1 Vide above, chap. ., canons I. D, and III. E, pp. lv & lxxxiii.

2 See instances given by M. C. de Perceval, vol.i. pp. 90 & 111.

3 Hamza mentions all ancient history of Yemen: but he means no doubt an ancient Mahometan history.

4 Il règne une profonde incertitude sur l'histoire des Sabéens issus de Yectan, appelés Cahtanides par les Arabes. Des traditions vagues, des listes do rois qui ne concordent pas toutes entres elles et offrent des lacunes manifestes, des généalogies interrompues ou douteuses, tels sont les documents que les écrivains orientax nou présentent. Avec d'aussi faibles éléments pour reconstiteur une histoire, ou ne peut espérer de parvenir à la vérité. Peut être, au moins, n'est-il pas impossible d 'attaindre a la vraisemblance. Je n'étends pas non prétentions an delà de ce terme." Vol. i. p.47. M. C. de Perceval does not pretend to give us from such doubtful materials the truth, but only a likely approximation thereto. He has fully realized these modest pretensions.

5 The names which connect the succession, or are of leading importance, are given in capitals. The same course will be observed throughout, especially as regarts the line of Mahomet's forefathers.

6 M. C. de Perceval calculates thirty-three years to a generation, excepting where the exact period is known by historical fact or synchronism; but he admits that thirty years would, in general, suffice for an Aral, generation. Vol.1. p.248, note 1. Sprenger allows three generations to 100 years; he admits that "this is somewhat too high in ordinary cases," but has adopted the calculation, because some of Mahomet's progenitors were begotten at an advanced age, which has raised the average interval between the successions immediately preceding. Asiatic Journal, ccxxi. p. 349.


8 M. C. de Perceval admits that from the imperfections of his materials he has frequently been obliged, by a reference to the genealogical lines of descent, to suppose lacunae in the reigns, and vice versa. Thus, about the time of Abd Shams II. the sixteenth prince of the line, a gap has been discovered of several names in the royal line, as we learn by comparing it with the genealogical trees.

The lines of Cahlan and Codhaa were presented memoriter, while the line of Himyar was recorded at least by inscriptions, and is likely therefore to be more complete.

9 The following passage from M. C. de Perceval is in complete accordance with this view - "Il parait point que, chez les premières il est existé aucune tradition nationals relative à la filiation de Cahtan. C'est depuis l'Islamisme seulement, quand les Arabes ont commencé a recueillir les souvenirs de leur histoire, et à les comparer avec les témoignages de la Bible que la plupart des ecrivains orientaux ont identifié Cahtin avec Yectan, fils d'Héber." Vol.1. p. 39. In the next page, however, he adds that, though the identity or Joktan with Cahtan is not demonstrable, it may yet be plausibly entertained, but only on the supposition that an indefinite number or unknown generations intervened between Cahtan and the descendants named by tradition as his sons. But it appears to me not only that the identity cannot be proved, but that it cannot be maintained as even possible. It is utterly incredible that the name of Yectan, belonging to a period twenty centuries before our era, should have survived so many ages, and been reproduced in the eighth century B.C. as that of an historical personage, while all that intervenes is blank. The dictum of Mahometan tradition on the subject is plainly of no more value than that of any speculator or scriptural harmonist of the present day. It is no better than that of the Medina party, who tried to prove that Cahtan was a descendant of Ishmael, and therefore had no connection with Yectin. Katib al Wackidi p. 262 ½.; M. C. de Perceval vol.1. p. 39.

10 Others attribute its construction to the Adites. (M. C. de Perceval vol. i pp. 16-53) in which case Abd Shams may only hare repaired it. In dealing with such remote facts, we cannot do more than conjecture. For an account of the ruins see the interesting Relation d'un voyage a Mareb (Sana) dans l'Arabie méridional entrepris en 1843, par M. Arnaud; Journal Asiatique, Fevr. Mars 1845; and the remarks of M. Fresnel, Ibid. September and October, 1845. The great dam is an hour's distance from Mareb, p. 242.

11 See Weil's Mohammed, p. 2; and M. C. de Perceval vol. 1. p.7, where the third, or aboriginal class given by the Arabs, viz. Ariba, is noted as consisting of indigenous tribe; such as the Amalica, Adites, Thamud, Jadis, Tasm; - who, it is held, became extinct, but more likely merged into the more powerful Mutariba and Mustariba tribes. The three terms are only different forms of the same word , the name of Cahtan's son, is from the same root. The Arabs may either be really called after an historical personage so named; or, which is likelie; the character and name may be mythological, symbolizing the received opinion of the descent of the various Arab tribes from a common ancestor who was thence styled by them Yarob.

12 See above, note 2, p. viii, chap. i.

13 The Himyarite was probably the indigenous tongue of all the races descended from Cahtan; but in the case of the tribes migrating northwards, it probably became assimilated with the Abrahamic Arabic from intercourse with the Abrahamic tribes. There are a variety of traditions regarding the prevalence or the two languages in Yemen. Cnf. M. C. de Perceval vol. i. pp. 8, 50, 56, 79. The Mahometan theory, that all the aborigines (Ariba) spoke Arabic, and that Yarob introduced it into Yemen, are evidently grounded on the etymological meaning of the words. A later king is said to have introduced the Himyar tongue into Yemen ”upon the Arabc" - as if the Arabic had been the vernacular. But the expression may refer to the court language of Mareb, which perhaps repeatedly changed at various times.

The fortuitous discovery of Himyar inscriptions at various places in a character hitherto unknown, and the fortunate recognition of an Arab MS. On the Himyar alphabet, give hopes that something may hereafter be deciphered from such monuments; but up to the present time little more has been identified than a few names, and those uncertainly. The lucubrations of Mr. Forster on this subject are ingenious but fanciful.

The usual mode of writing is from right to left; but sometimes the boustrophedon style is used. The letters are all separate, and the words disjoined by a vertical bar. Journal Asiatique, December 1838, and September and October 1845; M. C. de Perceval, vol i. p. 79. The Mahometans do not appear to have known much of the language: some saying that the writing was from left to right; some that the letters were disjoined, others connected. It is possible that there may have been a variety of styles; but the Mahometans are not remarkable for great exactness in such relations.

14 To illustrate the absurdity of the fictions which abound in the history of this line, it may be mentioned that the Arab writers have invented a story, in which a Persian king Menut Shahr; Shammir the grandson of Himyar; and Moses, are all three made to appear on the same stage! "Le 'synchronisme presénté par quelques historiens entra Chammir; Moïse, et un roi de Perse, Menontchehr ne mérite aucune attention. C'est une fausse conjecture, qui prend sa source dans l'idée très exagérée que so font les Arabes de l'antiquité des souverains da Yaman, dont on a conservé les noms." M. C. de Perceval, vol.1. p.56.

15 The origin of the name is doubtful. Some apply it to all Harith's successors; others to those of them only who ruled over the entire empire of Yemen, and did not divide its sovereignty with others. M. C. de Perceval, vol. i. p. 64. Their royal resitlences were successively Mareb or Saba, Tzafar; and Sana. Between the second and third centuries there were three renowned "Tobbas," known by that name par excellence.

16 M. C. de Perceval thinks that the Yemen empire may have become known by the title of Himyar from the date or this re-union. The first mention of it in classical authors tinder that appellation is by Strabo, in describing the expedition of AEius Gallus. M. C. de Perceval finds it difficult otherwise to account for the previous silence. But it would be still more difficult to believe that the name or so remote an ancestor as Himyar should have been then revived, and after the abeyance of so many centuries adopted as the distinguisluing title of the kingdom. I would attribute the silence rather to the ignorance or so distant a kingdom.

17 Coran, xviii. 85, et. seq. This fabulous wall has been identified with fortifications near the Caspian Sea made, as they say, by Alexander, and repaired by Yezdegird II. M. C. de Perceval, vol.1. p.66. Whatever Alexander may have done to stop the inroads of the barbarians, the Arab legend is too wild to be seriously considered. Possibly it originated in some grand construction or work hy Alexander, a magnified account of which reached the Arabs, and naturally in their hands would grow apace.

18 Yet the ancestor of one of these parties was but just now represented as contemporary with the remote descendant of the other i.e. Shammir, the thirteenth or fourteenth in ascent from Essab, as contemporary with Moses! Such is Mahometan criticism and chronology.

19 M. C. de Perceval is of opinion that the Mahometan writers have here confounded their idea of some ancient African Prince with Gregory the Patrician, who commanded in Africa when invaded by Othman. He well adds : "On voit là un exemple de peu de scrupule avec lequel l'ignorance de quelques écrvains orientaux rapproche les temps les plus éloignés: vol.i. p. 68.

He has also an ingenious theory that Africus may have been employed by Caesar in the war against Juba, and thence gained his African name. In the battle or Actium, the Arabs of Yemen are said to have fought for Antony, anf to have fled with Cleopatra.

Omnis Arabs, omnes vertebant terga Subaei. Aeneid, viii. 706.

It Is not more likely that this Africus made hostile incursions from Yemen into the Roman dependencies in Africa: and that these may have been at least one of the causes of the Roman expedition of Aelius Gallus, which followed shortly after?

20 M. C. de Perceval traces the legend to a poetical fiction in Ferdusi. "Si l'on en recherche l'origine, on s'apercoit, q'une vague tradition, on peut-être une pure fiction preséntée sous des formes indécises par le pöete Firdauci, qui florissait trois siècels apres l'hégire, a été arbitrairement arrangée par des écrivains postérieurs sous les traits précis d'un fait historique. Firdauci avait chanté une expédition de Caycaous contre le roi de Hamaweran, pays inconnu, fantastique, dont on a fait l'Arabie heureuse. Le pöete a avait pas nommé ce roi: on a imaginé qué c'état Dhou-l-Adhar." Vol. i. p. 72. He then shows that the Mahometan historians are utterly ignorant of the real history of Persia at the period supposed.

21 In the original but conjectured by M. Fresnel, with some likelihood to be a mistake for .

22 See Sura, xxvii. 24, et. seq. She is also styled by tradition Balcama or Yalcama; but no name is given in the Coran, where she is simply described as the Queen of Saba. "Mais les interprètes, ne trouvant pas dans la liste des souverains du Yaman, conservée par la tradition de reine plus ancienne que Belkis, n'ont pas hésité à déclarer que c'était elle qui avait fait le voyage de Jérusalem. Leur sentinient a été pieusement adopté par les chroniquers, et cette opinion, accréditée par la superstition et l'ignorance, est problablement, la cause principale qui a empeché les historiens de classer les rois du Yaman suivant un ordre chronologique raisonnable." M. C de Perceval; vol.1. p .77. I would hardly call this "the principal cause" for the departure of the Mahometan historians from a reasonable chronology. Their appetite for ancient dates had a far more important source. They longed to complete the chain of legendary tradition by connecting Adnan with Ishmael, and identifying Cahlan with the Joktan of the Mosaical record. The absurd antiquity thus imparted to modern names attached likewise to this Queen, and they were then free to deal with her as they pleased. The motive of identifying Belkis with the Queen of Sheba, is not of itself a sufficient one for the unsettlement of the chronology.

23 He is called Mozaikia, they say, from daily "rending" the garment of yesterday, which he always replaced by a new one; but more likely from "rending" the Azidites from their ancient settlements. But who can tell the thousand incidents from which a soubriquet may arise?

24 It is important to fix the chronology of this salient point in the history or Arabia. The Mahometan writers agree in placing the event between our Saviour and Mahomet, some six, some four centuries, prior to Islam. The Azdite genealogies, (such its those of the Aws and Khazraj of Medina,) place the birth of Amr Mozaikia about five centuries before that of Mahomet. These considerations combine to fix the emigration somewhere about 120 A.D. M. C. de Perceval thinks that the great prosperity ascribed to Mareb by Strabo anti Pliny argues that the calamity of the dam was posterior to the Christian era. I would draw the same conclusion rather from the fact that the altered stream of commerce would probably not have worked out its baneful effect upon the Himyarite State, till after the Christian era.

M. de Sacy conjectures, that the insecurity of the dam wan not the real cause of the emigration; but was invented by the later Azdites, to cover some less honourable cause; perhaps fear of defeat from Tobba al Akran. But the view given in the text appears more natural.

25 The author of the Periplus mentions Caribeal its reigning at Zhafar. This is supposed to have been about 200 A.D. Caribael may either have been this Abu Cariba-al Himyari, or his father Calay Cariba-al Himyari. C. de Perceval vol. i. p. 90.

26 The tale of the Jewish doctors is mingled with marvels and anticipations of Mahomet. The whole story is thus of such feeble authority that no safe inference as to the prevalence of Judaism can be built (as Lieut. Burton seems inclined to do) upon it. Pilgrimage to Medina and Mecca, vol. iii pp. 160 and 336.

27 The two expeditions are so confounded that many of the names belonging to the modern attack (as that of Ohaiha, who lived in the sixth century) are introduced by a patent anachronism into the ancient adventure. The later expedition will be farther considered when we come to Medina.

With reference to the ancient attack, the fact of the Aws and Khazraj being then at Yathreb (if it be a bona fide fact and not borrowed from the modern expedition,) would argue for its having occurred under the reign of Hassan Tobba the Less, and not under that of Tibban Asad Abu Carib: became those tribes did not settle at Medina till about 300 A.D., or a century after the reign of the latter prince. On the other hand, the introduction of Judaism into Yemen, if ready (as represented) a result of the present expedition, would favour the earlier date; because there is reason for thinking that Judaism was known there before 300 A.D.

The whole story is given at length by Hishami, pp. 7 et seq., and is common among the Mahometan historians. The reader will not fail to observe the ridiculous “foresinadowing" of Mahomet's flight to Medina. See Journal Asiatique, November, 1838, p. 414. Two valuable papers by M. Perron, in that and the previous number, may be consulted by the student, who wishes to see in greater detail the accounts of the Mahometan historians on the subject. See also M. C. de Perceval, vol. i. and vol. ii. P 647.

28 See Table at p. cxlix.

29 See Hashami, p. 5, and M. C. de Perceval, vol.i. pp. 96-100. The latter, with reason, regards the prophecy to be a fabrication, intended to cover a less reputable cause of emigration, perhaps fear of the arms of the Hinyarite monarch against whom, in the capacity of vassal, Rabia had rebelled. The Mahometan anxiety to ,discover or to fabricate foreshadowings of the coming Prophet, may have worked together with this motive.

30 M. C. de Perceval, p. 112; Philistorgius, Hist. Eccles. 1. iii. chap. 4-6. Gibbon gives a brief account or this embassy, Decline and Fall chap. xx. Philostorgius wrote his work in the first half of the first century.

31 M. C. de Perceval, vol. i. p. 114. The Greek Inscription at Axum, discovered by Salt, gives these titles as appertaining to the Axumite monarch Aeizannas. See the description of Axum, between Meroe and the sea-port Adule, in Heeren's Res. Africa, vol. i. p.460, &c.

32 The connexion is also marked by the fact that Sabbah, who reigned over Yemen 440 to 460 A.D. made a tour or Najd, to assure himself of the submission of the tribes of Central Arabia. M. C. de Perceval, vol. i. p. 116.

33 Hamza states that having visited Medina, one half of the inhabitants of which were then Jew's, Dzu Nowas was so well pleased with their religion, that he embrace it. But, as M. C. de Perceval shows (vol. i. p.122), it is much more likely that he became a Jew through the influence of the powerful and long established party in Yemen; and that he visited Medina in order to succour the Jews against the oppressive attacks of the Aws and Khazraj. This agrees with the history of Medina, and is in excellent keeping with the sectarian bias which led Dzu Nowas to the attack of Najran.

34 M. C. de Perceval, vol.1. p. 129; Hishami, p. 14. The details are briefly given by Gibbon at the close of the xlii. chap. of the Decline and Fall of the Roman Empire: and the subject is alluded to in the Coran, lxxxv. v. 4, et. scq., where those who perished in the trenches are styled As-hab al Okhdud.

35 We gather this from the Greek historians. The Arabs only tell us of the suppliant Dous, whom the Greeks do not mention.

36 The number of the force as given by the Arabs is probably exaggerated. An ecclesiastical work mentions that 600 Roman merchantmen were employed on the occasion by the Abyssinnian monarch: he had also 700 light transports. The Greek authorities state that the emperor wrote to the Patriarch of Alexandria to stir up the Negus or King of Axume, to avenge the massacre of his fellow Christians In Najran. This king is styled among the Arabs by the hereditary title of Najashi which is another form of Negus. The then prince is called by the Grecians Elesbans (Atz-beha), and by the Ethiopians Caleb or Amda. The former was probably his baptismal name. M. C. de Perceval, vol. 1, p 131.

37 Some Syrian and Greek writers place both the Abyssinian conquest and the massacre in Najran, within the year 523 A.D.. In Assemani (vol.1. p. 364), is given a letter of the Bishop Simeon, stating that tidings of the conquest or Najran reached the king of Hira early in Feb 524: it therefore occurred about the close of 523. Allowing time for the intervening events and preparations, the defeat of Dzu Nowas cannot well be placed earlier than the beginning of 525 A.D. M. C. de Perceval p.133

38 The account of these events is given in detail by Hishami, p. 19, et. seq. M. C. de Perceval, vol i, p. 146, et seq.

39 Weil objects to the story upon chronological grounds; but his objections appear to be removed by the explanation of M. C. de Perceval, who makes the Abyssinians to receive the first check and overthrow in 575, but not to be finally expelled till 597. Weils Mohammed, p. 8, note 1.