2 — The ancient Arabs, their genealogies and their history

and Africa; and that what is commonly held — the tracing of the Arabs to Ishmael and Qaḥṭān among the fathers of the Torah, and the descent of the tribes in the familiar form — is but a thing newly brought about, devised by men with ends to serve at a recent date not going back beyond the first century of the Hijrah, founded upon the register of the Imām ʻUmar ibn al-Khaṭṭāb in the matter of the Muslims' rights to the stipend as determined with regard to the tribes and their genealogies (page 6 of his book).

To support this view he began first with the establishing of mother-right among the Arabs, saying that the Arabs in ancient times had no family whose head was the father, nor did descent join to the fathers; rather marriage among them was after the manner of what obtains in the land of Tibet today and is known as Tibetan marriage — namely, that the woman marries two men or more, and her children are traced to none of them but are traced to the tribe and named by its totem, as has been set out. So he betook himself first to adducing the proofs establishing mother-right and its prevalence among the ancient Arabs; and when he supposed he had established it, he betook himself to establishing totemism, and spent the utmost of his effort in bringing out proofs and testimonies — all of which we shall set out in detail, showing wherein the error lies.

2 — The ancient Arabs, their genealogies and their history

Before advancing to the examination of Professor Smith's proofs, we say a word in summary upon the Arabs, their genealogies and their narrations, by way of preparing the ground for the inquiry:

Whoever peruses the view of the author of the totemism of the Arabs, and of those Orientalists who say what he says, perceives at the first glance that two things and no more brought them to it. The first is the weakness of their confidence in the sayings of the historians of the Arabs and in what has been preserved of their ancient legends. The second is the rising up of the men of the last century to challenge what had been established of the doctrine of ascent (irtiqāʼ, evolution) in the rules of the social order; for the spread of this doctrine in the middle of that century brought the men of letters of the Franks to refer everything back to natural causes, as Spencer did in referring the forms of worship and most of the usages to causes of this kind. Such too was the wish of the author of the totemism of the Arabs: for when he came upon what McLennan had written concerning the totem among the savage tribes — he being an Orientalist acquainted with the reports of the Arabs, ill-thinking of their Jāhilīyah, contemptuous of the sayings of their transmitters and their genealogists — and saw among the names of the fathers of the tribes and the sub-tribes what resembles the names of animals, it leapt at once into his fancy that these were of the traces of totemism among them. So he set this judgement before his eyes and took it upon himself to demonstrate it; and since totemism is built upon mother-right, he betook himself to establishing the latter, and produced weak proofs with which he passed the bounds of forced construction, and cited rarities from the reports of the Arabs that make the anomalous a rule, and left aside the established general rules upon which the genealogists and the transmitters are agreed — all of which is contrary to the principles of inquiry. This is strange in a scholar who has looked into the reports of the nations and their legends, and who knows that ancient history is for the most part taken from the legends handed down from the forefathers, which they sift clean of falsehoods, extracting the sound of them from the corrupt, so that they neither despise a legend nor reject a saying. For what is in the Iliad of Homer of the reports of the gods and their legends did not hinder the scholars from sifting it and distinguishing within it between history, religion and legend. The like is said of the reports of the ancient Indians from the time the company of the Aryans came down into the land of India, according to what is set down in their Sanskrit books. And so it ought to be said of the legends of the Arabs — the reports of ʻĀd and Thamūd, Ṭasm and Jadīs, the reports of the flood of al-ʻArim and the like; for these, far as they are from what is familiar to us, are not devoid of historical facts of consequence, the truth of many of which time has disclosed. We shall therefore bring forward fragments of this by way of example:

ʻĀd and Thamūd

The most deeply rooted of the legends of the Arabs in antiquity, and the furthest from the familiar, are the reports of the extinct tribes. Until a time not far past, investigators continued to reckon them among the legends fabricated shortly before Islam; others supposed them to belong to certain other nations, and that the Arabs had preserved them and attributed them to themselves. Then it became clear to them that they are not devoid of established fact, from what they found of mention of them in the books of the historians of the Greeks or of their ancient geographers, such as Strabo and Ptolemy and others. The most important of the extinct tribes are ʻĀd and Thamūd. As for ʻĀd, it had been supposed that they were not mentioned in the books of the Greeks, because no word resembling the name had been come upon among the names of the tribes of the Arabs; but we have shown, in an article of ours upon this subject (al-Hilāl 23, year 6), that the Greeks did mention them under the name «ʻĀd Iram», writing it Adramitae to distinguish it from Ḥaḍramawt, whose name with them is Xatramotitae; and we there judged it more likely that they and the tribe of Hadoram, mentioned in the Torah among the Arabs dwelling in the land of al-Yaman, are one and the same tribe.

As for Thamūd, it is mentioned repeatedly in the books of the Greeks and the Romans; they came upon its remains in the uplands of the Ḥijāz and deciphered some of what is inscribed upon its stones. For all that, they reckoned its history not to reach back in antiquity beyond the date of the Nativity save by a little — until the excavators came upon mention of it in the ruins of Assyria about the eighth century before the Nativity (1), in the course of reports of wars and conquests, which indicates that that tribe was of consequence in that age. Judge by this the rest of the reports of the extinct tribes, whose report was lost through the remoteness of its age, or whose name became doubtful with the Greeks through miscopying or the like, as befell the tribe «Jadīs»; for the Greeks wrote it Jolisitai, whereas the likelier original, in our belief, is Jodisitai, the dāl having been exchanged for a lām, the two being alike in the Greek language, since the lām is written thus Λ and the dhāl thus Δ. Judge the rest by this — and enough for you is what supports the reports of the Arabs and their genealogies from the texts of the Torah and from what they have come upon, and still come upon, among the remains of al-Yaman and elsewhere.

The Arab genealogists

If such is the case with the ancient legends of the Arabs, what then of their reports set down in books, upon which the genealogists of the first age of Islam were agreed — the transmitters of that day accepting no narration until it had been verified by a sound chain of authority, from their habit of verifying the Prophetic traditions and the like of the religious reports in that age? The Arabs are reckoned among the most exacting of nations in transmission and the most scrupulous in preserving what they transmit, especially in the first age of Islam, because of their reliance upon memory and their neglect of writing, for reasons we have set forth in the third part of our book The History of Islamic Civilisation.

Nor do we deny what runs through those narrations of things fabricated, or disputed, or unreasonable; but it is not reasonable that they should all be fabricated, since collusion to such a degree cannot come about. And were it allowed us to credit such collusion, we could not but ask about the time in which it took place: before Islam, or after it? If it be said, before Islam, then what called for its coming about? We know of no cause calling for that, nor do we suppose the author of the totemism of the Arabs knows of one. And if it be said, after Islam — and this is his view, for he has alleged that the genealogists devised the genealogies in the first age of Islam, dividing them into Qaḥṭānī and ʻAdnānī and dividing each of these into branches, and that the purpose of this division was to set forth the rights of the tribes with regard to the stipend which ʻUmar imposed — how can that be, when here are the poems of the Arabs of the Jāhilīyah, their sayings, their proverbs and their reports, all bearing witness to their guarding of descent and their concern to refer themselves back to their forefathers of Qaḥṭān and ʻAdnān? Nay, how can this be said when Islam, from its appearance to its spread, is built upon Qaḥṭānī and ʻAdnānī descent, and the caliphs were urging the Muslims to preserve their genealogies and to be exact in them? Among the sayings of ʻUmar ibn al-Khaṭṭāb is: «Learn genealogy, and be not like the Nabaṭ of the Sawād, one of whom, when he is asked about his origin, says: from such-and-such a village» (1)

Can this hold good, when the Arabs are totem tribes with no bond between them and no descent?

And if we suppose it to hold good, and that the genealogists devised these genealogies at the beginning of Islam for the sake of the stipend, then how were those tribes content whom the genealogists set far from the Prophetic line, so that their stipend was little or their rights weak? How did they not protest against it — nay, how is it that no scent of such protest is caught from the words of the historians? Moreover the collusion of the genealogists upon fabrication is far from possible, for they brought nothing from themselves; rather they used to go about the desert transmitting descent from the tongues of those who preserved it, setting it down or committing it to memory. A genealogist might gather his reports from the people of Najd, the Ḥijāz and al-Yaman by asking the trustworthy in those regions whose ends lie far apart: is their collusion upon this possible?

The Shuʻūbīyah and the genealogies of the Arabs

And if we grant its possibility, and that the Arabs showed no opposition out of respect for the caliph or fear of him, then how did the Shuʻūbīyah, and the Persians above all, keep silent about this divergence, given what the Arabs vaunted over them of the nobility of Arab descent, while the Shuʻūbīyah were searching for an argument by which to bring down the nobility of the Arabs that comes to them from their tracing themselves to Ishmael and Qaḥṭān? The Persians were bold enough in the first age of Islam to ascribe savagery to the Arabs, saying: «They are like ravening wolves and shying wild beasts; some of them devour others and some raid others; their men are bound fast in the rings of captivity and their women are captives mounted behind upon the packs of the camels» (1). Yet not one of them impugned their descent, neither by hint nor in plain words; and had they been able to do so, there would have been in it the strongest revenge for them. Nor can it be said that they kept silent about it out of negligence, or that they did not mark it; for they did impugn the Arabs' disagreement over descent, and their attaching to themselves of claimants, and the like of such matters as touch the genealogies. Buḥayr said, taunting the Arabs with their attaching of claimants:

You alleged that the Indians are the children of Khindif — and that there is kinship between you and the Berbers, And that Daylam is of the seed of the son of Ḍabbah the valiant — and Burjān of the children of ʻAmr ibn ʻĀmir. The Banū al-Aṣfar, the kings, are nobler than you — and worthier of our kinship are the kings of the Chosroes. Do you covet affinity with me, an open pretender — when you have seen no covering upon an open pretender, And you revile, in baseness, his kin-group and his tribe — and praise, in ignorance, Ṭāhir and the son of Ṭāhir? (2)

Yet for all that they did not touch upon the soundness or the corruption of their genealogies. The nation of the Persians reached the summit of its advancement centuries before Islam; the Arabs used to remove to them and to dwell among them, and there were wars and disputes between them before Islam; and the Persians took possession of al-Yaman and settled in the midst of the Arabs, consorting and mingling with them shortly before Islam — so that they are of all men the likeliest to know their circumstances in their Jāhilīyah. Had they found any doubt in the fixing of their genealogies, they would not have kept silent about it, for they began their rancour against them from the earliest part of the first century of the Hijrah. Stranger than this is that the genealogists themselves were for the most part of the non-Arabs: would they then devise something that would be a weapon against them in the hands of their enemies?

Divergence in some of the genealogies

All that we have of the reports of the Arabs goes back to the ordering of descent as they mentioned it in their books or transmitted it in their poems, and we have nothing that contradicts that ordering, neither by text nor by hint: how then is it allowed us to overturn it? And no weight attaches to what our author has mentioned of the genealogists' disagreement over the ascribing of certain tribes to Qaḥṭān or to ʻAdnān, or to Qays or to Kalb, or the like of that; for descent, as we have said, is in its origin transmitted from the mouths of men across differing regions — and man is the very title-page of error. Nor is it unlikely that the register of ʻUmar ibn al-Khaṭṭāb and the fixing of the stipend upon descent occasioned some confusion, and the attaching of certain sub-tribes to tribes not their own; and the exacting genealogists set forth the sound from the corrupt so far as their power reaches. But the existence of this disagreement does not indicate the corruption of descent from its foundation, just as the disagreement of transmitters over the details of one historical event does not indicate that it did not take place. For if a company disagreed over ʻAmr ibn al-ʻĀṣ's conquest of Egypt — one of them saying that he took it by treaty, others saying that he took it by force, others that he came to it with four thousand fighting men, others that he came rather with ten thousand, and others disagreeing over whether the Arabs came to it upon horses or upon camels — would that indicate that Egypt was not conquered? And if anyone said so, should we not ascribe him to eccentricity in his judgements?

Moreover the genealogists' disagreement may have its cause in the resemblance of the tribes in their names in utterance while they differ in sense; and this is frequent in their genealogies, and the genealogists have set down independent books upon it, such as the book The Divergent and the Concordant among the Tribes by Abū Jaʻfar Muḥammad ibn Ḥabīb, who died in the middle of the third century of the Hijrah, printed at Göttingen in the year 1850. Were you to consult the dictionaries of the tribes, you would see several of them under a single name, some of them of Qaḥṭān and others of ʻAdnān, and among them sub-tribes of the Yamanīyah and sub-tribes of the Qaysīyah. The Banū Asad are a sub-tribe of al-Azd of Kahlān of the Qaḥṭānīyah, and the Banū Asad are also a sub-tribe of Quḍāʻah of Ḥimyar; the Banū al-Aws are a sub-tribe of al-Azd of the Qaḥṭānīyah, and the Banū al-Aws are a sub-tribe of the ʻAdnānīyah; the Banū al-Ḥārith are several sub-tribes of differing tribes; the Banū Bakr are several sub-tribes, some of them of the ʻAdnānīyah and others of the Qaḥṭānīyah; the Banū Taghlib are a clan of Wāʼil ibn Rabīʻah of the ʻAdnānīyah, and the Banū Taghlib are a sub-tribe of Quḍāʻah of the Qaḥṭānīyah; the Banū Tamīm are of Ṭābikhah of the ʻAdnānīyah, and the Banū Tamīm are a sub-tribe of Hudhayl of the ʻAdnānīyah; the Banū Thaʻlabah are some ten and more sub-tribes of differing tribes (1); and the like of them are the Banū Rabīʻah, the Banū Sulaym, the Banū ʻĀmir, the Banū ʻAdī, the Banū Kaʻb and others. A single name, then, is shared by several sub-tribes going back to differing origins. They have found many sub-tribes under the name of the Banū Umayyah: in Quraysh, Umayyah ibn ʻAbd Shams ibn ʻAbd Manāf; in Iyād ibn Nizār, Umayyah ibn Ḥudhāfah; among the Anṣār, Umayyah ibn Zayd ibn Mālik of al-Aws; in Ṭayyiʼ, Umayyah ibn ʻAdī ibn Kinānah ibn Mālik; in Quḍāʻah, Umayyah ibn ʻAṣabah ibn Huṣayṣ — and judge the rest by this.

The names of the tribes may also resemble one another in written form while differing in utterance and in sense, such as Jassās with a doubled sīn and Jasās with a light sīn. Confusion is commonest in names alike in the shapes of their letters when the pointing is disregarded, and this was a great cause of confusion before Islam and in its first age. Thus in Madhḥij there is ʻAns with a nūn, son of Mālik ibn Udad; in Ghaṭafān, ʻAbs with a bāʼ, son of Baghīḍ; and in al-Azd, ʻAbs with a bāʼ, son of Hawāzin ibn Aslam. Judge ʻAnazah by this: it is so uttered in Rabīʻah, being ʻAnazah ibn Asad ibn Rabīʻah ibn Nizār; and in Khuzāʻah there is ʻAyrah with a yāʼ, which is also said ʻAnazah; and in al-Azd there is ʻAntarah ibn ʻAmr ibn ʻAwf ibn ʻAdī of al-Azd, and in it also ʻAbrah with a bāʼ, the ʻayn taking either ḍammah or fatḥah, and from it Ghayrah with a ghayn and a yāʼ, with differing vowels. Of this same kind are ʻAnz of Rabīʻah and ʻAtr of Rabīʻah likewise, and Ghubar the like of them. Judge by this Ajram and Akhzam and Aḥram, each of them of an origin other than the origin of the others (1)

These divergences in form and in utterance have occasioned some confusion in the genealogies of the tribes. The like is said of the fewness of the number of the fathers in relation to the span of time: its cause may be the loss of certain forefathers through forgetfulness or otherwise, or the reckoning of a forefather as a tribe in his own right and not as a single man, as is supposed of some of the forefathers of the Jews, the fathers of the Torah. This too is among the proofs of the antiquity of the genealogies from the age of the Jāhilīyah; for had a deviser devised them after that, he would have perfected the craft of forgery and multiplied the fathers, so that no room would be left for the counterfeit to show itself. But the genealogists brought nothing from themselves; they only transmitted what was current upon the tongues of the Arabs and preserved in their minds, with all its blemishes upon it.

Add to this that among the fundamental rules in the discrimination of rights is that «the presumption is freedom from liability» (al-aṣl barāʼat al-dhimmah). The presumption in the genealogies of the Arabs, then, is that they be taken as they have reached us; and it is not allowed us to object to them or to overturn them save by what is no less trustworthy than explicit texts and indications established by concurrent transmission (tawātur) or the like. As for relying upon rare sayings, or having recourse to stray reports and making the anomalous into rules — reliance upon that is not sound; or rather it is a defective induction, nay, it is nothing of induction at all, but is of the nature of arbitrary assertion contrary to the rule followed in inquiry and criticism. What comes nearest to right in establishing a proposition is that one should mount within it from the particulars to the universals, so that when the particulars are established the universals are established. Our author, however, assumed the universal proposition and set about establishing it, and did not lack among the scattered incidents of the reports of the Arabs something to take as a foundation upon which to build a weak structure that his skill makes appear sound.

Professor Robertson Smith, the author of the totemism of the Arabs, came upon McLennan's view of the totemism of the Indians of Australia and America and the like, and saw that certain tribes of the Arabs have the names of animals, and found the genealogists differing over the origins of certain tribes; so it came at once to his mind that these were the remnants of the totem, as we have said. He therefore laid down the universal proposition «that the Arabs were people of the totem», and then set about searching in their books for what supports this saying; and it is not hidden from you what is there of anomalous rarities and conflicting incidents. So he chose what he supposed supports his saying and left the rest aside. Were the proceeding upon this plan in the drawing of proof and demonstration allowable, we should not be unable to establish ⟨...⟩ proposition, and ⟨...⟩ it, however strange it might be. Were we minded to hold that woman in the Jāhilīyah had absolute freedom and a place of consequence in the social body like the place of ⟨...⟩ America today, we should not lack among the reports of the Arabs what would bear this saying up. And likewise if we said that she was dealt with among them as the beasts are dealt with, we should find what accords with what we alleged. But the rule in such an inquiry is that one look into the sum of the proofs and take the preponderant by consensus or by the greater number; and the Arabs have not agreed, in their reports or their legends or their poems or their histories or their usages, upon anything as they have agreed upon those genealogies. Shall we then deny them upon mere supposition? Is certainty removed by doubt — and shall we then turn to a view whose scent is caught neither in the reports of the Arabs, nor in their histories, nor in the histories of the rest of the Semitic nations?

Furthermore, those genealogies have reached us by succession from the genealogists to the historians, differing as these do in their places and their ages, and are for all that in agreement in most of their narrations: how should this agreement come together if their origin were not sound? And if it be said that that origin was devised after Islam, its deviser must needs have been a man of authority — and who, pray, is he? And how should his report stay hidden, given the multitude of the Arabs' enemies in that age?

The truth is that descent is ancient among the Arabs, as ancient as it is among the rest of the Semitic nations, and the Arabs hold to it the more firmly by reason of their bedouin life and their moving from place to place, their hands being empty of any other bond to refer to. They went to such lengths in the guarding of genealogies that they preserved the pedigrees of their horses through many generations, joining them to such of them as were famed for overtaking or for the race among the noble horses, such as Aʻwaj and al-Wajīh and Lāḥiq and al-Ghurāb and al-Yaḥmūm (1). Were you to consult what has reached us of the reports of the genealogists, you would marvel at their care in preserving the genealogies and their exactness in the fixing of them. When one of them traced a man's descent, he would follow that descent from his father to his kin-group, then to his clan, until he arrived at the tribe — or the contrary, from the tribe to the individual.

The Semitic peoples

The author of the totemism of the Arabs has held, in the introduction to his book «The Religions of the Semites» and in the book «The Genealogies of the Arabs» with which we are here concerned (page 178), that the Semites arose first in the Arabian peninsula and then branched out, the Hebrews and the Aramaeans going forth from it and peopling the lands about it, while the Arabs remained in it upon their bedouin life. Totemism, then, ought to have been among those two as it is among the Arabs. But he has not said this; and if he were to say it, we do not think he would light upon what would bear his saying up, even in outward show, as he lighted upon it in the totemism of the Arabs — for the Jews seldom named themselves with the names of animals, being far removed from the rough bedouin life, so that he would not find among the names of the tribes what would help him to this allegation. And grant that he has lighted upon the ⟨...⟩ the ⟨...⟩ in an article he published in the English Jewish review two years ago (2), such as Caleb and Jacob and ʻOreb — these are the names of persons, not the names of tribes, and it is not sound to have recourse to them in establishing totemism.

Yet had he left aside assumption and supposition and looked at the matter in its plainness, he would have seen these Semitic nations resembling one another in a real and evident matter in which there is no confusion — namely, the tracing of descent to the fathers of the Torah. The tracing of the Arabs to Ishmael and Qaḥṭān is established by what has come in the Torah of the genealogies of the nations; for it appears to whoever considers it that the genealogies of the Arabs are a branch of the genealogies of the Semites. This George Rawlinson verified and established in his book The Origin of Nations (3), and Eduard Glaser in his book The History and Geography of the Arabs (4); and we have an article upon the genealogies of the Arabs published in the twentieth number of al-Hilāl of the fifth year, in which we set forth the genealogies of the extinct tribes, besides those of the surviving tribes, upon the authority of the Torah and of the historians of the Arabs, and harmonised the two with the discovered remains and with the texts of the historians of the Greeks.

Arab descent, then, is established by the establishment of the genealogies of the Torah, allowing for what the critical among the investigators hold, that by the names of some of the first fathers tribes are meant and not persons. If we overturn these, nothing is left in our hands. And is it allowed us to disregard these genealogies, established through the succession of the centuries, and have recourse to a view that has no foundation

— Notes —

[10] (1) Glaser Sk. der Geschichte und Geographie Arabiens II. 259

[11] (1) Ibn Khaldūn 109, part 1

[12] (1) The History of Islamic Civilisation 135, part 4

[12] (2) Al-ʻIqd al-farīd 71, part 2

[13] (1) Nihāyat al-arab fī qabāʼil al-ʻArab (manuscript)

[14] (1) Mukhtalif al-qabāʼil wa-muʼtalifuhā

[16] (1) Al-Kāmil of al-Mubarrad 454

[16] (2) The Jewish Quarterly Review

[16] (3) Rawlinson's Origin of Nations 228

[16] (4) Glaser Gesch. & Geogr. Arabiens II. 266 & 424