It is well known that the Almohades instituted a politico-social system or hierarchy. In it the subjects of the Almohade state were divided into ranks or grades of aristocracy which do not appear to be closely connected with their administrative arrangements but may originally have been a battle-array. The broad lines of this system are fairly clear; it is the details that are often obscure. The Mahdī's, later the Caliph's, advisers, the ‘ cabinet’ of his government, were ten men known simply as ‘ the Ten’ or Ahl al-jamā'a. Below them came a council called ‘ the Fifty ’ {Ahl Khamsīn or Ayt Khamsīn). The mass of the people were divided thus in descending order (I follow here Ibn al-Qaṭṭān, later copied by the anonymous author of the Ḥulal al-mawshiyya): the Talaba, the Ḥuffāẓ, the Ahl al-dār, Hargha, Ahl Tīnmāl, Gadmīwa, Ganfīsa, Hintāta, the Jund, the Ghuzāt, and Rumāt. Other writers have additional categories and put them in a slightly different order. Gaudefroy-Demombynes has used the Kitāb al-ansāb for the light it throws on the description of Hafsid institutions given in his partial edition of ‘Umarl's Masālik al-abṣār, and more recently Terrasse has summarized its information in his Histoire du Maroc.

This paper attempts to deal with a short passage found at the end of the third chapter of the Huai Nan tzu book, a compendium of learning assembled in about 120 B.c. under the patronage of the prince Liu An I first became interested in it after reading the tantalizing reference in Needham (SCC, III, 224). (It occurs to me that the preceding sentence or some variant of it is likely to occur in learned journals with a high frequency for the next 50 years at least.) My first efforts to understand the original text led me to the conclusion that it was unique in a number of ways and had points of interest that did not fully appear in the version of Maspero, 1929, 348 ff. Professors A. C. Graham and D. C. Lau both gave very generously of their time in discussions of my first draft. Professor Graham raised the question of possible Mohist influences and brought to my notice a commentary by the Ch'ing scholar Ch'ien T'ang , Huai Nan tzu t'ien wen hsün pu chu c. 1788. I did not feel able to follow Ch'ien in all he wrote, but was relieved to find that we reached the same general conclusions about what the text was saying, although Ch'ien seems to miss much that is important. While there are still some obscurities of language in what appears to be a rather corrupt text, I offer the version given here with a fair degree of conviction that it does not substantially misrepresent the intentions of the unknown author. Any errors that I have obstinately retained despite the helpful advice given to me throughout the time I was engaged on this work are, of course, my responsibility alone.

Tigre, an Ethio-Semitic language spoken in Eritrea by a predominantly Muslim population, is known to have remained in intensive contact with Arabic since at least the nineteenth century. In the present article we attempt to survey the lexical and grammatical features of Tigre which are potentially attributable to Arabic influence. The genealogical proximity of Arabic and Ethio-Semitic complicates the task, as borrowed features and common retentions are not always easily distinguishable. In the lexical domain, the Arabic impact is undoubtedly conspicuous: thus, even in the core vocabulary Arabic loanwords are more prominent than in any other Ethio-Semitic language. Nevertheless, for many individual lexemes the borrowing hypothesis remains to be substantiated. Recent Arabic influence is probably responsible for a few relatively superficial features of Tigre phonetics and morphology. Several more deeply-rooted features may in principle be due to a more ancient and more intensive contact with Arabic: the specific influence of the ejectives on the vowels; the grammatical categories of singulative and diminutive; the diversity of verbal noun patterns in the basic stem; and several types of broken plural formation.

Bei der Durchsicht der Brāhmī-Handschriften der “ Mainzer Sammlung / [Mz]1 entdeckte einer der beiden Verfasser [Maue] mit Mz 639 ein Fragment, das neben Sanskrit eine Übersetzungssprache enthielt, die aufgrund einiger Indizien dem Iranischen, am ehesten dem Sogdischen, zuzuordnen war. Anhand einer ersten vorlä ufigen Umschrift konnte Professor R. E. Emmerick (Hamburg) das Sakische ausschlieβen. Der zweite Autor [Sims-Williams] erkannte darin zweifelsfrei das Sogdische. Somit ist nun auch dieser mir. Dialekt in Brā hmī belegt und erstmals ein sogd. Textstück, das definitiv auf eine Sanskritvorlage zurückgeführt werden kann. Der besondere linguistische Wert liegt in der durch die Schriftart erzwungenen vollen Vokalisierung. Das neu entdeckte Bruchstück nährt die Hoffnung auf weitere sogdische Materialien in Brāhmī.2 Die lesbaren und verständlichen Teile des Blattfragments lassen klar erkennen, daβ esp. sich um einen Heilkundetext handelt, vermutlich einen Abschnitt über Augenkrankheiten.3 Wir haben Reste von vier Rezepturen oder Paragraphen, wie die erhaltenen Zahlen beweisen, die es auch erlauben, Vorderund Rückseite zu bestimmen. Im übrigen aber ist das Blatt so sehr fragmentiert und die Schrift teilweise so stark abgerieben, daβ das gewohnte Schema (zu behandelnde Krankheit, Zusammensetzung des Medikaments und dessen Dosierung) an verschiedenen Stellen zwar noch durchscheint, aber im einzelnen nicht rekonstruiert werden kann. Dies beeintrachtigt die Sicherheit und das Verständnis der Lesungen. Hier kann der entscheidende Fortschritt nur vom Mūla-Text oder zumindest von Paralleltexten kommen Die Suche danach, an der sich freundlicherweise auch R. E. Emmerick beteiligte, hat bisher nur zu einem negativen Ergebnis geführt:

Unlike most ancient Arabic poetry, the poems attributed to Umayya b. Abī l-Ṣalt treat subjects that are also prominent in the Quran, such as creation, eschatology, and episodes from Biblical history. The authenticity of this corpus has, however, been the subject of some controversy. After a critical survey of previous scholarship, this article examines one particular passage from the Umayya corpus dealing with the destruction of the ancient tribe of Thamūd, which, it is argued, is likely to be pre-Quranic. The article then proceeds to highlight the crucial differences, both in content and in literary format, that exist between Umayya's retelling of the Thamūd narrative and its earliest Quranic version, and concludes with a number of general remarks on the Quran's religious milieu as reflected in Umayya's literary output.

Awngi is a central Cushitic language, the southernmost representative of the Agaw group. It is spoken in the Provinces of Agawmǝdǝr and Mätäkkäl, Governorate General of Gojjam, Ethiopia, by at least 50,000 persons. This dialect group has been called by Conti Rossini ‘Awiya’. This is a misnomer; the word awíya means ‘Agaw man’ (literally: ‘son of Agaw’, see §3.4). The language itself is called by its speakers awῃi, with the derivative -ῃi (Amharic -iñña) used for languages. In my earlier publications, I simply called it Southern Agaw, but following the preference of several of my colleagues, I adopted the self-denomination ‘Awngi’.

The people known to both Africans and Europeans as Mbugu (Wambugu, Vambugu) live in the western Usambara mountains in Tanzania among the Sambaa. Their language, long known as posing a linguistic problem, is usually-described as having a non-Bantu lexicon and a Bantu grammatical structure. However, the actual linguistic situation is somewhat more complicated than this, as our researches (A.N.T. 1965, M.A.B. 1959) now show.

The existence of a tradition of bow-playing is recorded in some of the earlies written accounts of the Zulu people. Captain Gardiner in 1836 (pp. 104—5 and p1. 1) notes the presence of a gourd-resonated musical bow. It is not, however, clear which of the two Zulu gourd-resonated bows Gardiner encountered. He provides an illustration of a simple bow with undivided string and gourd resonator attached near the centre of the stave but does not note its name. the instrument would seem to be a cross between the Zulu ugubhu, a simple bow with undivided string, but with the gourd resonator attached near the lower end of the stave, and the Zulu umakhweyana, a simple bow with the gourd resonator attached near the centre of the stave, but in which the string is divided by means of a loop which is secured within the centrally-located resonator. Angas (1849:p. 111 and p1. 25) also seems to have confused the two instruments. Plate 25 is an illustration of a young man playing what appears to be the ugubhu (although the gourd resonator is situated higher up the stave than is usual), but the drawing in the text (p. 111) in explanation of plate 25 shows a young man playing the umakhweyana. Thus, while the ugubhu is generally regarded as the older and more authentically Zulu of the two instuments (the umakhweyana being thought to have been borrowed from the Tsonga of Mozambique around the turn of the nineteenth century), it is apparent that, from at least the time of the first documentation of the Zulu, both instruments were in current usage.

IN a previous article, we noted that by the time of the death of Shāh Isma'īl I (19 Rajab 930/23 May 1524) it is possible to observe certain trends in the development of the principal offices of the Ṣafawid state. During the last decade of the reign of Isma'īl I, there had been a movement away from the theocratic form of government which obtained at the time of the establishment of the Ṣafawid state, and this was reflected in changes both in the scope and function and in the relative importance of the principal offices of state: there was a tendency to lay less emphasis on the paramount position of the wakīl as the vicegerent of the Shāh (wakīl-i nafs-i nafīs-i humāyūn), and to regard him rather as the head of the bureaucracy; there was a decline in the power of the ṣadr; and Isma'īl himself took steps to curb the power of the amīr al-umarā.

Rājasthānī in its historical aspect has been admirably served by Western scholarship—most notably by the work of Tessitori on Old Western Rājasthānī (OWR), and of Sir Ralph Turner on Gujarātī phonology (since OWR ‘represents the stage of comparative linguistic unity of Gujarātī and Rājasthānī’). Last year saw the unveiling in Bikaner of a memorial to Luigi Pio Tessitori; this year sees the seventieth anniversary of his birth, and the publication of the present volume in honour of Sir Ralph.

Those who wish to ascertain ‘le dernier état de ces questions’ will be grateful to Professor G. Dumézil who has now summed up for us in brief compass his immense labours of nearly a quarter of a century in the study of Indo-European religion. Although the book does not cover the whole wide range of his researches, it does deal comprehensively (and with full bibliographical notes) with the central dogma:

'La comparaison des plus vieilles traditions des divers peuples parlant des langues indo-europeennes a révélé un fond considérable d'éléments communs, et d'éléments non pas isolés, mais organises en structures complexes, dont les autres peuples de l'ancien monde n'offrent pas l'équivalent. L'xposé qu'on va lire est consacré à la plus importante de ces structures'(introduction, p. 5).

A Mongst the Kurds there is found a considerable variety of religions, which has caused unfavourable observers to regard them as idolatrous. Not only are the Christians of several dominations represented among them, but also the Muslims, who form the predominant millah or creed among the peoples of Kurdiâstan in addition to these there are several less important sects, of which by far the most important are the Qizilbâsh and the Yazîdî Kurds.

The Bornu quranic manuscripts with interlinear ajami glosses in a Kanuri dialect were discovered by A. D. H. Bivar in the late 1950s. The unique corpus of written evidence for the Kanuri language as spoken over 300 years ago has never been analysed in any depth. The first part of the paper contains a general description of the collection and a discussion of codicological and palaeographical characteristics of the manuscripts. The second part comprises a brief note on the Kanuri language and a description of the Bornu writing system used for quranic commentary in this language. The third part explores the phonological and morphological properties of selected examples of glosses. Finally, a preliminary view on the dialect affiliation of the language of the glosses is suggested.

Attaining just a glimmer of an understanding of Chinese or Japanese in the seventeenth century required prodigious feats of imagination and the abandonment of widely-held convictions about the nature of language. Progress towards a rational and sophisticated understanding was held up by persuasive but fantastical theories to which lifetimes were devoted in vain. In their very different ways both languages were subversive of contemporary notions of language, and the conceptual frameworks for adequate descriptions had to be generated from scratch. The difficulties can scarcely be overestimated: the Chinese language was vigorously attacked in 1678 as the language of the devil on the ground that its pictographic nature would occasion a breach of the Second Commandment if the name of God were written, and the following year Leibniz drew up a list of questions concerning Chinese which ask, among other things, ‘whether the Chinese language was artificially constructed, or whether it has grown and changed by usage like other languages’. Japanese attracted less interest, but the difficulty of the two languages was legendary: in 1708 the Dutch scholar Adrian Reland (1676–1718), who published numerous works on Persian, Jewish and Islamic studies, wrote of the immense numbers of characters to be learnt by anybody who wished to know Chinese or Japanese, with awe at the thought that ‘a man's life would scarcely suffice to attain perfect knowledge of one language’.

The present article may perhaps be regarded as supplementary to Professor Klingenheben's article, ‘Die Präfix- und die Suffixkonjugationen des Hamitosemitischen’, MIO, IV, 2, 1956, 210–77, in much the same way as that article followed on the article ‘Verbalbau und Verbalflexion in den semitohamitischen Sprachen’ by Dr. Otto Rössler, ZDMG, C, 2, 1950, (pub.) 1951, 461–514.

The chinese have held to the architectural standards of the past no less tenaciously than to other traditions of their ancient civilization. Buildings standing at the present day testify to this fact, and innumerable written records indicate a continuity of architectural practice lasting more than 2,000 years. The probability is that foreign importation has affected Chinese architecture least of all the arts.

The text and translation is given below of two documents concerning the grant of an allowance in the form of a soyūrghāl. The first document, dated Ṣafar 1067 A.h. (Nov.-Dec., 1656 A.d.), concerns the redistribution of a grant dated Ramaẓān 1047 A.h. (Jan.-Feb., A.d. 1638) among the heirs of the original grantee, Mīr Ni'matullāh ‘Abd al-Wahhābī, a sayyid. The second document, dated Sha'ban 1115 (Jan.-Feb., A.d. 1704), deals with the settlement of disputes among the heirs of the original grantee concerning their respective shares. These documents illustrate the complicated problems which arose with the fragmentation of grants by inheritance.

This article is an attempt to assemble and reassess the evidence relating to the problems of land tenure in the Middle Assyrian period, and to show that all ‘private’ land was normally held as a concession from the crown in return for the performance of ilku obligations. The subjects are treated in five main sections:

1: ilku and military service (pp. 496–502)

2: the Middle Assyrian laws, tablet A, § 45 (pp. 502–8)

3: allocation of tenures by the crown (pp. 508–12).

4: evidence from the land sale documents (pp. 512–17)

5: conclusions (pp. 517–19)

Vasumitra in his famous work‘Origin and doctrines of early Indián Buddhist schools’ (Samaya-bhedoparacanacakra)1 enumerates the following four doctrines as characteristic of the Sautrāntika school: '(i) The skandhas transmigrate from one world to the other: hence the name Saṃkrānti-vāda; (ii) There are mulāntika-skandhas and eka-rasa-skandhas; (iii) An average person (prthagjana) also possesses the potentiality of becoming a Buddha (lit. in the statě of average man there are also divine things, drya-dharmas); (iv) There are paramārtha-pudgalas '.2

The publication of the first volume of M. René Groussef's history of the Crusades, which is review'ed elsewhere in this issue, brings out again, and all the more vividly because of its wealth of detail and effort to present a complete and rounded-off picture, the very serious gaps in Orientalist research on this period. Whereas the study of the Western and Greek sources has progressed to a point at which it may be said that little more remains to be done, research on the Oriental sources is incredibly backward.