A GRAMMAR or THE ARABIC LANGUAGE, ‘TRANSLATED FROM THE GERMAN OF CASPARI, AND EDITED WITH NUMEROUS ADDITIONS AND CORRECTIONS ay W. WRIGHT, LLD., AYR PROFEASON OF AKABIO IN THE UNIVERSITY OF CAMRRIDOR, THIRD EDITION REVISED BY W. ROBERTSON SMITH, Lave PROFEASOR OF ABADIC IN THE UNIVERSITT OF CAMBRIDGE ax M., J. pz GOEJE, . ‘Paovasson OF ARABIG I -THR UNIVERSITY OF LEFDRN, VOLUME I. WITH A PREPACE AND ADDENDA ET CORRIGENDA BY PIERRE CACHIA LIBRAIRIE DU LIBAN BEIRUTLIBRAIRIE DU LIBAN Riad Solh Square - Beirut Associated companies, branches and representatives throughout the world. New Impression, 1974 New Impression, 1996 Printed in Lebanoi. by A. J PREFACE Wright's ARABIC GRAMMAR Generation after generation of English-speaking students of Arabic have found Wright's Grammar an indispensable stand-by. Its popularity is well-deserved and likely to endure yet longer, for itis a very practical work of reference. It does not concern itself much with the disputations of the theorists, but it does state clearly, and illustrate sufficiently, all the working rules of classical Arabic. It manages to do so within its modest bulk, not by ignoring fine distinctions, but by dint of very careful organization. ‘The new “Addenda et Corrigenda”’ in this edition are not the result of a systematic revision of the text, but have grown out of marginal notes made during nearly half a lifetime of constant reference to it. That they are so few is a tribute to the accuracy and comprehensiveness of the original. Not the least of the virtues of the Grammarare its indexes. A useful adjunct to them is P. Cachia’s The Monitor, A Dictionary of Arabic Grammatical Terms(Longman-LibrairieduLiban, 1973), whichcumulates the terminologies adopted by Wright and by M.S. Howell in his much more extensive Grammar of the Classical Arabic Language. Pierre CachiaADDENDA ET CORRIGENDA VOL. 1 5 Arread: *isiS5A5 the liquids. , which are pronounced with the extremity of the tongue ( ys). The term is sometimes, although dubiously, made synonymous with 3io,08 or uid, oF ‘gliosis Which include not only the three liquids above but also three ‘of ‘the labials ~-- . They appear to be so called by association not with any part of a vocal organ, but with the meaning of chastness or profuseness of speech, in that every quadriliteral or quinquiliteral root of genuinely Arabicderivationis said to contain at least one of these six letters. All letters that are Not 3.5, are called {21+ made silent, mute, or speechless. For other groupings of letters based on varying criteria, see Howell's Grammar, iv. 1725 ff. 191 Cadd: REM. j. On * and its fem. , both, a pair~ which have similarities with the dual - see Vol. ii § 83 (p. 212 D). REM. k. 2 is generally taken to be the dual of equal, like.There are ications, however, that it has sometimes been regarded as an indeclinable word. Most MSS of Béidawi's Commentaryon Kor'an v.31 read of, 538 . A line ascribed to ‘Abi Du'aib is quoted as phen ieey Ae, And they were two like cases that they should not send forth cattle to pasture or send him forth with them when the tracts were very dusty (Lane, p. 1480, col. 2, following the Muhkam of Tbn Sida); but the Lisén'has = instead of -< , and the verse most closely resembling it in Hell’s edition of the poet's Diwan has = . Itisnot without significance that in the colloquial, which generally gives currency only to the oblique form of all duals, the word occurs as i is not used in the Kor'an, nor - so far as has been ascertained ~ in the hadith. 255 C_ add to § 320: They are diptote, however, when used as abstract numbers (Comp. Vol. i, § 309 (E), p. 241 DJ. VOL. I interrogative i may, on the other hand, be the direct object of a subsequent verb: €.g." is!+.iyi Which, then, of the signs of God will ye deny (Kor'an XI. v.81). If, however, the verb has a pronominal suffix which falls back upon ( % or ==») the interrogative pronoun, the latter is either nominative absolute or accusative, as being the direct object of a suppressed verb. Pierre Cachia PREFACE TO THE THIRD EDITION. [THE Second Edition of Wright's Grammar of the Arabic language had been out of print long before the death of its author, but he was never able to find the leisure necessary for preparing a New Edition. The demand for it having become more and more pressing, Prof. W. Robertson Smith, who well deserved the honour of succeeding to Wright's chair, resolved to undertake this task, He began it with his usual ardour, but the illness which cut short his invaluable life soon interrupted the work. At his death 56 pages had been printed, whilst the revision had extended over 30 pages more. Robertson Smith had made use of some notes of mine, which he had marked with my is, and it was for this reason among others that the Syndics of the Cambridge University Prese invited me, through Prof. Bevan, to continue the revision, After‘earnest deliberation T consented, influenced chiefly by my respect for the excellent work of one of my dearest friends and by a desire to complete that which another dear friend had begun. Moreover Prof. Bevan promised his assistance in correcting the English style and in secing the book through the press T have of course adhered to the method followed by Robertson Smith in that part of the Grammar which he revised, Trifing corrections and additions and such suggestions as had already been made by A. Miller, Fleischer and other scholars, are given in square brackets. Only in those cases where it seemed necessary to take all the responsibility upon myself, have I added my initials, Besides the printed list of additions and corrections at the end of the Second Volume, Wright had noted here and there- PREFACE TO THE THIRD EDITION. on the margin of his own copy some new examples (chiefly from the Nakaid) which have been inserted, unless they seemed quite superfluous, without any distinctive sign, I have found but very few notes by Robertson Smith on the portion which he had not definitely revised ; almost all of these have been marked with his initials, Wright's own text has been altered in a comparatively small number of passages (for instance § 252, § 358), where I felt sure that he would have done it himself. Once or twice Wright hhaa noted on the margin “ wante revision.” ‘The notes bearing upon the Comparative Grammar of the Semitic languages have for the most part been replaced by references to Wright's Comperative Grammar, published after his death by Robertson Smith (1890). Thave to acknowledge my obligations to Mr Du Pré Thornton, who drew my attention to several omissions, But my warmest thanke must be given to my dear friend and colleague Prof. Bevan, who has not only taken upon himeelf all the trouble of seeing this revised edition through the press, but by many judicious remarks has contributed much to the improving of it. The Second Volume is now in the printers’ hands, M. J. pe GOEJE. Levoex, February, 1896. PREFACE TO THE SECOND EDITION. A, SECOND ition of my revised and enlarged translation of Caspari's Arabic Grammar having been called for, I have thought it my duty not simply to reprint the book, but to subject it again to a thorough revision. In fact, the present is almost a new work ; for there is hardly a section which has not undergone alteration, and much additional matter has been given, as the very size of thie volume (351 pages instead of 257) shows. In revising the book I have availed myself of the labours of ‘Arab Grammarians, both ancient and modern. Of the former I may mention in particular the "Aliya (4:03) of ‘Tbn Malik, with the Commentary of Ibn ‘Akil (ed. Dieterici, 1861, and the Beirat edition of 1872); the Mufaggal (aici) of ‘el-Zamahsari (ed. Broch, 1859); and the Lamfyatu ‘L'Af& (UST Ha) of ‘Ton Malik, with the Commentary of his son Badru ‘d-din (ed. Volek, 1888). OF recont native works I heve diligently used the Afibahu ‘LTakb fe Baht -Matalid (Qh ST tos (5 aot ci , that is, the Bahtu ‘l-Mafalib of the Maronite Gabriel Fabs, with the notes of Bufrus ‘el-Bistani (Beirdt, 1854); ‘el-Bist&ni emaller Grammar, founded upon the above, entitled Mijtahu "LMigbah , second edition, Beirit, 1867); and Nagif ‘el Vasigi's Faglu ‘1-Higad (ptheli 25, second edition, Beirat, 1866). ‘Among European Grammarians I have made constant use of the works of 8. de Sacy (Grammaire Arabe, 2de éd., 1881), Ewald (Grammatica Critiea Lingus Arabic, 1881-83), and Lumsden (A Grummar of the Arabic Language, vol j., 1818); which last,x PREFACE TO THE SECOND EDITION. however, is based on the system of the Arab Grammarians, and therefore but ill-edapted, apart from its bulk and rarity, for the ‘use of beginners. I have also consulted with advantage the grammar of Profesior Lagus of Helsingfors (Lirokurs i Arubiske Spriket, 1869), But I am indebted above all to the labours of Professor Fleischer of Leipzig, whose notes on the first volume of De Sacy’s Grammar (as far as p. 359) have appeared from time to time in the Berichte der Kénigl. Sdchsischen Gesellschaft der Wissenschaften (1863-64-66-70), in which periodical the student will also find the treatises of the same scholar Ueber einige Arten der Nominalapposition im Arabischen (1862) and Ueber das Verhiltnisa und die Construction der Sach- und Stoffwérter im Arabischen (1856). Tn the notes which touch upon the comparative grammar of the Semitic languages, I have not found much to alter, except in matters of detail. I have read, I belicve, nearly everything that has been published of late years upon this subject—the fanciful Tucubrations of Von Raumer and Raabe, as well as the learned and scholarly treatises of Néldeke, Philippi, and Tegnér. My standpoint remains, however, nearly the same as it formerly was. ‘The ancient Semitic languages—Arabic and Athiopic, Assyrian, Canaanitic (Phoonician and Hebrew), and Aramaic (so-called Chaldee and Syriac)—are as closely connected with each other as the Romance languages—Italian, Spanish, Portuguese, Pro- vengal, and French: they are all daughters of a deceased mother, standing to them in the relation of Latin to the other European languages just specified. In some points the north Semitic tongues, particularly the Hebrew, may bear the greatest re- semblance to this parent speech; but, on the whole, the south Semitic dialects, Arabic and Asthiopic,—but especially the former, —have, I still think, preserved a higher degree of likeness to the original Semitic language. The Hebrew of the Pentateuch, and the Assyrian®, as it appears in even the oldest inscriptions, seem © As regards Assyrian, I rely chiefly upon the well-known works cof Oppert, Sayce, and Schrader. PREFAOE TO THE SECOND EDITION. xi to me to have already attained nearly the same \- matical development (or decay) as the Saeaeiiaietan spoken language of mediseval and modern times. TU have to thank the Home Government of India for con tributing the sum of fifty pounds towards defraying the expenses of printing this work; and some of the local Goveraments for subscribing for a certain number of copies; namely, the Govern- ment of Bengal, twenty, and the Home Department (Fort William), twenty-five; the Government of Bombay, ten; of Madras, ten ; and of the Punjab, sixty copies. My friend and former school fellow, Mr D. Murray (of Adelaide, S. Australia), has also given pecuniary aid to the same extent as the India Office, and thereby laid me, and I hope I may say other Orientalists, under a fresh obligation. Professor Fleischer of Leipzig will, I trust, look upon the dedication as a mark of respect for the Oriental scholarship of Germany, whereof he is one of the worthiest representatives; and as a slight acknowledgment of much kindness and help, extending over a period of more than twenty years, from’ the publication of my first work in 1852 down to the present year, in which, amid the congratulations of numerous pupils and friends, he has cele- brated the fiftieth anniversary of his doctorate. W. WRIGHT. Caxpriver, Let July, 1874,: CONTENTS. PART FIRST. ORTHOGRAPHY AND ORTHOEPY. 1 Taw Lavvens as Comsomayts ©. IL. Tue Vowews awp Diratmonas. © 1 ee ee 7 IIL. Organ Ontuoomarato Sows... ww ee A Goma or Sukan 2 ee IBY Toate or OeMa eee C. Hémzs or Nobra rs soe 16 D. Wasla . . 19 B Madda or Matta 4 IV. Tw Same 2. ee 36 a7 38 PART SECOND. ETYMOLOGY OR THE PARTS OF SPEEOH. L THE VERB. A. GENERAL VIEW. 1, The Forms of the Triliteral Verb. 39 ‘The First Form : 30 ‘The Second Form a1 ‘The Third Form 38 Evy ‘The Fourth Formxiv sae L CONTENTS. ‘The Fifth Form. boo oO he ehth Oer ‘The Sevonth Form . oe ‘The Eighth Form oo ‘The Ninth and Hloventh Forme | ‘The Tenth Form The Twelfth, Thirteenth, Fourteenth, and Fifteenth Forms . ‘The Quadriliteral Verb and ita Forms. 9. ‘The Voices . 5600 ¢ ‘The States (Tenses) of the Verb D0 ‘The Moods. bo: ‘The Number, Persons, ond Genders |. B. THE STRONG VERB. The Active Voice of the First Form. - - + - . The Inflexion by Persons. . so 1, Separate Pronouns. 2. Sufled.Pronouna expressing the Nom 3. Prefixed Pronouns, expressing the Nominati b. Forms of the Tenses and Moods ‘The Imperfect Indicative . a ‘The Subjunctive and Jussive . - ; ‘The Energetic. + ‘The Imperative. - ‘The Pamive Voice of tho Fiat Form. ‘The Derived Forms of the Strong Verb ‘The Quadriliteral Verb. ‘Vorbe of which the Second and Third Radical ar Tdeotioa ©, THE WEAK VERB. Verba Hommta Yerbe which are moro eopecially called Weak Verba ‘A. Verba Primm Radicalis 5 ot us go 5 B, Verba Medim Radicalis 9 etys ss ©. Verba Tortiso Radicalis 9 et S ca Vorbs that are Doubly and Trebly Weak. Doubly Weak Verbs. 5 ‘Trebly Weak Verbs eee sarees 0 on a 68 8 81 ~~ CONTENTS. v Appenoix A. “ [i iheveho e 96 Il. Tho Verbs of Praise and Blame BO 97 TIL ‘The Forms exprossve of Surprise or Wonder . 98 Aveenpir B. ‘The Verbal Suffixes, which express the Accusative... 100 IL THE NOUN. A. THE NOUNS SUBSTANTIVE AND ADJECTIVE. 1. The Derivation of Nouns Substantive and Adjective, and ther different Forms. Boo 108 ‘&. Tho Deverbal Nouns. (@) The Nomina Vebi 9... 1 eee MO (B) The Nomina Vicis Se {y) The Nomina Speci . . . . 1 e+ 188 (8) The Nomina Loci et Tomporis . . . . «124 (©) The Nomina Instrumenti - 10 (© The Nomina Ageotis ot Patontis and other Verbal Adjectives 2 ee ee TD b. The Denominative Nouns. (@) The Nomina Unitatis a uy (6) The Nomina Abundentio vel Multitudinis | |. 18 () The Nomina Vass. oe (®) The Nomina Relative or Relative Adjectives |. 149 I. Changes of the Ausilinry Consonants =. ©. 161 IL Ciange of the Final Radicals 9 sod oS 188 IIL Changes in the Vocalisation ne) (@) The Abstract Nouns of Quality . +. 166 (©) The Diminutive . . . . ee ws 188 (q) Some other Nominal Forms. |. 175 2 The Gender of Nouns . Be ir Formation of the Feminine of Adjectives. fee 188 Forms which aro of both Genders 5. ss. 188, 3 The Numbers of Nous. 2... ee ee BT The Dol 2. eT The Pluralis Sanus. wwe 198 ‘The Pluralis Fractus Se 188)xvi" CONTENTS. nox 4. The Declension of Nouns : i 1. The Declension of Undefined Nouns. = Diptotes | IL. ‘The Declension of Defined Nouns ur Ars Provominal Buflse, which denote the Genitive 3a B. THE NUMERALS. 1. ‘The Cardinal Numbers - i 3% ‘The Ordinal Numbers . a 3. The remaining Clase of Numerals (©. THE NOMINA DEMONSTRATIVA AND CONJUNCTIVA. 1, ‘Tho Demonstrative Pronouns and the Article 2% Tho Conjunctive (Relative) and Interrogative Pronouns (@) The Conjunctive Pronouns . 5 : (@) The Interrogative Pronouns 3, The Indefinite Provouns * IIL THE PARTICLES. ‘A. THE PREPOSITIONS. ‘The Insoparable Prepositions ‘The Soparable Prepositions . B. THE ADVERBS. ‘The Inseparable Adverbial Particles ‘The Separuble Adverbial Particles ‘Adverbial Acousstives : ©. THE CONJUNCTIONS. ‘The Inseparable Conjunctions ‘The Separable Conjunctions . D, THE INTERJECTIONS PARADIGMS OF THE VERBS 270 370 a4 370 e228 PART FIRST. ORTHOGRAPHY AND ORTHOEPY. 1 THE LETTERS AS CONSONANTS. 1. Arabic, like Hebrew and Syriac, is written and read from A Fight to left. ‘The letters of the alphabet Wg Bye, Gast ret, BOY Sacti, or pk Sat) are twenty-cight in id are all consonants, though threo of them are also used see § 3). They vary in form, acoording as they are con- nected with a preceding or following letter, and, for the most part, terminate in a bold stroke, when they stand alone or at the end of a word. The following Table gives the letters in their usual order, along with their names and numerical values. FIGURE. ‘Connected. [wowenicat| VALUE,2 Parr First.—Orthography and Orthotpy. 1 lwommercat} mame, — | Ynoom ‘VALUE. nected. i) Re 2 a 200 i ay a Lae 1 be Sin. o e - Sj 60 Ser Sin. e e 2 + 300 Sle Sad. w we 2 2 90 3S Dad. we ry 2 a 800 se Ts. » a » a 9 305 Za. & & s & 900 oe ‘Ain. é t . a 10 Sb Gain. |g & é a 1000 3G Fa 3 a 4 a 80 Su Ket 3 q i a 100 SS Kat 2 4a |e 6S 20 29 lam | J J ’ 4 30 ane Mim. | # ” * «6 40 si Nan. o o 3 a 50 sik Ha. ‘ “ . vt 5 3i3 Waw. ’ ’ 6 1G Ya “ we t e 10 §1] I. The Letters as Consonants. 3 Rew. a, | in connection with a preceding .) forms the figures A ¥, 9, %. This combination called laméliy, and is generally reckoned @ twenty-ninth letter of the alphabet, and inserted before (g- The object of it is merely to distinguish 8lif as the long vowel @, § 3, from élif as the spiritus lenis (élif with hémea, 1, § 15). Rew. 6. The order of the letters ¢ and 9 is sometimes inverted. The Arabs of Northern Africa arrange the letters in a different sequence ; viz. EewwvusdIBbissoer error! SIs* Fes ‘They distinguish 5 from J by giving the former a single point below, and the latter one above, thus: 9.4 f, but $4 £* At the ond of a word these points are usually omitted, A, vs. ‘Rew. c. In manuscripts and elegantly printed books many of the letters are interwoven with one another, and form ligatures, of which the following are examples. Oh. = sh. bie é af dhe Ly. ¢ = Gh £4 atl Ih. em afk uh [These ligatures, in which one letter stands above another, are very inconvenient to printers, especially when, as in this book, English and Arabic are intermingled ; and most founts have some device to bring the letters into line. Thus ee appears as assim, or, in the fount used for this grammar, ss gem. The latter ‘method is a recent innovation, first introduced by Lane in his D Arabic Lexicon, and its extreme simplicity and convenience have caused it to be largely adopted in modern founta, not only in Europe but in the East. But in writing Arabic the student ought to use the old ligatures as they are shewn in Mss. or in the more elegant Eastern fount.) * This is not confined, in the earliest times, to African Mss. In some old Mss., on the other hand, & has the point below, 2, a, or even Bh4 Parr Frnsr.—Orthography and Orthoépy. $1 Rew. d. Those letters which are identical in form, and distin- guished from one another in writing only by the aid of the small dots usually called diacritical points (4643, plur. Li), are divided by the grammerians into ses, the loose or free, i.e. unpointed, letters, and Lom roii Gaymii, the bolted or fastened, i.e. pointed, letters, ‘To the former class belong 7, 3, 3, us ues & And gto theater @, 3,4, ss, Band g. The letters op, , Sand cg are generally distinguished as follows: we is called iss a the 4 with one point (3); BG a UEITIG, the » with two points above (3); So Mead See BELT AGH, the » with t100 pointe below (2)*; 2 EST AST, the » with three points (3). ‘The unpointed letters are sometimes still further distinguished from the pointed by various contrivances, such as writing the letter in a smaller size below the line, placing a point below, or an angular mark above, and the like; so that we find in carefully written manuscripts & ECTS Sai ee A ew si bs E* G5 ote. Also ¢ or * by way of distinction from 3, In some a Mes, .5 has only one point above, and then .» takes a point below. Rum. « The letters are also divided into the following classes, which take their names from the particular part of the vocal organs that is chiefly instrumental in producing their sounds, HALT Spe or edi, the labials (414 @ lip), 3» 9. ET Geol, the gingivals, 4 $ &, in uttering which the tongue is pressed against the guia ( che SEOST yp 0i, the si with the tip of the tongue ( * [With final cg the use of the two points below is optional, Some modern prints, especially those issued at Bairit, alw. them excopt when the (¢ represents 2if makyiira (§ 7, rem. 6): we gg but ws] §2] I. The Letters as Consonants. 5 agit Sajodt or nounced with the extremity of the tongue (Gui or 3.hi). Aye bi Sepa, the letters c A US, which are uttered through the open orifice of the lips ( (padi), ab Gye or Sabi, the lettera cs > be, which are uttered by pressing the tongue against the rough or corrugated portion of the palate (abi or alc, Edt Ga5i, the lettors Gand Y, in uttering which the B uvala (3GLi) is brought into play. HOT Saye or SALT Sapahi, tho guttarals, f ceee? The letters 1 5 «gare called sbi Sy3e or ahi S. ajo, the soft letters, and asi Baym, the weak letters, 2. The correct pronunciation of some of these letters, for ex- ample ¢ and ¢, it is scarcely possible for a European to noquire, except by long intercourse with natives. The following hints will, C however, enable the learner to approximate to thelr sounds, S with Aamza (i, 1, see § 15) ia the spiritus lenis of tho Grecks, the 8 of the Hebrews (as in Tx, ON, aban. It may be com- pared with the A in the French word Aomme or English hour, wv is our b. is the Italian dental, softer than our ¢, + is pronounced like the Greck 9, or ¢h in thing. Tho Turks and Persians usually convert it into the surd s, as in sing. [In Egypt it is commonly confounded with <, less often with V».] © corresponds to our g in gem. In Egypt and some parts of Arabia, however, it has the sound of the Heb. 3, or our g in get. C: the Heb. Mm, is @ very sharp but smooth guttural aspirate, stronger than «, but uot rough like g. Europeans, as well as Turks and Persians, rarely attain the correct pronunciation of it. ¢ has the sound of ch in the Scotch word loch, or the German Racke. 5 is the Italian dental, softer than our d, 5 bears the same relation to > that does tom. It is sounded D6 Pant First.—Orthography and. Orthoépy. (2 A like the 8 of the modern Greeks, or th in that, with, ‘The Turk Persians usually convert it into s [In Egypt it is sometimes z but oftener d.] ‘2 is in all positions » distinctly articulated lingual r, as in run, 3 is the English =. ‘> is the surd o in sit, mist; US, sh in shut. G+, the Heb. pis strongly articulated «, somewhat like as in iss, BV is an aspirated ¢, strongly articulated between the front part of the side of the tongue and the molar teeth (somewhat like #4 in this), ‘The Turks and Persians usvally pronounce it like 2. [In Egypt it is an emphatic d, without aspiration, more difficult to an English tongue than the true Bedouin .6.] ‘, the Heb. 0, is » strongly articulated palatal ¢. & bears, strictly speaking, the same relation to b that < and do toc and ». It is usually pronounced like a strongly articulated palatal z, though many of'the Arabs give it the same sound as yo © [with which it is often confounded in Mes.]. ‘The Turks and Persians change it into a common z. ‘To distinguish it from wo, & is some- times spoken of as SLT sili, The 2 ia then called Aoghi7 bi. , the Heb. yy, is a strong (but to [most] Buropeans, as well as ‘Tarks and Persians, unpronouncesble) guttural, related in its nature tog, with which it is sometimes confounded. It is described as produced by a smart compression of the upper part of the windpipe ‘and forcible emission of the breath. It is wrong to treat it, in any of the Semitic languages, as a mere vowel-letter, or (worse still) ax Da nasal n or ng. 2 i8 8 guttural g, accompanied by a grating or rattling sound, a8 in gargling, of which we have no example in English. The y of the modern Greeks, the Northumbrian r, and the French r grasséyé, are ‘approximations to it*. Sis our f. G, the Heb. p, is a strongly articulated guttural #; but in parts of Arabia, and throughout Northen Africa, it is pronounced as a ee cee ens neem concen * [Honco p is sometimes replaced by y ar in the Yemenite B+ for ees, Hamdant ed, Miiller 193, 17 ete., and often in Mss.—De G.] $34) IL, The Vowels and Diphthonge. 7 hard g; whilst in [Cairo-and some parta of] Syria it is vulgarly con- A founded with lif hémzatum, as ’ultu, yo'iilu, for Fultu, yakilu, 9, J, +, and gy, are exactly our &, J, m,n, When immediately followed by the letter , without any vowel coming between them, takes the sound of m: as ce gomb, 32:5 ‘ambar, AE samba’ u, not ganb, ‘anbar, dnba’u, + ia our A. Tt in distinctly aspirated at the end, as well as at the beginning, of syllable; eg. 2 Aum, QUAI ’ahlaka, In the grammatical termination 32, the dotted 3 [called tstif la] is pro- nounced like <>, #)*. B 9 and (g are precisely our w and y. ‘The Turks and Persians usually give 9 the eound of », I THE VOWELS AND DIPHTHONGS. 3. The Arabs had originally no signs for the short vowels, ‘To indicate the long vowels and diphthongs they made uso of the three consonants that come nearest to them in sound: viz. | (without Admza, see § 1, rem. a, and § 15) for d, ys for § and ai, » for dand au. B.g., Va, 3 Fi, GS hai, 95 da, 9) la, 4. Atallater period the following signs were invented to express the short vowels. (a) = foth (cH or fotha (dat5), a, 2 (a8 in pet), # (nearly the French « mut); 0g. Gide halaka, je idmsun, 4% kerimun. (0) > Rar (5-&) or Bera (F-S), i (as in pin), 3 (a dull, obscure i, resembling tho Welsh y, or tho #in bird); og. a biki, Lal afsfum. =D (©) 2 damm (25) or damma (425), u (as in bull, 0, 3 (neatly as 3 in Mértel, or the French eu in jeune); 0g. 2 laku, GDe haggetun, 522 ‘imrun. * In point of fact, this figure 3 is morely » compromise betwoon. the ancient 2 (Heb. N=, Nz), the old pausal ¢= (oA), and the modern + = (Heb. 77), in which last the ¢ is silent. .8 Parr Finst.— Orthography and Orthoépy. {55 A Raw. @. The distinction tetween the names féth, keer, damm, and fétha, k2era, damma, is that the former denote the sounds , 6, x the latter the marks =, ., 2. Compare tho Hebrew MMB, 2W, and Paap, The terms .~ai and Jj, commonly used of tho cxso- endings a, u, are sometimes applied to eT ay By y3c5 (Another name for damm is kabw, 525 aoe ¢, Sarr Rem. 5. A vowel is called 4550, a motion, plur. SiS; ; ite Aor Jykd. Rea. c. In tho oldest Mes. of the Kor’én, the vowels aro expressed by dots (usually red), one above for fetha, one below for késra, and one in the middle, or on the line, for damma, As re- gards the signs <, the third is a small y and the other two are probably derived from | and «4g or = respectively. 2 in othor positions; og. 5. Rules for the cases in which these vowel-marks retain their original sounds, a, i, u, and for those in which they are modified, through the influence of the stronger or weaker consonants, into 2, ¢, ©}, 0, or d, can scarcely be laid down with certainty; for the various dialects of the spoken Arabic differ from one another in these points; and besides, owing to the emphasis with which the consonants are uttered, the vowels are in general somewhat indistinctly enunciated. ‘The following rules may, however, be given for the guidance of the learner®, (a) When preceded or followed by the strong gutturals ¢ ¢ g 8, or the emphatic consonants Go us & & G, fetha is pronounced as a, though with the emphatic consonants ite sound becomes rather obscure, D approaching to that of the Swedish 8; e.g. pod Aamrun, 35 la‘bun, GH baRys, gddrun, Under the same circumstances késra is + [Learners whose ears and vocal organs are good, and who have an opportunity of hearing and practising the correct pronunciation of the consonanta, will find that the proper shades of sound in the three vowels come without effort when the consonants are spoken rightly and naturally. ‘The approximate rules for pronunciation hore given are mainly useful as a guide towards the right way of holding tho mouth in pronouncing the consonants as well as the vowels.) § 6] Il, The Vowels and Diphthongs. 9 pronounced as 3, eg. Je ‘mun, jam Sifrun, 325 Rérun; whilet A amma assumes the sound of an obscure o, inclining with the gutturals (especially ¢ and ¢) to 5; eg, Aki lafafa, ALLS lopfum, Ste boomum or fisnun, 8) ro‘bun, p28 ‘imrun, (&) In shut syllables in which there are neither guttural nor emphatic consonants,—and in open syllables which neither commence with, nor immediately precede, one of those letters,—fetha either has 8 weaker, less clear sound, approaching to that of a in the English words hat, cap, eg. Suid kutabta, j81 ‘akbaru; or it becomes a B simple 2 or ¢ (the latter expecially in short open allable followed by a long one), 6.8, J 8a, «ps mirkabun, Shes simdtun, Se seminun, 125 medindtun. It retains, however, its pure sound of « before and after r (which partakes of the nature of the emphatics), when that letter is doubled or follows a long @ or d, e.g. a: garratun, 52 marratun, 5 garatun, f5-0 siratun ; and also in general at the end of a word. . The long vowels a, i, a, are indicated by placing the marks C of the short vowels before the letters 1, (g, and, respectively, eg. SW bila, Bee B70, So sitkun ; in which case these letters are called SBT Soja, titeras productionis, “letters of prolongation.” ‘The com- binations (and y+ must always be pronounced i and @, not 4 and 4; though after the emphatic consonants y+ inclines to the sound of 3, and i to that of the French w or German i, 0g. 9, Gel, nearly forun, tinun, Rew, a, @ was at first more rarely marked than the other yy long vowels, and hence it happens that, at a later period, after the invention ofthe vowel: points it was indioated in some very common words merely by a fathe i Si, Kali, & exactly, however, the fath this case, so as to resemble og thi, Goel earl, Salt, Gazi, Us, 15, GA, Wisi More should ‘be written, perpendicularly in tog Bi, uta LS, Sy, ac2ali (the renurrection, to be carefully distinguished ‘trom ®. 210 Part Firat.—Orthography and Orthoépy. &7 A Keli wpimatu, price, oahu), Gy, Sub. ‘The words 395, 1554, , 7 ively £45, 13 and GSI, are also frequently written defectively 15, 333, GsHlS; and occasionally some other vocables, such ss dpi} and GS; SpE and LS; Gok, Dds, and other proper names ending in Giz; Gbehil; yas; Sut, Jus, 405i ii, ana other proper names of the forms Jeti and Jetiif; S435 ote. This is more common in Magribi Mes than in others —Tho long vowel Tis in a vory few instances written defectively at the end of a word, ong, Hadi, WHafs, otal, 0:20, Aol Go Mladen, Hodeifetu mu LYemant, for ial, gota, lea; waSi tor wohl. Ram, 5. Tho letter o, preceded by damma, is used by the Arabs of North Africa and Spain to indicate a final o in foreign words ; e.g. ASI, Carlo ; pba cgi, Don Pedro; 131 gly, the river Guadiaro, Rex. ¢, The sound of te inclines, in later times and in certain localities, from @ to 2, just os that of fetha does from @ to 2 (seo § 4, a, and § 5,6). This change is called ai i Urimala, the deflection” of the sound of « and 4 towards that of i andi Tho ‘Mogribt Arabs actually pronounce 4 in many cases aa i, Hence Sid, riba, 3S takin, DY ba, E15) liadn, aro sounded rikib, Takin, bb, lisin ; snd, conversely, the Spanish Beja, Jaen, Caniles, Lebrilla, are written ami, ters Ai ‘7. 1 corresponds to fetha, «g to ktera, and 9 to amma; whence D Vis called an$iit LL, the sistr of fatha, vs, sai eal, the sister of Hawa, and 9, 22407 Call the sistr of damma. Fetha before ug anc '5 forms the diphthongs ai and av, which retain thoir original clear ‘sound after the harder gutturals and the emphatics, 0.g. hee saifun, EE © [The omission of final gg in those cases is hardly mere ortho. graphical irregularity, but expresses a variant pronunciation in which the final § was shortened or dropped. See Naldeke, Geach. d. Qordn's, p. 261) _——— $7) Il, The Vowels and Diphthongs. n SL haufun; but after the other letters become nearly # (Heb. *=) A and 6 (Heb. §-), e.g. Cis ad¥fun, 252 m'otun (almost sifun, motun). Rex, a, After y at the end of a word, both when preceded by damma and by f&tha, | is often written, particularly in the plural of verbs ; e.g. Igya5, Iyay, '9j8). This |, in iteclf quite superfluous (Qlif otiosum), is intended to guard against the possibility of the preceding being separated from the body of the word to which it belongs, and so being mistaken for the conjunction 3 and, Tt is called 531 Gul, the guarding dif, or Dott GNI, the separating Ram, 6. (g at the ond of a word after a fetha is pronounced B Tike |, e.g. cg fatd, ge) rama, Jl da", and is called, like | itself in the same position (0.g. Lp Belinesd, Yik jaza), dyyak St BS the by that oan be abbreviated, in contradistinction to the lengthened Mis, Big ASI (see § 22 and § 23, rem. a), which is protected by haémza. It receives this name because, when it comes in contact with a Aémza conjunctionis (see § 19, rem. ), it is shortened in Pronunciation before the following consonant, as are the g and (gin gil and cyl before zes3hT (soe § 20, &)t. Re. ¢ If a pronominal suffix be added to » word ending g in .,<, the (g is sometimes retained according to old custom, as in 4¢2) OF 4eay, but it is commonly changed into |, as ol4;. * [But (=, with the mark gézma (s00 § 10), a8 in 5, (35 the diphthong ai.) ‘The diphthong ai, whon final, is often marked i old Mss, by the letters , whence the later 2, instead of the common £ or 4. In some old Mes. of the Kor‘én a small horizontal (red) stroke is used, 10. g anil», when they form a diphthong with Pthe, are marked with a geama, as Joi, 252, G, Codd; but when they stand for dif productionis they do not take this sign (eee § 7, rem. 6, c,d). Rew. In many nfnuscripte a gézma is placed even over the letters of’ prolongation, eg. J wes; and over the élif raakalra, 0g. ld, (Sa for gle, gah B. Tesdid or Sadda. 11. A consonant that is to be doubled, or, as the Arabs say, strengthened (334), without the interposition of a vowel (see rem. a), is written only once, but marked with the sign *, which is called14 Pant Finst.—Orthography and Orthoépy. gu A Sea83i, the atid (orentining al Meiulla, $2 kullan, A tevnmi, 7 amnin, PO temurra, 5 murrun. It corresponds ‘therefore to the Daghesh forte of the Hebrew. Rew, The solitary ‘exception to this rule, in the verbal forms gsh rtwita and gpd ubicta, instead of gS and Jyh, admits of an easy oxple (oee § 159).—When consonant is repeated in such @ manner that a vowel is interposed between its first and second occurrence, no doubling, properly so called, takes place, and consequently the tédid is not required ; eg. 5; mato, Port. of 33; S345, 3d pers. sing. fem. Perf. of the fifth form of J5. Raw. 5. A consonant can be doubled, and receive tdédtd, only when # vowel precedes and follows it, The cases treated of in § 14 form no exception to this rule Rew. c, All consonants whatsoever, not even lif bémzatum excepted, on of boing doubled and take thidid. ‘Hence we speak and write ok ra”asun, ok ea" dun, Pagun, Rew, d. + in an abbreviated ¥%, e first radical of the name Saad, or the firet lettor of the name 455, which the African Arabs ‘use instead of the other. Orit may stand for wi (from 3.24), since in the oldest and most carefully written manuscripts its form is. Tea opposite i war, i.e, wide (from GAL tightened, single); 0. we EM pe evoretty and openly Rem. ¢. TeSdid, in combination with 4,4, is placed be- tween the consonants and these vowelmarks, as may be seen from the above examples, In combination with -.the Egyptians write ‘4 instead of $5 but elsewhere, at least in old manuscripts, 7 may stand for 2 as well as 2. The African Arabs constantly write &, 3, % for 2, 7,4, In the oldest Mes. of the Kor'an, taédid is expresso] by ~ or v, which, when accompanied by ktera, ie some- times written, ae in African Mss., below the line In African. ‘Men. tho vowol is not always written with the éidda ; & alone may be = 4, do. * [The nomen unitatis is 3ayat3.—De G.] § 14] III. Other Orthographic Signe. DB. Teédid or Bddda. 15 12, ‘Teédid in either necessary or euphonic, 18. The necessary (26did, which always follows a vowel, whether af ne short (as in GLE) or long (as in 314), indicates a doubling upon which the signification of the word depends. ‘Thus yal (Amara) means he commanded, but jo! (ammara), he aygninted some one commander ; B (murrun) in bitter, but a word 5 (murun) does not exist in the language. Rem, The Arabs do not readily tolerate a syllnble containing a long vowel and terminating in a consonant. Consequently ¢2idid “necessarium scarcely ever follows the long vowels y and as in ShT 543, though it is sometimes found after |, as in SUR, (see § 25) Nor does it occur after the diphthongs $< and 5<,save in rare instances like Layyde, dayg5.and Sezss [see § 277]. 14, ‘The euphoric @édid always follows vowelless consonant, which, though expressed in writing, is, to avoid harshness of sound, passed over in pronunciation and assimilated to a following consonant, It is used (a) With the letters , 2, 9,3, 515, oh Ut Ge, 8, b, &, Us &» Gentals, sibilants, and liquids,) after the articta Ji; e.g. peal at-mru; Ssemegh “ar-rabmanu; Gabi sims ; Lbs ‘ag-golmu ; Sehi az-itu, or, in African and Spanish manuscripts, Jai, Rex, a These lottors are called S387 ype, the solar Lettere, because the word Gcue5, oun, happens to begin with one of thom ; and the other letters of the alphabet £487 Saya, the Tunar letters, because the word 545, moon, commences with one of them, Tew, This woimation ia extend by ome to the J of J and J, especially bofore y, as 2 Ja. (® With the letters 5, J, *, 9, os, after m with gam, og. Jd coe, JEG oF; and after the nunation, e.g, eT Ssegt pee Fitdbum mubinun, for ditaun mubinun, The n of the 6 coe mir radi A16 Pant Finsr—Orthography and Orthogpy. ($18 ‘A words Gus, cb2, oi, is often not written when they are combined with wT; eg coe for Sse or Ses ce, US for Uw, J for Tol. Raw, a. If to tho above lsttera we add cy itself, as C28 gf, the mnemonic word is Gpheyz- Rew, 5. J ois equally common with Si, but Cie, Ste, ey U5, are hardly over written separately; J cea, on tho contrary, always, Similarly we find i for Sop if not, Ut for Laut (; with B redundant la) and occasionally Usl for Lacs} (that, with redundant te). (© With the letter © after , 2, 3, 48, b, & (ental), in 215 pitty for Zt5 bite; Ssh certain parts of the verb; &. a weg on seratta for SM aradta ; 3551 ‘attakattum for SiH attabadtum ; “tbe basttum for 2384 basaitum. Many grammarians, however, Toject this kind of assimilation altogether, aud rightly, because the Abeorption of a strong radical consonant, such as a, 2 or b, by @ © weaker servile letter, like <, is an unnatural mutilation of an ess part of the word. Rem. a. Still more to be condemned are such assimilations as 3b for 338, Bid for Edad. ‘Rem, 6. If the verb ends in ©, it naturally unites with the nocond in the above cases, 0 that only one < is written, but the indicated by the t&sdtd ; as 223 for E23. union of the two i ©. Himza or Nebra. D. 18. Biif, when it is not a mere letter of prolongation, but a con- sonant, pronounced like the spiritus lenis, is distinguished by the mark t Kamza (Jab or Sj, compression, vit, of the upper part of the windpipe, see § 4, rem. a), whick is also sometimes called ndbra & leation); 0-8. SA, Se, 45, SA, ih, Sabi, ae, J, tea. Rem, a. In cases where an if conjunctionis (seo § 19, a, b, ¢, and rem. d, ¢) at the beginning of s word receives its own vowel, ‘the grammarians omit the hémza and write merely the vowel ; e.g. hs 525i praietelonge to God, 1, Sit, § 17] IIL Other Orthographic Signe. C. Himsa or Nebra. 17 Rex. &. £ is probably a small p, and indicates that the Bf is to A be pronounced almost as ‘ain, In African (and certain other) Mas, it is sometimes actually written $; og. $1, Ghat In the oldest ’ ‘Mas, of the Kor’sn, hémza ia indicated by doubling the vowel-points; 28 OTM GBI, oyiegell = Gategell, It is also marked in such Mss. by a large yellow or groen dot, varying in position accord- ing to the accowpanying vowel (see above, § 4, rem. ¢). Rina. ¢, Hémza is written between the 1 and the vowel that accompanies it, or the gezina (see the examples given above) ; but B we often find Getelé for Semld, x” for pos (see § 16), and occa- sionally (ini or Und. for thin, Si or Gi for Sy, JE tor Je or Je, and the like. i ° Rex. d, The effect of the hémza is most sensible to a European sensi mm ear at the commencement of a syllable in the middle of a word, precoded by a shut syllable; eg BU, masalatun (not ma-satatun) ONGH, Akor-ann (not sko-ranu). 16. (5 and 4 take htmza, when they stand in place of an eif0 Amzatum* (in which ease the two points of the letter (g are com- monly omitted); e.g. Coe for Sle, Getald for Gull, iy) for ete obey od 7 + ae wl, way for wah 217. Hema alone (+) is written instead of f, 1, (%, 5, in the fol- lowing cases. : (@) Always at the end of a worl, after a letter of prolongation or a consonant with gama, eg. # ise Wa, D 145935 £635, tabs, or more commonly Eb (sco § 8, rem, u); and in the middle of a word, after an dif productionis, pro- vided the hi » Ditsel (but for 2258) and LS iTaEk the Arabs usually write Sse! and £33024), Hae, ayy tye, sun; wa has the vowel fétha, as Cygi' Rew, Accusatives like U2 and Ejis are often written, though * [See below, $§ 131 seq]© 18 Pant Finst—Orthography and Orthoépy. 17 5, Liss; and in old Mos. we find such instances contrary to ru 1s i, for sy (W) Frequently in the middle of words, after the letters of pro- Tongation 5 and qs, or after @ consonant with glume, og igi for Byqpha, Stange for St Sy, 2935 for iss GES for GUS ; and also after Kia anit danma before the gs and. 3 of prolongation, 6: Goeth for Getbts, th for Sasgy Homan between 14, ai, a, , anda, is, however, more frequently, though improperly, place over the letter of Prlongation itself; e. ‘i Syl for bugjde, Had for Big or Babad, Sop) for sigh ot Seah}, which wortts must always be pronounced mukra'utua, bagVatun, r'isun. Rew, a. After a consonant with gézma, which is connected with 1 following letter, hinlaa and its vowel may be placed abovo the connoeting line; a» JE, for SU Rew, b. A hime precedad by w or é, and followed by « or dy to pure 9 or 4g 8s ier for yey Spo for au —If preceded by a or i, or the be changed into 9 be changed diphthong ai, the hiinza iay like 5 for See from fuyehs Rape for deyyy, from By ? for 53, C5 for ELEAF the héman hus gezma, it may [lose its consonautal power and] be changed into the letter of prolongation that is how gencous with the preceding vowel, ns uly for 7 os for Bh Jat for 25 necesnsily wy if the procding vonsonant bo wn éif with hemun, as Gall or Gol, Sesh Shap for all, ooh EGS, [This is called Soll Sight. Teen, ¢ ‘The name 39315 oF 3gil3, David, in often written 33, bout must nlways be pronounced Da'idn, $19] IIT. Other Orthographic Signe. D. Wasla. 19 D. Wala, so 18. When the vowels with hémza (1 11), at the commencement of A ‘a word, are absorbed by the final vowel of the preceding word, the ion of the spiritus lenis is marked by the sign =, written over the if, and called JLo3, or 2155, or Ale (see § 4, rem. a), i.e. union eg. MUST Se ‘abd Dmaliki for SUS raditu "bnaka for So} Sh ratitu ‘ibnaka, Rew. a. 7 scoms to be nn abbreviation of ye in Jog or XLa ; or rather, it is the word Ze itself. In the oldest Mas. of the B Kor'n the wagl is indicated by a stroke (usually red), which some- times varies in position, according to the preceding vowgt. In ancient Mugribi Mss. the stroke is used, with a point to indieate the original vowel of the elided tif; e.g., aL ~ igh i edi: 2 inet, ie _ Hence even in modern African Mss. wo find TPT L instend of the usual 7, Rew. 6, Though we have written in the above examples 431 and Ai, yet the student must not forget that the more correct orthography is SUG and JLj1. See § 15, rem. a, and § 19, rem. d. 19. This elision takes place in the following cases. (a) With the 4 of the article Jf; as 25931 g/l for yejgii gil, the Sather of the wrzir. “ = (8) With the 1 and 1 of the Imperatives of the first form of the regular verb; as oat JU for 2551 JU, he said, listen ; JEST Jub for D Si IU, he said, Kill. (©) With the | of the Perfect Active, Imperative, and Nomen actionis of the seventh and all the following forms of the verb (eee § 35), and the | of the Perfect Passiv aig for js! 54, he was put to flight ; Baily for eit and he wus appointed governor; j1a8H the being able (to do something) ; aba Si ill the dnonfull ot extinction,20 Parr Firen—Orthography and Orthoépy. $19 A (a) With the J of the following eight nouns: SA, and 21 or 28), a son. S41, « daughter. . oh two (mase.). Bs or So Eat, the anus. aman. ip, Je (rarely sl), a name Raw, a. With the articlo $21 and Hija) take, in classical B — Arabie, the form Jail and 3, Rew. &, ‘The hamza of csetl, oaths, is also clided after the assoverative particle Jj, and occasionally after the prepositions A and J (which then takes fBtha instoad of gdzma); a8 bi by God (it. by ta onthe of Gedy for which we may ‘alo write ahi G25) omitting the 1 altogether, or, in a contracted form, ai Rem. c. In the above words and forms, the vowel with hémza in in part original, but, has beon wenkened through constant uso (as the article, and in cad after J); in part merely prosthetic, that is to say, prefixed for the sake of euphony to words beginning with ‘a vowelless consonant, and consequently it vanishes as soon as a vowel precedes it, because it is then no longer necessary. See § 345. Reem. d, Tt is naturally an absurd error to write 1 at the begin- ning of a ed instead of @lif with hémza, as deed! instead ot Ztait ‘The Arabs themselves never do so, but, to indicate D__ that the dlif is an 24 conjunctionis (see rom. f), they omit the himnza and express only its aecompanying vowel, ax a Gomi. Seo § 15, rem. a, and § 18, rem, 0. Ram, ¢ In more modern Arabic the elision of the: 2if con- Junction (900 rem. f) is neglected, especially after the article, ss i, ab Sh aot AE, Anpath 33); but the gramma- brand this mm het Sats yal IS SE Cant §20) IIL. Other Orthographic Signe. D. Wasla. 21 Rew. f The dlif which takes wasla is called gesil Hf ai or yk A veal Buf or hemea conjunctionis, the connective alit; the opposite being ear i, if egjunctionis or separationia, the disjunctive alit, 20. The aif conjunctionis may be preceded either by a short vowel, a long vowel, a diphthong, or # consonant with gtama. To these different cases the following rules apply. (a) A short vowel simply absorbs the alif conjunctionis with its vowel; see § 19, 6 and c, B (®) A Tong vowel is shortened in pronunciation, according to the rule laid down in § 25; e.g uti S fi “andi, among men; 25331 oil ‘abis "Lwbciri, the father of the wizir, for fi and ‘aba. ‘Thin abbreviation of the naturally long vowel is retained even when the Jam of the article no longer closes the syllable containing that vowel, but begins the next syllable, in consequence of the elision of a following lif (either according to § 19 or by poetic license). Cat, e of these examples the ; in the other two it is an alif soparationis, but hae been changed for the sake of the metre into an alif conjunctionis, ‘The suffixes of the Ist pers. tat. way, instead of ot equally admissible. (© A diphthong is resolved into two simple vowel, accon- ing to the law stated in § 25, vi, ai into ai, aud au into ai; as ui ee ay 9B ALS shia “Lhauma, foar the people abit 52% mustasai ‘Mahi, the elect of God, for af sigas. The silent alif (§ 7, rem. a) does not prevent the resolution of the diph-22 Part First.—Orthography and Orthoépy. _—_[§ 20 Athong, a8 Glpeil tye ramil Thipdrata, they threw the stones; Ea CL fa-lamma ra'ti’n-s2gma, and after they saw the star. But 31 and gi take Mora, as dal gh or his name ; Sokal 95 if he went to meet. . (@) A. consonant with gtzma either takes its original vowel, if it had one; or assumes that which belongs to the élif conjunctioni or adopts the lightest of the three vowels, which in ite nature ap- proaches nearest to the gézma, vir. ktsra. Hence the pronouns of the Bed and 34 pers. phir. mase., 251 you, and 6 they, the pronominal suffixes of the samé pronouns, x2 yonir, you (nccus:), and sh thei them, and the verbal termination of the 2d pers. plur. mase. Perf. 13, take damma (in which they originally ended); as GogiSi8i ABI ye are the Hare; abit 72355 may God curse them! Jangit 222 ye have seen the man. The same is the cake with Js, since, from which time forth, because it is contracted for Ls. The preposition cee, from, takes C fatha before the article, but in other cases A2sra; as deat ony ‘ai, All other words ending in a consonant with fézma take A2sra ; vie. nouns having the tnwin, as cgiil Aeaee Mokammadun ’n-ndbiyu: mani "Lkaddilu ; verbal forms like the pronoun 2, as ptdSt aie as agp CdS hatalati ’r-Riomu; and particles, such a8 32, St, Ji, 35, Ja, 55, ete. D Rew. a. In certain cases where ts becomes ies (soo § 185, rem, 5) the wasl may be made either with damma or ktera, a» or eh Rew. 5. If the vowel of a prosthetic lif be damma, the wasl is sometimes effected by throwing it back upon the preceding vowel oe ane kn eek cea typist JS gett Sits 5 Wyhbst 2S ectamunu 'dhula. Rew. c. ‘The final gy of the second Energetic of verbs (see § 97) is rejected, s0 that the wasl is effected by the preceding fétha; as § 21] IIL. Other Orthographic Signe. D. Wasla 23 BGT SayBi j 1a tautriba "bnaka, and not YOM cripbi °§ la tactribani A "naka : 21. Tis altogether omitted in the following cases. (a) In the solemn introductory formula abi," for abi_pG, in the name of God, DYTONT DWI. As a compensation for the omission of the |, the copyists of Mas. are accustomed to prolong the upward stroke of tho letter wy, thus: J. @ Iw the word Sot when the name of the son precedes, and that of his father follows in the genitive ; provided always that the said series, as a whole, forms part cither of the the predicate of a sentence. For exainple, AN oe yh BE SE SRS ak SF SG 5 Zale, the son of Hid, struck Sa'd, the som uf ‘Auf, the son of ‘Abdu "lah. (Cf. §315, rem. 6.) But if the second noun be not in apposition to the frat, but form part of the predicate, so that the two together make » complete sen- tence, then the 1 is retained; as iT a5 Zoid (és) the som of © ‘Amr, pUacst Sei 528 ‘Omar (is) the am of e-Lattab. is retained, if that |, son, in a gencalogical series, that is to say, B Rex. a, Even in the first case the Fof word happens to stand at the beginning of a line, Rex. 6, If the name following cyl be that of the mother or grandfather, the T is retained ; as Jip Gul cgtee, Jesus the son of Mary; yale SHIGE, Ammar the (grand)won of Mangir, Like- wise, if the following name be not the real name of the father, but a [) surname or nickname; as 35-50 Gil Stade, Mikdad the son of lewad (the real name of 8’Amwad, “the black,” being ‘Amr, ‘gyeé). Or if the series be interrupted in any way, as by the interposition of an adjective ; 0.6: Cpeee ST nT Getty Vols Ue noble, the son of Méimin; sige Gil csipli& «gisy Ridba (pronounced like the word girba) the aun of Misa. (© In the article Jf, when it is proceded : (©) by the preposition J to, as larg 10 the man, for glaz%).% Pant Finsr—Orthegraphy and Orthotpy. [§ 22 ALE the first letter of the noun be J, then the J of the article is also omitted, as Aigh to the night, for 2c, and that for 09. (B) by the affirmative particle J truly, wrily, as Sau, tor dave, @ In nouns, verbs, and the article Jj, when preceded by the interrogative particle Ff, ie thy om? STE ie it (fom.) broken? SB3LH, for ISLET, save yo receivad? BAGH, for AST, is the water—? "the alif of the article may however be retained, so that 4,1 with the interrogative | is often written i Rem. a, In this last case, according to some, when the ead Gif has fétha, the two élifs may blend into one with middda (see below); a Soke GLH, ia &Husan in thy house ? for (mail; CAE S258, ke of Kort on he of Tabi? Legs abi Sati, iv thy oath ‘by God"? (aoe § 19, rem. 6) for abil Cnet c Rew. 5, The prosthetic Bf of the Imperative of Jt, to ask is frequently omitted, in Mss. of the Kor'an, after the conjunction 5; 08 JUS, for JUG, [oF § 140, rem. a.) for E. Madda or Matta. 22, When lif with hemea and a simple vowel or tonwin (|, etc.) is preceded by an élif of prolongation (I=), then a mere hémza is written instead of the former, and the sign of prolongation, = madda or matta D (34, 355 or AE:, i.o. lengthening, extension), is placed over the Inter; eg. 105 sema’un, she gaa, SCS yatasd'atina, for fc, ie Oss Rem. a. As mentioned above (§ 17, a, rem.), we find in old tee ke Ps ‘Mes, such forma as iG, iiyy, for slp, yy. Rem. 6. In the am and best Men, to form of the médda is . [Note also the cases, in poetry, cited in § 958, rem. ¢ further the contracted tribal names sais), cyl for gait 94%, Soe 94 —DeG.} § 28] IIL. Other Orthographic Signs. E, Médda or Matta, 25 this is but rarely written. In some old Mes. of the Kor’an médda A is oxpressed by a horizontal yellow line = 23. When, at the beginning of a syllable, an alif with hémza and fétha (1) is followed by an elif of prolongation or an lif with t hhomea and gézma (1), then the two are commonly represented in writ ingle alif with médda; e.g. St} for ML, Gghet for Split, acts a * for Gait (see § 17, rem. B). In this ease it is not usual to write either the hémza, or the vowel, slong with the médda. [But wo some- B times find &, see §174.] Raw, a. Tis called Ssgaeeil GNI, the lengthened or long Rif, in opposition to y-akt GAN, the Rif that can be abbreviated or shortened (§ 7, rem. b). Rex. 6. Occasionally a long élif at the beginning of a word is written with héinza and a perpendicular fetha, instead of with mada (see § 6, rem. a); © instead of ai or ath Rem. c. Médda is sometimes placed over the other letters of © prolongation, 9 and 4g, when followed by an 8lif hémzatum, only the hémza being written (§ 17, 4); a8 15%, \igmea. Also over the final vowels of the pronominal forms 23, 4, @ 0 and the verbal termination U3, when they are used a8 Tong in poetry; &g. Bi, os Riu, d. The mark +, often written over abbreviations of words, has nothing in common with médda but the form. So J for JUS, He (God) is exalted above all; 42 for Sait ale, peace be D upon him / gslleo tor as sole Abi Lo, God biees him and grant hhim peace ! aby for 23a BF 5, may Gab ‘well pleased with, or gracious to, him! Ey oF aby for ahi inj, may God have merey spon him Qh tor pick gh or Uy GS for Base, he narrated to ws lor U for scat, he informed us; E for tiym, then. —The letters » » aro written over words or Yorses that have been erroneously transposed in » manuscript, for ™ 4 Fash, fo the ond of it, i.e. ete;26 Parr Finst.—Orthography and Orthoépy. p24 A singe, t0 be placed last, and Sais, to be placed first.—On the margin of Mas. wo often find words with the lottors ¢, cy and ge over them, The first of these indicates a variant, and stands for 43, @ copy, another manuscript ; the second means that a word has been indistinctly written in the toxt, and ia repentod more clearly on the margin, lg, explanation ; the third implies that the marginal roading, and not thet of the text, is, in the writer's opinion, the correct one go, it ie correct, or Cremmal, correction, emendaton B Written over a word in the text, po stands for @o, and denotes that the word is correct, though there may be something peculiar in its form or vocalisation.—Again lee (ie. (es, together) is written over a word with double vocalization to indicate that both vowels are correct, ala) over a word on the margin implies & conjectural emendation ala), perhaps it is, IV! THE SYLLABLE. C a The vowel of a syllable that terminates in a vowel, which wo call an open or simple syllable, may be either long or short; as JE kita. 25. The vowel of a syllable that terminates in a consonant, which we call a shut or compound syllable, is almost always short ; as JB bil, not Job (Heb. by). Generally speaking, it is only in pause, where the final short vowels are suppressed, that the ancient ‘Arabic admita of such syllables as in, im, dm, ete. D Rex. Before » double consonant @ is however not infrequent (200 § 15, rem). [Such « long & preceding consonant with gézma sometimes receives a médda, as (gid) 26. A syllable cannot begin with two consonants, the first of which is destitute of a vowel, as ¢f or fr. Foreign words, which com- rence with a syllable of this sort, on passing into the Arabic lang take an additional vowel, usually before the fist consonant a8 ‘) expen; yb Th, urdu; Fi, the Fronts (Buropeans); soob, an elizir, +3 Eqpdv (medicamentum siccum). 27. A syllable cannot end in two consonants, which are not either separated or followed by a vowel (except in pause). §31) V. The Accent. 27 V. THE ACCENT. A 28. The last syllable of a word consisting of two or more syllables does not take the accent. Exceptions are : (a) The pausal forma of § 29 and § 30, in which the accent remains unaltered ; as yo 7Al, ki-n6n, mu'-mi-nfn, ka-ti-bdt, f-rind, ‘aka, ma-fér, ku-béil, billéur, bu-ndi. Rie But words ending in (>, 3» i. or HS, ty" and 46> throw back the accent far as powsible in their passa forma; Eh Koratt yun becomes Kora, dos nebtanun, nd: BS eat. wun, Geli; $Uabl "ietindun, ‘tend; pgm hamrd'u, hme ; B tyke mak-rd-tun, mék-ri; tela) bugésn, bof. () Monosyllables in com! The only exception to this rule in old Arabic is the ve enclitic 4; as 44 bi-ma, a lima, in contrast with C Uy bimd, Us timd, Soe § 351, rem. 29. ‘The penult takes the accent when it is long by nature, i.e. is an open ayllale containing a loug vowel; a8 JU ki-la, J9it ya-kt-lu, S530 fa-nt-nun, Seueye mu-mi-ni-na, Ste ka-ti-bd-tun. 30. The penult has likewise the accent when it is a shut syllable and consequently long by position; as Sod fal-bun, > df-bun, ty buir-un, het G- i-rin-dun, ost ‘a-kal-lu, ja ma-fér-D run, Sigh yo-ki-lén-na, Seed ku-bdicla, 39h, billdw-ran, <4 bu-ndi- yun, 31, When the penult is short, the accent falls upon the ante- ponult, provided that the word has not more than three syllables, or, if it has four or more syllables, that the antepenult is long by nature or position; as (36 hi-ta-ba, 336 Ki-ta-bat, \ypit ké-ta- bi, LIS kA-ti-bun, Lib pila-bun, GEE 'éi-na-ma ; WphilgS tard28 Pant Finst.—Orthography and Orthoépy. ($32 In other cases the Asati, dS bé-nt-n-hum, accent is thrown as far back as possible ; as USS ké-ta-ba-td, ME més-a-la-tun, Gili2 més'a-la-twhd, Uiead bé-sa-ba-tu-hu-ma. Rex. On deviations from these principles of accentuation, in ‘Egypt and among the Badawin, see Lane in the Journal of the German Oriental Society. vol. iv., pp. 183-6, and Wallin in the same journs], vol. xi 812-0. ata Spitta, Gram, des arab. Vul- gariialectes von Aegyp''s ‘13E0}, p 59 0qq-] B VI. THE NUMBERS. 32. To express numbers the Arabs use sometimes the letters of tke alphabet, at other times peculiar signs. In the former case, the numerical value of the letters accords with the more ancient order of the Hebrew and Aramaic alphabets (see § 1). ‘They are written from right to left, and usually distinguished from the surrounding words by 9 stroke placed over them, as saaé, 1874. This arrange- ment of the alphabet is called the "Aduged or 'Abgfad, and is con- (osherwise pronounced fis Sad Sts Gis or, as usual in North Africa: fis bL3 Sib Bins GS (he yey al ‘The special numerical figures, ten in number, have boon adopted by the Arabs from the Indians, and are therefore called Gost Si, the Indian notation. They are the same that we Buropeans make use of, calling them Arabian, because we took them from the Arabs. ‘Their form, however, differs considerably from that which our ciphers have gradually assumed, as the following table shows, Indian: {RRB eee Ambic: | oc fee OS VY AA Kup: 12 8 4 5 6 789 0 ‘They are compounded in exactly the same Way as our numerals e.g. 14¥4, 1874. PART SECOND. ETYMOLOGY OR THE PARTS OF SPEECH. ‘THE VERB, Jaa, A A. Genenat View. 1. Tho Forms of the Triliteral Verb. 4 33. The great majority of the Arabic verbs are triliteral (oe, that is to say, contain three radical letters, though quadriliteral 35 (Gets) verbs are by no means rare. 34. From the first or ground-form of the triliteral and quadri- B literal verbs are derived in different ways several other forms, which express various modifications of the idea conveyed by tho fire. 35. The derived forms of the triliteral verb are usually reckoned fifteen in number, but the learner may pass over the last four, because (with the exception of the twelfth) they are of very rare occurrence. Subs x1. Je05 VI. Jai 1. esa xu. aid vu. Jai mL pads XIII. JS vn Jeb OL c Seah XIV. Jat mx. Sah iv. tah xv. Jaz x. Jali Vv. Rum. a. The 3d pers. sing. mase. Porf., being the of the verb, is commonly used as paradigm, but for shortness’ sake we always render it into English by the infinitive; JS to wi, instead of he has killed.30 Parr Seconn.—Etymology or the Parts of Speech. (§ 36 A Ras, 5, Tho Arab grammarians use the verb Jad (bpp) a paradigm, whence the first radical of the triliteral verb is called by them iii the fa, the second Cyeall the ‘ain, and the third, ahi the Lam. Rex. ¢, Aa the above order and numbering of the conjugations ‘are those adopted in all the European Lexicons, the learner should note them carefully. 36. Tho first or ground-form is generally transitive (Gass) or intransitive (ais eb or 259) in signification, according to the vowel which accompanies its second radical. B 37. The vowel of the second radical is a in most of the transitive, ‘and not a few of the intransitive verbs; e.g. $42 to beat, 2% to write, JE to kill, Lb} to give; LAS to yo away, 345 to go the right way, wala to sit, 38. The vowel i in the same position has generally an intransitive signification, u invariably no. ‘Tho distinction between them is, that é indicates a temporary state or condition, or a merely accidental quality in persons or things; whilst w indicates a permanent state, or a C naturally inherent quality. E.g. oe or Jas to be glad, Se to be sorry, 53% ot sla to bw proud and insolent, 43h to become whitish, ors to becoms gray, ik to be safe and sound, aya to be sick, obo to become old, (gyé to be blind; but Gynt to bo beautiful, Ef to be ugly, A310 b0 heavy, G5 to be high or noble*, Ji to be low or mean, 2 to be large, zhao to be small. D Rau. a. Many vorbs of the form jad are transitive according to our way of thinking, and therefore govern the accusative, eg. Ae to know (ecire), Gaus to think, my to pity or have mercy spon, fog” to Ber. © [0r, 40 become noble, for the form with w of the recond radical often means to become what one was not before, Kamil, p. 415.—De G.] § 41] 1. The Verb. A. General View. 1. Forms of Triliteral Verb. 31 Raw, }. The same threo forms occur in Hebrew and Aramaic, A though the distinction in in these languages no longer 90 clearly marked. [See Comp. Gr. p. 165 saq.] is formed from the first (Jad) by 39. The second form ( doubling the second radical. 40. The si intensive (aii ification agrees with the form in respect of being ) oF extensive (SSW). Originally it implies that an act is done with great violence (intensive), or during a long time (temporally extensive), or to or by a number of individuals (numerically extensive), or repeatedly (iterative or frequentative). Eg. beat, 5-5 to beat violently ; > to break, p> to break in pieces; AS to cut, ELS t0 cut in pisces; 55 to separate, 555 to disperse ; (SS to kill, J t0 massucre ; Sle or Hb to go round, Son or Sys t0 go round much or often ; 5&) to weep, «gi to weep much; SUIT Ss the cattle died off rapidly or in great numbers (Ste to die); Dy Jeet the cumet kneeled down, aii Sz) the (whole drove of) camels kneeled down. 41. From this original intensive meaning arises the more usual C causative or factitive signification. Verbs that are intransitive in the firat form become transitiee in the second; a8 gyi to be glad, oe to gladden ; to be weak, 525 to weaken, Those that are transitive in the first become doubly transitive or causative in the wecond ; an 15 to know, 216 to teach ; Bd to write, 5d to teach to write; Som to carry, Jem to make carry. Rew.a, The causative or factitive signification is common to D second and fourth forms, the apparent difference being that it is original in the latter, but derived in the former, Rew. . The second form is often rather declarative or estimative than factitive in the strict sense of the torm; as pd to lie, pid 0 think or call one a liar ; Gio to tell the truth, Sno to think that cone tella the truth, to believe him.A D 82 Parr Seconp.—Etymology or the Parts of Speech. [§ 42 Rex. c. The second form in frequently denominative, and ex- pressos with various modifications the making or doing of, or being occupied with, the thing expressed by the noun from which it is derived ; e.g. cede to pitch @ tent (Xjek), to droell in a place, Arm to collect an army (Ei), 2k) to pave with marble (airy), asi to become bent like a bow (ug8), pe to nurse the sik (Heys), Slee’ to skin an animal, to Bind a book (he. the akin, compare our ‘to stone fruit” and ‘to stone s person”), 3y to clean an animal of ticks (3458), cgi to take a mote (G48) out of the eye. Compare in Hebrew 873, 331, Tht, Spo,:ete. Similarly, aba he said to him 3 ELS. (may thy nose, or the like, be cut off), sles he said to him BH WEE (may God prolong thy life), ae A he said to him LIE LAG (pence be upon thee), jedo s shouted the Moslem swar-ery, CBI abi), FA Mb JES Go te who entera (the city of) Zafar, must epenk Himyaritic (the language of Himyar, sem). Sometimes, like the fourth form, it expresses movement towards a place; as 425 to set out in any direction (a3), pS to go to the east (3,5), ph to go to the went (ili) . oy-05 csalell Ub Se he toent up and down the valley. Rew. d. Jad corresponds in form, as well as in signification, to the Heb. bygp and Aram, Sep, WG.d.—[see Comp. Gr. p. 198 so] 42. The third form (Jet) is formed from the first (Sas) by lengthening the vowel-sound 2 after the first radical, as is indicated by the 2Ui/ productionis. 43. It modifies the signification of the ground-form in the follow- ing ways, (a) When jad denotes an act that immediately affects an object (direct object or accusative), JéU expresses the efort or attempt to perform that act upon the object, in which case the idea of reciprocity (dBi) is added when the effort is necessarily or accidentally a mutual one. B.g. 465 he Filled him, a6 he (tried to Kill him or) fought with him ; 25ap he beat him, sole he fought with him ; ape § 48) 1. The Verb. A. General View. 1. Forms of Trititeral Verb. 38 he threw him down, 62; he serestled with him ; SLi he overcame him, A Sb he tried to overcome him ; aii he outran him, ai he ran a race with him; 44;5 he surpassed him in rank, ail he strove to dom; AB he strove to dom, he vied with him in rank and glory; 45a5 he excelled him in composing poetry, S215 he competed swith kim in doing s0; Sad. he got the better of him in a lawsuit, ayolén he went to law with him. () When the first or fourth form denotes an act, the relation B of which to an object is expressed by means of a preposition (indirect object), the third form omverts that indirect object into the immediate or direct object of the act (xecusative). The idea of reciprucity is here, asin the former case, more or less distinctly implied. Bg. cg) i The sorte (a letter) to the king, SAS Ste he wrate to the king, U5 he said to him (something), sig\5 he Ubi nT Fy J25t he sent (a message) to the sultan, C saga ed Sle LG he sat beside the Pome mander of the Faithful, Sasa Feth SIR do. sty Bs he Soll upon shim, attacked him, 43515 do. ; she jUL4 he advised him, sulted with him. : (c) When jad denotes » quality or state, JEU indicates that one person makes use of that quality towards another and affects kim thereby, or brings him into that state, B.g. Gebde to be rough or harsh, aNd he treated him harshly; S—s to be good or kind, jhe he treated him Kindly; G9) to be soft or gentle, 42295 he treated him gently; 5 to be hard, AAS he hardened himself against him or it ; Ai85 or 5 to lead a comfortable life, A225 he procured him the means of doing 30, Rew. a. The third form is sometimes donominative, but. tho sgt he con- ideas of offort and reciprocity are always more or less clearly implied. Bug. ELS to double, from is the like or equal ; jit w, 534 Part Seconn.—Ftymology or the Parts of Speech. [§ 44 A to double, fold (3,6) on fold; abt UIs may God keep thee safe and wall, from £58 robuat health: JHC t0 go om a journey (he). Rew. 5, JeU5 corresponds in form and signification to the Heb. Sup (Arab. a= Heb. 4); s00 Comp. Gr. p. 202 sy [Raw « Ina fow vorbs the third form is used in the sense of thefourth. ‘Thus Jats, SLs (G1. Geog.s.v. Jas). Zamabiari, Paik, 4. 197 cites sash, sbtp, ole for vassl etc. Alo @ll= aul, B Aghant xiii, 52.—De G.) 44.) Tho fourth form (Sail) is formed by prefixing to the root the ayllable f, in consequence of which the first radical loses its vowel. 48. Its signification in fuctitive or causative (Hass). If the intransitive in the first form, it becomes transitive in the ; if transitive in the firat, it becomes doubly transitive in the fourth. Bg. esse 0 run, (ggert to make run; Le to sit dnon, Crbeth to bid one ot dem GAIT SEI he ate Dread, Sat ET he gave him bread to eat ; 45230 gly he enw the thing, nigl 85) he showed him the thing. Reu. a. When both the second and fourth forms of a verb ‘are causative (§ 41, rem. a), they have in some cases different significations, in others the same. E.g. 16 to know, he to teach, Et to inform ona of a thing ; lai to escape, (goes and «gepil to eet fat liberty, to lat 9. Comp. jai and jee, D —_Rra. 5, The fourth form is sometimes declarative or estimative, like the second as aiji de thought him, or found him to be, niggardly ; agai he thought him, or found him to br, cowardly ; 35624 he found him, or it, to be praiseworthy or commendable ; SDT GA he found the district abounding in fresh herbage. * Rew. ¢. ‘The fourth form comprises a great: number of denomi- natives, many of which are apparently intransitive, because the Arabs often regard as an act what we view asa state. Such verba combine with the idea of the noun, from which they are derived, that of a transitive verb, of which it ia the direct object. E.g. Jas eis * Comp. 551. § 45) L The Verb. A. General View. 1. Forms of Triliteral Verb. 35 to produce herbage (Jj, 531 to put out leaves (393) paSl to bear A fruit (343), jlaal to give or yield rain (5la%); Comeil to beget a noble son, Szdoil, S57, ahs bors a male or a female child, S215 she bore tins (compare ‘to Hower,” “to seed,” “ to ealve,” “to lamb”); Git eat dont, EDN 0 eat with party and crea Mi to give a Proof ) of his prowese in battle; Gael, sisi, (0 act well or ith fault or error, wptal to do or say what is right; Lagi to bs alow or B tardy Erte mate haste; a to run with outetretched neck ; ST to become fullgrown (from & 4 tooth) ; SUSI to dwell or remain in a place—Another class of these denominatives indicates move- ment towards @ place (compare “to make for a place”), the entering ‘upon a period of time (being, doing, or suffering something therein), getting into a state or condition, acquiring a quality, obtaining or having something, or becoming something, of a certain kind*. E.g. Jail 10 advance, 5231 to retire (“reculer”), 251 to go on boldly C (compare, in Hebrow, |" to go to the right, and Symi, to (sail), Sap5I to go to 8-Megd (Sm), 234 t0 go to Tihama ua, pti to go to Utrak (Styli), 5301 to enter the Jaram or sacred tarvitory 5 col adi, it to mie spon the tine of morning ua, mid-day Cra ) camely, Gch to abound in beats of prey oF to, have onde flocks evoured by them, Lit wo abound in lisards (G25) oF to be foagy people) or to be dry (of a season), I to become penniless (wo be * [Hence in a few cases IV. serves (instead of VII. or VIII.) tw the p gts of I. Thus &B he threw him on hia face, St he fell on his face, Seapets he held him back, papert he drow back, he retired.)36 Part SEcOND.—Etymology or the Parts of Speech. [§ 46 A. reduced to the last farthing, 55), 2321, j96, to be reduced to uller want; abl to become cloudy, iki to become worn out (of a garment) ; Jai to become dubious or confused; &N9\ to become plain or clear; (S21 10 become possible—Another shade of meaning (CALI, deprivation) may be exemplified by such words as idl, (0 break onde compact seth a person % 65H ko remowe ons conse AEST Sabi he pointed (the text of) the book, literally, B took away its deat, obscurity or want of clearness. Rex. d. Jail corresponds in form and signification to the Heb. bropn, Phan. beep (xn, Aram. bupx, Gof, Seo Comp. Gr. p. 204 seq. ‘The Hebrew, it vill be observed, has 7] as the prefix, instead of the feebler Arabic and Aramaic X- Some traces of the A are still discoverable in Arabic; as cis for Eni to give reat to, t0 let rest ; syittor SN to wish ; Gish for Si} to pour out for Jt to mark a cloth; Aa give, for oA (rad. i, TINK, 0 come); WON fo believe, Forms like Gla are treated in Arabio as quadriliterals (sce §§ 67, 68, and 118) impert. Sent or Sem rents nom. patient, Shes or Shea, 46. ‘Tho fifth form (Jad5) is formed from the second (Jad) by prefixing the syllable S. D 47. This form annexes to the significations of the second the reflerive force of the syllable Ss; it is the sth’ of the second form, that is to say, it expresses the state into which the object of the action denoted by the second form is brought by that action, as efecto result, In English it must cften be rendered by the passive. E.g. 5-55 to be broken in pieces, Spi5 to be dispersed, CEE to be out in pieces, 27-35 to be moved or agitated ; to be afraid (Syd to terrify), dies SIGS he girt on li odd he git sword upon him—another person); 35, 4835, to be proud; Hel, 355, to side + B21 in properly to listen, to givs attention to a complaint § 47] 1. The Verb. A. General View. 1. Forms of Triliteral Verb. 37 with Kais or Nizar, 6385, ib GS) or of the Sta eee. 35 to call oneself an Arab, 5d to brome a Jew (Gos, Sa to become a fresorshipper (Sse), 5555 to Derome a Christian (Si5p25), CS to gine oneself out as. prophet (tug), Sol5 tc become as Dold or flerce as a lion (Sch, KS to become as savage as a leopard (593); he5 to try to acquire, to fect, cemeney, BoB, ya, to affect courage or manliness, B to endure with patience a, Co, 985, 5, to abstain from, oF edi to avoid blame. Rest a. ‘The idea of intensiveness may be traced even in cases i, at first sight, to have wholly disappeared, leaving form apparently identical in meaning with the eighth. wor crime, ae it Thos LT Gris and GAT F75 are both translated the people dispersed, bat. 573) oxpresses the mere separation, JpA5 the separa- tion into a great many groups or in various directions. c Rem. 6. ‘The idea of especially in such verbs as govern an accusative; e.g. oJ lo pursue step by step (literally, to make oneself, or turn oneself into, a pursuer of something), GLb5 to seck earnestly, pai to try 0 rriveness is often,not very prominent, understand, Gad to examine or study thing carefully, so that it be quite clear, o ascertein a thing for certain, may be yi 0 investigate thoroughly, ets to smell leisurely and cargfully, 5-5 to look at loug or repeatedly, to exumine or study, eS to hear D OF listen to, IK 10 apeat, J ta have charge of to discharge the duties of, es to mwaltow by mouthfulé, on S515 to milk or suck at intervals, S555 to ynavo, ke take under one’s arm, Ju 93 to put under one's head as a pillow, 5235 to take as an abode, .s5 to adopt as @ sont. © [In oome cases the difference between II.and V. entirely disap, ‘Thus for 95, am-5 (§ 41, rem. c) we may substitute yw without change of sense.)38 Pant Seconn.—Etymology or the Parts of Spesch. [§ 48 A Raa. c. The above examples show that the subject of the fifth form ia sometimes the direct object of the act (accusative), sometimes the indirect object (dati 48. Out of the original reflexive signitication arises a second, which is even more common, namely the effective. It differs from the passive in this—that the pawive indicates that « porson ia the object of, or experiences the effect of, the action of another ; whereas the afectice implies that an act is done to person, or a state pro- duced in him, whether it be caused by another or by himself. Eg. Bielé to know, fe to teach, 2165 to become learned, to learn, quite diferent from ale (passive of 12) to be taught, We can say as ae, he was taught (received instruction), but did not learn (become learned)*. Again, SN) to be separate, distinct, clear, Gsei to make clear, explain, Se3 to appear clear or certain ; Saa-5 to become, or prove to be, the reality oF, fact. Rem. a. Such of these verbs as govern an accusative admit not only of an impersonal, but also of a personal passive; 0g. hat AT he learned the art of medicine, lit Las the art of medicine was learned, Raw. 6. jaiS sometimes assumes the form Jai @® 11), whence we see ite identity with the Heb Sean or Sepnn and the Aram. Sepnw, dz) (e00 § 41, rom. a). D 49. The sizth form }) is formed from the third (jc), likewine by prefixing the eyllable &. 50. Ibis the Gyths (see § 47) of the third form, as Se 2324, I kept him aloof and he kept (or said) aloof—Tho idea of effort and attempt, which is transitive in the third form, becomes reflecive in the sixth ; ©.g. (seh3 to throw oneself down at full length, JS to be off one's guard, to neglect » thing, uighly O55 to draw @ gud omen from * Using » Scoticism, we might say, he was Learned (= taught), but did not learn, §50) I. The Verb. A. General View. 1. Forms of Triliteral Verb. 89 the thing, 255 to pretend to be dead, 52105 t0 pretend to be blind, A BlES to pretend to squint, SUS to pretend to ory, wasted to foign sickness, Jats to feign ignorance, g&3 to pretend to be deceived. Further, the possible reciprocity (44;(2.31) of the third form becomes a necessary reciprocity, inasmuch ax the sixth form includes the object of the third among the subjects that exercise an influence AGU he fought with him, NS the tro spoke to him, \peiiS they (the three) B 1 OST QU, be tried to pull away the garment from him, 2531 Gi25 the too pulled the garment to and fro between them; Lyactt i365 he conversed or argued with them, Egat ye they conversed together ot argued with one another ; SbkgT 1G he tried to make him forget the hatred between them, s forgot their mutual hatred; whence in the passive, 51 Syed, Leaeli p94, and EAGT oy. c Rem. a When used in speaking of God, the assertory (not optative) perfects Ij65 and (,ita5 are examples of the reflexive signification of this form: abl Jj(3 God has made Himself (is become of and through Himself) Blessed, o perfect, above alt; ee 5 God has made Himself uplifted, or exalted, above all ; Hy IG BT JE God (Blessed and enalted is He above all) Jas said (ef. vol. ii. §1f. rem. Somewhat similarly, oT a-Stei D the thing made iteef (Became oF 100s) too great, or dificult, for hin ; Boo Gta FFF 5h ie ta ow matter than which nothing maker iteelf greater (or more important), with which nothing can vis in importance. Rem. 6, As the reciprocal signification requires at least two subjects, the singular of the sixth form is in this case always 5 Ut 4g ELF the people heard of it from one upon one another with one anoth40 Part Szconn.—Etymology or the Parts of Speech. [§ 51 A Fs | (ie tribe of) Kortié came to him, all of them, following one Rew, The iden of reciprocity may be confined to the parts of one and the same cohserentes,” to be of compact and firm build; Hyait 55165 the woman became middl-aged and corpulent (each part of her body, su it wore, supporting, and so strengthening, the others); gé1a3 HST che building cracked and threatened to fall (as if its parts Bava B called on one another to do 80; compare gas ale (6155 the ony gas, NS E55 the cloud tightened and thundered from every quarter) [Hence this form is appropriate to actions that take place bit by Dit, or by successive (and painful) efforts, as to fell picce by piece, JalL5 to carry onceelf with dificulty (gE A. in walking®).] c Rew. d. JUS sometimes assumes the form féti5t (§ 111), and is consequontly identical with the Heb. Oxsipnin (aco §43, rem. 6). B1. The seventh form (Sadi) i is formed from the first (Jad) by prefixing a ¢», before which is added a prosthetic | to facilitate the pronunciation (see § 26). Rew. For the cases in which this | becomes 7, and why, see {§ 18 and 19, , with rom. ¢; and as to the orthography 1 instead of $819, rem. a. 7 D 82. The seventh form has also originally, as @ stie of the fit, a middle or reflexive signification, It must be aaa however, (@) that the reflexive pronoun contained in it is never the indirect object (dative), to which may be added another direct object (accu- sative), but always the direct object itself; and (8) that it never ascumes the reciprocal signification. By these two points the seventh form is distinguished fom the eighth, and approaches more nearly enemy adoanced againat him from every si © [Bee OL Geog. 0.0. Jem, Hamdsa p. 20 frst va and comm.— De] §54] I. The Verb. A. General View, 1. Forms of Triliterat Verb, 41 to a passive, At tho same time, the aféctive signification ia often A developed in it out of the reflexive, Eg. G1 to open (of a flower), lit, to split itself; jt to break (intrans.), to be broken ; Aleit! to be out off, to be ended, to end ; AEE to be uncovered, to be made manifest, to appear ; last to become broken, to break into pieces; SMEI to be uttered or spoken, 53. Sometimes the seventh form implies that s person allows B an act to be done in reference to him, or an effect to be produced upon him; eg. deceived ; put to let oneself be drawn or dragged. Rem. a. Hence it is clear that such words as Gor, to be stupid or foolish ; ASAI to be non-existent or missing, wot to be found, from ss not to have; gg), from (sgh to sink C down, to fall; 31331 to be repeuted, from 316 to return; SLbit to be in strate of distress, from GUS to be narrow ; are incorrectly formed, though in actual use, especially in more recent times, Rem. 6. Sometimes, particularly in modern Arabic, the seventh form serves as the ens of the fourth; eg. to be bolted, 1 to be extinguished, from Ub to extinguish » AGI to be put to rights, from ao to put to rights, (Similarly p CH Sibi, Qs, the last in a tradition, and so anciont, Faike i 63.—De @] Rau. ¢. fail) corresponds to the Heb, bpp: see Comp. Gr. p. 215 0g. 54, ‘Tho cighth form (ai is formed from the first (Jad) by inserting the syllable 2 between the first and second radicala. The first radical in consequence loses its vowel, and it becomes necessary to prefix the prosthetic | (§ 51, rem.). Wy 642 Pant Sxoonp.—Ktymolegy or the Parts of Speech. [$55 A Ren. One would expect < to be placed before the first radical, asin the fifth and sixth forms, and in the Aramaic reflexive Wpod}. 8 possible explanation of the actual form see Comp. Gr. p. 208.) eighth form is properly the reflexive or middle voice (231K ofthe fiat, The reflex object i ether (a) the direct object or accusative, as Gob to divide, 55h 10 go asunder, to part ; ays to place (something) before one, 735+ to put oneself in the way, to oppose ; Bip 10 beat, pled! to move oneself to and fro, to be agitated (com- pare the French battre and se débatire); or (8) the indirect object or dative, implying for oneself, for one's own advantage, an 35 to tear a prey in pieces, so53\ do.; Gnd t0 touch, G5 to feel about for a thing, to seek for it ; takes t2édid, and the gézma, with which the third radioal ought properly to be marked, is omitted. T have served ; Edajy tor Edis, thow hast i ye have taken. On this assimilation see Thus, Saad tor 5. bound ; s3ic4 tor 23 §l4e. Rex. ¢. When the third radioal is cy it unites with the gy of the suffixes into o single gy with téédid; as Gal they (women) believed, wwe believed, for pial and Cal. Rem, d. For a view of the Inflexion of the Perfect and Imperfect Indicative in Hebrew and Aramaic as compared with Arabic see Comp. Gr. p. 165 seg. }, FORMS OF TITE TENSES AND MoODE, 91. Whon the second radical of the Perfect has /étha, it may take either gamma or A?sra in the Imperfect SSS to writ ts eS to strike, yb; Fhe to sit down, sd. Many verbs admit of both forms ; as Uuiaé to sneeze, Las and abate; LS to remove the hair by scalding, bets or bent 5S) to stick upright into the ground, 3252, Rex. a, Verbs, of which the second or third radical is a guttural letter, are an exception to the rule, for they commonly retain in the Tmpertect the fetia of the Perfect ; as Sei “ do, ides ab 10 ous abks Eis to hinder, a 15) to create, bess SE to, SUS; SAS to go awoy, 05; Let t0 took at, Lats Cb to throw, Crbi. Not = few, however, conform to the rule, Forti i the sooond radical ix @ or Bas we to sit, oad; to transpierce, to ascend, i lls oe to be sound, right, good, 2425; LF to atiain to, reach, to blow, GBs HE to for, eatbe 2 fe turn Bais @ 25S to draw or pull away, 95%; j25 to bray, inti. Some verbs have two we 8 A B DA 58 Part SzcoND.—Ktymology or the Parts of Speech. [§ 92 forms; as a5 to oroak, aie; CS to give as @ present, Be men ci EAS wn i ES ton Eis fang dona with, byt; and oven thee, a E25 cut or hew, enh; Ei H Hntnn (ot wnat fs linens ee gush out, Gr Rew, b. Verbs of the form (jab denoting superiority, iad aia ue Jar (see § 43, a), always have damma (the grammarian Ri alone editing tt wth » gator, an 5 he end 3 abo SL ha ecalad him in goodness, bes Has he eurpamd hin in shooting with arrows, sease- Rew. ¢, Excessively rare are cases like (255 to incline to, lean 272, which is probably a combination of the two forms 5S), GBS, and 455, GE. Soo 4175, rom. 6. 92. When the second radioal of the Perf. has Kiera, the Imperf. takes f2tha ; as 212 to know, tite; Say’ to drink, oti; Gye to be sorrowful, Sj: ibys to bo sick, izet; ih to be enfe, ah, Rex. a. A few verbs may retain in the Imperf. the era of the Peet, an at to think oF euppont, anahs OF Seti as to be green and flourishing, 2a ; FE or oy. Bal Pan § 94] 1. The Verb, B. The Strong Verb. 39 or innocent of, 3304 oF ips The most common example of this kind A is. verb med. 9, viz. SU, to dia (for ys, lat p. sing, Bort. 2), bei —Bimilar cases in Syriac and Hebrew, Comp. Gr. p. 180°. ©3. When the second radical of the Perf. has amma, that ‘vowel is retained in the Imperf.; as (pum to be beautiful, Gyno; GS to be high, noble, S42; 31 to be dull or stupid, AL. Rea, With the above forms compare the Heb, 3/3) e ‘Tap. In Heb., however, verbs in o usually take a in the Impert., as Sogn TOpY, whereas in Arabic instances like E51 became arn i 1 became bad, 53, hare very rare, Some authorities admit the forms sell, 23t, yeh 04. Tho difference between the Perf. and Imperf. in-regard to their infexion is, that the marks of the numbers, genders, and persons, are only sufized to the Perf.; whereas they are both sufficed aud prefized to the Imperf., more generally the latter. c Rex, a. In the Perf. the act is placed conspicuously in the foreground, becauro completed ; in the Impert. the agent, because still occupied in the act (see § 77, rem. «). If we look upon the root Jz as primarily conveying the abstract idea of “killing,” we as meaning “killingof-me” (i. done by me), «7 have killed ;” and 51 as meaning “T-killing,” ="Tam killing.” ** Rex, , In the Import, the pronominal profes mark the state or tense, and to some extent the gender; whilst the eufizes serve solely to indicate the gender. ‘Thus, the 2d pers. sing, mate. 35 ia sufficiently distinguished from the Sd pere. sing. mase. —-i% by the form of the temporal prefix ; but to distinguish the 2d pers. sing, mase. from its fom. a suffix is necessary, and accordingly we got mano, LSS, fom. Seg. ~ [Anbart, Nozhat &Lalibba p, 459 states from pervonal observation in Yeoman wad Higis that in some dialects every verb Jab makes ** Fleischer, KZ, Schr. i, 368 considers the root as a concrete noun.A Cc 60 Part Seconn—Ktymology or the Parts of Speech. [§ 95 Raw. c. In tho active voice of the Grat form, the prefixes of the Imperfect are pronounced with 2th. But a pronunciation with aay instoad of foth is regarded as admissible and was used by some of the old Arabs with any of the proformatives excopt «6s, save in the case where the next consonant has damma (verbs med. 5). ‘That is, one must not say sgh, Aydt, 25) for Jyll ete, nor sity for Ji%; but on tho other hand the pronunciation Jai and 0 are recognized a8 Coens in Stra i. 4, and Soe] in Sore xxx, legitimate dialectic variations of the usual Jai ete, In one case, Std} for tei, 7 suppose, the pronunciation with kier is generally preferred. The tribe of Kalb used Her even withthe prefix (¢ (Aik). Dialectically, too, the vowel of the prefix might be assimilated to » following damo, as in 333 for Jad. 95. ‘The Indicative of the Imperf. is distinguished by the third radical having gamma, the Subjunctive by ite having /2he; as Indie. CHS, Subj. CHG. ‘The Jussive is denoted by the absence of any vowel with the third radical, as 76); whenco it is: eometimes called the apocopated Imperfect. Rew. a. The damma and fétha of the Indicat, and Subjunct. Impert. in the verb, correspond to the damma and fétha of the Nom. and Accus, in the noun (see § 308); for the Imperf. is closely ‘akin to the noun, and its government in the Subjunct. falls under the same category with the government of the noun in the Accus. Honce the technical name of the Inport, g,lbeif, because it resembles the noun, [The Indizative is called gy3p4ii, the Subjune- tive wogatsi, and the Jussive saja-ii] Rau, 8. Tho peculiar meaning of the Jussive has brought along with it the rejection of the final vowel, which soeins originally to have been i. At least the pocts make use of tho form Ja%; in shyme, [Cf vol, ii. § 247.) ‘The forms of the Indicat, which end in cy and 5 reject these syllables in the Subjunct. and Jussive, because the genders, numbers, and persons are distinctly indicated even after their omission. ‘The § 98] 1. The Verb. B. The Strong Verb. él ‘Qd and 8d pers, plur. fem. are exceptions, for in them ¢ is retained, A because it ie absolutely necessary in order to mark the gender. Com- 5 eB, with Noni As, GER, with CSB ; and Sei, 5, with uid, 97. The Energetic is formed by adding the termination Gs or 5 (called by the grammarians Sz5ci" yall, or the corroborative ») to the Jussive, If the Jussive ends in 7 or a, the fetha of GH or = is elided, and the long vowel of the verbal form shortened, because it is in a shut eyllabl 5, SS, SS, SS, from 1385 ete. In the dual, the first fetha of 4 is absorbed by the != B of the termination, and the second weakened into a kesra through the influence of the same long vowel GL In the 2d and 3d pers. plur. fem. the fetha of the verb unites with the initial Atha of GK into a long @, and in consequence the second fetha of Ex becomes kesra: U236 (3) from Gath (3). Rew. a. Tho syllable cy of the second Energetic is appended only to those persons which have, in the first Bnergetic, a short vowel before nd not to the dual, because its forms would then Cc coincide with those of the singular, nor to the fem. plur., apparently because the sound of the syllable 5 (C233) was disagreeable to the ear. Raw, 6. Before an alifu "Laongl (§ 19) the n of the termination Ex is rejected (§ 20, rom. c), a8 jill Gad 9, despise not the poor, for dregs from Gylal, TV, of Gt. Res. c. The syllable is often written 12, and pronounced D in pause 1 Compare the Hebrew Energetic or Cohortative in ty, Comp. Gr. p. 194. 98. ‘The Imperative (;-23\ the order or command) may be described ‘us formed from the Jussive by rejecting the prefix of the 2d pers, sing. Hence it has always the samo characteristic vowel as the Jussive; but, since it begins with two consonants, it takes, according to § 26, a short62 Pant Sxcon.—Etymology or the Parts of Speech. (§ 99 ‘Approsthetic vowel. When the second radical is pronounced with faehs or Bera, this vowel is Hera; when with damma, it is gamma, B.g. Sah, Syd, i Rix, a, Regarding the elision of the prosthetic vowel (1), see $19, 8; and on the orthography | and j, in cases where that elision does not take place, § 19, rem. d. Rex. 5. Fotha is never employed as a prosthetic vowel. B Ram. c As an Imperative the Arabs also use the indeclinable form JUS; as jLbe bs present ! Ie beware / JB alight / ee listen! a5 let alone! sp creep along! jks announce the death f—! from (4 This corresponds to the Hebrew /nfinitive abvotute Sigp (6 for a, and the final short vowel dropped), which ix also used in the same way; as “}3} remember /* In quadri- literals this form iy very rare, the only examples mentioned being C538 555, tet chy dhunder crash, and ese, come and play the game called ‘ar‘ara, Occasionally it seems to take its meaning from one of the derived conjugations, as oe Bring owt! Bhs overtake ! = \geuyéel, \yoy31, Tmper. IV. ‘The same remarks apply to the energetic forms of the Im- erative as to those of the Imperf. (§ 97). D [Ram. The common phrase 4222 Upsl, strike off hiv head, is sometimes pointed without ttnwin (ipl) and is then explained by Py the grammarians as a dual used in an intensive senso (ge a2: aeSgiil, of. vol. i, § 35, a, rem, 6) in addressing o single person. Similarly Kor'an 1, 23, Gal with a various reading txdi_pe a.) © [And again the phrase GUE Guide Tab, i, 1842, 1. 15 in parallel to the Hebrew use of the Inf, Abs. with the finite verb,— De G.] § 106) I. The Verb. B. The Strong Vorb. 63 2. The Passive Voice of the First Form in the Strong A Vorb.—Table II. 100. The Perf. and Imperf, Passive sre distinguished from the corresponding tenses of the Active by a change of vowels. In the Perf. Pass, the first radical has damma, and the second radical Hera, In tho Imperf. Pass. the prafizes take damma, and the second radical Setha, Rem. The vocalisation of the Passive remains always the ‘same, whatever be the vowel of the second radical in the Perf. and Imperf. Active. 101. There ia no special form to express the Imperative Passive, B the Jussive being used instead. 3. The Derived Forms of the Strong Verb.—Table IIT. 102. ‘The second radical of the Perf. Act. is pronounced with Sitka in all the derived forms. 103. The second radical of the Imperf. Act, is pronounced with ‘ftha in the fifth and sizth forms, with era in the rest- Rew. The Imperfects of the ninth and eleventh forms, Jay and Jaks, are contractions for Jui; soon from the Jussives JUS; and J Sat ana Jn Si. Soe §§ 106 and 120. 104. In tho second, third, and fourth forms, the prefives of the Imperf. Act. are pronounced with damma, in the rest with /atha. 108. Tho characteristic lif of the fourth form disappears when D another letter is prefixed ; as Jsiy, not JS, from [Rex. But we find S3y2 335, a pot set on the fire, and also 31455), Sibawdih, i. 9, 1.21, whore the 1 is treated like the » of Gus $118, rem, &.—De G.) 106. The ninth and eleventh forms were originally ist and Jiast. But, by » rule of the language (see § 120), if the last radical , and the Imperatives64 Parr Seconn—Etymology or the Parts of Speech. (§107 ‘Ain such words has s vowel, the preceding radical loses the two are combined into one letter with teédid ; og, pho for jpit, JiS5 for jyhbe. If tho Inst radical has no vowel, the word remains uncontracted ; a8 Siysiuo, jyhuas, jyot (eee § 120). 107. The formation of the Perf. ‘and Imperf. Passive in the derived forms is oxactly analogous to that in the gromd-form. Rem. a, The Imperfects Pass. of the first ‘and fourth forms are identical. B Rem. 6. The Imperfects Pass. of the fifth and sixth forms are distinguished from their Imperfects Act. only by the vowel of the prefixes, which is damma instead of fetha. vowel, and 108. Since the iden of the Perf. Pass. is expressed by pronouncing the first radical with damma, and the idea of the third form by lengthening the vowel of the first radical, there results in the Passive of the third form (in which both ideas are united) the form S955 and hence in the Pass. of the sixth, © 209. In the Perf. Pass, of the fifth and sixth forms, not only is the fothe of the first radical changed into daimina, but also the fetha of the characteristic S (which expresses the reflexive iden of these forms); eg J, i533, In like mannor, in the Perf: Pass. of the seventh, eighth, and tenth forms, not only is the first radical, or the Characteristic <, pronounced with dainma, but also the prosthetic elif eg. Jost, , SEELI. Compare § 98 and rem. 6. D120. ‘The ninth and eleventh (to fifteenth] forms, being neutral in their signification, have of course no passive (see § 78). 111, When the verbal root begins with <3, 2, @. 9:35:04 Le, us, b, or &, the characteristic © of the fifth ‘and sixth forms Sccasionally (in the Kor'an frequently) loses its vowel, and unites with the first radical to form a double letter, ‘The forms thus originated take & prosthetic if, when they happen to commence with two consonants (compare § 54). Bg. gill, JOU, 55, Sa. g114) 1. The Verb, B. The Strong Verb. 65 , seb. ‘The language in its later stages admits this in all verbs of the fifth and sixth forms, merely rejecting the vowel of the preformative 2; as AAS! for 15, to take breath. Soo § 48, rem, 5, and compare such Hebrew forms os IM, ENB; Comp. Gr. p. 110 seq. 122. ‘The & of the fifth and sixth forms is sometimes omitted B in those persons of the Imperf. Act. to which is prefixed (2d pers. ing. du. and plur. masc. and fem., 34 pers. sing. and du. fem.); eg. Cope, for pak, JLRS, EUS, Gye [and necessarily Qi iS i. 130)—De @.]- These shortened forms are sufficiently distinguished by the fethas of the prefixed < and of the second radical from the samo persons in the active voice of the second and third forms (;, det); and by the f&tha of the prefixed < from the same persons in the passive of C the second and third forms (=, 33), 113. Verbs of which the first radical is 7, 9, 6, 3, J. oF o, have no seventh form in classical Arabic, but use the fifth or eighth, or the passive of the first, instead. In the (so fur as we know) solitary example of the seventh form from a verb beginning with ¢,—namely AS, to tio conceated,—tho characteristic cy is united by taédid to the first radical. Rem. a. Some grammarians regard 1 as being of the eighth D form, by assimilation for 251. Rew, 6. Tn modern Arabic such forme as Lit, 5661 (Kamit, p. 569, note i), Spit, pall, imei. 35H, aro of common oecurrence. 114. If the first radical is», the characteristic < of ‘the seventh form often unites with it into 2; a8 Gaol! or Gast from Ja-*, Ww 966 Parr Seconn.—Etymology or the Parts of Speech. [§ 118 bee, CLG or Ao or at from ia, bite or baa ALE) from AG, Yall or Gaitl from Lali Rex, These forms are sometimes assigned to the eighth form ; 1, gol for gota, bal for b; GLA for 118. If the first radical be or ©, the characteristic — of the eighth form unites with the i itial = into %, with the a into Sor b, Bg. Ql, for A531, from ad iti or 561, for 331, from 385 5h or 5551 from 53; 5451 or ii from 545. B Rex. The same assimilation is sometimes extended to the a8 Totter ye, as Aen for QoiMt, from ay 116. If the first radical be 3, 3, or j, the characteristic < of the letter 3, a8 Cail, for Sisjt, from & The letter © is sometimes changed into » after an gg} L. The Ver. B. The Strong Verb. 67 117. If the first radical be Go, us, b, or 4, tho characteristic A ia changed into &, which unites with initial b into O, with initial & into & or &, and occasionally with initial Ga into V5. Bg. es, ihc! bas Ciksy tom 5, OG, CS Er st, Jas, ay, or pllabt, gs; sheds, abs, from 33, ZL. Rew, a. The letter ye sometimes assimilates the following be; B Rew, 6. From Qa the form [227*. If the second radical be < the characteristic < of the eighth form may lose its vowel and unite with it. The first radical then necessarily assumes a vowel, either a or i, and the helping vowel 1 is unnecessary and disappears, ‘Thus for or j= Impert. =, et or or even to the first radical); Part. act. s-0 (jin4); Ink rem. a). Similar forms from verbs whose second radical ia», 3, uo oF ‘ occur (or are recorded as variants) in the Kor'an (Sir. x. 36, ix. 91, ii. 19, xxxvi, 49).] 4. The Quadriliteral Verb.—Table IV. 118. ‘The four forms of the quadriliteral verb follow throughout D their inflesion the second, fifth, seventh and ninth forms of the triliteral (eee §§ 69—72). which is prefixed to certain persons in the Tmperf, Act., is omitted in the second form of the quadriliteral verb, just as in the fifth form of the triliteral (eee sm) )), 10 pour out, and Cyery (0 beliews, are eee68 Pant Seconp.—Etymology or the Parts of Speech. [§ 119 ‘A. The lattor is inflected exactly like jla45, but the former is irregular: Impert. Sani Imperat. os Nom ect SSija, Port. Pass. Gaps The form Sipah, Import. Sxpyd it also used —The tenth form of ee vie Qube, to obey, is sometimes shortened into git or ee Impert. genet or fod and then converted ee ge Iperf. an [Also, in verse, wo find jue for 3 ] B 5. Verbs of which the Second and Third Radicals ure Identical —Table V. 119. ‘These verbs aro usually called verba media ot secunder radicalis geminatee (¥’y). ‘The Arab grammarians uaine them 225i, the solid verb, or HELE Juli, the doubled verb. 190, ‘They differ from other strong verbs in two points (@) When both the first and third radicals have vowels, the second radical rejects its vowel, and unites with the third, so as to form a double letter, which is marked with taédid. Eg. 55 to fle, for 338; GE to aplit or cleave, for G&S; 2 to touch, for me sak to amell, for ng; Ss to become dear (to one), for am; 3 to become wise or intelligent, for <2. (0) If the third radical has a vowel, but the first is without one, tho second radical throws back its vowel upon the first, and then p combines with the thir, ao as to form a double letter. Eg. Jap for Maes, ed 325 for 303, Sei for JZ. But if the third radical has no vowel, the a retains its vowel, and no contraction takes place ; as 2555, Shs, S05, syle, Sk. Rex, a. Transitive verbs of this class, of the form jai, have damma in the Tmportoct, with the exception of six, which also smi Kara; vis £44 to sever or mparate entirely, make decisive or to make hard or firm, tie firmly, Jé to §120] -L. The Vers. B. 5. The Geminate Vorb. 69 seater (camela) a second tne, 5 to ‘read abroad or divulge scrlly, 3h to abhor, detest, Trport. Saty or Sag, ota. Ono verb has only kaarn, vie Ec to love (instead of the common IV, al), Impert. on Rex, 8. Uncontracted verbs of the forms Jad and Jab some- times occur ; a8 IS5 to be bnock-knead or weak in the hock, bs tohave < swelling [splint] on {he poster (of horse), oi to amell badly, Serb to abound in Uisarde (eb), Ea’d to be sre (ot the eye), B Lieb to be curly, LCE to have it hoof worn atthe edges (of » hore, ete. SA to be wie o intelligent, ad to be ugly, jzh to be bad, LE to be silly, in one’s dotage, jj to have narrow orifices of the teats (of @ she-camel, ewe, ete.), Rex. ¢, Forms like 2333, S34, CLs, are, however, some- ‘The second radical is inserted before the ‘oneal suffix. This diphthong GL as Spa8 for Saeal, £45221 for form which is not uncommon in the fifth conjugation, as 2955, ols, CLES, CLG, for Sop, BGS, SA8H, SAD (compare in the Hebrew Imperf. 13°3DA for M3231 or (b) the long vowel 12, a8 Side for Sas (compare in Hebrew IBD, whore 0=a). ‘The form described under 2a is the usual one in modern Arabic, but in N. Africa ai becomes § as reddit for 25 Such forms as Sgcatl for S12 also occur —Comp. Gr. p- 297 00g. [feo De Goeje, Gloss. to Ibn al-Fakth av. Jj]10 Panr Seconp.—Etymology or the Parts of Speech. [§ 121 ‘A. 481, In the Jussive, however, the second radical not unfrequently throws back its vowel upon the first, and combines with the third, in hich case the doubled letter necesssrily takes a supplemental vowel ($27). In verbs that have « or i in the Imperf, this vowel may bo either /dtha or Alora; in those that have u, it may be any ax See or KG, for three vowels, Eg. Gay or eaeeiel . eee Pia oF 3h, wea or at, for iyhe, eghts ets Sees OF ets does de oes oF 3ye, for Sasi S338". B 122. Those persons of the Imperative in which the third radical has a vowel (sing. fom, dual, and plur. mase.), sometimes do not follow the rule given in § 120, b, but keep the second radical apart from the third; 05 GsyzSt, Wish, NajzSt, When the usual contraction takes place, the prosthetic alif is obviously no longer necessary, and therefore the Arabe aay <6), 94, lys—not i, G3), yd —instond of sy ete. ‘The masc. sing. undergoes exactly the same contraction as the Jussive (§ 121), rejecting at the same time the prosthetic 1; eg. wae for ALLL, 3 for it, 32 for 3 c Reem. If the verb has « suffix, the choice of the supplemental vowel depends to some extent upon that of the suffix ; say 3 (ey 255 (ad), but thy, GE, not Ws, GbE In tho woul (§ 20) say sii 5 or 43 > 193. ‘The samo rules that apply to the Active of the first form, apply also to ita Passive, and to the third, fourth, sixth, seventh, D eighth, and tenth forms, But in the second, fifth, ninth, and eleventh, the second or third radical cannot be united with the other, because it is already doubled, Consequently 335, yi, contraction. (But cf. § 120, rem. ¢, for Conj. V.] nd 51551, undergo no ' [Tho uncontracted forms are said to belong to the dialect of Hiigts, the contracted to that of Tamim, Faik ii, 566—DeG. Cf. Sibaweih ii, 443.) 128) I. The Verb. ©. The Weuk Verb. n Rew, In the Passive some of the Arabs substituted kenra for A amma, at fo for Jie (contracted from Jn), whilst others gave the vowel of the first radical a sound between those of keera and damma (technically called abs, giving the one vowel @ scent or “flavour of the other), a Sy, 3, riddda, didda (with the Gorman it or French u), instead of rudda, fudda, 124. In the third, sixth, and eleventh forms, a long vowel, namely @, precedes the double consonant, which is sllowed in the case of fatha alone (§ 25, rem.). However, the uncontracted forms, B uch on Ge, at, gett, GBB, COU, C4, yh, dts, Ht, not unfrequently occur, Porms like 335, 3,943, and Me ‘are not contracted. 15. The Jussive of the derived forms may undergo exactly the same contraction as the Jussive of the ground-form, by throwing back the vowel of the second radical upon the first, combining the second radical with the third, and giving the double letter an auxiliary vowel. Bog. dst for i$, chsh for aus, the 1st per. sing. Juss, of the fourth & form of Jj and 55 C, Tar Weax Vere. 126. Weak Verbs (verbo infirma) are those in which one of the radicals is subject, on account of its weakness, to transformation or rejection ; and which consequently differ more or less, in some parts of their inflexion, from strong verbs (see §§ 82 and 83). D 127. Tho weak letters are, 5, and qs. 128. There are two sorts of weak verbs. (a) Those that have among their radicals a moveable dlif or hémzn, the weakest of the gutturals, ‘These aro called verb Kimsata, (®) Those that have among their radicals one of tho weak con- sonanta y and cg, which approach very nearly in their nature to the yowel-sounds # aud i. ‘These are more particularly called weak verbs.72 Panr Seconn.—Etymology or the Parte of Speech. (§ 129 A Rex, The Arab grammarians do not reckon the verba hémzate among the weak verbs, restricting this appellation to those that contain # 9 or g (§ 83, rem.). 4220, In a root there may be tro, oF even three weak letters ; a5 sh, oid, il. Verbs that have two weak radicals are said to be doubly weak; those that have three, to be trebly weak. These may be reckoned as forming a third class of weak verbs. 1, Verbs that have a Himza among their Radicals (Verba ‘Hemzata).—Tables VI., VII, VIII. B ; 130. These are divided into three classes, according a the hémza is the fat, second, or third radical (vrba prime, medi, ultime radi calis hémzate). The following sections point out wherein they differ from the strong verbs. , 29%. Ifthe lf with himen and gbzma, at the end of a silable (, be preceded by one of the heterogeneous vowels damma and késra, it is converted, after the damma, into with hénza (5); after the kesra, into cg with hime (f6). Hence 25) for Sis, tat per, sing. Pert 132. The 9 and (6 represent in these cases the sound to which the hémza inclines through the influence of the preceding vowel®. + [This is « convenient formula, and cannot well be improved upon without reference to the history of the Arabic language and writing, a D consideration that lay quite beyond the scope of the native systematic grammarians, to whose method of exposition this work, for good practical reasons, is closely conformed. But from an historical point of view, when we consider the cases when Jdmea is expressed by $, ¢ or by + alone without « kurst, or vupporting lottor, we must distinguish between two pronunciations—that indicated by the consonants alone, which in the oldest times were written without any supplementary signs, and that indicated by the later pointe, such as. Tt is known § 132] I. The Verb. Verba Hémzata, 13 ‘The hémza is retained, not only to show their origin from , but also to A i remind us that the syllables 5~ and (s~ are not to be confounded in pronunciation with 5~, @, and «¢- ‘The damma and késra remain short, whilst $ and q's are pronounced like { itself; that is to say, at the commencement of a syllable, with the spiritus lenis between the preceding ayllable and the vowel that accompanies the hemza (08 543, danw-'a, vot danu-wa); at the end of a syllable, with a slight emphasis and resting of the voice upon the soft breathing (as S55, duni’-ta, not éani-ta). Res. a, In modern Arabic, hémza in the middle and at the B end of words has so completely disappeared, that § and i, when preceded and followed by vowels, become and (3 except when tho former has dana (3) and tho Tater kése (6), ax explained in et $§ 198-4, The modern Arab also pronounces 5 and slike the long vowels 5” @ and g— 7 Even in the ancient, Innguage, ‘especially among the poots, we find traces of n softer pronunciati or total rejection, of the himza (3pey! ead 5 $17, b, vem. b]; ‘and hence the custom, at the present day, of resolving the verba C tart, end, hbmantie into verba Certio: yt, ax 638 for IS fo rend, 2455 for HS, yk tor 1385, ‘This chango has already begun in Hlbrew, and is almost universal in Aramaic the poople of the Higiz in the time of Mohammed gave up the original guttural sound of /émsa in very many cases where the other ‘Arabs still presorved it. Now the rules of Arabic orthography were snainly fixed by tho Kor'in, which was originally written down in the p Higia in accordance with the local pronunciation. ‘This pronunciation did not ultimately prevail over the Arabic area, but the old ortho- graphy could not lightly bo tampered with, having the character of crered tradition. ‘The first scribes wrote yng! xem le because they said bawmusa, gita, §aka (or nearly so). The pronunciation that prevailed, however, was ba'usa, §i'ta, $@aka ond this ‘was oxpressed, without touching the old consonanta, by writing ws, Edy, dite. Roles for writing hémza as $, fg or are therefore really rules for preserving the old guttural ’, in cnses where it was already lost or transformed by the first soriber of the Kor’in.} w 1074 Pant Seconn.—Ktymology or the Parts of Speech. (§ 182 A Rew, 5. The himes gézmatum over y and s falls away after 4m Bif hamzatum, bosause of the impeeiility of pronouncing it (17, 8, rom. 8). Hence past, not juft, Imporat. of 521; Get not Cyilt, Tmporat. of Gils si, not ash, Imperat. of Jah; not 5231, 34 pors, sing. Perl, Act, VIII. of ya; 5s), not Galil, 3a pore, sing. Port. Pass, VIII. of Gal, all with ait B_ conjunctionis asi Hla); Sash not Suagh, 8 porn sing. Pert bie IV. of bf Lg, not a Infin. IV. of bd ah not conjunctionis, comes into the wal, the 8lif conjunctionis falls away in pronunciation, though it may be retained in writing. In Im- peratives, when precoled by jor 3, and, it is vnsaly rejected; Segib, tm other canes it in retained, and the radical hamza is left in its altered form (¢5, 3); 08 Sjj80 freasaras, nicl Sa datdarilyin, CST gigi whudireina, GSB Syks yakatardan, wai EAST (also written SugBT (sail) Mladvtumina, In lator timos the pronunciation was softened in some of these cases by rejecting the hémza and lengthening the preceding vowel; og. Whudating, yokilidan, Uaditumina (as if written UII, Solsli Seed. D Raw. c | is always retained after fotha in the ancient lan- guage, on jul; but in modem Arabic it passes into the dif of prolongation, an 324, JSU, for jb, SU. [And 20 oven of old in Mecca, Nildeke Geach d. Qordns, p. 250, 257, whence with serplio difetioa 6, vom. 0) tach variations ax 22; for 25h Stra xlix, 14,] ‘Those who used the form 143 (see § 94, rem. c) also ssid 285 for 22%, from 3h § 135) I. The Verb. Verba Hémeata. 15 138. In the same way, f passes into $ or ¢g, when it is pronounced A with damma or késra and preceded by fetha, or with fetha and pre- ceded by damma or ktsra; and into ¢%, when it is Pronounced with Kedsra and preceded by damma (see § 17, 8). Big. Gest, for Gut, tobe SE or s)G, it agrees with, Tmperf. IIL. of 4 agree with, bo reconciled to, Imperat. VIL of 35; 95, 3852, for 538, an impression is made, was asked, for SU, Pert Paws. of St; 3} peace is made (betoeen them), for a}s), Perf. Paso. IIL. of 24. Rex. At the end of « word, j, pronounced with damma and procded by lth, i usually loft unchanged $5 trom from ¢yié, gh Import. Pass. II. of (iy, inst of 5733, But the latter form is commonly used before the accusative suffixes, aa ays. 134. Finally, 1 pronounced with damma or késra dor Pp, bec comes § or ¢ at the beginning of « syllable which is preceded by 8 gyllable ending in a consonant. Eg Cenie for 515 Saga, for Jgl%s, Pane Pasticip. of Ste; 22%, for at Imperf. of 2 fo groan, to twang atk, for 2h, Ie cts stngily and meanly, Impert. IV. of si SEA, put on armour, Imperat, X of 5. Ram, {at the beginning of » word remains unchanged, except in the oaos stated in § 135. Bg. 33, 30, D 135, At the beginning of a word, if an aif productionis follows the radical 1, the two alifs are combined into one, which is written either with médda alone, or with médda accompanied by a himza to tho right of the alif, or somotimes with hémsa and a perpendicular Reha (eee § 6, rem. a); a8 521, or pal, for sll, 40 consult, Il. of76 Paur Seconp.—Etymology or the Parte of Speech. [§ 136 Aj toorder. "The same thing takes place when a radical f with gézma (}) is preceded by an 8lif hémaatum with fecha (compare § 182, rew. B); a J, 3S, or 33, for g3H, co profer, IV. of 3. Iu old Mi wo often find atl, 33H, 136. In s more modem stage of the language, aif hémztum with fecha passes into 3, when preceded by fetha and followed by an B lif of prolongation (compare § 17, 0 rom. 6); as Igpalg5, for typailt or 135, they deliberated together, 84 pers. plur. Perf. Act. VI. of 155, for (Geil or (2G, the two became intimate friends, ftom bl (for 544), Rem, The same change sometimes takes place even with the initial if of the third form; a1 yal 10 be intimate with, 65g to be opposite or parallel to, .5\9 to console, Jy to eat along with, © for (gEtl, ote. It commenced, of course, in the Imperf. and the Nomina agentis and actionis, where, according to $17, 6, rom. b, iy. 'g took the place of 35 a8 cgelgts uals, and 3 187. The verbs Lil to take, ja) to order, and GS to cat, reject ‘the first radical in the Imperat., making 4b, 50, and JS. 138. When preceded by 2 or 3, and the Imperative > gene- rally recovers its radical lif, AS or 3455 but not so Jd and JS, p which make only 25, J. For the rule as regards other verbs prim, rad. hémz,, seo § 132, rem. 6; aud on the Imperative of i, 0 come, see also § 175, rem. a. 139. The first radical of J} is assimilated in the eighth form to the characteristic < of that form; Ai, for Judit (§ 192, rem. 6), to take for oneself. Bex, a, ‘Tho same assimilation sometimes takes place in bh 0 put on one the article of dress called S{it, and ject 0 give wages, which makes to put on an ‘izar, and janis! or jet to § 141] I. The Verb, Verba Hémsata, 7 give alms, to receive wages; still more rarely in SS, for GE, 1 trust oF confide in, and Jal, to marry, Jost, for St, do—The tenth form of Jaci muy also lose ita aif and be written Rew, &. From the above assimilated forms are derived the secondary radicals to take, and j—3, to trade (eee § 148, 8), Compare in Syriac p22, HZ), ths and with B Skz, tbe}, if from the rad. pal. 2140. Verba med, hémzate are occasionally inflected like verba med. ral. 9 et og ($149, ete,), and take an &lif of prolongation instead of the radical hémza with fetha. ‘This is particularly the case with the verb SE to ask, which has JU for JL, ad pers. sing. m. EL (not £54), SS for JEG, SS tor SUS, Jo for JE1 (imperat.), Perf. Pass Jey. —Sometimes the aif hemzatum is elided, its vowel being trans- ferred to the preceding (previously vowelless) consonant. Eg. i230 for JUG, from SEs 65s for Loi, from ch; to see ;* it to send, for SSF, whence BL, for SS, an angel (NOB). Re. a. The Imperative J makes in the fem. ig, du he, plur. tpl, not (J, ete. When preceded by band J, wo may gay SUT or Jis5, 1SLG, USES G21, a, rom. 8), or Ighid, Rem. 8, The clision of the élif occasionally happens in Hebrew, and in Syriac it is the rule; see Comp. Gr. p. 46, p. 282, D 2. Verbs which are more especially called Weak Verbs G 128, 0). 141, Those likewive fall into threo classes, according as tho letter 5 or «f is the first, eecond, or third radical (verba prime, seoundw, terti rad. 9 et 6).78 Parr Seconn.—Etymology or the Parts of Speech. [§ 142 A. Verbs of whick the Fira Radicul is 9 or 5 (verbs prima rad. 9 et ($)—Table IX. 142, ‘Those verbs primm rad. 9, which have Hera as the characte- ristic vowel of the Imperf. and Imperat,, reject the y in these forms. E.g. 335 to bear children, Import, 54 for Sigé, Imper. 30 for JAet (315); 55 w promise, Impert, dag for 452, Impor. de for da! (33!) Res. a, Eight verbs primm md. , of the form (fai, have in the Imperf. adj instead of Saiz (contrary to the riile Iaid down in § 92), and hence elide their first radical ; 533 to trust or confide in, G5 Syy to inherit, Sh to abstain from (what te un- lawful), ees to evel 3 5 Syy to be firm and hard (of fat), Uszii By to be in good condition and handsome, Gs Gp be near, to be in charge of, hts bas 10 love, OF those G49 has ‘leo dialectically the form 552, and a few more admit both forms; e.g: jung to be angry with, full of hatred of, sant seat He 10 be rough and broken (of ground), 323, 32925 349 to be hot, angry, ae Sixes Uy to bo aupafid with grigh to be melancholy, ay dines Jay to be cowardly, to forget, Jy. Sis Rew, 6. Tho Imporat. ‘yc in the phrases Wo 46 good morning / Ai Le good evening | seems Sn ets ba in reality from ja, Imperf. 2a%, o be happy, comfortable, This is the solitary instance in Arabic of the loss of the initial n in tho Imperat of verbs {"B, which is 90 common in Heb. and Aram. 143. But those verbs prime rad. y, which have fetha or damma as the characteristic vowel of the Imperf. and Imperat., retain the 9 in these forms. B.g. je9 or lena to be afraid, 3-53 or Jasi, sarc! or are! (For ja3h or clasts Gang to be im priv, Bangi; Jeng to stick §144y 1. The Verb. Verba pr. rad. 9 et us. 0 in the mud, Jase; Gx to perish, isis 923 to be visited by the mur- A rain, 532 J, to be unwholesome ot insalubrious, Justi 9-89 £0 be clean and fair, }533. ‘The same is the case with thore verbs which are at once prime rad. 9 and medio rad. geminate ; ax 39 (for 339) to Sat for 355). Rex. In verbs prime rad. 9, of which the second and third radicals are strong, and in which the Imperf. has fétha, some Arabic dialeota change the y into lor ys. Eg. Jet and Jeet for Jasis B Qs for aoe from eps (0 be sm pain s at; and is for sige from say, to make @ mistake, Others even use the forms James, Qamez, and vet: 144, In a few verbs, of which the eight following are those that ‘most commonly occur, the initial y is dropped in the Imperf. and Imporat., notwithstanding that the characteristic vowel of these forms C is tha, G3 10 let alone, é 385 to let alone, 3 B55 to vera é Ed 10 be wide or spacious, & E55 10 put down or place, B52, &. Gb} to trample up, ty, 85 sal, em ow AS to give, wi et Rem. a. The reason why the 9 is elided in thoso verbs probably is, that the fétha of the Imperf. and Inperat. owes its existence only to the fact of the second or third radical being in each case a guttural or somiguttural (,). Raw, 6, gg and ‘339 are not used in the Perf.80 Panr Szconp.—Ktymology or the Parts of Speech. [§ 145 ‘A. 145. In those forms in which » késra or damma precedes » vowelless , the » is changed into .¢ or 9 productionis, according to the preceding vowel. Hence Jaret, 34:1, for Je: 3 Sie for Ssh, Pert. XL; gta, glbetA, for gl TV, and Xj Sopa for Sopst, Import, Act. IV.; Sool, ee geet, Port. Pass, IV. and X. Rau, In the Passive of verbs prime rad, 5, the 9 is sometimes changed into f, on account of a certain repugnance of the Arabs to 1B the sound of the ayllable 5; eg. SSL, for S53, it is fio or wept determined (ot time); Goch for Gagy it ia revealed. 146. Verbs primm rad. cg are inflected in almost all their forms like the strong verbs; e.g. pu to pluy at huzard, or to be gentle, easy, is to ascend (a Lill), to be grown np, ae 3 Ga t be awake, do, WEES; £5 to be i c eam. Lagi, fo be dry, has aw or UNG or UAE. Seo § 92, rem. a. Dialectic varieties aro 21, for and yut, fo despair, as Be ere eee iti, and oily or Gait, for vat. Seo § 145, rem. 147. In those forms in which » ktsra or damma precedes a vowelless cs, the ¢ is changed into «¢ or 9 productionis, according to the preceding vowel. Hence pau} for joi), Imperat, L; ja} and for 34} and jt-e541, Infin, IV, and X, spt, Bist, for peer, p Saez, Import. Act. IV. of jo2 and 1 teristic ©, prodncing & for 2 and <3; as S831, for a&5s1, (S855)), to receive a promise ; 5-31, for 55) (j—H), to play at hazard. Rem, a, Sometimes, however, although many grammarians dis- approve of it, 9 and «g are not assimilated to the or cy in the pro- nominal suis v9 ax to form & =, aud & Bg. 33. a for 22s, (ope) to from 1 (cou) fo Be eparate Bee § 90, eae a, be. u In the Passive of the third and sixth forms of verba med. rad. 3, the 9 productionis (§ 108) does not coalesce with the second radical into 3D for, if it did, the peculiar feature of these forms would be effaced, and they would become 1 in appearance with the second and fifth (jg5 and Jpid), Henve wo write Syed, gsi, not p Ss, Sh. For the same reason, no coalition takes place in the same forms of verba medise rad. c¢, which are always written, for example, Qior and Gig. Seo § 11, rom. a. 160. Some verba medise rad. 3, and a few medi rad. cg, of the form Jad, are inflected throughout like strong verbs ; as igi to be curved or bent, Impert. 350; Sys to be Black, Impert. 353, 1V. gf; Jy to bo one-eyed, Import. 3983, IV. 5Ht; Gye to be woolly, lmperf. § 163} L._ The Verb, Verba med. rad. » et 6. 87 Spt; Sgt to squint, Impert. Syma, IV. igé to be wanting, A Tmperf. to have a particular disease (Seo, the slander), aid of «camel VTmpert, 3-23; dee to have a long, slender ‘neck, Impert. Sart; gh to be tender and flerible, Imperf. sok ; ae to have a slender waist, Imperf. ays. 161. Some verba mediee rad. 5 et cg follow in the fourth form either the strong or the weak inflection. Eg El or S551, t0 reward, from <6 to return ; ol or oh to perceive the smell or odour of & B thing, from 1) do.; at8h or Jabl, to be cloudy, from stb do; Set or Sebt, to watch a rain-cloud, from Sie. 162. A few verba medim rad. have only the strong inflection in the eighth form, used to denote reciprocity ; a8 j5:m! to be neighbours, from the rad. jlo ; cB to pair, to marry or intermarry, from the rad. ey 1 to borrow, from the rad. 3; C96! to help one another, from the rad. le. 163. Many verba medie rad. 5 admit in the tenth form of either inflection, but they generally prefer the weak, with the exception of fow, (chiefly denominatives}, which slmost always adopt the strong. Bg. E01 or SogmpSal, fo give an answer, grant a prayer, from the ye D imilarly, from verba med. rad. «s, umes! to Become 1 to become like an elephant (Se). Rex. a. On gartor e621, shortened from 61 be able to do, X. of gle, and on tho sooondary 61 rem. b. Rew. b. On the formation of the nomina agentis et patientis of the first form from verba med. rad. » ot «6, see §§ 240-1. Rex. c. For the inflection of verbs }'y and ¥’y in Hebrew and Aramaic see Comp. Gr. p. 242 209.A cy 88 Part Srconn.—Etymology or the Parts of Speech. [§ 164 ©. Verbs of which the Third Radical is or .g (verba tertios radicalia Ss watt Sabi the defective vert)—Tables XIV.—X VII. 104. These verbs are of five kinds ; namely :— (a) Verba tertiz rad. of the form Ja ; as jd to make a foray or raid, for a5é (§ 167, a, B, 8). (b) Verbs tertin rad. ug of the form Ja3; as (43 f0 throw, for 54 G 167, a, B, 8). (6) Verba terti rad. of the form Jai; as (58s to be pleased with, for 5-5 (§ 166, a). (a) Verbs tertize rad. gg of the form Jad; as Copde to be ashamed. (© Verbs tertie rad. 9 of the form Jai ; as go> to be noble. 165. There are three things to be noticed regarding the third radical of these verbs; namely, that it retains its power as a consonant, or it resolves iteelf into vowel, or itis elided. 166. At the commencement of a syllable, one of two things takes place, Namely :— (a) The third redical maintains its power as a consonant between the vowels 3d (I, U2), ia (52), 4 (152), a (Ge), a GS) as sleo when the preceding syllable ends with » consonant. Eg. 8s Ss, Re, Os Gh. as, Ga, eg Bie, 55, ENpby. The letter » between the vowels 7d () and ia (55) always passes into yg; a8 Gos, Gh, for 5-55, 48. The letter W is nover found between the vowels id, i—a. Riu, In the first and second classes, the Sd pers. fem. sing. and dual of the Perf, Act. I. and II. might have been 2334, W332, 2445, (G5, ote, after the analogy of 2255, Lj, and 5455; but the “Arabs followed in the sing, the mass. forme 133, «gy (§ 167, 0, 8, 8), and, not being able to say (compare § 137, and the Byrise form 12}, which, ft the end of sentence, is written 43. ‘The same thing holds good in pause of all imperatives that consist of only one letter; a3) for B Satrom (gly toe (§ 176); Sb for Gf, from iy lo keep faith (§177). Rex. 8, The vorb «sit, imperf. Gi, is an oxample of the rare forms mentioned in § 91, rem. c. Lexicogrephers mention the forms Ke, cs, and «seh 5 94 rom. cat being occasionally used 176. The Blif hdmzatum of the verb (sly is almost always elided in the Imporf. and Imperat. Imperf. Indicative. ¢ 2m. Le a 8. sl D. be ed 4 8. aD D. P. a Imperative. &. m. jor 5 6175, rom. a) (555 Deo js Bm sf Sie Rew, a. Tho Perf. Act. of (6; almost alway retains the hémza, which may however be tranapoood, fj; (for z i : 5 nome say <3} ‘Tho Tupert, (s1jj and the Imporat, 51 are used dialectically.94 Pant Seconp.—Ktymology or the Parte of Speech. [§ 177 A Raw. 5, The Pert. Pass. is 33; (like 54) or, by transposition, 443°. In the Import. Pass. the hémza is elided, just as in the Active Rex. c. In the fourth form, when it signifies o show, the hbmza sti ges gts gh Otherwise it ia in always clided: (65, 3 retained. B 177. Of the second clase, in which 5 or .s occurs twice, there are two sorts: (a) those in which » or «g is the first and third radical, a5 35 0 guard, (3 to be near, (gaps t0 be sorefooted (of horse); and (8) those in which » or cg is the second and third radical, a8 e555 to roast, (Sg5 to be strong, (sy (for jem) 10 live, Gx6 to have an impediment in one's speech. 178. The first sort follows in its inflection the verbs of the two © classes prime and tertio rad. yor (s. Bg. ren Gor 45 (§ 175, rem. a). od. ‘ Gn. 179. In the second sort, the second radical undergoes no change whatever, Eg. 2, Viki, not cseord, (ger to prevent the union of two cg, and also, in tue former caso, to distinguish the Import. of [qm from the proper name «seat Yahya (John). Raw. 5 Gym admite (1) of the contraction of the two 46, (a) in ‘© [A more reoant form is «$33; ste the Gloss, to Tabarl, D. G.] $181) I. The Verb. Doubly Weak Verbs. 95 CF for oe ) in the Import 108 Se GAs Gedts (yin the nomen actionis IT. ($§ 80 and 202, rem.), Be for Higa; (2) of the tlision of the second «g in the Perf. and Imperf. X., when it ne nifies f0 feel shame, a8 got, wri. for Geist, yeti Ge slo adaita of being contracted into Ef, and Git into Jad —Tho forms (yo and oe are eaid to occur (compare § 123, rem., and §153, rem). 180. Trebly weak verbs are divisible into two classes; namely (@) those in which one radical is hémza and the other two 9 or 4g; and (6) those in which all the three radicals are 3 or (5. Rew. We pass over the second class, as it seems to consist of only one verb, which is hardly ever used; viz. Uz fo write the letter (5. 181. Verbs of the firat class are of two sorts, namely (a) those in which the hémza is the first radical, as (63! ¢o betake oneself to, to repair to; aol (B) those in which the hbmzs isthe second radical, ‘a8 (sly to promise. ‘The —s are inflected like a and (555 (§ 179), 8 GI. SN, E55), vagh, gels the latter like UL and «ip (§ 178), eg. Perfect, 3. m. BE Le . 1 2 ls Ng aly “ Us P. hh Sal ay ‘ weB 96 Pant Sxconp.—Etymology or the Parts of Speech. [§ 182 Tussive. Bf 2m. af Le 8. 6 8 rou A D. ws us wa . . . P. ont 6 & Imperative, 8, m. J oof 175, rem. a), as}s D-o. Gs Pm Aah, Coal APPENDIX A. I. The Verb ed. 182. ‘The negative substantive verb o~2), he és not, has no Imperf. or Imperat., and is inflected like verba medim rad. 9 et (5. BE. n af Le. 8 D. Po lgedd roma wi Rew, a. Lod in compounded of 5, not, and the unused 21 Hob. wr, he is, Aramaic ‘TR. MN, tu); originally a substantive, signifying being, existence, an in the phrase yd Ge SH S32 5, he doce not know what is from what is not. (Therefore the eufix of the Ist porson sing, added to 47 is not only (ind (and Bysd, Dut also cgutd (§ 185, rem. a, Vol. fi, § 186, e)] ‘The Asnyrian D seems alsq to have the word isu, with its negative /a iow, in the double sense of ‘to be’ and ‘to have’.** Rau, b. Instead of wo find occasionally [as in the Kor'inic phrase yotis Cee S55] the indectinable 25, which corresponds to the Aramaic’ Aa, myo, mb, compounded of wd and mys Le. 4 [Or rather .y according to Noldeke, Mand. Gramm. p. 293, note 6.] 6 and 3 from below has deni opp. denebi. §183] I. The Verb. Appendis A. Verbs of Praise ¢ Blame. oT Il, The Verbs of Praise and Blame. 188. The verbs of praise and blame (23315 ot Jui are as, to be good, and y-H, to be bad, ‘They are used as exclamations, d are generally indeclinable, though the fem. and Sf, (ood, it is said, the dual and the pur. typed, Spaad) occur. ["The following noun must bo defined by the article or a dependent woke sa, 2a the companion Zaid, or ele the indefinite accusative must gonitive, aa an excellent companion, lit. excellent be used gj Us sai, ancellent as a companion is Zrid.] Rex. a, Instead of a5 we may say a3, 225, and Jai, which tor pa5, 25% for 234,— contraction which is sometimes extended to the passive Jud, an (hrs for (sles (from tha). Hence a and $5 become a5 ‘and (il,, (2) If the second radical be guttural, its’ vowel, instead of being elided, may be transferred to the first radical j an Sy? for Spl, py for Gal. Hence 123 for ead, for J}. (3) The form ja, which has boon thus atcained, may take nn additional késra to lighten the pronunciation (jas); as Sed, ay Hence Joa [or rather, according to Comp. Gr. p. 166, ‘becomes oe by assimilation of the vowels, and the latter may then be shortened to ay, a8 the former may be shortened w. 13 A D98 Part Seoon.—Etymology or the Parts of Speech. [§ 184 to Jy5}—These observations cast light on the peculiar form of intransitive verbs in Atthiopic ; as gabra (“to do”) for gabir .com- pare dee), ‘and, when the second radical ia guttural, mzhra (“to have pity upon,” say) for makira, stkna (“to be hot,” by for achina or salina, ace Comp, Gr. 1.0. Rew. c. Other verbs of thi ans are sess OF wet tobe pleasing or charming (contracted from ~-a)®; je, usually contracted into Cyne, to be good or excellent ; 5, to be bad or evil ; Jay, commonly contracted into GA}, to be far off; and G5 oF G5, to be quick (contracted from e) The first of these ia frequently combined with the demonstrative I§ (§ 340), and forms the indeclinable Iie, Jovely, charming, or excellent, is On such forms as 54; i, see above, note to p. 30, IIL The Forms expressive of Surprise or Wouder. G4. The Arabic langunge, yoassses two forms of expression, C called by the native grammarinus past tail or verbs of surprise or wonder, ‘The one is the 3d pers sing. mnsc, Perf. Act. 1V., eos eae preceded hy Us (degmadil We the mi expressive of suprise), and followed by the nocusative of the object that causes surprise; as 1daj Labi what an excellent man Ztid is! The other is the 2 pers, sing, mase, Imperat. IV,, followed by the preposition 4» with 1s 2oz Lal, with the same signification a before. Rem. a. The first formula literally means: what has made Zéid eacellent? can anything make him more excellent than he is? The second : make Z. excellent (if you can,—you cannot make hin more excellent than he is); or, more literally: éry (your ability at) making excellent upon (op) Zeid. They are, of course, indectinable. delete the remark in square brackets, Res. 5, Verbs of surprise are, generally speaking, formed only from triliteral verbs in the active voice, which are capable of being fully inflected, and express an act or state iti which one person may the goniti © [You soy UE Gj Gt and, more commonly, Csi] aiiy Cot, how beloved Zeid is tous! D. G.) § 184] I. The Vorb. Appondis A. Verbs of Surprise. 99 vie with or surpass another, They cannot be formed from the A passive voice; nor from quadriliterala; nor from verbe that are defective in inflection, like Jai and ¥-4, or in meaning, like the substantive verb (td to be (from SUF 335 tb, Ztid was standing ‘up, we cannot say GSU 1agj eu Gor OO diy SySb; nor from verbs like SLs to die and (5b to perish, expressing an act or state in which one agent cannot excel another ; nor from negatived verbs (a8 gaily Ele U, he did not heed tha medicine); nor from vorbn signifying colours and defects, whence are derived adjectives B of the form abl (as 39 to be black, B51; Sgt to aquing, Spent. ‘The grammarians add that verbs of surprise cannot be formed from the derived forms of the triliteral verb ; but neither this limitation, nor that with respect to the passive voice is strictly observed (com- pare § 295). We find, for example, from the Passive aliti U, how much he ia busied | trom Jab to be busy UG, ote proud or on He te) from aj to be proud or oan ole tf ha isto me | from Ets to be hated ; and from derived forms, eepecially the fourth, suas! Us, how liberal he ia! from (gltl to syive, IV. of Und to take in the hand; gyro) 931, how tiberat ha ia in bestowing gratuities | trom .)3\ to bestow, IV. of (5 to be JF, or Skef Li, how wily he és from JiEat to practise un artifice or wile, VIII. of JUL to be shified or changed; pediv, how short, or shortened, it is ! from jothh to be shortened or abridged, D passive of VILL from the rad. jose. ‘The rule with regard to vorbs expressing colours or defects is violated, for example, by aint Us, how stupid he ia 1 from Gen to be stupid, Sool, 5ST he Gaail G, how white this pises of cloth ia! from ici to be whit soi. Rex, ¢. When formed from verbs med. rad. gemin. or tert. rad. 1. 46, the verbs of surprise fllow the infection ofthese classes; a bad, how strong his father is! Sac} \, how how rich he ia! But if formed from verba n.