In the of the Fuad I University, vol. 9, part 1, May, 1947, pp. 25–27, Dr. K. Y. Nami has published an interesting boundary inscription now in the possession of that University. In his treatment of the inscription, however, he has overlooked the fact that a parallel text is to be found in the Corpus Inscriptionum Semiticarum, iv, 949. This is unfortunate, since a correlation of the two texts is most helpful in understanding the formula contained in them.

The following are the two texts, with the renderings proposed by Nami for his inscription and by Rhodokanakis for the Corpus text.

The word(pi. ) appears in most of the lexicons which I have consulted, but it does not seem to be known in the sense of a supernatural creature, although this meaning is familiar to Yemenite folklore and dialect.

I

IT is well known that the Ṱabaqāt al-shu‘arā’ of Muḥammad b. Sallāam al-Jumahi is one of our most important source-books for the history of Arabic poetry and poetic theory; it is equally well known that the text as at present available is both deficient, because of lacunae in the manuscripts hitherto discovered and used, and otherwise unsatisfactory, on account of inadequate editing. Despite the very numerous emendations to Joseph Hell's edition (Leiden, 1916) proposed by several eminent scholars, the book still remains lamentably short of perfection; though there can be few texts in the whole of Arabic literature which it would be more desirable to have in good and proper shape.

It has now been my unusual fortune to study in Mr. Chester Beatty's library an exceedingly fine old copy of the Ṱabaqāt al-shu‘arā’, undated but certainly of the fourth/tenth century. This hitherto unknown manuscript not only supplies a great part of the major lacuna in Hell's text (p. 19), but also contains other extensive additions summing up to a considerable fraction of the whole work; it presents moreover a very reliable recension of the book, and its numerous variants will make it possible at last to establish a satisfactory edition.

In the present paper a collation is offered of all the passages contained in Hell's text for which new readings are provided by the Chester Beatty manu-script. It is hoped in a subsequent paper to publish the additional material for which that manuscript constitutes the sole authority. Scholars will be interested to note how remarkably this fresh evidence confirms many of the brilliant emendations proposed by the famous Arabists who have from time to time devoted themselves to the improvement of Hell's edition.

IT is a commonplace that the New Testament never speaks of the ordination of bishops. Elders or presbyters with powers of administration and instruction were the only ordained ministers. The church at Jerusalem was in the care of “the apostles and the elders”. When St. Paul visited Ephesus, it was the elders (presbyters) whom he summoned and addressed. It is not a little strange that he should tell these elders that the Holy Ghost had made them “bishops” to feed the church of God. The natural inference (unless the word episcopus is used in a general sense of overseer) would be that the office of elder and bishop could be held by the same person. St. Paul left Titus in Crete with the express purpose of completing the organization of the church there by appointing elders. But here, too, the bishop and the elder seem to be one and the same if any meaning is to be given to γὰρ in v. 7. Again, 1 Peter v, 2, if the reading ἐπισκποῦντ∊ς be allowed to stand, points in the same direction; if it is to be omitted then the epistle recognizes no bishop but Jesus Christ, and then only in a verbal, not a real sense. Moreover, though there are presbyters in heaven, there are no bishops: at any rate the apocalyptist of Patmos saw none !

On the other hand St. Paul, in addressing the whole congregation of the faithful at Philippi, adds “together with the bishops and deacons”, making no mention of presbyters.

MAYYĀFĀRIQĪN, a small town situated on one of the left tributaries of the Tigrīs, at 70 km. to the north-east of Āmid (Diyārbakr), owed its importance to its situation on a short road connecting Armenia (Mush) with Upper Mesopotamia. It is probable that the ancient capital of Armenia, Tigranocerta, built by Tigran II circa 80 B.C., stood in the immediate neighbourhood of MAYYĀFĀRIQĪN.

In Islamic times Mayyāfāriqīn had a historian, Ahmad b. Yusuf b. ‘All ibn al-Azraq al-Fāriql, who wrote shortly after 572/1176. The only two copies of this curious work belong to the British Museum. The detailed description of the work and the first systematic presentation of its contents belong to that accurate British historian H. F. Amedroz, who has so considerably increased our knowledge of the medieval Arabic sources for the Near East. Numerous passages from Ibn al-Azraq are quoted by Amedroz in the footnotes of his edition of Ibn al-Qalanisi (1908). In more recent years M. Canard has published six passages of the history of Mayyāfārīqīn relative to Sayf al-daula and Claude Cahen has summed up its rich information on the early Artuqids.

These preparatory works will greatly help the future editor of the Mayyāfārīqīn chronicle. His task will not be easy, for the two versions are defective and divergent, and the best plan will be to print them in parallel columns. The script of Or. 6310 is very cursive and devoid of dots; that of Or. 5803 is defaced towards the end. The scribes were negligent even in geographical and personal names. The grammar of the author (or of his copyists) is lax and may occupy the attention of some student of vulgar Arabic in Upper Mesopotamia.

THIS paper is the outcome of a short study of the pronunciation of an Ossete born in Dzaeusegi-hseu (Wladikavkaz). The Digoron dialect of Ossetic is my informant's mother tongue, but as he left his homeland more than twenty years ago and has in the interval probably had more occasion to speak Kussian and Polish than his own language, the possibility that his pronunciation has been thereby affected cannot be discounted. At the suggestion of Professor H. W. Bailey a short investigation of his pronunciation of Ossetic was undertaken. Some of the material used was taken from notes put at my disposal by Professor Bailey, to whom I am greatly indebted for his generous help and for much invaluable information and criticism. Other material was found in a collection of folk-stories, the Pamyatniki narodnogo tvorchestva Osetin, or was supplied by my informant himself. The tentative nature of any linguistic analysis based on a short study of the pronunciation of one speaker is obvious, and it will be understood that the findings are only valid within these limitations.

The typographical conventions adopted in this paper are as follows:—

Italic type is used to show the roman spelling of Digoron as used in the Pamyatniki or by my informant.

Heavy type is used for the phonetic transcription of my informant's pronunciation.

Heavy type in single quotes is used to indicate the units of what may be called “unuttered speech”.

The use of all three conventions is demonstrated by the following statement:— “nivәrun niværun is the realization of the junction of ‘ni’ and әvәrun’”.

Several years before the war a photograph of a stone inscription in Aramaic letters was published in the Persian language periodical Kabul (vol. ii, 1932, p. 413). The following note (in Persian), printed below the photograph, constitutes the whole of our information on the material, history, and provenance of the inscription:ߞ

“A historical inscribed stone: A few years ago a broken stone tablet (lauhe) with ancient Sanskrit writing was procured in the neighbourhood of Pul-i Darunta, Laghmān, and entrusted to Kabul Museum. So far its contents have not been read and understood. The original of the above (depicted) stone is in Kabul Museum.”

Laghmān, older Lamyān (from Lambayān<*Lampakāna), is the name of a district on the left (northern) bank of the Kabul river, a little above Jalalabad; it comprises the valleys of the Lower Alingār and Ališang. This district, whose name is familiar to Sanskritists as Lampāka (also Lambāka), was traditionally regarded as part of the Indian borderlands, the ultima Thule of Jambuduīpa. Cf. Mémoires de Hiouen-thsang, i, 55, “en partant de ce royaume ( = Kāpiśī), il… franchit les montagnes noires, entra dans les frontières de l'lnde du nord, et arriva au royaume de Lan-po”. Cf. also Lampāka in the Yaksa catalogue of the Mahāmāyūrī, and further H. Lüders, SbPAW., 1930, 43, 48, 51, 63. One may presume that the traditional view reflects the conditions prevailing under the Maurya dynasty.

The inscription was made known in Europe by Professor Morgenstierne, who, on finding that the alphabet was Aramaic, asked H. Birkeland to publish it.

1. Jayhdn¯'s Lost Work

ABŪ ‘ABDILLĀH MUḤAMMAD IBN AḤMAD AL-JAYHĀNĪ, who in 302/914 became vazir to the infant Samanid king of Bukhara, was the author of the celebrated geographical work Kitāb al-mas¯lik wal-mamālīk. According to Muqaddasi, this opus magnum was in seven volumes and in it one found incorporated “ the whole of Ibn Khurdādhbih's original work ”. Muqaddasi himself saw it in Fars in ‘Adud al-daula's library. He also saw in Khorasan an abridgment of this geography, one copy of which was attributed to I. Khurdādhbih, and the other (somewhat expanded) to Jayhānī.

Jayhānī's work has not come down to us, probably on account of its bulkiness, and we know it only through the references found in numerous authors belonging chiefly to the eastern part of the Islamic world. These quotations—both those acknowledged by the borrowers, and those which are unacknowledged—have been closely scrutinized and now it would seem about time that they should be collected in a special treatise, similar to the theses published on classical authors who are known to us only in excerpts.

In 1923, in the library attached to the sanctuary of Mashhad, A. Z. Validi found a new copy of Ibn al-Faqīh's Kitāb al-buldān, more complete than the abridgment which de Goeje published in BGA., v. The discovery in itself was highly interesting and the discoverer published a responsible account of it in 1924. Yet two years later rumors, based on mere hearsay, announced the discovery of the original Jayhani. In point of fact, the Mashhad MS. (212 ff.) proved to contain only the second, and still abridged, part of Ibn al-Faqīh (roughly corresponding to pp. 173–330 of the printed edition), and of Jayhānī the only real news was that in 309/921 Ibn Fadlān was his guest in Bukhara, as recorded in the traveller's own introduction to his journey.

Introduction

The (Kara-)Kirghiz language is spoken on the territory of the Autonomous Republic of the Kirghiz by about 650,000–700,000 individuals. My Kirghiz informant was Odajarov Abdilla (20 years old), of a village near Frunze (Purunza, 42° 50" north latitude; 75° east latitude), the capital of Kirghizistan. He and his parents had always lived in that village, where they had worked as agricultural labourers. He had been to school for four years and spoke a little Russian apart from Kirghiz. For the literary language I had a grammar at my disposal (Qьrgьz Tilinin elementardьq gramatikesi = “Elementary Grammar of the Kirghiz language”, by Şaвdan uulu Açьman and I. A. Batmanov), from which I had taken most of the examples.

THE early history of Central Asia, the background of the Turkish expansion in the ninth and tenth centuries, and of the Mongol eruption in the thirteenth, presents a large number of problems which the slowly progressing study of documents made accessible to us in particular by the late M. Aurel Stein and Paul Pelliot is bringing nearer to solution. Among these documents those from the kingdoms of Khotan and Krorayina are of outstanding importance for the early period. It seems desirable to treat of some of these problems here.

1. The vocabulary of Khotan and the Iranian words in the Krorayina documents

A large part of the vocabulary of the Khotanese language is now known, much of it already accessible in print, and it is possible to confront it with the Iranian words found in Krorayina. It is at once noticeable that in the two vocabularies, while part is identical, as in lāstana- ‘quarrel’, corresponding to Krorayina lastana, or prahona- ‘garment’ to Krorayina prahuni, yet marked differences appear which are unlikely to be due only to difference of time.

I would notice here the following ten cases.

1. In an article Recent Work in ‘Tokharian’ (Trans. Phil. Soc. 1947, 149–150), I called attention to the difference between Krorayina guśura, Kuchean Sanskrit gauśura, and the Khotan *bisīviraa-, attested in the plural bīsīvīrā of high birth ‘(Suvarnabhāsa-sūtra, Khotanese Texts I p. 248, 73 r 1), and the related bäsīvärāssai, translating Sanskrit kulaputra, a person of good family. The treatment of the initial υ- shows that here two dialects are concerned.

Among a bewildering mass of archaeological reports and learned papers the small book of the late Dr. Ernest Mackay, The Indus Civilization, published in 1935, has provided an excellent concise outline of the subject. It has now been reissued under a significantly amended title, with revisions and additions by the author's widow. New maps and illustrations have also been added.

Since 1935 further researches have considerably modified our earlier view of the Indus culture. The excavations at Chanhu-daro, conducted by Dr. Mackay for the American School of Indie and Iranian Studies have shown that the culture of which the Harappā remains are the type was superseded, in this part of Sind at any rate, by later intrusive cultures, those of Jhukar and Jhangar (p. 3). Advances in Mesopotamian chronology have permitted a closer estimate of the date of the abandonment of the Indus cities. This, it appears, took place at about the same time as the end of the first Babylonian dynasty, which, on the most recent evidence, must be placed as late as the sixteenth century b.c. (p. 157). It has been conclusively shown that we may no longer retain the rather Utopian picture suggested by reports of earlier excavations, which seemed to reveal an affluent commercial people, living in comfort and security in unfortified cities, and free from the oppressions of a theocratic government such as ruled in contemporary Sumer and Egypt.

Dr. Mackay and all serious archæologists doubted the accuracy of this picture. In the first edition of The Indies Civilization it was pointed out that the control of such well-planned cities as Mohenjo-daro and Harappā demanded a highly organized government, that the excavations had not been completed, and that further work on the sites might reveal fortresses and royal palaces.

III. Apokopa or Poinai Them

The first and westernmost of the seven Indian “mountains” mentioned by Ptolemy (vii, 1, 19) bears the curious name τὰ ’άπίκιπακαλεîται Ποιναὶ Θευ. Its extremities are stated to be situated at 116° E. 23° N. and 121° (v.l. 124°) E. 20° 30′ (v.l. 26°) N. In Ptolemy's idea it was therefore a mountain-range of considerable length running either in a south-easterly or north-easterly direction.

Lassen points out that according to Ptolemy the Apokopa was situated to the west and south of the Vindhya. This, he argues, is impossible as the latter extends almost as far west as the coast so as to leave no room for another mountain range. He then proceeds to correct Ptolemy's error with the aid of Megasthenes and the Māhābharata. In a fragment of the Greek ambassador's work on India, preserved by Pliny, mention is made of a high mountain named Capitalia. This name Lassen renders by ” Capital Punishment ” (“ Todes-strafe ”) and assumes it to be the equivalent of Ptolemy's Poinai Theon, i.e. ” Punishments of the gods ” (“ Strafen der Götter ”). He believes it to originate from a legend relating how the gods punished a crime committed on this spot and furthermore detects a reference to such a legend in the name Apokopa to which he assigns the meaning of ” hewing down, chopping off ” (“ das Abhauen ”). An allusion to the supposed legend Lassen discovers in the Māhābharata (Bombay ed., iii, 82,55–6). Here in the course of an enumeration of firthas a visit to the mountain Arbuda is recommended and it is briefly stated that here a cleft (chidra) of the earth is found and the hermitage of Vasistha celebrated in the three worlds.

Although Geiger's Grammar of the Sinhalese Language (1938) has cleared up admirably the main line in the historical development of this language, there still remains much to be done. Especially in the morphology of the verb there are many obscure points and in the following lines we shall try to unravel some of them.

1 The Sinhalese has three conjugations (paradigms:labanavā “to get”, badinavā “to fry”, tävenavā “to be heated”). Geiger explains the third conjugation from the MInd, passive: tävenavā < *tāvīya- < Skr. (causative passive) tāpyate (Geiger, § 140, 3), which coincides well with the fact that nearly all the verbs belonging to this conjugation are intransitive. As for the development ī > e compare elu: name of the old Sinhalese language < Sīhala, neranavā “to set aside, put out of the way ” < P. nīharati (Geiger, § 21, 3). So there can hardly be any doubt that this explanation is in the main correct. The -lya- is not always the passive-morpheme; thus Geiger mentions ālenavā “to adhere to” from Skr. āliyate; another case is pipenavā “to expand, open (of flowers)” from the denominative Skr. puspyati > *pupphīya- (other examples later in this article). Geiger derives vädenavā “to affect, concern, to be struck by an arrow” (Clough's dict.) from Skr. vrajati (Et. Gl.; Geiger gives the meaning “to strike against, to enter, intrude ”); I should, if the meaning given by Geiger is correct, prefer the derivation from āpadyate with loss of the initial vowel as in bisev “royal unction ” from abhiseka-, etc. (Grammar, § 30, 2); but it must be admitted that all the examples mentioned by Geiger contain short initial vowel.

1. GEIGER, in his SghG., § 21, remarks “Vowel-levelling rests on the same tendency of making the vowels of a word uniform as the vowel-assimilation. It is the equalization in two successive syllables of a —i (e) and i (e) —a to e —e, and of a — u (o) and u (o) — a to o — o.”

This term “Vowel-levelling ” is redundant, for if this term is to be used it should apply equally well to vowel-assimilation. It is clear that, for instance, a — e: e — e or a — o: o — oisas much vowel-assimilation as, for instance, i — u: u —u or u —i:i —i. It is also difficult to reconcile the fact that a — i, which was shown to become i — i [cf. § 14], or i —a, which was shown to become a —a [cf. § 18], should also become e — e, and again that a —u, which was shown to become u — u [cf. § 15], or u —a, which was shown to become a —a [cf. § 21], should also become o — o. This e —e and o —o are not direct developments in Sgh. but are the result of contaminations which are discussed under the various categories given below, and therefore Geiger's theory of “Vowel-levelling ” falls to the ground. These contaminations are due on the one hand to the influence of loan-words and on the other to the instability in the development of Sk. s and Sk. h in Sgh.

In an article on the Malay Founder of Malacca (Bulletin of the School of Oriental and African Studies, vol. xii, pts. 3 and 4, 1948) I used the Suma Oriental of Tomé Pires to advance the view that the Portuguese and Chinese were wrong in failing to identify the Parameswara with Megat Iskandar Shah, the Mukans-autirsha or, better, Mou-wa-kan-sa-yu-li-eul-cha of the Chinese.

I suggested that all the evidence went to show they were not father and son but the same ruler with the different titles he bore before and after conversion to Islam, and that this founder of Malacca died in 1424.

In a posthumous article in Toung Pao (vol. xxxviii) Paul Pelliot (p. 168) notes the further point (which is not apparent from Groeneveldt's Notes: Essays on Indochina, 2nd ser., vol. 1) that, according to the Ming-chan tsang and Ming Che, Megat Iskandar Shah visited China in 1414 to announce the death of his father and ask for investiture; but that according to the Chou-yu tcheoutseu lou (8,136) and Ming-chan tsang it was in 1419. This discrepancy confirms my doubt as to the accuracy of the Chinese chroniclers, who seem to have construed every Malay visit as an act of homage.

Paul Pelliot (p. 169) gives another case where the Chinese sources are obviously inaccurate. The Ming Che (325, 3a-4b) records events of 1474, then of 1481, and continues that “some time after” two envoys Lin Jong and Houang K'ien-heng were sent to invest “the son of the king, Ma-ha-mou-cha” (Groeneveldt's “Mahamusa”).

WHILST the tonal languages spoken in West Africa have latterly received W a good deal of attention, it has not been generally realized that such languages are found also in parts of New Guinea. In New Guinea there are in fact two types of tonal languages. In one, the tones may be described as ” ornamental”, i.e. though they exist they do not seem to have semantic value, but rather to be connected with a certain type of sentence rhythm; in the other the tone systems are more definitely akin to those of Western Africa, and in such languages tones do possess semantic value. It is interesting that one at least of the latter group of languages tends also to be monosyllabic. It is hoped to provide a study of the former class of languages—at least one representative of which is also found in Northern Australia—at a later date. The present paper is occupied with the languages in which tones do have semantic value.

These languages, so far as has been observed yet, are two in number, and they are spoken along the shores of Huon Gulf, in north-eastern New Guinea. The map on p. 186 will show their locations. Both belong to the group of languages known as Melanesian, and this makes the occurrence of tones even less expected. A considerable number of other languages are spoken round about this area, and though some of these also are Melanesian, they do not seem to have developed tones. Moreover, it would seem that the use of tones in these two languages antedates the coming of the Melanesian languages, for reasons that will be given towards the end of this paper.

FOR years now the question of orthography in the South African Bantu languages has kept the authorities busy. It flares up periodically in different parts of the country, and local committees are called from time to time to deal with the conflagration.

In the past, unfortunately, these committees have too often concerned themselves with the problems of one language only, and apparently ignored both the effect their decisions would have on literature exchange and the attempts of previous committees to solve similar problems in related languages. Thus we have Zulu and Xhosa, very closely related languages, with considerable orthographical differences, while North Sotho and Tswana now differ from each other and from South Sotho, which (perhaps wisely) has set its face against orthographic change since 1906.

Until Dr. Jacob Nhlapo launched his campaign for Southern Bantu linguistic unity, few people had seriously considered the possibility of an orthographic system which would cover both the Nguni and Sotho groups, the feeling being that such unity lay in the realms of wishful thinking in view of the virtual impossibility of achieving unity within the groups.

The present article is an attempt to combine the experience of past efforts into a system that would actually be applicable to these two language groups. The acceptance of such a system would naturally involve great sacrifices of tradition and prestige from both parties. This aspect is not to be ignored, but at the same time can have little connection with the scientific elaboration of an alphabet. Consequently the argument “The Sotho (or the Nguni) would never accept such a solution” must be relegated to the sphere of practical politics.