Lectures on the Comparative Grammar of the Semitic Languages: With a New Introduction by Patrick Bennett 9781463207922

Citation preview

GORGIAS REPRINT SERIES Volume 28

Lectures on the Comparative Grammar of the Semitic Languages

WILLIAM WRIGHT

LECTURES ON THE COMPARATIVE GRAMMAR OF THE SEMITIC LANGUAGES WITH A GENERAL SURVEY OF THE SEMITIC LANGUAGES AND THEIR DIFFUSION AND OF THE SEMITIC ALPHABET, ORIGIN AND WRITING EDITED WITH A PREFACE AND ADDITIONAL NOTES BY WILLIAM ROBERTSON SMITH

WITH A NEW INTRODUCTION BY PATRICK BENNETT

GORGIAS PRESS 2002

First Gorgias Press Edition. Copyright (C) 2002 by Gorgias Press LLC. All rights reserved under International and Pan-American Copyright Conventions. Published in the United States of America by Gorgias Press LLC, New Jersey, from the edition published by Cambridge University Press in 1890.

ISBN 1-931956-12-X

A

w

GORGIAS PRESS 46 Orris Ave., Piscataway, NJ 08854 USA www.gorgiaspress.com

Printed in the United States of America

Introduction By Patrick Bennett Principal, The Jerome Institute; Emeritus Professor, African Languages and Uterature, University of Wisconsin-Madison Why read Wright? William Wright's "short course of elementary lectures" began in 1877. The posthumously published version appeared in 1890. The book was a handy size, just short of three hundred pages. It has neither a bibliography nor an index, which would have been a desirable feature in a work as crammed with examples as this. But it was - is — easy to use; the principal topics are arranged in a very logical order (as one might expect of a course of lectures), and the table of contents gives detailed chapter outlines. The book assumes certain things about those who consult it. First, it is hardly usable without access to the Arabic, Hebrew and Syriac scripts. Akkadian always appears in transcription, Ge'ez generally appears with transcription, but nearly every page is laden with examples without transcription. Secondly, the reader needs a background in the grammar of these three languages and its terminology. Wright assumes that his audience will understand his reference to "the so-called construct state" in his discussion of the feminine *-at, p. 133. But that should not be taken to mean that this is a book for the expert. As W. Robertson Smith says in his preface, p. v, "It was not his design to produce a complete system of the Comparative Grammar of the Semitic Languages or to give a complete account of all recent researches and discussions, but to do through the press for a wider circle of students what he had done by the oral delivery of the lectures for his Cambridge pupils." If he assumes some background in Arabic, Hebrew and Syriac, he does not assume wide reading in Comparative Semitics. Accordingly, it is not exhaustive. We are dealing here with the basic structures of Semitic in their broad outlines, set forth for undergraduates, with little time for minutiae. Which is not to say that only the novice will find anything of value. In a book so

crammed with data and careful analysis, there is something for everyone. But that raises the question, what is the value of these lectures? Since their time numerous studies comparative Semitic studies, some much more comprehensive that Wright's, have been published. Our knowledge of the Semitic languages has widened and deepened. There have been great changes in linguistics, which have had serious impact on the field of comparative linguistics. Have Wright's lectures any more than antiquarian value? What is their relevance today? Wright's status in the Semitic field cannot be denied. The lectures are the one serious Anglophone contribution to comparative Semitic grammar at a time when the field was dominated by German works. His Grammar of the Arabic Language was still the reference of choice in at least one Arabic course a century after it was written. A reading of the Lectures reveals a scholar well read in the Semitic scholarship of his day and well able to draw his own conclusions. We see this at the very outset, in the discussion of the original Semitic homeland at the end of Chapter I. He runs through the views of von Kremer, Guidi, Hommel, Sayce, Springer, Schrader. And De Goeje "to mention only a very few names", with specific quotations. "Which of these scholars is in the right," he concludes (p.9) "we shall be better able to judge by and by. Meanwhile I will only say that I range myself on the Arabic side with Schrader and De Goeje." The Lectures reveal an experienced scholar, closely familiar with the Semitic languages available to him, with a wealth of data at his fingertips. In Chapter V, "THE VOWELS AND THEIR PERMUTATIONS" (a topic he handles "in a somewhat superficial manner"!), he cites Assyrian, Classical Arabic, Colloquial Arabic, Maltese, Hebrew, Phoenician, Biblical Aramaic, Mandaic, Christian Palestinian Aramaic, Classical Syriac, Modern Aramaic, Old English, Modern English, Scots, Gothic, German, Dutch, Classical Greek, Modern Greek, Sanskrit, Bengali, Farsi, and Turkish. Vowel changes are traced across time and space, and parallels to developments in IndoEuropean languages pointed out. They reveal a cautious scholar. He appeals to regular phonetic correspondence, "any deviation from which must be regarded with a careful scrutiny before we accept the relationship of the words in question (p.73). When he draws his conclusions, he does not use

m

the language of certainty. Speaking to the issue of the Semitic verb as derived from nominal sentences, he says "I have dwelt for a little while on these classes of nouns, because I believe that they really lie at the root of the inflection of the verb in the Semitic languages... (p. 164). That "believe" sets Wright apart from many a comparativist. Finally, they reveal an imaginative scholar, who can and will look beyond the surface of the data and speculate as to origins. In his discussion of the Semitic case markings, he suggests origins in a demonstrative hâ for the accusative in *-a, in the 3sm independent pronoun h û for the nominative in *-u, and in the nisbe suffix *-iyfor the genitive in *-i (p. 143). Wright does not simply possess the arguably negative virtues of being balanced between the extremes of sloppiness and rigidity, shallowness and overimagination. One of the fascinating aspects of books that began as series of lectures is that one can get a feeling for the person behind the printed words. I, for one, feel that I would have enjoyed discussing Comparative Semitics with him. Not, by any means, that I find myself in agreement on every point. Far from it; I would certainly take issue with him on the suggested origins of the case markings, and would want to debate his statement that the Semitic languages are "as intimately connected with one another as old Norse, Gothic, old High German, and old English, on the one hand; or as Italian, Spanish, Portuguese, Provençal, French and Wallachian, on the other (p.l). But there are far too many scholars with whom I do not feel I could have a pleasant and productive debate. But is that enough? Wright's is not a timely book. Packed with data, yes — but the vast majority of those data are from the linguistically, culturally, and geographically close neighbours Arabic, Hebrew, and Syriac. Yes, Akkadian and Ge'ez are well represented, and there is coverage of modern dialects of Arabic and Aramaic.. But we do not see the contributions from Ugaritic, Eblaitic, Modern South Arabian, and the fringes of Ethiopian Semitic which we would expect were Wright to resume lecturing today. He is further handicapped, in my opinion, in that he is restricted to looking at Semitic as an isolated language group. He cautiously refuses to deal with proposed relationships between Semitic and Egyptian on the one hand, Semitic and IndoEuropean on the other. Working today, attention to other AfroAsiatic groups

[ill]

such as Egyptian, Berber, and Cushitic would affect certain conclusions, such as his suggested derivation of both Akkadian 3rd person pronouns in *s and West Semitic pronouns in *h from a demonstrative formative in *t (p. 105). And, of course, we may often disagree with him. Not his facts; his data are accurate as well as voluminous. Not his attitudes and methods; while I have known comparativists who would wish him to go farther, and others who would urge greater caution, I doubt any of them would seriously fault his careful comparison. But consider his somewhat strange etymological suggestions for the case markings (p 143), or his statement on Semitic gender (P-131): The vivid imagination of the Semite conceived all objects, even those that are apparently lifeless, as endowed with life and personality. Hence for him there are but two genders, as there exist in nature but two sexes. Consider the restriction of the discussion to phonology and morphology, leaving out syntax and semantics; consider the lack of systematic formal reconstruction; consider Wright's failure to trace in detail the changes over time in Arabic and Aramaic. But all this is beside the point. The bottom line: Wright's lectures were intended to impart a basic understanding of the principal similarities and differences in the Semitic languages to 19th century Cambridge undergraduates. There is, I feel, adequate evidence that the lectures were very well designed to do just that. They hold up very well beside comparative linguistic lectures I attended in the 1960s. How can they serve us today? There is, of course, antiquarian interest. In itself, this is not a great incentive. To the degree that Wright's presentation has been superseded by more current scholarship, it has diminished in value. We do not train physicians with textbooks published in 1890. While the history of science is a valid and valuable field in itself, it is not, surely, a prime concern to the Semitist. But there is another aspect to history. As Santayana said, those who cannot remember the past are condemned to repeat it. One of my greatest pleasures teaching diachronic linguistics is showing how often our latest discoveries prove to have been published a

[iv]

century ago. We reinvent the wheel at regular intervals — and in most cases never realize it. Wright's lectures are packed not only with data, but with carefully drawn conclusions. It would behoove us to take these into account before drawing our own. Can it profitably serve its original purpose and stand as a textbook today, for today's elementary courses in Comparative Semitic Grammar? Not easily at many universities. The majority of students who have an adequate Hebrew background lack the Arabic, and vice versa. But given the necessary background, today's students could indeed use it with profit. True, one would wish to supplement it with data from, for instance, Modern South Arabian, and to build in supplementary discussion of such areas as syntax and the AfroAsiatic connections. But the basic field is well covered. In sum, I for one am pleased to see Wright's lectures made available again. We need contact with his thorough scholarship and careful insight.

M

PREFACE. H E Lectures printed in this volume were composed and delivered for the instruction of students in the University of Cambridge, and with special reference to the Examination for the Semitic Languages Tripos. It appears from the Cambridge University Reporter that Professor Wright began " a short course of elementary lectures " on the Comparative Grammar of Hebrew, Syriac and Arabic in the Easter Term of 1877, and he continued to lecture on the subject at intervals till he was withdrawn from work by his last illness. The manuscript from which this volume is printed represents the form which the Lectures ultimately assumed, after they had passed through repeated and sedulous revision. They were never redelivered without being retouched, and in parts rewritten; and the whole manuscript, except a few pages at the end, was so carefully prepared as to be practically ready to go to press. It was Professor Wright's intention that the lectures should one day be printed, and during his last illness he often spoke of this intention in such a way as to make it clear that he meant to publish them without any substantial modification or addition. It was not his design to produce a complete system of the Comparative Grammar of the W. L.

vi

PREFACE.

Semitic Languages or to give a complete account of all recent researches and discussions, but to do through the press for a wider circle of students what he had done by the oral delivery of the lectures for his Cambridge pupils. Under these circumstances the task of editing the book for publication has been very simple. I have divided the text into chapters, for the convenience of the reader, but have printed it for the most part word for word as it stood in the manuscript. In a very few places I have removed repetitions or other slight inconcinnities of form, but in such cases I have been careful to introduce nothing of my own, and to limit myself to what would certainly have been done by the author's own hand if he had lived to see the book through the press. Occasionally I have thought it necessary to add a few words [within square brackets] to complete a reference or preclude a possible misconception, and I have also added a few notes where the statements in the text seemed to call for supplement or modification in view of facts or arguments which had not yet come under the writer's notice when the lectures were last revised. So long as his health allowed, Professor Wright closely followed all that was done in Semitic learning, and incorporated with his manuscript, from time to time, references to everything that he deemed important for the practical object of the lectures. But it was no part of his plan to give a complete view of the literature of the subject; as a rule he only referred to essays which he wished to encourage his hearers to read in connexion with the lectures. Bearing this in mind, I have been very sparing in the introduction of additional references

PREFACE.

vii

to books and papers ; but, on the other hand, I have borne in mind that every written lecture must occasionally be supplemented in delivery by unwritten remarks or explanations, and a few of the notes may be regarded as taking the place of such remarks. I have, for example, occasionally thought it necessary to warn the reader that certain words cited in the text are loan-words. In all questions of phonetics this is a point of importance, and I am informed by those who heard the lectures that Professor Wright was careful to distinguish loan-words as such in his teaching, in cases where the fact is not noted in his manuscript. A considerable number of the notes are due to the suggestion of the author's old and intimate friend Professor Noldeke, of Strassburg, who has kindly read the lectures in proof, and the notes signed N. or Nold. are directly taken from his observations. Some of these, which were not communicated to me till the book was in page, have been necessarily placed among the Additional Notes and Corrections, to which I desire to call the special attention of the reader. It will be observed that the Lectures do not embrace any systematic discussion or classification of the forms of nouns in the Semitic languages; nor can I find any indication that the author intended to add a section on this important and difficult subject. He seems to have regarded it as lying beyond the region that could be conveniently covered in a course of lectures to undergraduates ; and he did not live to read the recent works of his old and valued friend Professor de Lagarde (Uebersicht iiber die im Aramaischen, Arabischen und Hebr'dischen ilbliche Bildung der NominaGottingen 1889 : Abh. der k. G. d. W., Bd. xxxv), and of Professor

viii

PREFACE.

Barth (Die Nominalbildung in den Sem. Sprachen, iste Hälfte, i., Leipzig 1889). On the other hand he doubtless intended to complete the subject of verbal inflexion, and I have therefore thought it right to make a few additions to the rough sketch of the derived forms of verbs whose third radical is 1 or \ with which the manuscript ended, and also to supply, by way of appendix, a short section on verbs one of whose radicals is an N. Here also I have derived great advantage from Prof. Nöldeke's suggestions. T h e printing of the volume, necessarily slow from the nature of the work, has been still further retarded by a prolonged illness, which fell upon me after the early sheets were printed off, and which would have caused still more delay had not Mr A . Ashley Bevan, of Trinity College, kindly undertaken to read the proofs during my enforced absence from Cambridge. I have to thank Mr Bevan not only for this service but for suggesting several useful notes.

W. ROBERTSON CHRIST'S C O L L E G E ,

June,

1890.

CAMBRIDGE,

SMITH.

CONTENTS.

PAGE

C H A P T E R I. INTRODUCTORY REMARKS. THE TERM SEMITIC. DIFFUSION AND ORIGINAL HOME OF THE SEMITES. . .

I

T h e subject I — Founders of Semitic philology 2 — B o o k s recommended 3—Distribution of the Semitic races 4 — T h e i r original seat 5.

CHAPTER

II.

GENERAL SURVEY OF THE SEMITIC LANGUAGES .

10

Northern and southern Semites 10 — Northern dialects: Babylonian and Assyrian 1 2 — A r a m e a n group 1 4 - - W e s t e r n and Eastern Aramaic 1 9 — M o d e r n Aramaic dialects ibid.— Canaanites 21—Phoenicians 22—Hittites 2 3 — H e b r e w s ibid.— Moabites 25—Southern dialects : Arabic 26—Himyaritic 2 8 — Ge'ez or Ethiopic 29—Relation of the Semitic languages to the Indo-European 30—to the Egyptian 33.

CHAPTER

III.

SEMITIC WRITING

35

Literature 3 5 — E g y p t i a n origin of the alphabet 36—Oldest monuments of Semitic writing 3 7 — O l d Hebrew alphabet of the Siloam inscription 38—Square character 39—Aramaic alphabet ibid.—Alphabets of the southern Semites 40—Inadequacy of the Semitic alphabets 41.

C H A P T E R I V . THE LETTERS OF THE SEMITIC ALPHABET AND THE CHANGES THEY UNDERGO T h e gutturals 4 2 — T h e palatals 5 0 — T h e dentals 52—Aspirated dentals 5 5 — T h e sibilants 5 7 — T h e labials 6 4 — T h e liquids 6 7 — T h e weak letters w and_y 6 9 — T a b l e of permutations 73.

42

CONTENTS.

X

PAGE C H A P T E R

V.

T H E V O W E L S A N D THEIR

PERMUTATIONS

.

.

75

The original vowel-system 75 — Modifications of the vowels in Arabic 76—The short vowels in Hebrew and Aramaic 78— under the influence of the tone in Hebrew 8 1 — T h e long vowels in Hebrew and Aramaic 84—The diphthongs 87— Supplementary vowels, especially in Hebrew 91—Prosthetic vowels 93.

C H A P T E R

VI.

THE

PRONOUNS

95

The personal pronouns : suffix forms 95—Separate forms : 1st person 98—2nd person 101—3rd person 103—The demonstrative pronouns 106—The sign of the definite pronominal accusative 112—The definite article 114—The relative pronouns 116—The Hebrew relative 118—Certain possessive pronouns 119—The interrogative pronouns 120—as indefinites 125—The reflexive pronouns 127—Expression of the reflexive by the aid of substantives 128.

CHAPTER

VII.

THE NOUN

131

Gender 131—The fem. termination t {at, ah) 132—used adverbially 135—with increment (te) 136—Other feminine terminations 138—Cases 139—in Arabic and Ethiopic 140—Traces of case-endings in Hebrew 141—The nouns 3N. I"IK> Dn 142— Origin of the case-endings 143—Nunation 143—Mimation 144— Origin of these ibid.— The plural 145—The feminine plural 147— Broken plurals 148—The dual 149—Construct dual and plural 151—The feminine plural in Aramaic 152—The emphatic state ¿¿¿¿—Pronominal suffixes to the noun 152.

C H A P T E R VIII.

THE VERB

161

Nominal forms underlying the inflexion of the verb 161— Intensive forms 162—Segholates 163. I. T H E P E R F E C T 165—Forms with characteristic a, i, u ibid. —Inflexion : 3 sing. fem. 167—3 pi- masc. 168—3 P1- fem. 169— Accentuation of 3rd pers. 170—2nd pers. 171—2 sing. masc. ibid.—2 sing. fem. 172—2 pi. masc. 173—2 pi. fem. 174—1 sing. 1 75— 1 pl- 177—Dual ¿¿/¿.—Later analogies to the formation of the Perfect 178. II.

THE

IMPERFECT

179—3

sing, masc.: the preformative

CONTENTS.

xi

ya 182—Other preformatives 183—3 sing. fem. 184 —3 pi. masc. ibid.—3 pi. fem. 185—2 sing. m. and f. ibid.—2 pi. m. and f. 186 — 1 sing, and pi. 187—Dual ibid.—Accentuation 188. III. IV.

THE

IMPERATIVE

VARIATIONS

188—Plural 189.

OF IMPERFECT

AND

IMPERATIVE

191—

Moods of the Imperfect in Arabic ibid.—in Hebrew : Jussive forms 192—Energetic forms 193—Moods of the Imperative 195. V.

VI.

THE THE

INFINITIVE 1 9 5 . PARTICIPLES

196—Passive participle 197.

VII. D E R I V E D CONJUGATIONS. A. First group 198—1. The intensive and iterative stem 198—2. The conative stem 202—3. The factitive or causative stems 204—B. Second group, Reflexive stems 207—1. Reflexive of the simple stem ibid.— 2. Reflexive of the intensive 209—3. Reflexive of the conative 212—Infinitives of B. 2 and B. 3 213—4. Reflexive of the causative 214—C. Third group (with characteristic syllable no) 215— 1. Niph'al and its congeners ibid— 2. Ethiopic forms with preformative an 217—3. Nithpa"el 218—4. Arabic forms with infixed n ibid.—D. Fourth group, Reduplicated stems 218—r. Pi'lel and its congeners ibid.—2. Pe'al'al 219—3. Pilpel and its congeners 220—4. Arabic Conj. XII. ibid.—J. Forms of the type katlaya 221—E. Passive Forms 222—in Arabic ibid.— in Hebrew 223—in Aramaic 224.

CHAPTER

IX.

THE

227

IRREGULAR V E R B

I. VERBS JTJJ 227—in Arabic ibid.—in Hebrew and Syriac : simple stem 228—Derived conjugations 232. II. A. VERBS i"a AND *"Q 234—1. Verbs with original - Sai. jJm> t—alai, J1m, ^Jaj,

all of which convey the idea of " cutting " in some form or other. PI or fl are found in J j , jjj-

^Jj.

^Jj,

^Jj,

all meaning "cleave" or "divide."

jjj,

^Jj,

j]j,

H, k are the basis

of ppn, npH, "IpH, Of which the original signification is also " slit" or " cut." jJli,

¿U,

Ph or fh ^aj,

^¿j,

are the essential constituents of meaning

"blow,"

"puff."

When

Semitic philology has advanced so far as to have discovered the laws by which the original biliterals (assuming their separate existence) were converted into triliterals; when we are able to account for the position and to explain the function of each variable constituent of the triliteral roots; then, and not till then, may we venture to think of comparing the primitive IndoEuropean and Semitic vocabularies. Meantime, to assert the identity of such a word as ¡"113 " he built" with pono, or of l y S TT

~

T

" he burned up " with nrvp, is little better than sheer folly. And why ? Because the comparison is not that of original forms, but of an original form (or what is very nearly so) with a comparatively late development. ¡"03 was originally b&n&yá; pdno is a softening of posno, as we learn from its perfect and supine, and includes a suffix and a pronominal element. IJJS originally sounded ba'ara; ~

T

irvp is stated to be a contraction of -rrvip, which probably stands for an original *pavar, and comes from a radical pu, in Sanskrit " to be bright," " to purify," plus a derivative suffix. If such comparisons as these could be upheld, they would prove that Hebrew and Arabic were not merely connected with, but actually derived from Sanskrit or Greek or Latin. What has been written on this subject by Fiirst and by the elder Delitzsch in his Jesurun (1838) is absolutely worthless; as are also the lucubrations of von Raumer and Raabe. The best that can be

IL]

33

ROOTS.

said about it you will find in the younger Delitzsch's Studien über Indogermanisch-Semitische Wurzelverwandtschaft (Leipzig 1873) and in McCurdy's Aryo-Semitic Speech (1881). A s to the affinity of the Egyptian language with the Semitic stock, that is also a question which is as yet sub judice. Benfey, in his well-known work Ueber das Verhältniss der ägyptischen Sprache zum semitischen Sprachstamm (Leipzig 1844), sought to establish this affinity by various considerations, grammatical and lexicographical; and the conclusion to which he came was, that the Semites are only one branch of a great family, which includes not only the Egyptians but also all the other languages of Africa. His views have been combated by Pott, Renan, and other scholars; and certainly in this unrestricted form they seem to land us in almost Turanian absurdities. But with regard to the ancient Egyptian and the Coptic, Egyptologists seem gradually ^to be arriving at conclusions similar to those of Benfey. De Rougö, Ebers, and above all Brugsch, in the introduction to his Hieroglyphic Dictionary, have declared their belief in the descent of the Egyptian from the same stock as the Semitic languages. A n examination of the Coptic alone readily suggests several considerations in support of this view. For example, there is the marvellous similarity, almost amounting to identity, of the personal pronouns, both separate and suffixed—a class of words which languages of radically different families are not apt to borrow from one another. " I " in Coptic is knoK, " Th0U "

It TOR, HT&.K.

" H e " n-e-oq, etc. " She " iwKJc, etc. " W e " kiton, kn&.n " Y e "

TI-ATOTEN, TITTOTTT, JITA-TII

" They " N-E-ioou', J I T O O T , JITA.-*The suffix pronouns I give as they appear in connexion with the preposition na. " to." " to me " WHI, N«.I u " to thee," m. n&.u f. ne " to him " n&.q " to her " ita^c

" to us " iw>n " to you " moTen, imten " to them

"

NWOTS-, N&.T

COPTIC.

34

[CHAP. II.

Again, there is the curious resemblance in the forms of some of the simplest numerals; e.g. 1, masc. o-s-^i, ovt>, otoot ; fem. oti, o-s-ei, ottoot 2, masc. cn&T, fem. cenTe, cnou-^ 7, masc. gj^ajq, Cfc.njq; fem. aj&.ujqi, ca-ujcje 8, masc. oj-mhii, ujMoim; fem. ujMHni, uj-uoime. In the verb, the formation of the present tense presents a remarkable analogy to that of the Semitic imperfect or, as some still prefer to call it, future,—I mean the form sing. I.

't- TtojU I am joining, adhering;

2. m . K. T(OM,

TIOM

E.g.

pi. I. Ten. twm 2. TCTett. TCOM

f. TC. TIOM

3- m. q. TIOM

3. CC. Tlx)M

f. C. TOMl

Analogies like these seem to favour the idea of a genetic relationship between the Semitic languages and the Egyptian; or at least of a closer affinity than can be said to subsist between the Semitic and the Indo-European. To discover any connexion between the two latter, we must endeavour to work our way back to the very earliest stage of their history—to a period before Semitic really was Semitic; we must try to disintegrate the triliteral Semitic root; to extract from it the biliteral, which alone can be compared with the Indo-European radical. And if haply we succeed in this, it is apparently the utmost that we can hope for; their subsequent developments, the growth of their grammatical systems, are wholly distinct and discordant. But the connexion between the Semitic and the Egyptian languages seems to be of a somewhat nearer kind. It is true that we are met by the old difficulty with regard to the form of the Egyptian roots, the majority of which are monosyllabic, and certainly do not exhibit Semitic triliterality; but, on the other hand, we have not a few structural affinities, which may perhaps be thought sufficient to justify those linguists who hold that Egyptian is a relic of the earliest age of Semitism, of Semitic speech as it was before it passed into the peculiar form in which we may be said to know it historically.

CHAPTER SEMITIC

III.

WRITING.

A F T E R these preliminary investigations and surveys, there remains yet another subject on which it is desirable to say a few words before we address ourselves to the special object of these lectures, the comparative grammar of the Semitic languages. That subject is—the origin and history of Semitic writing. My account of this interesting topic must, however, be very brief and sketchy ; the more so as I hope to treat it more fully in a subsequent course of lectures. Meantime I would refer those of you who seek further details to the treatise of the Vicomte de Rougé, Mémoire sur l'origine égyptienne de Valphabet phénicien, 1874; to the work of Lenormant, Essai sur la propagation de Falphabet phénicien dans Fancien monde, of which the first part appeared in 1872, and two more have since been added, though the book must now unhappily remain unfinished ; to the Mélanges d'Archéologie orientale of the Cte de Vogué, 1868 ; and to Mr Isaac Taylor's excellent book The Alphabet [London, 1883], especially vol. i.

All writing—Chinese, Assyrian, Egyptian—was originally pictorial. The next stage was that of the ideogram. Each picture received a fixed, often symbolic, value, and was always used in the same way. In Egyptian the figure of a tongue meant " to speak " ; two hands holding a shield and spear meant " to fight " ; and so on. The third step—a great one—was to make a particular sign stand in all cases for one and the same syllabic sound ; e.g., the figure of a mouth for ro, the Egyptian for " mouth " ; the figure of a hand for tot ; the figure of an eye for iri. The last and greatest step was to divide the syllable into its component parts or letters, and to represent

36

ORIGIN OF

[CHAP.

each of these by a special figure. Here the ancient Egyptians happily lighted upon what has been called the "acrophanic" principle ; that is to say, they designated each letter by the picture of an object, the name of which began with the sound which the letter was to represent. For example, the picture of a lion, would mean the letter /, because the word labo, A^&oi, begins with that sound ; the picture of an owl the letter m, because the word mUlag, begins with that sound ; the picture of a mouth the letter r, because the word ro, po, begins with r. T o this stage the Egyptians attained at a very early period ; but, like the inventors of the cuneiform characters, they did not avail themselves fully of their great discovery. On the contrary, they mixed up the two principles, the ideographic and the phonetic, in a manner that is extremely puzzling to the reader. T o an Egyptian the figure of a lion might actually mean " a lion " ; or it might, as an ideogram, be a symbolic sign, meaning " preeminence," " sovereignty " ; or it might, as a mere letter, designate the sound /. T o an Assyrian a certain combination of wedges might convey the idea of " the earth " ; but phonetically it might express the syllable ki. Hence the mass of determinative signs of various kinds employed in writing by the Egyptians, Assyrians and Chinese. Of course, in process of time, the picture gradually faded away. Details were neglected ; a few bold strokes sufficed to depict the object intended ; and, in the end, the form of the letter often bore little or no resemblance to the thing from which it was derived. T h e group of wedges, the hieratic or demotic character, and the modern Chinese sign, are, in most cases, wholly unlike any object in heaven or earth. The Egyptians, in addition to the stiff pictorial hieroglyphs, had two sorts of more current or cursive characters, called thè hieratic and the demotic. T h e former, used (as the name indicates) by the priests, was employed for sacred writings only, the latter, used by the people, served for all ordinary secular purposes. It was of the former that the inventors or adapters of the Semitic alphabet appear to have availed themselves. They used the forms which are found in papyri anterior to the eighteenth dynasty, belonging, roughly speaking, to the period between 2100 and 1500 B.c. De Rougé endeavours to show

III.]

THE

ALPHABET.

37

that out of the twenty-two Phoenician letters, fifteen are beyond doubt directly derived from Egyptian models, whilst only one, the 'ayin, is clearly of Semitic invention. It may be that the " spoiling of the Egyptians" went so far; that the plundering Semites appropriated not only the idea of a written alphabet, but the very forms which the letters were to take. However, I cannot profess myself entirely convinced, not even by Mr Isaac Taylor's argumentation. If they did so, the Semites both remodelled and renamed their acquisitions. Out of the Egyptian eagle or vulture they made the head and horns of an ox, SpK; the throne, ~~Zu, became the head and neck of a camel, group of lotus plants growing out of the water, a set of teeth, W , and so on1. Deecke's attempt to derive the forms of the Semitic alphabet from the Assyrian, I must regard as an utter failure. You will find his views stated in an article in the ZDMG., vol. xxxi. p. 102. The remodelled Egyptian alphabet has been, in the hands of the Phoenicians and other Semites, the parent of nearly all the systems of writing used by the nations of Europe and Western Asia. The Greeks received it from the Phoenicians, and having again remodelled it, passed it on to the Etruscans, the Romans, and the Copts. The sacred books of the Persians are written with an alphabet of Aramaic origin. The Uigur Tatars [and through them the Mongols] acknowledge a similar obligation. And even the Sanskrit alphabet, with all its Asiatic offshoots, has been traced to a South Semitic source. The oldest monument of Semitic writing as yet discovered, with what we may call a certain date, is the inscription of Mesha', king of Moab, which we may place about B C. 890s. Here we find already a carefully developed system of orthography and punctuation, which contrasts favourably with those of Phoenician inscriptions of later date by several centuries. Final vowels are expressed by the letters s (i), 1 (ü) and H (