It is well known that, according to Pliny, the originating terminal for caravans conveying frankincense from South Arabia to the Mediterranean was at Shabwa; and there has been much talk of an ‘incense road’ followed by those caravans. It is, however, very unlikely that there was any single ‘road’ invariably followed. More probably, there was a multiplicity of routes, any one of which might be chosen according to prevailing circumstances. One such route runs directly from Shabwa to the eastern end of the Wādī Jawf, across a gravel corridor bisecting the Sayhad sand-desert, surveyed by Philby in 1936. The possibility of this route has been disregarded mainly because of a widely prevalent but tendentious translation of Pliny's remark, ‘evehi non potest nisi per Gebbanitas’, as ‘[the frankincense] cannot be exported [from Shabwa] except through the land of the Gebbanites’, where the text says nothing about a ‘land’ but merely ‘by the Gebbanites’. Other possible routes are one skirting the southern edge of the sand-desert and passing through Tumna', and one reaching the upper Wādī Bayḥān by way of Nisab and the Wadi Markha. West of Bayhan, the passes of Nagd. Marqad and Mablaqa clearly represent alternative routes.

On the hills in the angle formed by Assam and Bhutan2live the Hruso, called by their neighbours Aka or Angha, a band of “ bold and daring robbers and cut-throats ” who exert great influence on the powerful and numerous clans of the Midzi tribe on the southern slopes of the Himalayas. The Hruso act as intermediaries for the Midzi in contacting the peoples of the valley of Assam.

Numerous Arab geographers and Persian literary texts refer with a variety of detail to an important desert caravanserai known as Dayr-i gachīn (Ar. Dayr al-jiss), situated to the south of Ray. The word dayr, meaning in Arabic ‘monastery’, or perhaps ‘hospice’, seems a natural term for a caravanserai. That the term could also mean a domed building and a fire temple is suggested by both the Burhān-i qāti' and the Farhang-i Ānandrāj. The Burhān-i qāti', under ‘Khwarnaq’, states:

Khwarnaq is the arabized word for Khuwarna, and this was a most excellent building which was built for Bahrām-i Gūr by Nu'mān b. Mundhir … and the other palace, which had three domes, one inside the other, and was built as a temple and a place of worship, they called Si-dayr since in Pahlavi they call a dome ‘dayr’.

IN the literary world of medieval Christendom, few works rival the popularity of the edifying tale of Barlaam and Ioasaph. As is well known, the work purports to relate the conversion of the Indians by the Christian sage and ascetic Barlaam. In reality, it is a Christianized version of episodes from the life of Buddha. In its Western form, it consists of a set of apologues or fables woven into a narrative framework, and interspersed with quotations from the Scriptures and from the Greek Fathers. All versions in European languages derive from the Greek Barlaam, which became well-known from the 11th century onwards.

Being rather inconspicuous among the many beautiful and important exhibits of the Benaki Museum in Athens, the small bowl described in the present paper has so far received but scant attention and has not been published before.

For 300 years after Vasco da Gama touched Calicut generations of traders, merchants, missionaries, soldiers, and other emissaries from at least five different nations of Europe took their turn in India, pursuing their interests at a respectful distance, making no obtrusive efforts to scrape acquaintance with Sanskrit culture. Such advances were socially difficult, and would not have been welcomed. Moreover, our early associations were with Dravidian India, and very few cultured Brahmins sought membership of Christian Churches.

Between 1985 and 1987 the present writer took part in a multi-disciplinary study of the Wahiba Sands organized by the Royal Geographical Society. During this time I was able to travel extensively through and around the Sands to visit many bedouin families of the Āl Wahība, who inhabit the greater part of the area, and other tribes including the Āl ،Amr, Āl Bū ،Īsa, Ḣikmān, Hishm and Janaba who are found in certain enclaves, particularly to the east and north of the main sand desert. The purpose of my research was to make a social and ethnographic survey rather than a purely linguistic one, but in the process I was able to record some dialect speech including the text presented below.

Very little was known about the bedouin of this part of Oman prior to the Wahiba Sands Project. Holes has recently published examples of some related dialects and has proposed a classification scheme as the basis for a dialect geography of Northern Oman (Holes, 1989). Bedouin dialects of central and eastern Arabia have been more extensively described and classified (Holes, 1983; Ingham, 1982(a), 1982(b), 1986; Johnstone, 1961, 1964, 1965, 1967). Comparative studies of the distribution of certain features of present-day dialects have been shown to shed light on the population movements of the past, often supporting the written histories and oral traditions of the area.

The 1990s bore witness to significant advances in the study of Chinese migration history. This paper explores ways in which these new insights into population and migration patterns in the lower Yangtze watershed throw light on the history of dialects currently assigned to the Jiāng-Huái family by the Language atlas of China. It is found that the lexical corpora of many of these dialects were laid down as the result of two migration waves, one during the late Táng period and the other after the fall of the Northern Sòng. Material from the first of these waves is now found in bái or colloquial pronunciations of certain syllables. Material from the second wave constitutes the primary lexical component of these dialects and also contributes those wén or literary readings which pair directly with the colloquial pronunciations in each dialect. Interaction with several contiguous dialect groups is also noted. In particular, interlocking or commensurate wén/bái correspondences with dialects of the Northern Wú group are discussed and exemplified.

While working in the Algemeen Rijksarchief, The Hague, in 1973, this writer had his attention drawn by Dr. G. V. Smith to an envelope in Koloniaal Archief 982 (Overgekomen Brieven en Papieren 1620) which apparently contained several early Javanese letters. Upon investigation these proved to be original correspondence between the dignitaries of Bantěn and the Dutch under Jan Pietersz. Coen after the Dutch conquest of Batavia in 1619. They are written not in Javanese but in Malay, five of the letters employing Javanese script and one Arabic script. The letters are of interest as evidence of the Bantěn opinion of the Dutch (although the reader would be wrong to accept the innocent tone of the letters at face value) and as early examples of pasar Malay, the lingua franca of maritime South-East Asia.

“Dhāraṇī”, with its many cognates in Chinese and other languages, is one of the most complex terms in Buddhism. Particular shadings of the tradition's notions of memory, meaning, and meditation, as well as specific kinds of incantations and mnemonic devices, all fall within what can seem a rather bewildering semantic range. However, the logic of the term is consistent over a wide range of sources, though Western treatments have often been misleading. It is usually claimed that the basic practical significance of “dhāraṇī” is either memory or incantation. Yet Chinese sources make clear that each of these understandings overly privileges a narrow band of the term's usage. Understanding that the basic practical sense of “dhāraṇī” was “grasp”, not memory or spells, illuminates connections within a range of sources, from doctrinal treatments to injunctions to “hold” dhāraṇī-incantations in mind and, indeed, on the body encountered in texts of various kinds.

This article is one of a series that reassesses the dhāraṇī texts of Mahāyāna Buddhism. The article seeks to examine dhāraṇī texts by using the linguistic tools of pragmatics, especially historical pragmatics, to assist the understanding of their statements. Rather than the meaning of the term dhāraṇī as a subject term, the domain of truth-conditional semantics, this paper examines statements in texts labelled dhāraṇī. Pragmatics examines meaning in context, and the categories of speech acts developed by Searle has been especially helpful in mapping out differences within such texts and the formalization of statements across texts. The grammaticalization of specific speech elements, especially interjections, in the context of mantra-dhāraṇīs is also discussed.

Received wisdom has always held that the Ottomans took little interest in events beyond their borders except when they were likely to affect them. Previous scholars have suggested that it was only when French revolutionary forces occupied the Eastern Mediterranean that the Ottomans took an interest in and then condemned the revolution. From the despatches and reports of Ebubekir Ratib Efendi, ambassador to Vienna, we discover that at least one Ottoman diplomat was sending detailed accounts of events in Paris and the reactions of governments throughout Europe. Ratib Efendi's diplomatic activities would suggest that reforms were already taking place in 1793, at least in the field of gathering intelligence. This signals a fundamental change in the psyche of the Ottoman political order.

During a journey in the Near East I spent some days in Urfa, the ancient Edessa, in the south-east of Turkey, and from there I visited Sumatar Harabesi in the Tektek mountains on 27 and 29 January 1971. The object was to study all known Syriac inscriptions in Urfa and its surroundings, which date from the first three centuries of our era, in situ for their republication in a small corpus. In Sumatar Harabesi my attention was primarily directed upon the inscriptions published by Segal, all situated upon the so-called Central Mount, clearly an ancient cult site. I also visited the cave with bas-reliefs and inscriptions discovered by Pognon. It is now inhabited by a Bedouin family, but in spite of the inconveniences caused by this circumstance, it was possible to read the inscriptions, blackened as they were, by the light of a pocket torch and an oil-lamp, and it was impressive to observe once more how able an epigrapher this French consul was.

It is many years since at an Oriental Congress in Paris Dr. C. O. Blagden started the comparative study of the genealogy of the Malay rulers of medieval Malacca, comparing the account in the then best known version of the Sejarah Melayuor “ Malay Annals ” with the list of rulers recorded in Chinese sources (Essays Relating to Indochina, 2nd series, vol. 1, Notes on Malay Archipelago and Malacca, W. P. Groeneveldt) and in d′Albuquerque′s Commentaries (Hakluyt Soc, 1927). Since Dr. Blagden wrote his paper, there have been pubKshed under my name two other Malay sources for the genealogy of the Malacca dynasty: a chapter from a MS. of the Bustan al-Salatin by Shaikh Nur al-din (JRAS., Straits Branch, 1920, No. 81, pp. 39–47) and an earlier version of the “ Malay Annals ” (JRAS., Malayan Branch, 1938, No. 16). There is also new material in the Suma Orientalis of Tomé Pires (Hakluyt Soc., 1944, vol. ii, pp. 229–259). And there is a valuable paper on the Founder of Malacca by P. V. van Stein Callenfels (JRASMB., 1937, pt. ii, pp. 160–6).

Several verbal forms reconstructed for proto-Semitic strongly resemble reconstructed forms in proto-Berber: compare Semitic yV-PaRRaS to Berber y-əFăRRăS, Semitic yV-PRaS to Berber y-əFRăS, and Semitic yV-PRuS and yV-PRiS to Berber y-ăFRəS. We suggest that these forms are historically related and sketch a line of development from the reconstructed meanings to their attested uses. yVPaRRaS, originally imperfective, retains that value in both Berber and Semitic. yVPRas, originally stative, gained a perfective meaning in Berber and Semitic; the stative meaning is retained in Berber, but was largely lost in Semitic. yVPRus/yVPRiS, originally perfective, retained that meaning in Semitic, merging with the newly perfective yVPRas forms; in Berber, yVPRaS completely replaced perfective yVPRuS/yVPRiS, relegating the latter to non-aspectual uses. We conclude by considering the quality of the first vowel; the alternation seen in Berber y-əFRăS and y-ăFRəS supports reconstructions as yiPRaS and yaPRuS/yaPRiS, conforming to the Barth–Ginsberg Law of Semitic.

The name of Aḥmad Bābā first became known to European scholarship through the articles of the French scholar M. A. Cherbonneau in the years 1854 and 1855. In 1857, when Heinrich Barth published his Travels and discoveries, Aḥmad Bābā achieved a spurious fame which lasted for over 40 years as the supposed author of the Ta'rīkh al-Sūdān. It was not until 1897 that the learned German's attribution was scornfully refuted by the French journalist Felix Dubois, in his Tombouctou la mystérieuse, and the work was correctly assigned to al-Sa'dī the Timbuctoo scholar who died in 1656. During the twentieth century the name of Aḥmad Bābā has frequently been mentioned by writers about the medieval Western Sudan, usually as the symbol of all that was finest in sub-Saharan Islamic learning in the Middle Ages.

The most northerly section of the African Rift Valley is inhabited by a pastoral people who call themselves 'Afar. Their present territory extends eastward from the scarp of the Ethiopian plateau and as far north and east as the Red Sea. The railway line between Addis Ababa and Jibuti appears to coincide with the southernmost fringes of 'Afar population. The area is mainly desert or semi-desert, and contains numerous tokens of its geological past in the form of volcanoes, lava fields, and hot springs. Survival is not easy in this harsh environment, so that to some extent most 'Afars are obliged to live a seasonally nomadic existence. Certain areas do have permanent habitations, but 'Afars have never taken seriously to agriculture and those who live in villages along the coast and engage in fishing probably exemplify the most radical departure from pastoralism. Most of the terrain described lies within Ethiopia, but there are a great many 'Afars in the contiguous Territoire Français des Afars et des Issas. As far as the 'Afar are concerned this political boundary does not underscore any tribal divisions of note and there is a good deal of contact between folk occupying areas on different sides of the border.

(a) Material for the present article was obtained before and during a recent period of study-leave in Cyrenaica. My particular informant, who also accompanied me during the tour, was Mr. Idris ‘Abdallah, from Gyrene, a member of the Ḥāsa tribe. The dialect illustrated may be termed the Bedouin dialect of the Jebel.

Examples are given throughout in Arabic script and follow the method used by my informant for writing his dialect. Thus presented, they are immediately intelligible to a literate Cyrenaican, who is accustomed to written colloquialisms, for example in letters, petitions, agreements, and vernacular (usually Bedouin) poems appearing in the local Press. The phonetic transcription is intended to suggest a pronunciation for the examples.

In an appendix to the article ‘Chinese and Indo-Europeans’ it was shown that, contrary to the view put forward but never demonstrated by the eighteenth-century scholar Ts'ui Shih, and adopted by the eminent sinologists Paul Pelliot and G. Haloun, chapter 123 of the Shih-chi on ‘Ta-yüan’ could not have been reconstituted from parallel passages in the Han shu but must have been the source for the latter—as one would naturally assume if doubts had never been raised. In summing up the evidence I noted that the most important discrepancies between Shih-chi, 123, and the Han shu concerned the Wu-sun and Sakas and I argued that one must assume that the Han shu had grafted separate, additional, material on to the account in the Shih-chi. I did not, however, go into the matter in detail. This is the burden of the present article.

Shaykh Khaḍir al-Mihrānī, the Ṣ؛fī guide and soothsayer of the Maml؛k sultan al-Ẓāhir Baybars (regn. 658–76/1260–77), was the subject of an article by Louis Pouzet, ‘Haḍir ibn Abî Bakr al-Mihrānî’ (Bulletin d'Études Orientales, XXX, 1978, 173–83), which is based almost entirely on the account of Shaykh Khaḍir in al-Nahj al-sadīd by the Coptic chronicler al-Mufaḍḍal b. Abi'l-Faḍā'il. Pouzet was apparently unaware that an almost identical account of Shaykh Khaḍir appears in the eighth volume of Ibn al-Dawādārī's Kanz al-durar wa-jāmi' al-ghurara. The editor of this volume, Ulrich Haarmann, discusses at length in his introduciton (pp.24–9) the relationship between the works of Ibn al-Dawādārī and Ibn Abi'l-Faḍā'il, and demonstrates that they drew independently on a source derived at one or two removes form the still largely unpublished Ḥawādith al-zamān of the Syrian chronicler al-Jazarī (658–739/1260–1338). Since Ibn al-Dawādārī completed this volume, according to the colophon, in 734/1334, while the unique (and probably autography) MS of al-Nahj al-sadīd is dated 759/1358, the version in Kanz al-durar may reasonably be assumed to be the older. A collation of the two versions, however reveals that they agree almost entirely, both verbally and in content. Two points of detail regarding Blochet's edition and translation of Ibn Abi'l-Faḍā'il, as used by Pouzet, may be noted in passing. Blochet, p. 459, line 2, reads kanīsat al-fuh؛d (tr. ‘l'église des Panthères’), a textual error for kanīsat al-Yah؛d, Șthe synagogue of the Jews’. Pouzet here tacitly corrects his source (p. 178). On the other hand, he accepts Blochet's curious rendering of bābā as ‘intendant’ (Pouzet, pp. 174–5 and n. 9), whereas it is surely obvious from the context that it has here, as often, a dervish connotation.