence is placed in apposition (see Syntax op Sentenobs). 
 
 The verb distinguishes neither genders nor numbers : ti-ni, 
 " he, she, or they called." 
 
 The gerund used as present, future, and conditional is invari- 
 able. It is also used as precative. 
 
 The infinitives and participles are treated as nouns and 
 adjectives. 
 
 The word ^an, which has the meaning'of a kind of preposition, 
 must be considered as an adjective governing a noun, and its 
 use becomes quite regular according to the Vannic grammar. 
 
 SYNTAX OF SENTENCES. 
 
 The Vannic language has been strongly influenced by 
 Assyrian, and the order of the words in the sentence is nearly 
 the same as in the latter. 
 
 The verb is placed at the end of the sentence, the object 
 follows the subject, the name of manner, that is the comple- 
 ment, precedes immediately the verb ; the reason of the action 
 is placed either at the head of the sentence before the subject, 
 especially if it is a proper name, or follows the verb (as in 
 Assyrian) if it is a name of action. The determinatives of time, 
 place, and state are placed at the head of the sentence. The 
 indirect object is confounded with the reason of the action; 
 when both are expressed in the same sentence, the reason of 
 the action is considered to be a qualificative of the indirect 
 regimen. 
 
SYNTAX OF SENTENCES. 79 
 
 Kudaia-di pane Eia Hu^ani uBa-hi, " on departing from 
 the land of Bias, I conquered the land of Husas.'' 
 
 Haldie halt xviipahini, " to (or for) Haldis for sacrifice seven- 
 teen oxen (I gave)." 
 
 ffaldie eurie ini pulusie Menua^ huguni, "to Hadis the 
 lord Menuas wrote this inscription." 
 
 Complex sentences are rare. As a rule the sentences follow 
 one another without being connected in any grammatical way. 
 Two sentences may be connected by the copula ui, more rarely 
 with the copula cdi. When sentences S follow one another 
 without copula, they may be considered as being in apposition. 
 When several sentences follow one another, especially in the 
 case of an enumeration, the verb given in the first one is not 
 repeated in the others. 
 
 When an incidental sentence marks the time or state, repre- 
 sents in fact the determinative, it is rendered by the infinitive 
 with the suffix li (compare Akkadian). When two sentences are 
 placed in apposition^ the order of the members of the second 
 are often reversed as in Assyrian. After the verb " to say " or 
 " to speak " the incidental sentence is sometimes introduced by 
 the adverb iu, " thus,'' following the verb, but more often this 
 adverb is understood. 
 
 Haldinini Menua^ armani athuli 'iidiMuali, Haldinini 
 alsunini Menua^ ali : alu^ ini armanili tulie, aluS pitulie, altiS 
 ainiei inili dulie, alu^ uli^ tiulie, ie% zadub turinini, ffaldiS 
 mani, pirn mei, arhiurulini mei, inani mei nara aue ululie. 
 " Menuas has restored the destroyed tablets for the Haldians, 
 to the numerous Haldians Menuas says : ' whoever will carry 
 away these tablets, whoever will deface (them), whoever will 
 thrust (them) in the dust, whoever will undo that I did on the 
 rock, may Haldis consign to fire (and) water, {i.e., burn and 
 drown or obliterate), himself, his name, his family, (and) his 
 town.' " 
 
 Haldinini Menua^ alie iu: Manaidi ehania tuhi, amaUuM 
 ihuTcani huradinili. " Menuas spoke to the Haldians thus : 
 'in approaching the land of the Minni, I carried away the 
 population, I plundered the goods, the camp.' " 
 
8o VANNIC GRAMMAR. 
 
 Ebania tuhi, inanie dbiliduhi, iiimoclxx tar^uame aliki 
 zatguhi, aliki ^ehira aguM, clxx nauuh paruhi, lxii dituni, 
 nMCCccxi pahini, vimclx &&. " I carried off the population, 
 I burned the cities, of 3270 soldiers I slew some, I took some 
 alive, I carried off 170 horses, 62 camels, 241 1 oxen, 6140 sheep." 
 
 Nu aliisi ikukaniedini ^aue, nanuli mei ahi piwrtani, 
 haidiani terihinie saridurini tini, saridurini uduUei aue, 
 8ehaulie uni urulie kurienini uiduS suni. " (When) the 
 king (had been) assembling the wealthy people of there, he 
 chose the site for his temple, he named the workmen working 
 for Sariduris, Uidus gratuitously provided those (working) for 
 Sariduris at the monument with water from Sahauli {i.e., the 
 place of Sehaus), and with food from (his) household." 
 
 These few sentences will serve as illustrations of simple and 
 complex periods. The last one is especially interesting, because 
 it is not imitated from those found on the Assyrian monuments. 
 From it it appears that the king Sariduris, wishing to have a 
 temple built, imposed taxes on the wealthy people, and task on 
 the poorer class of the neighbourhood, and that a certain man, 
 Uidus, provided to the working men water, brought from a 
 place called Sehauli, and food from his own household; for 
 this reason the king thought that he deserved to have his name 
 recorded. 
 
 It is to be noticed that in the last sentence of the period, 
 which may be considered as the principal, the subject, Uidu^, is 
 placed immediately before the verb ; perhaps we have here the 
 primitive order of the sentence in Vannic before it was in- 
 fluenced by Assyrian. The expression kurienini suni "pro- 
 vided gratuitously " might be translated " gave as a gift ; " what 
 is meant is that Uidus did not charge anything. 
 
MEDIC GRAMMAR. 
 
 PHONETIC. 
 
 The Medic alphabet is the same as in Akkadian, except for the 
 three consonants Tc, s, and t, which are not represented. 
 
 Among the vowels aa represents d; e and i are easily con- 
 founded ; the letter u was perhaps pronounced sometimes as ii. 
 There are two signs for this vowel u isolated, and it is supposed 
 that one of them represented o, but this is doubtful. 
 
 The consonants offer the same confusion as in Akkadian at 
 the end of a closed syllable, the same sign representing ab and 
 ap, &c. The consonants seem also to have often been con- 
 founded at the beginning. It is supposed with some reason 
 that all the letters of the same order were confounded. As 
 certain sounds cannot be expressed, it has been supposed that in 
 certain cases the vowel was not to be pronounced ; for instance, 
 ffi-ud standing for gid, hu-ir for kur, &c. This must have been 
 the case in some instances, but according to the rule we have 
 taken, we give the spelling given by the inscriptions, and write 
 Icuir, although it may have been pronounced kur or Mr} 
 
 rOEMS OF THE WOEDS. 
 
 Nouns and Adjectives. 
 
 There are two genders and two numbers ; the genders are not 
 distinguished by any special forms, and the plural is often the 
 same as the singular. 
 
 ^ Many words are preceded by a straight wedge or a determinative 
 prefix aphone ; this is indicated in our transcription by a capital letter. 
 
S2 MEDIC GRAMMAR 
 
 All stems, nominal and verbal, end in a vowel. The words 
 aje built from these stems by means of suffixes, and their re- 
 lations are also indicated in the same way. 
 
 The bare stem is used to form certain compounds with other 
 words, and for this reason represents the dependent case of 
 Vannic- or the construct of the Semitic tongues. The suffixes 
 are added to this dependent case, and the stems with a suffix 
 constitute a new word which can take other suffixes. 
 
 The suffixes are : — 
 
 (i.) r, ri, or ra (primitively a demonstrative) forms nouns 
 and adjectives, or is simply added to the noun to individualise 
 its meaning: ir^a, " great ;" zVSa-ra, "chief." 
 
 (2.) n, ni, or na (primitively a demonstrative, cf. Yannic 
 enclitic ni) forms principally adjectives of manner, but also 
 substantives: niman, "belonging to-" KuraSna, "of Cyrus, 
 Cyrusian." 
 
 (3.) h, Id, or lea forms adjectives and substantives, especially 
 abstracts: miini-'ka, "bad;" tidu-Jc, "falsehood;" hari-lci, 
 "scarce." 
 
 R. The consonant of these three suffixes may be doubled, and 
 we have ir^arra, harikhi, <fec. 
 
 (4.) da forms adjectives and substantives, but also is used as 
 definite article: marri-da, "all;" mar-da, "the right;" adda-da, 
 " the father." 
 
 (5.) ti or tu, perhaps the same as the preceding, used especially 
 to form name of country: zi-tu, "so," from zi, "to see;" Apir- 
 ti (part of Susiana), " Apirtia." 
 
 (6.) ma^ (perhaps compound of ma and f) forms abstracts : 
 Ain-mai, "royalty,;" duki-umma^ (iiora du, "to be"), "cause." 
 
 (7.) ma (primitively " place" or "country") forms the loca- 
 tive of the nouns. 
 
 (8.) ^ (abbreviated form of ma^, or more probably primitive 
 suffix forming this one with the precedent) forms a few adverbs : 
 ukda-S, " above," really " the above." 
 
 Some words appear also to be formed by prefixes; a very 
 characteristic one is ha, which implies futurity : ha-hudda, "to 
 become," from hudda, " to do." 
 
NUMERALS. 83 
 
 As stated above, the suffixes can be superposed ; so from ma 
 and r we get mar, used to express a kind of ablative. The 
 combination of ka, ma, and r gives Itamar, meaning " far from.'' 
 
 A certain number of words (nouns, adjectives, and adverbs) 
 are used after nouns in the dependent case, and have been 
 considered as postpositions, but these words are really used as 
 abstracts, and form, with the preceding word, compounds : 
 Barsir-Mr, lit. "Persian-one," i.e., "a Persian." 
 
 The use of idaka, really a substantive meaning the action 
 of standing, but employed as postposition to express "with," is 
 to be explained in the same way, as also the locative postfix ma. 
 
 The plural is formed by the suffix pi or bi, added to the bare 
 fprm of the singular ; sometimes the final vowel is dropped, in 
 other cases the j> or & is doubled. The suffixes are added to 
 the plural as to the singular. 
 
 When a new stem is formed by means of a suffix the charac- 
 teristic of the plural is added after it. 
 
 With the ideograms the plural is indicated by the sign of the 
 plural read perhaps meS. 
 
 NUMERALS. 
 
 The numerals are always written by means of ciphers, their 
 pronunciation is therefore unknown. 
 
 We only find in the texts Jcir used as a postfix to express 
 the indefinite article " a," which may be also the number "one." 
 The expression ^amakmar, " twice," is also found ; the number 
 " two " may have been ^ama or ^amak. 
 
 The ordinals are formed by adding umma^ to the cardinals : 
 v-ummas, " sth." 
 
84 MEDIC GRAMMAR. 
 
 DEMONSTEATIVES. 
 
 The demonstrative mostly used with, the noun is hi; this form 
 is really the objective case, but the nominative hu is found only 
 in composition to form demonstrative pronouns : hu-pirri, " this 
 one," plural hupibi, "these ones." 
 
 From the same stem is derived hubi, " that," used only for 
 things (as the French cela), and always in the objective case. 
 Other forms are obtained from these derivatives. From hupihi, 
 " these," we have with the suffix na an adjectival form hupibi- 
 na, " the, theirs "or " theirs ; " from hubi we have hubi-da, " that 
 thing." 
 
 Another demonstrative pronoun is akJca, "this,'' used also 
 as interrogative and relative pronoun. The plural is ahhabi. 
 
 The relative pronoun is appa, invariable. 
 
 PERSONAL PEONOUNS. 
 The personal pronouns are : 
 
 ist pers. sing. U^ plur. Nuku. 
 
 2nd pers. sing. Nu plur. manka. 
 
 3rd pers. sing. hupirri plur. Appi. 
 
 The pronouns of the first and second persons in the singular 
 were used as nominatives or objectives, but to mark the. 
 objective case an n might be added ; U becomes then un (written 
 U-un), and Nu Nuin. 
 
 The pronoun of the third person was no doubt originally 
 irri or m'but took for support a demonstrative and became 
 hupi-in-i, but the original form is preserved in the objective 
 case, which is ir. 
 
 The objective case of the first and second persons of the 
 plural is not known, but in the third person it is regularly 
 formed, appin or apin, " them." 
 
 * Found once written Hu. 
 
VERBS. 85 
 
 These pronominal stems can take the suffixes like the 
 nominal stems ; so we have U-ni-na, " the mine ; " appi-ni, " the 
 theirs;" U-iJcM, "thel," ie., "myself," used as dative; ap-ir 
 or appi-ir, "the they," used for "to them," U-ikki-mar ; "from 
 me," &c. 
 
 Suffixes derived from the pronouns serve to express the posses- 
 sive: mi (wi perhaps for ui), "my;" ni, "thy;" and ri, "his."' 
 
 The forms for the plural are not known. 
 
 Eor the first person u is also used as a possessive, but is 
 prefixed to tlie noun : karpi-mi or U-karpi, "my hand." 
 
 There are other forms — really nominal or adjectival forms — 
 which are used to express possession or connection, as Nultami 
 "the ours." Of the same class is nudami (also written once 
 (laminu), which refers to the third person, though it would rather 
 appear to be a derivative of the second. 
 
 VEKBS. 
 
 All the verbal stems end in one of the three vowels a, i, and 
 11. This distinction forms three conjugations : ist conj. dana, 
 "to give," 2nd conj. tiri, " to say," 3rd conj. iazu, "to pray." 
 
 From each stem voices are formed by adding certain suffixes : 
 
 nunu for the desiderative. 
 manlu for the reciprocal. 
 na for the causative. 
 da for the emphatic. 
 
 New derivatives may be formed from these voices ; so we 
 have the desiderative of the causative formed by adding nunu 
 to na : nanun-A, &c. 
 
 Besides these voices the verbs are divided into active and 
 neuter, taking different sets of suffixes to form the persons of the 
 tenses. When the same stem takes the two forms, the neuter 
 has the force of passive. 
 
 For the active and transitive verb there arc two principal 
 
86 
 
 MEDIC GRAMMAR. 
 
 tenses, wliich serve to form the others. The present forms the 
 future and the imperative. The past forms the imperfect, 
 pluperfect, and preoative. 
 
 The personal suffixes of the present are : — 
 
 1st pers. 
 -man, 
 
 Singular. 
 2nd pers. 
 
 -mainti, 
 
 3rd pers. 
 -manra or -manri. 
 
 1st pers. 
 -manun, 
 
 Plural. 
 
 2nd pers. 
 •maintip, 
 
 3rd pers. 
 -mampi. 
 
 The future is formed by dropping the syllable ma, so wo 
 have ist pers. n, 2nd pers. inti, &c. 
 
 R. The n of the ist person in these two tenses is sometimes 
 replaced by the syllable ra : mara for man. 
 
 The imperative is formed by changing man of the ist pers. 
 of the present into i for the 2nd pers. sing, and plur., the 
 only one used. 
 
 The personal suffixes of the past are :— 
 
 1st pers. 
 a. 
 
 Singular. 
 
 2nd pers. 
 ki. 
 
 3rd pers. 
 
 1st pers. 
 
 ud, 
 
 Plural. 
 2nd pers. 
 
 3rd pers. 
 ai or maM. 
 
 JR. The a characteristic of the first person is often absorbed 
 by the ending vowel of the verbal stem, so that the first person 
 is confounded in form with the latter. To prevent this con- 
 fusion, the characteristic a is sometimes united to the stem by 
 the semi-vowel corresponding to that of the stem, we have then 
 
VEKBS. 87 
 
 with a >stem in a, 'a (the aspirate is not written cuz for a' a) ; 
 with a stem in i, ia ; with a stem in u, ma {loa). 
 
 The neuter verbs have only one set of personal suffixes : 
 
 Singular. 
 
 1st pers. 
 giud, 
 
 2nd pers. 
 iMi, 
 
 Plural. 
 
 3rd pars. 
 ik. 
 
 1st pers. 
 giud. 
 
 2nd pers. 
 [ildip ?] 
 
 3rd pers. 
 p, 01 ppi. 
 
 The present takes these suffixes with the syllable ma ; for the 
 past they are joined to the stem directly. For verbs of this ■ 
 kind the present and future have the same form. 
 
 The other tenses are formed for the active and neuter verbs 
 by suffixes added to the personal suffixes. These suffixes are : 
 
 ra for the imperfect. 
 da for the pluperfect. 
 ni iov the precative. 
 
 The inscriptions do not of course give examples of all the 
 forms with the same stem. 
 
 From hudda, " to do," we find hudda, "I did ; " huddal, "he 
 did ; " huddaud and hudtiud, " we did ; " huddai, " they did ; " 
 huddara, "I have done;" huddada, "I had done;" huddaMa, 
 " he had done ; " huddudda " we had done ; " hvddalt, " he was 
 done ; " huddamara, "I do ; " huddi^, " do ; " huddimanra, " he 
 will do ;" huddinunuba, " they wished to do." 
 
 From hudda, "to flee;" buddaJca, "he fled;" buddana, "I 
 caused to flee." 
 
 From tida, " to lie ; " tidinra, " he will lie." 
 
 From tiri, "to say, to call," tiri and tiria, "I say;" tiri^, 
 " he says ; " tirlai, " they say ; " tirimanpi, " they are called ; " 
 tirimanun, "we are called;" tiritti, "they were called," &c. 
 
 There are few verbs which appear to have abnormal forms, 
 but these are in some cases archaic, contracted or decayed. It 
 must also be borne in mind that the flexions of the verb have 
 
88 MEDIC GRAMMAR. 
 
 grown out of the nominal and adjectival suffixes, and forms, 
 which at one period were regularly employed with all the 
 verbal stems, have maintained themselves with some verbs, 
 when with others different suffixes have been preferred. 
 
 The verb da, " to do, to be, or to send," i has preserved for 
 the first person singular the archaic form da', " I did ; " but for 
 the third we have dai^ and daS, "he did." In the pluperfect 
 the second person is a nominal formation with the pluperfect 
 suffix da-s-da, lit. (thou) doer (wast). 
 
 The verb " to be " was primitively gin, but the g passed little 
 by little into y, and disappeared, changing only the i into e. 
 The g of this verb is the one preserved in the first persons 
 of the neuter verbs gi-ud. We find it without the suffix in 
 na-n-gi, " I say ;" the other persons are regular : na-inti, " thou 
 sayest;" na-nri, "he says." 
 
 The persons of the verb " to be " are very irregular, several 
 suffixes being superposed to compensate the weakening. We 
 have ennugiud, " I was ; " en-ri and en-ri-ir, " he was ; " enMb, 
 en-ri-p, and en-ri-pi, "they were." 
 
 To daMda may be compared biMa, "he did;" the same root 
 is perhaps to be found in innuihhida, " as long as thou wilt be 
 able," which is difficult to explain. 
 
 dumane, "he was or he did," appears to be an adjectival 
 formation from da. 
 
 ADVEEBS. 
 
 The adverbs are formed from substantives, adjectives, and 
 pronouns, with the suffixes iklci and da ; but a great many other 
 terminations are found, and nouns, adjectives, and pronouns 
 may be used adverbially, without taking any particular endint^. 
 
 From the demonstrative hu or hi are derived numerous 
 adverbs: Mma, "here;" hami, "there;" hamimar, "thence;" 
 
 ^ Thus three notions seem to have been connected in the minds of those 
 who spoke this language. 
 
SYNTAX. 89 
 
 kamaSir, " then ; " hupamaUr, " now ; " hamdlc, " how ; " with 
 zitu, "so J" hizitu, lit. "this so;" hupamtuikkiummaS, "conse- 
 quently," &c. 
 
 iaM^ata, " previously ;" mo^la/ca, "latterly." 
 
 The negative is inni. The form annu or anu is used before 
 verbs to express the prohibitive. 
 
 The word badar, which has been considered as a preposition, 
 is no doubt an adverb or a noun : taken adverbially, it means 
 " thoroughly " or " at last." 
 
 CONJUNCTIONS. 
 
 There are two words to express the copula "aiid," dk and 
 Jtiidda. 
 
 Other conjunctions are : 
 
 kui^, "while;" anka, "if;" sap, "because," &c. 
 
 SYNTAX. 
 
 The Medic syntax offers many analogies with the Vannic. 
 The genders are not distinguished by any special form. 
 
 The possessive or genitive is expressed in two ways. The 
 governed noun, without flexion, is placed before the regent, 
 forming with it a kind of compound word, and the regent takes 
 the determinative suffix ri: Kura% $ak-ri, "the son of Cyrus;" 
 or the regent stands first without flexion, and the governed noun 
 takes the adjectival suffix na : $ak Kura^-na, lit. " son Cyrusian." 
 
 The adjective, or the noun used as such, follows the word 
 qualified : it is reaUy placed in apposition, and for this reason 
 sometimes takes the same suffix as the noun it qualifies, but 
 also may take others : Nap ir^arra, " a great god ; " ukkurarra 
 irtanna, " the great universe." 
 
 When an adjective follows a plural it takes the plural suffix : 
 
go MEDIC GRAMMAR. 
 
 Eu'me^ harikkip, "a few men." But the plurals taken as 
 abstracts are followed by the adjectives in the singular. 
 
 The words governed by an adjective or a participle precede it. 
 
 Two words, nouns or a noun and adjective, may be joined 
 by the relative appa, "who, which:" Da^^utum appa Patipna, 
 " the rebel people,'' lit. " the people who (are) rebels." 
 
 The cardinal numerals precede the word in the plural : ix. 
 Ain-ip, "nine kings ;" xxiii. Dadau^, "twenty-three provinces." 
 
 The ordinals follow the noun : 
 
 U xs.-um.ma%, "I the gth." 
 
 To give the date the number — apparently the cardinal — 
 precedes the word for " day," then comes the ideogram for 
 " month," followed by the name of the month. As we have 
 no document really national, but only those written by order 
 of the Persian kings, these names of the months are the tran- 
 scription of the Persian. 
 
 The demonstrative hi is invariable and placed after, rarely 
 before, the noun; hupirri a\.s,o follows or precedes the noun, 
 but often has not the force of a demonstrative, and is used 
 simply as an article. 
 
 The demonstratives derived from hu may be used abstractly 
 without a noun and take the suffixes as the nouns. 
 
 The personal pronouns are considered as nouns and take the 
 suffixes accordingly. That of the first person U may be used 
 instead of Un as objective ; Uda^, " is to me." The objective case 
 is used to express the direct or indirect object (accusative or 
 dative). 
 
 Sometimes when the object of the verb is expressed a 
 pronoun representing it is however placed before the verb, as 
 in Akkadian. In this case the pronoun may be considered to 
 be incorporated, and if the verb is a compound, the pronoun 
 will be placed between the two elements of the word : 
 
 ema^ap-duMa, " he took them." 
 
 The tenses are not rigorously used, the present and future 
 are often confounded. 
 
 Nouns and adjectives are with or without suffixes used 
 
SYNTAX OF SENTENCES. 91 
 
 adverbially, that is, to express the complement of the verb or 
 the determinative. 
 
 The negative is always placed before the verb, but the 
 objective pronoun remains, -when expressed, prefixed to the 
 verb : the negative is therefore placed before it. 
 
 Of the two words meaning " and " ah is used especially at 
 the beginning of a sentence and may be translated by " also ; " 
 the two words may be used together : ah Tcudda, " and also." 
 
 The conjunctions are placed at the beginning of the sentence. 
 
 Syntax of Sentences. 
 
 Long periods are avoided ; they are, as a rule, cut into short 
 sentences. 
 
 The order of the word in the sentence is r-s-d-o-i-c-v, but this 
 order is not always observed. The reason of the action and the 
 determinative of state are often confounded. The subject is 
 often placed after the object, especially when the former is a 
 pronoun. When the object is a personal pronoun it is always 
 placed near the verb. 
 
 UramaMa hi Ain-maM U dunuii, " Ormazd gave me this 
 royalty." 
 
 mahii Pirrumartit hupirri teulnuip hariltldp Ualca huddulcka, 
 " then Phraortes, he having a few horsemen {i.e. with a few 
 horsemen) fled." , 
 
 Zaumin TJramatdana DaMwtum appa Hunina DaUutum appa 
 Bitipna irhikld halpit, "by the favour of Ormazd the people 
 who (were) mine killed in great (number) the people who (were) 
 rebels." 
 
 ma&ni Nutiudbiul hupirri U Bapilu halpHa, "then I killed 
 this Nidintabel in Babylon." 
 
 Two sentences may be united by a copula dk, " and," or 
 laidda. When the first sentence finishes with a verbal form 
 ending in a consonant, this one takes the vowel a after being 
 doubled to connect the second sentence. In that case (as with 
 the enclitic ma in Assyrian) it implies that the action expressed 
 
92 MEDIC GRAMMAR. 
 
 by the verb of the second sentence is the consequence or result 
 of the first. 
 
 Incidental sentences are introduced generally by the relatives 
 appa or akha ; sometimes the relative is understood, and the 
 sentence is really placed in apposition explaining one word. 
 In some cases words which might be introduced as an incidental 
 sentence are treated as forming a long compound word ; such a 
 combination is often preceded by the demonstrative tii. The 
 incidental relative or qualifioative sentences are generally placed 
 after the words they explain, being thus in the middle of the 
 principal sentence. 
 
 The secondary sentences marking the time, the place or the 
 state, are generally introduced by a conjunction (" when, as, if, 
 while," (%o.), and are placed before the principal. 
 
 dk iarak ii-ummaM-maBitip pirruir ^airraib-ba ^abarrakummaS 
 Mimana idaka, " and the second time the rebels gathered to- 
 gether (so that they fought) a battle with Mimana (lit. having 
 Mimana)." 
 
 Impirri Da^^utum Apir tituk-ka nanri, " this one lied to them, 
 tlie people, (so that) he said." 
 
 Ddiaui^ MarkuiS hih U-ikldmar biptibba Ru' kir Pirrada 
 hiie Markuii-irra hupirri Ain appini ir huddai^ ah maMni U 
 DadurU^ Idte Barsir kir U-Lubaruri Sak^aba manama^ BaMii 
 huddoM huttik hupirrikki tingia nangi, " the province, called 
 Margiania, revolted (plur.) against me (so that) a man called 
 Phrada, a Margian (rose and) this one they made king, and then 
 I commissioned to the Persian, called Dadarses, my servant 
 (who) had the satrapy of Bactriane, a messenger saying : " 
 
 ah anka DIppi hi innakkanuma hi zarinti inni kukirti UramaM- 
 da Nuin halpi^ni dk kudda Numanme^ni anu kitinti dk appa 
 huddainti apin UramaSda rippiSni. ' ' And if thou destroyest 
 this inscription (and) these images (or figures), (if) thou dost 
 not protect (them), may Ormazd kill thee, and also mayest thou 
 have no ofi'springs, and may Ormazd curse (these things) which 
 thou wilt do." 
 
SUSIAN AND APIEIAN. 93 
 
 SUSIAN AND APIEIAN. 
 
 Susian (i.e. the language of the inscriptions of the old 
 Elamite kings) and Apirian {i.e. that of the inscriptions of 
 Mai-Amir) may be considered simply as dialects of Medic ; and 
 the latter being better known has been for convenience' sake 
 taken as standard. The vocabulary and the grammar are to a 
 great extent the same in the three dialects. Susian and Apirian 
 have naturally in many cases more archaic forms ; in others, 
 however, they give examples of a more advanced phonetical 
 decay. It may also be concluded, from the use of parallel 
 forms employed promiscuously in the very same texts, that 
 Susian and Apirian never were systematised and fixed as 
 literary languages ; Medio, on the contrary, appears to have 
 passed through such a process, probably at the court of the suc- 
 cessors of Cyaxares, and no doubt under the influence of the 
 mother tongue of the Aryan conquerors. 
 
 The phonetic changes between the three dialects are almost 
 nil. Sometimes Apirian preserves the harder sounds, as in 
 dinIm = 'NLedic, dingi, "to restore;" and sometimes has weaker 
 forms, as gani = 'M.. Imni ; itni — M.. matni, "then;" Mma = M. 
 huma, " to create." Apirian has often the two forms hard and 
 weakened : Mr and ir = M. ir, " him." The changes of 6 and 
 p, t and d, are difficult to register on account of the uncertainty 
 of these sounds in Medic. Susian offers the same changes and 
 variations as Apirian. 
 
 The nominal and adjectival formation is the same in the 
 three dialects, and the suffixes are the same. The pronouns 
 are also the same, as far as they are known ; in Apirian the 
 stronger form of the first person pronoun hu has been reserved 
 to express the nominative " I " and u the accusative " me." 
 The aspirate has been lost in Susian as in Medio. 
 
 For the verbal forms Susian and Apirian give also the same 
 paradigm as Medic. But both dialects preserve in the greatest 
 number of verbs the primitive aspirate to mark the first person 
 of the first singular in the perfect sari' = M. sari, " I destroyed ; " 
 
94 MEDIC GRAMMAR. 
 
 parih = 'M.. pari, "I -went." In Susiau we find nali "I am" 
 from the verb en, "to be " with loss of the initial vowel. 
 
 The suffix of third person ri is often replaced by ni in Apirian, 
 and this appears to be almost exclusively used in Susian. 
 
 Besides the three derivative tenses in -ra, -da, and -ni (the two 
 first being often ri and ti), Apirian has a kind of correlative 
 formed by the suffix ba. 
 
 The syntax is the same in the three dialects. 
 
 Examples of Susian. 
 
 nah Sutrulc-Nahhunte Sak TJalladu^ anin hMnah gilt mnldlc 
 an7M,n ^u^unha erizuib tibuhi. 
 
 " I am Sutruk-Nahhunte, son of Halladus, the Susian king, 
 powerful ruler of Anzan, I have united together the Susian 
 warriors.'' 
 
 ina&mi Urmana ^arrahi, " here I placed the foundation." 
 
 Examples of the Apirian. 
 
 almuma^ Tirutur hu ulhi dihra tibba ^alhuba i%ni xl tartulc 
 umi Napir hupaklcir Dipii huddanba hiimei sirmuinni ardan. 
 
 " Having gathered together the images of Tirutur, protector 
 of this family (and) then having caused to be made there 40 
 (images) in all of the chiefs gods (and) of Dipti, I caused the 
 enclosure for them to be established." 
 
 kime^ rukku Rappitum uliku Pirni ^arara, Salzi undannamip 
 ea. "I will place a great curse on these transgressors, unless 
 they honour the house." 
 
 N.B. — The Medic and Apirian inscriptions are written by 
 means of a few ideograms and a special syllabary, derived from 
 later Babylonian soon after the fall of Nineveh, but Susian is 
 written with characters hardly modified borrowed from old 
 Babylonian. 
 
OLD PERSIAN GRAMMAR. 
 
 PHONETIC. 
 
 The alphabet of old Persian differs from that used for the 
 other languages of the cuneiform inscription. It has 24 letters : 
 
 The three vowels, a, i, and u. 
 The consonants are : — 
 
 Two aspirates, h, h. 
 
 Two gutturals, g, k. 
 
 Two'palatals, c', / 
 
 Four labials, b, p, f, in. 
 
 Five liquids, n, I, v, r, y. 
 
 Three sibilants, s, $, 2. 
 
 Three dentals, d, t, t. 
 
 The sounds of these letters are known to us only from the 
 comparison of proper names and words found in the other lan- 
 guages, and are therefore to a certain extent uncertain.. The 
 letter I is especially doubtful, as it is only found in two proper 
 names, and does not exist otherwise in Old Persian. 
 
 Those who studied first the Persian inscriptions have classified 
 the letters in the following manner, and this order has been 
 adopted in the glossaries which accompany their works : 
 
 gutturals palatals dentals labials nasals half-vowels sibilants 
 Hard k c' 
 
 Aspirated h — 
 Soft g j 
 
 The simple aspirate h is left out and placed last. 
 
 t 
 
 p 
 
 n 
 
 y 
 
 t 
 
 f 
 
 m 
 
 r,l 
 
 d 
 
 b 
 
 — 
 
 V 
 
96 OLD PERSIAN GRAMMAR. 
 
 If the classification of certain Indianists is followed we would 
 have : — 
 
 hard asp. soft nas. half-vow. spirants 
 
 Gut. k h g — • — h 
 
 Pal. c' — j — ■ r, I i 
 
 Dent, t t d n y s 
 
 Lab. p f h m v z 
 
 Note. It must be remembered that the Old Persian is written 
 by means of syllabic characters expressing one consonant and 
 one vowel, the consonant being always first. This syllabary 
 does not possess all the possible combinations, and such syllables 
 as ki, hi, c'i, cu', &o., could not be expressed ; closed syllables, 
 as ah, ig, ar, &c., could not either be written. It has been 
 supposed that when we had such a group as /co-j, it was to be 
 read Id, and that the final vowels are not to be read in many 
 cases, a-da-ma being written for adam. The peculiarities of the 
 syllabary may, however, be the result of the phonology of the 
 language. It is more rational to suppose that if the writers 
 of the Persian inscriptions had required the sounds Id, hi, dk, 
 ig, &c., characters would have been invented for them, and it is 
 more practical to take the language — the real pronunciation of 
 which we cannot know for certain — such as it is given in the 
 inscriptions. For the same reason have been rejected the ih 
 and n, which were supposed to exist by some scholars, though 
 Tiev&r written. 
 
 The vowel could be long or short. As initials they are always 
 written simple when short or long ; after a consonant the long 
 vowel is expressed by repeating it after the syllabic sign : mama 
 is written ma,-a-ma, dlta is written di-i-'ia, (fee. 
 
 When the vowels are strengthened — or gunated as it is called 
 by the Indian grammarians — a remains a, i becomes ai, and 
 u au. 
 
 Final consonants of words or syllables are dropped : paitd for 
 paitar, napd for napat, Icalaujlyd for Jcambaujiyd. Two con- 
 sonants are therefore never found together ; the only exception 
 is for the group tra, for which there is a special character. 
 
FORMATION OF WORDS. 97 
 
 There seems to have existed in Old Persian a kind of con- 
 sonantal harmony, the letter changes are consequently very 
 numerous ; consonants influenced one another even over vowels 
 and other consonants. The rules of these changes are very 
 difficult to establish, on account of the small number of texts ; 
 and the exceptions to the rules which can be detected appear to 
 have been numerous. A few of the changes are : 
 
 s becomes h after a and is often dropped altogether ; it often 
 reappears before t (see further on the forms of the verb " to be ") ; 
 i> becomes S after i and u (see the genitive formation). 
 
 t and d become s before t : basata past. part, from bada, before 
 c' they become ^ : avaiac'aiya for avadac'aiya. 
 
 k becomes c before t : c'aratanaiya from the stem kara. 
 
 j becomes h before t : duruhatama from the stem duruja. 
 
 All the consonants generally become aspirated before r, m, 
 n, S, and v. 
 
 But as already stated there are numerous exceptions. 
 
 Formation of Words. 
 
 The great majority of the words are formed by means of 
 suffixes. All words and stems end in a vowel, generally a ; 
 the suffixes can be superposed, that is, a word formed by a 
 suffix may receive another to form a new word. The formative 
 suffixes are : 
 
 (i.) a in mdragava} "a Margian," from maragu, "Mar- 
 giana ; " added to the a of the stem or of any word it gives us 
 a, which is the mark of the feminine, and is also found in some 
 words : duvard, " door." 
 
 (2.) at, only in napat, "grandson,'' written napd in the texts. 
 
 (3.) ana, in varakdna, "Hyranian" (from varaka "wolf"). 
 
 (5.) i, in great many proper names: faravaratai, "Phra- 
 ortes;" is a suffix of feminine names: baurm, "the earth;" 
 apai, "water." 
 
 (6.) iya, in patronymics, names of populations and others: 
 
 ^ 4s proper names are not in the inscriptions preceded by any deter- 
 minative we do not use capital letters. 
 
98 OLD PERSIAN GRAMMAR. 
 
 Tiahdmanai^aiya, "Achoemenid;" habairumya, "Babylonian;'' 
 marataiya, " a man, i.e. a mortal." 
 
 (7.) i^a, in manaiia, found in hahdmanai^a, " Achsemenes." 
 
 (8.) i^ata, superlative formative suflSx. 
 
 (9.) u, in vahau, found in the proper name vaJiauJta; the u 
 absorbs easily the other vowels, so this suf&x may be the for- 
 mative of margu, "Margiana." 
 
 (10.) Mia forms adjectives of matter : atagaina, "of stone." 
 
 (11.) A;a forms adjectives : vazaraka, "great, large." 
 
 (12.) ta, formative suffix of the past participle passive and 
 old disused form of the ablative found in parumyata, and 
 preserved as adverb formative suffix : tdkata, " exactly." 
 
 (13.) tama, superlative formative suffix. 
 
 (14.) td (for tar), va. paitd, "father." 
 
 (15.) tar a, comparative formative suffix. 
 
 (16.) tarana, in dumtdtarana. 
 
 {17.) td, adverb formative suffix : c'aitd, "so long;" forms 
 also other particles as conjunctions : yatd, " while." 
 
 (18.) tdda, with the final a assimilated in dhaifara^atadiya, 
 locative, " in perdition." 
 
 (19.) tana, formative suffix of the infinitive, always found in 
 the locative : tasatanaiya, "for speaking.'' 
 
 (20.) tai, in ^aiyatai, "goodwill." 
 
 (21.) tai, in ufaratai, "Euphrates.'' 
 
 (22.) ta, same as ta, the old ablative, forms adverbs : amuta, 
 " thence." 
 
 (23.) ta, perhaps same as td, forms abstracts : gaitd, "posses- 
 sion," and also adverbs and particles: avatd, "there," yatd, 
 " when." 
 
 (24.) tai, same as tai, in duvaratai, "gate-way." 
 
 (25.) tau, same as tau, m gdtau, "place.'' 
 
 (26.) tra, in ha^atra, " &mpiie ;" paufra, "son." 
 
 (27.) dd, another form of the old ablative, forms adverbs and 
 particles: idd, "here ;" y add, "when." 
 
 (28.) wa, forms adjectives, as osawa, "near;" and neuter nouns, 
 as satdna, " place ; " sometimes the inner vowels of the stems 
 are strengthened: daraujana, "liar," from duruja, "to lie." 
 
FORMATION OF WORDS. 99 
 
 (29.) nd, in faramand, "law." 
 
 (30.) ma, in navama, "nintli;" darasama, "strong, much." 
 (31.) md, in taumd, "family." 
 (32.) mdna, in asamdna, "heaven" or "sky." 
 (33.) ya, in araiya, "Aryan," perhaps same as iya. 
 (34.) yau, in maraMaiyauta found in composition uvdmaraSai- 
 yau^a, "self -killed." 
 
 (35.) ra forms adjectives and substantives of ho th genders : 
 zaura, " violence ; " dura, " distant." 
 
 (36.) vd (for van), in hoMatrapdvd, "satrap." 
 (37.) vata (for vanta), in harauvatai^a, "the Arachorian." 
 This list might be extended, but the other suffixes, as well 
 as some of the above, may have possibly the same derivation. 
 
 Particles, especially prepositions used as prefixes, form many 
 words, not only verbs, but also nouns. The principal ones 
 found in the texts are : 
 
 a or ava, augmentative in amdtd, " trusted " (participle pas- 
 sive plural) ; avahanama, " dwelling-place." 
 a, privative in andmaha, " nameless." 
 apa, "awa,j," in apagaudaydhaya, "thou wilt hide away." 
 aparaiya, " according," in aparaiydyd, " they conducted them- 
 selves according," i.e. " obeyed." 
 
 u, "well, good," in ufarasata, past participle, "having done 
 well," before a vowel it becomes uv: uvasapa, "having good 
 horses." 
 
 itda, " out," in vdaptatd, " he arose." 
 upa, "over," in upasatd, "help." 
 
 nai, "down," in naiyaMatayama, "I ordered, imposed." 
 naija, "out," in naijdyama, "I went out." 
 pard, "away," in pardbara, "he took away." 
 faro, " forwards," in farabara, " he gave, offered, presented." 
 hama, " with," in hamatahaMaiya, " I laboured," often 
 abridged in ha ; hamdtd, " of same mother." 
 m, " over," in myataraydma, " we went over." 
 The formation of compound words is very free. These are, 
 as in Sanskrit : 
 
 Copulative compounds, as in garamapada, name of a month. 
 
loo OLD PERSIAN GRAMMAR. 
 
 Determinative compounds, as in aradasatana, "high place." 
 Dependent compounds, as in tahahara, " who wears a crown ; " 
 
 asabara, "one carried by a horse," i.e. "horseman." 
 
 Possessive compounds, as in paruzana, " having many 
 
 branches ; " taliamasapada, " strong-hearted." 
 
 POEMS OP THE WORDS. 
 
 Nouns and Adjectives. 
 
 Nouns have seven cases formed by means of suffixes added 
 to the nominal stem ; for the singular we have : 
 
 (i.) Nominative formed by the suffix ^a : kwH-^a, " Cyrus ; " 
 faravaratai-^a, "Phraortes." 
 
 With the stems ending in a, the ^ of the suffix is dropped, the 
 a of the stem is not always lengthened : paraMa, "a Persian;" 
 auramazadd, " Ormazd." 
 
 With the stems ending primitively in a consonant, both 
 consonants of the suffix and of the word are dropped, and the 
 final vowel of the stem often lengthened : paitd (stem paitar), 
 "father;" napd {siam. napat), "grandson." 
 
 (2.) Vocative, in wliich the vowel of the stem is merely 
 lengthened : marataiyd, "man ! " 
 
 (3.) Genitive formed by the suffix & added to the stem, the 
 final vowel of which is strengthened : kurau-^a, " of Cyrus." 
 
 The stems ending primitively in a consonant, retain this 
 consonant followed by a, but the suffix is dropped : paitra, " of 
 the father." 
 
 With the stems ending in a the genitive is formed in lia or 
 haya : auramazaddha, " of Ormazd ; " hoMdyataiyahaija, " of 
 the king." 
 
 (4.) Accusative formed by the suffix ma added to the stem : 
 auramazadd-ma, "Ormazd;" hdji-ma, "the tribute;" bdhairu- 
 ma, "Babylon." 
 
 (5.) Ablative. This case is formed by the suffix d, but it 
 is supposed that it primitively was ata or ada, which is still 
 found as an adverb formative ; it is confounded in form with 
 the instrumental, and cannot be distinguished from it. 
 
NOUNS AND ADJECTIVES. loi 
 
 (6.) Instrumental, formed by the suffix a : Jcdra, " witli the 
 army." 
 
 (7.) Locative. This case appears to be formed very irregu- 
 larly : for the stem ending in a, we have aiya and di/a : ara- 
 manaiyaiya from aramanaiya, "Armenia," arabairdya from 
 arabaira "Arbela;" for the stems in i we have yd: apaiyd 
 from apai "water;" for the stems in u we have auva: hdbai- 
 rauva, from habairu, ''Babylon." 
 
 For the plural the examples found are still fewer : 
 
 (i.) The nominative is, for the stems in a, aha, sometimes 
 abridged in d : bagaha, from baga, " god ; " haMdyataiyd, from 
 ha^dyafaiya, " king ; " for the stems in i, it was probably iy'a, 
 but there is no example ; for the stems in u it is va : ddhaydva, 
 horn, dah'aydu, "province." 
 
 (2.) The vocative is as the nominative. 
 
 (3.) The genitive is formed by the suffix ndma, rarely dma, 
 added to the stem : bagdndma, " of the gods ; " daJiayau-ndma, 
 " of the provinces." 
 
 (4.) The accusative is the same as the nominative. 
 
 (5.) The ablative is not found, but was probably the same as 
 the instrumental. 
 
 (6.) The instrumental is formed by the suffix ibai^a : haga- 
 ihai^a, "by the gods;" vitaibai^a, "by the tribes." With the 
 stems in i, the initial i of the suffix is not repeated : asabdra- 
 ibai^a, "with horsemen," from asabdrai ; it is rarely dropped 
 after a : rauc'a, " day," makes rauc'abai^a. 
 
 (7.) The locative is formed by the suffix ^auvd added to the 
 stem : dahayau-iauvd, " in the provinces," With the stems in 
 a, the ^ of the suffix is dropped : anaiyduva, "in the others." 
 
 The feminine nouns have no special flexions, but take the 
 same as the masculine. The neuter nouns are not distin- 
 guished either, the only peculiarity is that the nominative and 
 the accusative are always similar: ranc'a, "day," nom. and 
 ace. ; hamaranama, " battle," nom. and ace. 
 
 This dual is found only in a few words, and especially those 
 designing pairs as " ears," " hands," &c. The dual follows for 
 the nominative and accusative the formation of the plural, but 
 
102 OLD PERSIAN GRAMMAR 
 
 the locative is formed as in the singular : gavMd, " the two 
 ears," nom. and ace. ; dasatayd, " in the two hands." 
 
 Sometimes the same word appears to follow indifferently the 
 formation of several classes ; for instance, we find the accusative 
 haumlma and baumama, the first being formed from the stem 
 bauml, and the second from the stem hauma. 
 
 Some words have no singular, as duvard, " gate ; " loc. duva- 
 rayd, "at the gate," which is in the dual, a gate being con- 
 sidered as a pair composed of two pieces. 
 
 Adjectives are not distinguished by special flexions, but take 
 the same as the noun ; the genders are shown only by the for- 
 mation of the stems, as in the nouns : for instance, in vazaralca, 
 "great," the feminine is distinguished by the long final vowel 
 of the stem : fem. vazarakd, gen. vazarahdya. 
 
 The degrees of comparison are formed by means of suffixes, 
 tara for the comparative, and tama for the superlative ; the 
 first is found in apatara, "the other," or "the more distant," 
 and the second in faratama, "the first" or "foremost." Two 
 other suffixes appear to have been in use by the side of the two 
 first for the same purpose : iya^a for the comparative, and i^ata 
 for the superlative ; iyaMa is preserved under the form of yaza 
 (? having become z), in a proper name vahayaza-ddta, " better- 
 gift," and Uata in matai^ata, "the uppermost." The words 
 formed with these suffixes take the case-endings like the stems 
 ending in a. 
 
 NUMEEALS. 
 
 The cardinal numerals being always written in ciphers, their 
 pronunciation is unknown. 
 
 Only a few of the ordinal numbers are known : 
 
 faratama, " the first." 
 dvmtaiya, "the second." 
 traitaiya, "the third." 
 navama, " the ninth." 
 
PRONOUNS. 103 
 
 From this last one it might be inferred that the ordinals 
 ■were formed by the suffix ma. 
 
 "We have also aiva, " the sole," or " the only," supposed to be 
 derived from the word for "one." 
 
 PEONOUNS. 
 
 The personal pronouns are not completely known. We have : 
 First person : 
 
 adama, "I," ace. mama, gen. mand. 
 vayama, " we," gen. amdhama. 
 
 There are, besides the enclitic forms, Tnaiya for the gen., and 
 ma for the ablative of the singular ; the ace. mMraa is also used 
 as enclitic. 
 
 Second person : 
 
 tuvama, " thou '' (really a vocative), ace. tauvdma. The 
 enclitic form is for the genitive taiya, written once 
 taya. 
 
 Third person : 
 
 The emphatic pronoun of the third person is expressed by 
 
 the demonstratives ; there are, however, special enclitic 
 
 forms : 
 Sing. ace. Saima, "him," gen. ^aiya. 
 Plur. ace. iai^a, " them," gen. idma. 
 
 • There are several demonstratives : 
 
 Hauva, "this one," found only in the nominative for the 
 masc. and fem. 
 
 From ava, " that," we have neuter nom. and ace. ava, masc. 
 ace. avama, gen. avdhayd, plur. nom. and ace. masc. avaiya, 
 fem. ava, neut. ava, gen. avai^dma. 
 
 From a stem ima, " this," we have : nom., masc. and fem. 
 iyama, neut. ima; ace. masc. imama, fem. imdma, neut. ima; 
 
I04 OLD PERSIAN GEAMMAE. 
 
 gen. f em. ahaydyd ; also written ahaiydyd ; inst. and ; plur. 
 nom. and ace. masc. imaiya, fern, imd, neut. imd. 
 
 From the stem ai we have the neuter aita, " this.'' 
 
 Of the indefinite pronouns we have : 
 
 Nom. masc, anaiya; neut., anaiya ; ace. masc, anaiyama, 
 gen. anaiyahayd, ahl. anaiyand ; plur. nom., and ace. masc. 
 and fem. anaiya, loc. anaiyduvd. 
 
 The relative pronoun, which is better known, is : 
 
 Neut. 
 > taya. 
 
 
 Masc. 
 
 Fem. 
 
 Sing. Nom. 
 
 . haya, 
 
 hayd, 
 
 Aec 
 
 tayama. 
 
 taydma, 
 
 Inst. 
 
 tayand, 
 
 
 STom. and Aec. 
 
 tayaiya, 
 
 iayd. 
 
 Gen. 
 
 tayai^ama. 
 
 
 Plur. Nom. and Aec. tayaiya, taya, taya. 
 
 The interrogative is kd for the masc, and probably for the fem. 
 also, and c'ai for the neuter. 
 
 There is a demonstrative enclitic di, which is at the ace. 
 dima in the sing., and di^a in the plural. 
 
 The neuter of the interrogative seems to have formed an 
 enclitic, e'aiya, which is added to other pronouns to form new 
 ones. It is to be noticed that before it the pronouns take their 
 primitive forms : anaiyaSa-c'aiya (ioi anaiyada-c' aiya), "some- 
 thing;" avaSa^o'aiya (for avada-c'aiya), "whatever;" IcaMa- 
 c'aiya, "somebody;" c'aHa-c'aiya (for c'aida-c'aiya), "some- 
 thing." 
 
 VERBS. 
 
 The classification and the system adopted in the Sanscrit 
 grammar being equally applicable to the verbs of Old Persian, it 
 is advisable to follow them as close as possible, as our texts are 
 so few that many forms would be inexplicable, if we had not 
 Sanscrit and Zend. 
 
VERBS. 
 
 los 
 
 There are three persons, the first, the second, and the third, 
 and two numbers, the singular and the plural ; if the dual ex- 
 isted, no example has come down to us. The genders are not 
 distinguished like they a»e in the Semitic verbs. 
 
 There are two different sets of personal suffixes for the active 
 and the middle or intransitive voices. They are, as far as 
 known to us from the texts — 
 
 Full forms : 
 
 
 
 
 Sing. 
 
 Plur. 
 
 Sing. 
 
 Plur. 
 
 Act., 1st pers. miya, 
 
 mdhaya, 
 
 Middle aiya, 
 
 mahaya. 
 
 2nd pers. haya, 
 
 — 
 
 — 
 
 — 
 
 3rd pers. taiya, 
 
 taiya. 
 
 taiya. 
 
 — 
 
 Short forms : 
 
 
 
 
 Act., ist pers. ma, 
 
 nid, 
 
 Middle i. 
 
 — 
 
 2nd pers. {ha), 
 
 — 
 
 — 
 
 — 
 
 ■ 3rd pers. &, 
 
 ia, 
 
 td, 
 
 td. 
 
 R. The suffix of the 2nd pers. of the short forms often drops 
 altogether. 
 
 R. With the verbal stem ending in a, the 2' of the 3rd pers. 
 sing, and plur. becoming h, is dropped altogether and the ter- 
 mination is reduced to a. 
 
 Special endings are used to form the imperative : 
 
 Act., 2nd pers. sing, diya ; plur. td ; middle sing. uvd. 
 „ 3rd ,, „ tuva : „ — „ tdma. 
 
 There are four tenses : present, imperfect, aorist, and pluper- 
 fect ; and four moods : indicative, conjunctive, potential, and 
 imperative. 
 
 The present is formed by adding to the stem the full forms 
 of the personal suffixes : ddraya-mlya, " I hold," from daraya 
 (root dara). 
 
 The imperfect is formed by adding to the stem the short 
 forms of the suffixes and prefixing to it the vowel a, which is 
 called the augment : dbavama, " I was," from hava (root hau). 
 
 When the verbal stem begins with a vowel, the augment forms 
 
io6 OLD PEESIAN GRAMMAR. 
 
 a diphthong witli it or is elided. The augment is always placed 
 before the verb, even if the verb is preceded by a preposition, 
 forming with it a compound, and the augment is also in this 
 case fused with the final vowel of' the preposition : avdjanama, 
 "I was killing," from the preposition ava, and the stem jaraa. 
 
 These two tenses are called special tenses, because the stem 
 of most verbs assumes a special form in them, and these special 
 formations have been used as bases of the classification of the 
 verbal stems by the Indian grammarians. 
 
 The aorist is formed by adding to the unmodified verbal root 
 the short forms of the suffixes and prefixing the augment : add, 
 "he gave" (for add^a), from the root dd. 
 
 The pluperfect is formed by doubling the root of the verb, 
 but as in Greek the consonant is weakened, k becomes c'/ the 
 only example known is the third person of the potential 
 caharaiyd, from the very irregular verb Icara, "to do.'' 
 
 The conjunctive mood is formed by adding a to the stem 
 before the personal suffixes : havdtaiya, 3rd pers. of the pre- 
 sent of the conjunctive, from hava (root hau). The character- 
 istic a merely lengthens the final a of the stem. 
 
 The potential is formed by adding ya to the stem before the 
 personal sufiixes : avdjanaiyd, 3rd pers. aorist from the compound 
 verb avOrjana, " to kill." 
 
 As examples of the imperative we may quote : pdd^ya and 
 jidtuva, from the root pa, "to protect." 
 
 Besides these moods and the middle voice, the Old Persian 
 verb had also two other forms — considered by the Indian gram- 
 marians as simple derivatives — the passive and causative. 
 
 The passive is formed by adding ya to the verbal stem, and 
 takes the personal suffixes of the middle voice, from which it 
 is distinguished only in the special tenses: tdhaydmahaya, "we 
 are called;" asaraiyatd, "he was broken," and several other 
 examples taking the suffixes of the middle voice. 
 
 The causative, of which we have but few examples, is formed 
 by adding aya to the stem, and it takes the personal suffixes of 
 the active : avdsatdyama, " 1 caused to be placed," from ava- 
 satd, " to stand." 
 
VERBS. 107 
 
 All verbs are supposed to be derived from a root, and they 
 may be divided into ten classes according to the modification 
 experienced by the root in the two special tenses. 
 
 (i.) The verbs of the first class are those the vowel of which 
 is strengthened or lengthened, or as it is called in the Sanscrit, 
 gunated, a of the root becomes a, i ai or ay, u au or av, and 
 takes the vowel of union a; the root hau, "to be," makes 
 hava in cAava 3rd. pers. sing, imperfect ; guha, " to speak,'' 
 makes gauha in gauhataiya 3rd. per. present. 
 
 (2.) In the second class the root does not suffer any change, 
 but takes directly the personal suflSxes': hara, " to carry," 
 makes abqrama ist pers. imperfect. 
 
 (3.) The verbs of the third class are those in which the first 
 consonant is doubled : the root da, " to give," makes adadd 3rd 
 pers. imperfect. 
 
 (4.) In the fourth class the suffix iya is added to the root, 
 the passive derivatives belong to this class : the root, mora, 
 " to die," makes maraiya in amaraiyata 3rd pers. imperfect. 
 
 (5.) The verbs of the fifth class add nava to the root : the 
 root vara, " to declare," makes varanava in varanavdtaiya 3rd 
 pers. present of the potential. 
 
 (6.) The sixth class includes the verbs which take the per- 
 sonal sufiixes with the vowel of union a; but as all words 
 must end in a vowel in Old Persian, this class is practically the 
 same as the second : jiva, " to live," makes jlvahaya 2nd pers. 
 present. 
 
 (7.) The seventh class, which is composed in Sanscrit with 
 the verb inserting a nasal before the last consonant of the root, 
 practically does not exist in Old Persian, as a nasal at the end 
 of a closed syllable is always dropped. The Sanscrit hand is 
 hada, " to bind," in Old Persian. See, however, further on, the 
 form dkunau&a from the root hu. 
 
 (8.) The verbs of the eighth class add uva to the root : dana, 
 " to flow," makes danauva in danauvataiya 3rd pers. present. 
 
 (9.) In the ninth class the verbs add na to the root : dl, " to 
 take away," makes dina in aMna 3rd pers. imperfect. 
 ' (10.) The verbs of the tenth class add aija to the root, the 
 
io8 OLD PERSIAN GRAMMAR. 
 
 vowel of the root is besides gunated : guda, " to hide," makes 
 gaudaya in apa-gaudaydhaya 2nd pers. present of the con- 
 junctive. 
 
 The irregularities observed in certain verbs come from the 
 fact that the same verb sometimes follows two or more classes 
 in some of its forms, or that one verb is only used in certain 
 tenses or persons, the others being supplied by another stem, as 
 in our verb " to go," past " went." As a specimen, the verb 
 " to be " might be given as far as it is known from the inscrip- 
 tions : the root as in Sanscrit has become aha, but the h is 
 generally dropped altogether, the primitive s reappears in the 
 third person of the singular. 
 
 Perfect 
 
 [g., istpers. 
 
 army a, plur. amahaya. 
 
 2nd,, 
 
 ahaya, ,, — 
 
 3rd „ 
 
 asataiya, „ hataiya. 
 
 ist „ 
 
 ahama, „ — 
 
 2nd,, 
 
 aha, ,, alia 
 
 Imperfect 
 
 Conjunctive sing., 3rd. pers. ahataiya, „ — 
 
 We have besides a third person plural with the sufiSx of the 
 middle voice, ahatd, written once ahata. 
 
 From the root bau are taken the following forms : 
 
 Imperfect ist pers. a&awa?n. 
 
 2nd „ abava, plur. abava. 
 Conjunctive 3rd „ bavdtaiya. 
 Potential aorist, 3rd pers. baiyd, with 'the change of 
 u into i. 
 
 The verb Itara, "to do," gives us another example. Most 
 of its forms are taken from the root ku, according to the fifth 
 class, i.e. with the suffix nava reduced sometimes to na (ninth 
 class), or according to the second class. The forms found in 
 the texts are : 
 
 Present conjunctive, 2nd pers. sing, kunavdhaya. 
 Imperfect, ist pers. sing, akunavama, plur. akuma. 
 Imperfect, 3rd pers. sing. akunavMa and akunala, plur. aku- 
 
ADVERBS. 109 
 
 The form aJcunauSa must be brought back to the seventh 
 class, with insertion of na in the root, the n not being dropped, 
 because the stem ends in a double voweL 
 
 Pluperfect potential, 3rd pers. c'aharaiyd, in which the 7c 
 has become aspirated into h. 
 
 Of the middle voice we have : 
 
 Imperfect, 3rd pers. aJcunavatd; plural, aMlnavatd and 
 dlmtd. 
 
 Of the passive derivation we have : 
 
 Imperfect, 3rd pers. sing, ahunavayatd. 
 
 The texts give us also the infinitive, c aratanaiya, where 
 the k is weakened into c', and the past participle karata 
 or karatd. 
 
 ADVEEBS. 
 
 Nouns in the accusative, locative, and instrumental are used 
 as adverbs; nurama, "now," is no doubt the accusative of 
 a disused noun. 
 
 The special formative of adverb was td or da, which is 
 found in takatd, "exactly, immediately;" ada, "then;" idd, 
 " here," &c. 
 
 The negative is naiya. Another negative is md, which forms, 
 with the verbs, kinds of prohibitives. After this particle the 
 augment of the verb is dropped : md tarasama (for md atara- 
 sama), "that I may not fear." Properly speaking this prohi- 
 bitive particle may be considered as a conjunction. 
 
no OLD PERSIAN GRAMMAR. 
 
 CONJUjSTCTIONS. 
 
 The copula is expressed by utd, " and," and the two enclitics 
 va and c'd. 
 
 The principal conjunctions are : 
 
 taya, "that;" caitd . . . ydtd, "till," or "as long . . . 
 as ; " yadlya, " when ; " mataya, " lest," &c. 
 
 Many of the adverbs are also used as conjunctions. 
 
 PEEPOSITIONS. 
 
 The prepositions, which play, as we have seen, an important 
 part in the formation of words, have a nominal origin, and are, 
 when used isolated, found with the flexions : so we have upd, 
 "on," instrumental of wpa; the genitive patai^a, "to, against," 
 and the locative pataiya, "for, at," from patai. 
 
 In the syntax will be given the list of the prepositions found 
 in the texts, not in composition, with their meaning, their use, 
 and the cases they govern. It need only be mentioned here 
 that some prepositions are found only in composition, and that 
 such a preposition as pataiya is used as postposition as well as 
 preposition, and even as enclitic postposition, in which case it 
 really forms a kind of flexion or case. 
 
 para is used as a postposition with pronouns : ava-pard, 
 "thereto," and with the flexion of the accusative, hayd-para, 
 " towards them," taken adverbially. 
 
 SYNTAX. 
 
 A noun in the singular may be used as a collective, taking 
 the verb in the plural : Jiayd amdhama hoMayataiya aha, " our 
 (family) were kings." 
 
 In a few cases the same word designates the country and its 
 
SYNTAX. Ill 
 
 inhabitants : pdrasa, " a Persian " and " Persia ; " mdda, " a 
 Made " and " Media." Sometimes a word is used indifferently 
 in the singular or plural to designate a country : mvdardya and • 
 mudardyd, " Egypt." The names of people are generally in 
 the plural, ku^aiyd, " the Cossoeans." 
 
 The dual is rare and generally replaced by the plural ; it is 
 preserved only to designate pairs : gau^d, "the two ears." 
 
 The force and meaning of the flexions had been to a great 
 extent lost, and they are often replaced by prepositions. 
 
 The nominative marks the subject. 
 
 The genitive is used without preposition. The noun in the 
 genitive may precede or follow its regent : kurwuMa paulra, 
 " son of Cyrus ; " vaMand auramazaddha, " by the favour of 
 Ormazd." The dative being lost in the nouns, it is expressed by 
 the genitive : hdrahaya avata ataha, " he spoke so to the army." 
 
 The accusative marks the direct object, without preposition : 
 avama matai^atama alcimavatd, " they made him leader." 
 Without a preposition, or with the preposition dbaiya, " to," it 
 marks the direction towards a place, there can therefore be two 
 accusatives : avama adama fardiiayama armlnama, " I sent him 
 to Armenia ; " ahaiya avama naditabairama aMaiyava, " it went 
 over to this Nidintabel." In this case, however, properly speak- 
 ing, it marks the dative of direction. 
 
 The instrumental marks the manner : va^and, " by the 
 favour." 
 
 The ablative is never used without a preposition (see preposi- 
 tions). 
 
 The locative is often used : apaiyd, " in water ; " ndviyd, 
 "in boat;" dahayav^auva, "in the provinces." 
 
 The adjective agrees with the substantive it quahfies, in 
 gender, number, and case : hoMdyataiya dhaydya humiyd vaza- 
 rakdyd, "king of this great earth." 
 
 Often the adjective is united to the noun by the relative pro- 
 noun, which may be translated by the article : Icdrama tayama 
 hamitraiyama, " the people the rebellious." 
 
 There is no agreement when the words are merely in apposi- 
 tion. 
 
112 OLD PERSIAN GRAMMAR. 
 
 In a comparison the ■word compared to is placed after the 
 comparative with the preposition had a: wpalarama hac'apdrasd, 
 "other than Persia." After the superlative of comparison the 
 genitive is used : matai^atd hugdndma, " the greatest of the 
 gods.'' 
 
 The cardinal numerals are always written in ciphers, and are 
 placed before the noun : xix hamarana, "19 battles." 
 
 When the number is the total of a summation, it follows the 
 noun : faraharavama dahaydva xxiii, "in all 23 provinces." 
 
 Cardinal numbers may be taken abstractly : viii ha^d- 
 yataiyd aha, " 8 were kings." 
 
 With the day giving the date the cardinal numbers are given, 
 not the ordinal ; the day is placed in the instrumental, preceded 
 by the month in the genitive, and followed by the adverb 
 " exactly : " andmakahaya mdhaya 11 rauc'abaiia takatd, " on 
 the 2nd day (lit. 2 days) of the month of Anamaka exactly." 
 
 The ordinals are real adjectives : pataiya traitaiyama, " the 
 third time." 
 
 The personal pronouns are used emphatically with the verbs. 
 
 As there are no possessive pronouns, these are expressed by 
 the genitive of the personal pronouns : mand paitd, " the father 
 of me." 
 
 The enclitic pronouns are attached to nouns, pronouns, and 
 particles, but not to verbs, except the enclitic di, which is also 
 attached to verbs. This union of enclitics is, after all, only a 
 question of prosody : these words having no accent, were not 
 considered as words in Old Persian, but written merely to the 
 preceding word which happened to be before them. The 
 enclitic has, therefore, not necessarily any connection with the 
 word to which it is joined. 
 
 The relative pronoun agrees with the word it represents : 
 imd dahaydva tayd mand pataiydiia, " these provinces which 
 pay homage to me." 
 
 The relative is used as a kind of article to unite words placed 
 in apposition : gaumdtama tayama maguma, " Gaumata the 
 Magian " (accusative). 
 
 The distinction between the active and middle voices is 
 
SYNTAX. 113 
 
 always made, though some of the forms of the two voices are 
 confused. The middle and passive voices are much more oftea 
 confounded. The middle voice is neuter or reflective : hauva 
 udapatata, " he rebelled ; " IpaMtrama hauva agarabdyata, " he 
 seized (for himself) the empire." For the passive : ava aJcunor 
 vayatd, " this was done ; " hauva atrina basata anayatd abaiya 
 mama, "this Atrina was brought prisoner to me," these two 
 examples will be sufficient. 
 
 The pronoun subject of the verb is often expressed even 
 when the noun is already given. In the last example, for 
 instance, it is really : " he, Atrina, was," &o. 
 The present requires no explanation. 
 
 The imperfect answers exactly to the English past : it is the 
 historical tense. It expresses also, like the imperfect of the 
 Latin tongues, a. lasting action prolonged sometimes even till 
 the present: adama-^aima avdjanama, " I killed him ; '' avahaya 
 kabaujhjahayd bardtd aha, "a brother was to this Kambyses ;" 
 adama-Sdma haMdyatauja aham, " I was (and still am) their 
 king ; " Icdra haya nadltabairahayd taigardma addraya, " the 
 army of Nidintabel was holding the Tigris." 
 
 The imperfect has sometimes, especially after yata, the force 
 of pluperfect : " I worked ; " yata gaumdta vitama taydma 
 amdhama naiya pardbara, " so that Gaumata might not have 
 superseded our clan." 
 
 The aorist answers to the past perfect, and expresses an 
 action quite finished : haya imdma baumlma add, " who has 
 given this earth." The aorist also may have the force of 
 a pluperfect : hara hac'd yaddyd faratarata, hauva haed-ma 
 hamltraiya abava, " the army had separated itself from duty, 
 it became rebel against me." 
 
 The conjunctive mood is used to express the future : ada- 
 taiya azadd bavdtaiya, "then knowledge will be to thee." 
 
 The potential expresses a doubt or a wish, and is used for 
 conditional : auramazadd-taya jdtd utd-taiya taumd md baiyd, 
 " may Ormazd be thy enemy and no family be to thee." 
 
 The imperative expresses an order or a wish. Often a sen- 
 tence begins in the potential and finishes in the imperative : 
 
 H 
 
114 OLD PERSIAN GRAMMAR. 
 
 utd-iaiyi taumd vasaiya haiyd utd daragama jlva, " and may 
 a numerous family be to thee and live long." 
 
 The infinitive, always in the locative, is used as in our 
 modern tongues after another verb, hut without any preposition : 
 ai^a hadd Jcdrd patai^a mama Jiamaranama c'aratanaiya, "he 
 came to do battle against me with an army." 
 
 The negative is generally placed before the verb : haya mand 
 naiya gauhataiya, "who does not call himself mine." It is 
 also placed before noun : adama naiya haradiya amlya, " I 
 am not Bardes." And it is repeated before every noun and 
 even adjective : naiya alia marataiya naiya pdrasa naiya 
 amdhama taumdya, " there was no man, neither Persian, nor of 
 our family." 
 
 The conjunction taya unites simply two phrases ; the verb of 
 the second may be in the indicative : taya adam amvya, " that 
 I am." 
 
 After yatd the imperfect is used when it means " when," and 
 the conjunctive when it means "in order that.'' Yadiya 
 governs the conjunctive in the conditional sentences. When it 
 means " when " it is followed by the imperfect. 
 
 The prepositions found in the texts are : 
 
 ahaiya, " to, against," with the accusative. 
 
 abai^a, " on," with the locative. 
 
 aniiva, " along, on," with the locative. 
 
 atara, " in," with the accusative. 
 
 ayasatd, " with," with the accusative. 
 
 uj>d, " for," with the accusative. 
 
 uparaiya, " over," with the accusative (in two doubtful pas- 
 sages only). 
 
 pataiya alone, or with the ordinal number in the neuter, it 
 is used for " again " or " time : " jpataiya traitaiyama, " the 
 third time ; " in these cases it may be a neuter noun,'used alone 
 or with an adjective as adverbial locution. "With nouns this 
 particle is postposed, with the instrumental "in, by:" mta- 
 pataiya, "in clan," with the locative "on :" uzamayd-pataiya, 
 " on a cross." 
 
SYNTAX OF SENTENCES. 115 
 
 patai^a, " against," with the accusative. 
 paraiya, " on, about," with the accusative. 
 pasd, "behind," with the genitive. 
 hadd, " with, by means of," with the instrumental. 
 hac'd, " by, from," with the ablative. 
 
 Syntax of Sentences. 
 
 Old Persian is rather free in its construction. As a rule the 
 subject stands first and the verb is thrown to the end of the 
 sentence, immediately before the verb is placed the complement, 
 and before this one the object ; the indirect object precedes the 
 direct object : the formula is therefore s-i-o-c-v. The determina- 
 tive of time, place, or state is generally placed at the head of 
 the sentence before the subject : pasdva Jcdra hahairuviya Jiac'd- 
 ma hamltraiya abava, "then the Babylonian people became 
 rebel to me." 
 
 When there is an emphatic statement, the verb is placed 
 first : tdtaiya darayavauta haMdyataiya, " the king Darius 
 said ; " naiya aha marataiya, " there was not a man." 
 
 The reason of the action follows generally the verb ; the in- 
 direct object, being confounded sometimes with it, is also thrown 
 back after the verb in some cases : pasdva Tiarmiraiya hagamatd 
 paraitd patai^a dddara^aima hamaranama c'aratanaiya, ." then 
 the enemies gathered together (and) came to make battle 
 against Dadarses ; " hamltraiya abava hacd-md, " they became 
 rebel against me." 
 
 The complement and the object may be also placed after 
 the verb : va^and auramazadalia hara haya mand avama 
 kdrama tayama hamltraiyama aja vasaiya, " by the favour of 
 Ormazd my army killed the rebel army in great number;" 
 adama hdrama faraiiayama babairuma, " I sent an army to 
 Babylon." 
 
 When the subject is expressed emphatically it is sometimes 
 placed near the verb after the object : babairuma hauva agara- 
 hdxjata, " he took Babylon for himself." 
 
ii6 OLD PERSIAN GEAMMAE. 
 
 The mention of a date is a determinative, and placed at the 
 lieginning of the sentence : atraiyadiydliaya mahaya xvii. 
 raudahaita talcaid alia, "it was exactly on the loth (lit. the 
 lo days) of the month of Atriyadiya.'' 
 
 When the verb has two objects one is often placed after it : 
 adain hadd liamanaihaita maraiaiyaibaiia avama magum avcija- 
 nama utd tayai-iaiya faratamd marataiyd arm^iyd aliatd, " I 
 with faithful men killed this Magian and his chief followers 
 who were with him." 
 
 The complex period offers no difficulty, as every one of its 
 members is considered as a compound expression, and placed in 
 the same order as the simple elements of an ordinary sentence. 
 
 The sentences beginning by '■ if, when," &c., are considered 
 as determinative and placed before the principal sentence : ydtd 
 adamahdbairuva ahama imd ddliaydva taydhac'd-md liamltraiyd 
 abava pdrasa uvaja, &o., "when I was at Babylon, these 
 provinces, which became rebel against me, (were) Persia, 
 Susiana," &c. 
 
 The following sentences will show that the construction is 
 very regular : 
 
 naiya aha marataiya naiya pdrasa naiya mdda naiya amd- 
 hama taumdyd IcaMc'aiya haya avama gaumdtama tayama 
 maguma ha^airama dltama c'aharaiyd, " there was not a man, 
 neither Persian, nor Medic, nor any of our family whoever, who 
 could have taken the power from Gaumata the Magian." 
 
 We have in this last sentence an example of double accusa- 
 tives : ha^airama dltama is really the complement of the verb : 
 literally it is "who could have made Gaumata, the power 
 taken away." 
 
 Mrama vasaiya avajanaiya haya paranama laradiyama 
 addnd avahayaradiya Mrama avajanaiya mdtaya-mama ha^and- 
 sdtaiya taya adama naiya baradiya amiya haya Jmrau^a pautra, 
 "he killed many people who had known the old Smerdis, for 
 which reason did he kill the people : ' lest they should know 
 me that I am not Smerdis who (is) the son of Cyrus.'" 
 
 koMac'aixja naiya adard^anauU c'aiiac'aiya tasatanaiya 
 paraiya gaumatama tayama magvma yatd adama arasama 
 
SYNTAX OF SENTENCES. 117 
 
 pasdva adama auramazaddma pataiydvdiiaiya, auramazadd- 
 maiya upasatdma abara, " any one dared not say anything 
 about Gaumata the Magian till I came, afterwards I prayed to 
 Ormazd, Ormazd brought me help." 
 
 pasdva hauva Jcdra Jiaya mand Jcapada namd daliayduta 
 madaiya avada mama c'aitd amdnaya yatd arasama mddama, 
 " afterwards this army which (is) mine waited for me there, a 
 province of Media called Kampada, till I reached Media." 
 
 THE END. 
 
 PEIMTED BY EALLANTYNE, HANSON AND CO. 
 EDINBURGH AND LONDON. 
 
* . 
 
 **n