When the first volume of Otto Franke's great Geschichte des chinesischen Reiches was published in 1930 it bore the descriptive subtitle ‘An account of its origin, nature, and development up to most recent timesx2019;. But alas, in spite of the author's prodigious efforts, the times were unpropitious, and the great design was not, in the end, to be fully realized. The first volume (Das Altertum und das Werden des konfuzianischen Staates) took the story from the earliest age down to the fall of the Han. The second volume (Der konfuzianische Staat, I, Der Aufstieg zur Weltmacht) appeared in 1936 and carried the account of China's history on through the period of division between North and South (Universalismus und völkische Krafte) to the centuries of re-unification under the Sui and T'ang dynasties (Der Sieg des Universalismus: die konfuzianische Weltmacht). In the following year came volume three, with the notes and supplementary materials to the first two.

Mr. Tukumi-Syunsuke, the author of the article of which a translation is here offered, is a member of the Sisoo-no-Kagaku Kenkyuukai, or Institute for the Science of Thought, founded in Tookyoo in 1946. The article appeared in the first number of its journal. ‘Study group’ would be a more exact rendering of kenkyuukai than ‘Institute’, but the English title is that used by the members of the group themselves. A survey of the Institute's work has been published by a colleague of mine.

An Lu-shan's rebellion marks a turning point in practically every aspect of the history of the T'ang dynasty. The strong centralized empire which reached the apogee of its magnificence under the Emperor Hsiian Tsüng (713–56) was able, after seven years of exertion, to put down the rebellion. But in doing so it was forced to compromise with the very principle of provincial military autonomy of which An Lu-shan was the arch-representative. The empire which remained in the hands of the T'ang imperial family deteriorated by degrees into a state of provincial autonomy which differed from the actual subdivision of China during the ensuing Five Dynasties period more in name than in reality.

Lilias E. Aemsteong in her article ‘The phonetic structure of Somali’ has shown that there is a correlation in this dialect of Somali between tone patterns and grammatical gender in certain types of Nouns. Moreover, at many points in her article she implies by her tone markings that there is also a correlation between other grammatical categories and tone patterns. She states, for example, that ‘plurals of both masculine and feminine nouns are pronounced with the mid-level tones in all syllables … exceptions are the feminine words … ending in in the singular …‘.

There are, broadly speaking, two uses of language, assertive and expressive. ‘The Marunouti building is in Tookyoo’, ‘1 and 1 are 2’, are assertive statements; their truth can be ascertained by experiment or by logic. ‘Go over there‘, ‘~ is good’, are expressive statements; they are designed to influence the recipient through expressing a state of the utterer. The wartime statement ‘America and Britain are inhuman monsters’ is a pseudo-assertive one.

In an article ‘The Ōi river poems and preface’, Asia Major, N.S., III, 1, 1952, 65–106, 1 examined the Japanese preface written by Ki no Tsurayuki, and quoted (p. 74) a passage by Fujiwara no Kiyosuke which stated that the Ōi river poems had a Japanese preface written by Mibu no Tadamine.

Various writers on the language of the Fulani, or, have included in their accounts a description of the three Voices of the verbal system, and a brief statement of their role in the operation of the language. But no detailed analysis of Voice in Fula has yet been published. Such an analysis is attempted in this article, on the basis of the speech of two informants from Gombe Division in Northern Nigeria.

2. The term Voice is here applied to each of the three series of verbal affixes which are found in Fula. The tense affixes fall into three series, differing in distribution and behaviour, and to some extent in meaning. These series cut across the tenses, so that each tense (except for the Negative of Quality and the Imperative) has three different tense-signs associated with it, each belonging to a different series. Thus in the two sets of sentences.

The phonological structure of the Vietnamese word as used in the system of Vietnamese word classes established in this paper may be stated in terms of syllables. The syllable is denned as the element of Vietnamese having the prosody of one, but not more than one, of the system of six tones set up for the analysis of the Vietnamese language at the phonological level. The syllabic unit is used to refer to an element of text at the grammatical level of analysis corresponding to the syllable at the phonological level. A study of Vietnamese texts enables the following statements to be made:

In the notes here published an attempt is made to explore some of the peculiar linguistic problems involved in the interpretation of epigraphic texts. As a subject for study these have certain advantages for the linguist: the material on which he bases his conclusions is, at any given time, complete, reducing the chances of his being afterwards confronted with evidence which he has overlooked; its nature limits the directions which his inquiries can take; and since the text is irrevocably committed to stone, he is not delayed by the possibility of infinitely refining his phonetic and phonological observations. It is a cardinal principle of the present investigations that such texts in dead languages carry an implication of utterance; it is equally cardinal that what that implication is cannot be determined, and is not to be educed.

In the first part of this study we followed the development of the Ṣūfī ṭarīqa organization in Somaliland after the introduction of Islam and examined the functions exercised by Ṣūfī communities in the social structure. We interpreted the closer social and genealogical assimilation of jamā‘as among the nomadic tribes in terms of lack of arable land available for the foundation of independent settlements. It was argued that the genealogical idiom in which social relations are normally described, especially amongst the nomads, is extended to the jamā‘as by virtue of their identification with the Arabian genealogies of their sheiks. The incorporation of such Ṣūfi genealogies was held to explain how the Somali lineage system is in its furthest extension extrapolated to the Qurayshitic lineage of the Prophet.

Since Bleek first introduced the method of classifying nominals in Bantu languages by means of their concord prefixes, and of assigning a number to each class so constituted, this has been the accepted practice in Bantu linguistic studies.

The numbering of the classes is, however, more precisely a means of labelling the different sets of concord prefixes that operate the grammatical agreement in any given language. Nevertheless, as it is in one type of word only, the independent nominal, that the class is an integral characteristic of the word, it is in respect to these nominals that the assorting into classes is important.

IN my book Witchcraft, oracles and magic among the Azande (1937) there emerged from the treatment of these three topics a picture of some features of the Zande mind, revealing itself often as a suspicious, distrustful, hostile mind. Another facet of this mentality is revealed in the Zande concept of sanza (pi. asanza).

Baule is one of the most widely spoken languages in the French Ivory Coast; the Handbook of African Languages, Part II, gives the number of speakers as 373,000. It is, according to the Handbook classification, a member of the Anyi-Baule dialect cluster of the Akan language group. Because of its importance, and the fact that no description of it has so far appeared, it is considered worth while to bring forward the present phonological analysis even though it is only provisional; a more definitive statement must await further study in the field.

A RECONSIDERATION of the material used for my article ‘The monosyllable in Szechuanese’1 in the light of the technique of prosodic analysis being developed in the Department of Phonetics of the School of Oriental and African Studies 2 has led me to believe that it is susceptible of a more rigorous and satisfying treatment, in which some of the apparently eccentric features take a normal place.

In connexion with Professor Wittek's article ‘Devshirme and sharī'a’ in BSOAS, xyn, 2, pp. 271–8, a passage in Idrīs Bitlīsī's Hasht bihisht is of interest, for there the author, far from feeling any incompatibility between the devshirme and the sharī'a, even shows the former as justified by the latter.

An essential part of the prosodic approach is the abstraction of those features that may be regarded as syntagmatic. One feature of this kind is ‘vowel harmony’; prosodic analysis is well equipped to deal with this in terms of x2018;frontness’ and ‘backness’, or ‘openness’ and ‘closeness’, these being treated as characteristic of the entire word or of a considerable part of it.

In the phonological analysis of Tigre a prosodic feature of this kind may be abstracted, the relevant phonetic observation being that there are sequences of open front vowels, and that within those sequences there are no half open central vowels. It is this feature that is referred to in the title as ‘openness’.