The final expulsion of Moriscos from Spain in 1609 brought a new wave of Spanish Muslims to Tunis to join that substantial community of their fellow-countrymen which had increased steadily in number from the fall of Naṣid Granada onwards. Among these refugees and immigrants were some with considerable technical expertise: architects, military men, and skilled craftsmen whose arrival on Muslim territory caused a minor technological renaissance.

The peculiarity of the Druze law of divorce does not lie in the mechanism of dissolution, which is not materially different from that of the Muslim sharī'a, but in its effects: a dissolved marriage cannot be restored; the repudiation of the wife by itself creates an absolute bar to a remarriage between the parties. Article 11 of the Lebanese Druze Law of Personal Status of 1948 (hereinafter referred to as ‘the Lebanese Druze Law’), which was adopted by the Israeli Druze Community in 1961, provides, in conformity with Druze religious law, that a man must not remarry his divorced wife. Unlike Islamic law, Druze law does not distinguish between revocable (raj'ī) divorce, which does not immediately sever the marital bond but permits reinstating the wife during the waiting-period ('idda) by express utterance or significant conduct, and irrevocable (bā'in) divorce, which takes immediate effect; reinstatement in the latter case requires the conclusion of a new marriage with all that this implies (dower, etc.). A Druze divorce is absolute under any circumstances; the wife cannot be reinstated by way of taḥlīl, i.e. an intermediate marriage to another man and divorce from him so as to become again permitted to her first husband; this course is adopted in Islamic law after the third repudiation. Although the Lebanese Druze Law does not say so expressly, it seems that the legislator intended to block the reinstatement of the wife and restoration of the marriage by every possible means. In fact, religious law goes so far as to forbid divorced spouses to meet under the same roof even if separated by many partitions. The ban on taking back a divorced wife dates from the early days of the Druze religion.

Stimulated by Z. Harris's ‘Discourse analysis’ which raised the problem of continuing linguistic analysis in a particular connected discourse beyond the limits of a single sentence and by K. L. Pike's hints at the existence of a formal and semantic supra-sentence unit, such as the paragraph, American linguists produced some systematic and coherent descriptions of larger stretches of discourse, pointing towards certain grammatical, lexical, and stylistic relations reaching across sentence boundaries. Although most of these works deal with Amerindian languages, the method of analysis and the terminology they use can readily be applied to other languages, as was done, for instance in English, by V. Waterhouse in ‘Independent and dependent sentences’. This author draws a parallel between independent and dependent morphemes, phrases, and clauses on the one hand, and independent and dependent sentences, on the other, pointing out that ‘not all sentences are independent’ and that ‘there are in fact grammatical constructions which indicate that some sentences are dependent and therefore to be included in some larger linguistic form such as paragraph or discourse’.

Emperor Menelik's reign (1889–1913) opens a new era in the kind of sources that the historian has at his disposal for the analysis of modern Ethiopian history. During his reign printing presses were set up in the country and spurred a gradual growth in the more widespread use of Amharic, not just as the spoken but also the written language of the imperial court. This is not to say that the Gә'әz literary tradition in Ethiopia disappeared altogether, for some chronicles in Gә'әz continued to be written after 1935, often very similar in form and content to those which have survived for the Gondarine and earlier periods of Ethiopian history. However, Menelik's reign, and the official chronicle of it by Gäbrä Sәllase, mark a significant departure, not least because the chronicle was written in Amharic and not in Gә'әz. There are a few earlier published literary works in Amharic, the songs of the kings of the fourteenth century being the most significant, but it should be emphasized that the Gә'әz tradition continued parallel to the Amharic in the form of tarikä nägäst, i.e the history of kings. Some of these are now preserved at the Institute of Ethiopian Studies at Addis Ababa, like the Gondarine tarikä nägäst which belonged to Nәgus Wäldä Giyorgis, the valuable biography of the grandson of Emperor Tewodros (1855–68), Mäšäša, and a published biography of Ras Makonnen.

In two previous studies we have sought to advance the process of understanding the complex linguistic character of the Āḍi Granth (AG) by isolating and describing two special stylistic varieties of language, which significantly differ from the normal idiom of the book as a whole. These were shown to be based respectively on the extensive use of elements drawn from the dialects of the South-Western Panjab, and of loan-items from Persian: but it was further emphasized that they were stylistic varieties of the usual AG poetic language, and were in no sense to be regarded as straightforward examples of contemporary Multani or Indo-Persian.

Throughout most of pre-Han literature we find a particle hu which appears to occupy the same syntactic position as the directive preposition yü. Indeed, one particle often substitutes for the other in parallel passages.

‘The old and the weak are rolled into the gutters’ (of the dead in time of famine; the parallel in Me. 2B/4 has yü for hu).

It is an inherent feature of the adjectives ‘many’ (‘much’) and ‘few’ that they qualify nominal concepts implying ‘plurality’ so that the individuality of the units of which the entities underlying the specific nominal concepts are composed is clearly conspicuous: ‘Many (or: few) men (or: people)’, designating the plurality of the concept ‘man’; ‘much money’, designating the plurality of the concept' coin' (or: ‘bank-note’); ‘a (great) many clergy’, designating the plurality of the concept ‘clergyman’; etc. Such a semantic type, in which the epithets ‘many (much)’, ‘few’ are used, exists beside a semantic type of a related sense, as exemplified by the following instances: ‘a large mass (or: number, or: group) (of men, people, soldiers, etc.)’; ‘a small amount (of money)’; etc.

Of the various appellations by which ‘Alī b. Abī Ṭālib is known, none has evoked more interest and controversy than that of ‘Abū (al-)Turāb’. This by-name is attested in early sources, both Muslim and non-Muslim; an example for the latter is the letter allegedly sent by the Byzantine Emperor Leo III (717–40) to ‘Umar b. ‘Abd al-’Azīz (99–101/717–20), in which ‘Abū Turāb’ appears twice.

A detailed though somewhat unsophisticated account of Dóngola is that of Armbruster (1960). During 1975–6 I had the opportunity of working for short periods with a speaker of Dóngola, Mr. Ibrahim Abdel Karim, and there was little doubt that his speech was the dialect described by Armbruster. I paid particular attention to the possessive pronouns and found that both my observations and my interpretation differed in some important ways from those of Armbruster.

In my review of Dr. Shaban's book Islamic history: a new interpretation. 2. A.D. 750–1055 (A.H. 132–448) in BSOAS, XL, 3, 1977, 622, I stated that the rule of ‘Aḍud al-Dawla was not mentioned by Dr. Shaban. He has drawn my attention to the fact that he devoted three pages (183–6) to the reign of Fanā Khusraw (‘Aḍud al-Dawla), to which, in fact, I referred in my review. My apologies for this mistake are due to Dr. Shaban.