¹ Hourani, p. 303. (All references to the book under review will be given by contributor.)

² Cf.the ideal type definitions given by the editors, p. 4; Tibi, p. 131.

³ Tapper, , p. 63.Google Scholar

⁴ Crone, P., “The tribe and the state”, in Hall, J. A. (ed.), States in History (Oxford, 1986).Google Scholar

⁵ Schneider, D. M., A Critique of the Study of Kinship (Ann Arbor, 1984), p. 112.CrossRefGoogle Scholar

⁶ Tapper, , p. 63.Google Scholar

⁷ Barfield, , pp. 156ff.Google Scholar, notes that the kinship is real at the lower levels of segmentation, to become increasingly in the nature of political statements the higher up we go (similarly Crone, , “The tribe and the state”, p. 53, n. 25Google Scholar). Beck stresses its political nature at the higher levels (p. 194): Lapidus (p. 26), Tibi (p. 131) and Hourani (p. 304) describe the postulated descent at the higher levels as mythical, while Dresch speaks of it as a language (p. 283, n. 15). The question raised by my reformulation (in terms nicely analysed by Schneider, , Critique, pp. 50ff.Google Scholar) is whether we can equate genuine with constitutive kinship on the one hand and postulated with purely representative kinship on the other. Is the postulate of common descent only an idiom? (Is modern citizenship only a language?) My own answer would be negative. Both Barfield (pp. 155ff.) and Tapper (p. 58, on Lindner) could be read as subscribing to the same view, but whether they actually do is another question; Beck, Lapidus, Hourani and Dresch clearly do not.

⁸ In common with most of the contributors, I do not distinguish descent from kinship, but rather see it as an aspect of it.

⁹ Crone, , “The tribe and the state”, p. 66;Google Scholar Tapper, , p. 62.Google Scholar

¹⁰ Cf. Schneider, , Critique, pp. 165ff.Google Scholar

¹¹ Tapper, , pp. 50, 52;Google Scholar Tibi, , pp. 137ff.Google Scholar; Beck, , p. 196;Google Scholar Crone, , “The tribe and the state”, pp. 52ff.Google Scholar

¹² Back, , p. 198;Google Scholar

¹³ Back, , pp. 193ff.Google Scholar

¹⁴ Hartley, J. G.,The Political Organization of an Arab Tribe of the Hadhramaut, unpublished PhD thesis, London, 1961, p. 56.Google Scholar

¹⁵ Back, , p. 194.Google Scholar I should have taken account of such cases in my own paper (Crone, , “The tribe and the state”, p. 51Google Scholar).

¹⁶ Khoury, and Kostiner, , p. 5,Google Scholar quoting Tapper, R. (ed.), The Conflict of Tribe and State in Iran and Afghanistan (London, Canberra and New York, 1983), p. 9;Google Scholar Back, , pp. 190, 200, 203.Google Scholar

¹⁷ Tapper, , p. 51Google Scholar, quoting Leach, E. R., Political Systems of Highland Burma (London, 1954)Google Scholar (without page reference); Back, , p. 189.Google Scholar Would Dresch also agree? (cf. his Tribes, Government, and History in Yemen (Oxford, 1989), p. 75).Google Scholar One cannot tell from his paper.

¹⁸ Gellner, , p. 110.Google Scholar

¹⁹ Back, , p. 189.Google Scholar

²⁰ Musil, A., The Manners and Customs of the Rwala Bedouins (New York, 1928);Google Scholar cf. Crone, , “The tribe and the state”, p. 55Google Scholar.

²¹ Raswan, C., The Black Tents of Arabia (London, 1935), pp. 36, 100Google Scholar (where the Rwala migrate in a body due to warlike conditions) and photograph facing p. 9 (where 7,000 tents are encamped at a watering place). Raswan estimated the Rwala at a total of 7,000 tents or 35,000 souls (pp. 24, 248), presumably sharing his source with the French mandate government which also put them at 35,000 in 1935 (Lancaster, W., The Rwala Bedouin Today (Cambridge, 1981), p. 8Google Scholar). But Lancaster estimates them at a quarter to half a million in the 1960s on the basis of a reputed congregation of 12,000 tents that could only have represented part of the tribe. (This is presumably the source for Barfield's figure, p. 161.) If Lancaster is right, the congregations witnessed by Raswan can hardly have been complete either. But the problem is probably beyond resolution by now.

²² Tapper, , pp. 53, 63.Google Scholar

²³ Tapper does not say what size he credits them with. Though they were certainly one of the largest tribes, Raswan (Black Tents, pp. 248.Google Scholar) credits the Shammar with the same size (7,000 tents) and the Atayban with almost the same (6,000 tents), closely followed by the Harb (5,600 tents), so “exceptional” seems too strong a word even though many tribes were smaller or even minute, as Tapper says. At any rate, mere size has no bearing on this particular problem, though the ability of the Rwala to congregate (unlike the Āl Murra dispersed in the Empty Corner) probably does.

²⁴ Crone, , “The tribe and the state”, p. 55.Google Scholar Compare Khazanov, A. M., Nomads and the Outside World (Cambridge, 1983), p. 139Google Scholar (the assertion that the principle of descent is the most important factor in the structuring of social and political organisation is only applicable to pure nomads on the whole).

²⁵ Tapper, , p. 63;Google Scholar Dresch, , p. 254Google Scholar and n. 8 thereto. Given that I have argued against the “easy equation of tribes with pastoral nomads” myself, I find this accusation particularly annoying (cf. Crone, P.,Meccan Trade and the Rise of Islam (Princeton and Oxford, 1987), p.236Google Scholar). But I concede that the formulation in my article invited misunderstanding.

²⁶ Khoury, and Kostiner, , p. 5;Google Scholar above, n. 7; Caton, , p. 91;Google Scholar

²⁷ Tapper cited by Khoury and Kostiner, p. 7; Beck, , pp. 187, 189;Google Scholar Dresch, , p. 277;Google Scholar

²⁸ Crone, P., Pre-Industrial Societies (Oxford, 1989), p. 125.Google Scholar

²⁹ Cf. Tapper, pp. 59, 65, who first castigates Lindner for overstressing the role of “shared interest” in recruitment to tribal groups at the cost of “kinship morality” and next stresses that “the ability to unite usually rested on the hope of material gain…as much as, if not more than, on any tribal ‘asabiyya”.

³⁰ A historical dimension is lacking in the work of those who wish to deny kinship any constitutive social role, relegating it to the status of an “idiom” used to bargain for personal power, as Lindholm, C. rightly observes in his “Kinship structure and political authority: the Middle East and Central Asia”, Comparative Studies in History and Society, XXVIII (1986), p. 348.Google Scholar

³¹ Gellner, , pp. 109ff.Google Scholar, Caton, , pp. 92ff.Google Scholar

³² Lapidus, , pp. 29ff.Google Scholar It is of course true that the great confederations of Qays/Muḍar, Rabīa and Yemen in which the Arabs were united after the conquests were post-conquest creations, but this does not mean that the tribes which went together in these super-units had no pre-Islamic existence. It is groups such as Asad, 'Āmir, Fazāra, 'Abs, Sa'd, Ḥanẓala, 'Atīk, 'Akk, Daws, Kalb and so on which are supposed to have been, or consisted of, tribes in the sense of large-scale genealogical units, and how does Lapidus propose to revise them out of existence? Further, the fact that Ghassān and Lakhm were monarchies while Quraysh had a tribal council does not mean that large-scale organisation was “conciliar or monarchical” as opposed to “patriarchal or genealogical”: all these tribes had genealogical articulation too. Both Lakhm and Ghassān survived as tribes after they had lost their kings, and the tribal council of Quraysh is scarcely incompatible with a tribal definition of Quraysh. Lapidus's claims are so extraordinary that one can only conclude that he wrote his paper in a hurry. Compare his surprising invocation of Gibb, who unsurprisingly fails to figure in the relevant note (which only contains a global reference to Shaban); cf. also the fact that his spelling is astray: commitatis for comitatus (p. 29), madhdhahib for madhahib (p. 27) and even “Ummayad” for Umayyad (p. 37)

³³ al-Iṣbahānī, Abū 'l-Faraj, Kitāb al-aghānī (Cairo, 1927–1974), v, pp. 45, 61f.Google Scholar For a modern real-life story of a very different kind illustrating the same unambiguous and pre-determined nature of kinship roles, see Gellner, E., Thought and Change (London, 1964), p. 156.Google Scholar

³⁴ Back, , p. 196;Google Scholar cf. Tapper, , Conflict, p. 9,Google Scholar quoted by the editors, p. 5.

³⁵ One could make the same point by saying that people who mostly speak Iranian languages other than Persian even though they know Persian too are members of ethnic minorities, whereas people who do the equivalent in the Arabophone world are merely normal.

³⁶ Back, , pp. 199, 203.Google Scholar

³⁷ Back, , pp. 189, 197ff.Google Scholar; Caton, , pp. 81, 99ff.Google Scholar; Barfield, , p. 176.Google Scholar

³⁸ Crone, , “The tribe and the state”, p. 54, n. 31;Google Scholar Tapper, , p. 63,Google Scholar where he discovers that he is wrong and reacts by accusing me of shifting my ground.

³⁹ Tapper, , p. 63;Google Scholar above, n, 11. Beck makes her point with reference to Iran alone. Only I make the point that tribal Arabia is ethnically homogeneous whereas East Africa is not, so this is where Tapper chooses to find fault, but I cannot see that he actually disagrees.

⁴⁰ Cf. Crone, P., Slaves on Horses (Cambridge, 1980), ch. 2;CrossRefGoogle Scholar Lindholm, “Kinship structure and political authority” (above, n. 30)

⁴¹ Cf. Lapidus, , pp. 29, 33.Google Scholar Barfield claims that “The debate about the nature of tribe-state relations in the Middle East has often generated dispute because it is assumed that all Islamic tribal peoples are fundamentally similar” (p. 180), but it is not clear whom he has in mind. (He also suffers from the curious misconception that the Qur'ān recommends patrilineal first-cousin marriage, p. 170.)

⁴² Dresch, , p. 252.Google Scholar

⁴³ Khoury, and Kostiner, , pp. 4, 7.Google Scholar

⁴⁴ Khoury, and Kostiner, , pp. 6.Google Scholar

⁴⁵ Cohen in Cohen, R. and Service, E. R. (eds), Origins of the State (Philadelphia, 1978), p. 2,Google Scholar with the comment that lumping together all known political forms does nothing to explain the differences among them. He examines three further definitions before concluding that the diagnostic feature of the state is its ability to counteract fission (p. 4).

⁴⁶ Khoury, and Kostiner, , p. 7.Google Scholar

⁴⁷ Hourani, , p. 306.Google Scholar

⁴⁸ Gellner, , p. 126;Google Scholar Tibi, , pp. 147f.Google Scholar Tibi restates Migdal's “strong societies/weak states” as “segmentary fragmented societies/rtificial imposed states”, a wording which could be taken to suggest that he thinks of Muslim society as weak too; but that would go beyond restatement.

⁴⁹ Cf. Crone, P. and Honds, M., Cod's Caliph (Cambridge, 1986), pp. 108ff.;Google Scholar Crone, , “The tribe and the state”, pp. 75ff.Google Scholar

⁵⁰ Khoury, and Kostiner, , p. 19.Google Scholar Between them, the editors say some odd things that neither seems to be in the habit of saying on his own.

⁵¹ Tapper, , p. 64;Google Scholar contrast Crone, , “The tribe and the state”, pp. 74ff.Google Scholar

⁵² Tapper, , p. 49;Google Scholar Dresch, , p. 253.Google Scholar

⁵³ Fried, M. H., “On the evolution of social stratification and the state”, in Diamond, S. (ed.), Culture in History (New York, 1960), p. 713.Google Scholar Secondary states are those “whose origins can be attributed to pressures from already existing states and which often used parts or all of the organization of some prior state as the model for emulation or improvement” (id., “The state, the chicken and the egg”, in Cohen and Service, Origins, p. 37Google Scholar); and “the processes by which a state-level society emerges sui generis are necessarily quite different from those by which states emerge when other states have preceded them” (id., “Tribe to state or state to tribe in ancient China?”, in Keightley, D. N. (ed.), The Origins of Chinese Civilization (Berkeley, Los Angeles and London, 1983), p. 470.Google Scholar Apparently, later scholars have added tertiary etc. state formation, thus spoiling the beauty of Fried's distinction (cf. Dresch, , p. 284, n. 24Google Scholar).

⁵⁴ Lapidus, , pp. 27ff.Google Scholar

⁵⁵ Crone, , “The tribe and the state”, pp. 58ff.Google Scholar Dresch (p. 284, n. 24) apparently takes this part of my article to be a comparison of the Sumerians and the Islamic world!

⁵⁶ Fried, M. H., The Notion of Tribe (Menlo Park, California, 1975), p. 98Google Scholar (I am indebted to John Hall for obtaining a xerox of this book for me). Compare his examination of the Chinese case (“Tribe to state or state to tribe in ancient China?”, above, n. 53).

⁵⁷ Tapper, , p. 64;Google Scholar Dresch, , p. 281.Google Scholar

⁵⁸ Cited in Fried, , Notion of Tribe, p. 89.Google Scholar

⁵⁹ Cf. Dresch, , pp. 253 and 285,Google Scholar n. 35, on Caneiro. Dresch takes it for granted that evolution is synonymous with progress and thus entails value judgement: whether the fall of Rome is evolution or not thus depends on whether one is Boethius or Charlemagne. (Compare p. 280, where most of us have renounced evolutionary schemes so as not to rank ourselves against other cultures.) But Sahlins and Service are not that simplistic.

⁶⁰ Khoury, and Kostiner, , pp. 1ff;Google Scholar Barfield, , p. 153.Google Scholar

⁶¹ Cf. Dresch, , p. 253,Google Scholar where the Yemen is compared to a problematic fossil sturgeon with the suggestion that just as the sturgeon was discomfortable to schemes of natural evolution, so the Yemen casts doubts on most schemes of social evolution. But the sturgeon in question has not caused natural historians to abandon the theory of evolution. (For the crocodiles, see for example, Guggisberg, C. A. W., Crocodiles (Newton Abbot, 1972), pp. 68ff.)Google Scholar

⁶² Barfield, , p. 160.Google Scholar Hourani tries to explain the longevity of tribal groups such as Ḥāshid and Bakīl, Qays and Yemen, but not the longevity of tribal structures as such (pp. 304ff.).

⁶³ Crone, , Slaves on Horses, ch. 2.Google Scholar

⁶⁴ Tapper, , p. 69Google Scholar (cf. Beck, , p. 192Google Scholar); Crone, , “The tribe and the state”, p. 50.Google Scholar

⁶⁵ Lapidus, , pp. 28ff.Google Scholar

⁶⁶ Barfield, , p. 175;Google Scholar contrast Lapidus, , pp. 33ff.Google Scholar

⁶⁷ Dresch, , p. 274.Google Scholar

⁶⁸ That the political traditions of the Central Asian tribes played a role in this development was suggested by Lewis, B. as far back as 1968 (“The Mongols, the Turks and the Muslim polity”, reprinted in his Islam in History, London, 1973).Google Scholar Whether Barfield has read this article or not, he appears to have a similar impression.

⁶⁹ Lapidus, , pp. 39–42;Google Scholar Gellner, , pp. 115f.Google Scholar; Barfield, , pp. 171, 172.Google Scholar For Iran, see also Beck, , p. 192;Google Scholar Caton, , pp. 99ff.Google Scholar

⁷⁰ Hourani, , p. 307.Google Scholar

⁷¹ Dresch, , p. 274.Google Scholar

⁷² Gellner, , pp. 109, 113, 124.Google Scholar

⁷³ Tapper, , p. 49.Google Scholar

⁷⁴ Fried, M. H., “On the concepts of ‘tribe’ and ‘tribal society’”, in Helm, J. (ed.), Essays on the Problem of Tribe (Seattle and London, 1968);Google Scholar Fried, , Notion of Tribe, pp. 88ff.;Google Scholar and the references given above, nn. 53, 56. It is for this reason that he denies the very possibility that the first states in history developed out of tribes properly speaking, cf. above, II (a).

⁷⁵ Crone, , “The tribe and the state”, pp. 68ff.Google Scholar I had not been able to locate his Notion of Tribe when I wrote the article and hoped that he would solve the Polynesian problem in that book (cf. “The tribe and the state”, nn. 95, 99). But he does not.

⁷⁶ Tapper, , p. 52.Google Scholar

⁷⁷ Tapper, , p. 55;Google Scholar Beck, , p. 204;Google Scholar cf. Caton, , pp. 81ff., 99ff., 102;Google Scholar Lapidus, , pp. 42ff.Google Scholar; Barfield, , p. 177.Google Scholar

⁷⁸ He merely stresses that “at least in recent centuries” no tribe has been totally unaffected by the state (p. 52). Contrast Barth, F. in Pridham, B. R. (ed.), Oman (London, 1987), p. 23:Google Scholar “ribalism represents the response of a local population to the presence of state organizations: tribalism is not a pre-state formation”. How can he be so sure?

⁷⁹ Caton, , p. 102.Google Scholar

⁸⁰ Tapper, , pp. 51, 68;Google Scholar similarly Tapper, in his Conflict, p. 4.Google Scholar

⁸¹ Beck, , p. 192.Google Scholar

⁸² Khoury, and Kostiner, , p. 7;Google Scholar similarly Caton, , p. 102, on Beck.Google Scholar

⁸³ Cf. Crone, , Meccan Trade, pp. 246ff.Google Scholar

⁸⁴ Compare Khazanov's comments on the supposed symbiosis between nomads and agriculturalists (Nomads and the Outside World, p. 35).

⁸⁵ Cf. Barfield, T. J., The Perilous Frontier, Nomadic Empires and China (Oxford, 1989),Google Scholar which does in a sense describe China and its barbarians as part of a single political system, but without any attempt to mask the discordant nature of the parts that made it up.

⁸⁶ Lapidus, , pp. 34f.Google Scholar; Hourani, , p. 308;Google Scholar cf. also Barfield, p. 177.

⁸⁷ Gellner, , p. 113.Google Scholar

⁸⁸ Lapidus, , p. 35.Google Scholar has the ‘Abbāsids start by using Arab troops, presumably by simplification of Shaban's view that Arabs predominated in the ‘Abbāsid revolution.

⁸⁹ Beck, , p. 192.Google Scholar

⁹⁰ Unlike Tapper, Beck makes no claim about tribes as such, only about those of Iran; but the assertion that they have formed a single system with states through history is a bit strong for a paper in which history is basically modern. What evidence can be adduced in respect of Achaemenid, Parthian, Sasanid or early Islamic Iran? The question how far the Turkish influx from the Seljuks onwards changed the balance between tribal (especially nomadic) and non-tribal populations in Iran still has not found its historian.

⁹¹ Hourani, , p. 309.Google Scholar

⁹² Anderson, , p. 290.Google Scholar

⁹³ Cf. Barfield, , p. 172.Google Scholar Fried did not of course deny that the state replicated itself at the cost of simpler societies. As he puts it, the state would impose its form on less complexly structured societies that could bear it, while those that could not sustain it would become tribes (“Tribe to state or state to tribe in ancient China ?” (above, n. 53), p. 479, cf. p. 473). He does not however concern himself with the further fate of the tribes once formed.

⁹⁴ Tapper, , p. 63.Google Scholar

⁹⁵ Crone, , “The tribe and the state”, p. 71;Google Scholar cf. above, n. 45.

⁹⁶ Tibi, , pp. 132, 143.Google Scholar

⁹⁷ Anderson, , p. 290.Google Scholar

⁹⁸ Kostiner, , p. 246.Google Scholar

⁹⁹ Anderson, , p. 292.CrossRefGoogle Scholar

¹⁰⁰ Hourani, , pp. 307, 309, 310.Google Scholar

¹⁰¹ Lapidus, , pp. 28ff., 33, 34.Google Scholar

¹⁰² Cf. Tibi, , p. 133.Google Scholar

¹⁰³ What we need in his view are analyses of “face-to-face interactions of significant individuals and their audiences, who are communicating their culturally constituted aims through the use of politically laden signs”, that is of the “production of ideology through the use of signs in concrete, social acts of communication by charismatic personalities”, or “the semiotics of power” for short (pp. 75, 103).

¹⁰⁴ Cf. Crone, , Pre-industrial Societies, ch. 7 (which does not claim to be saying anything new).Google Scholar

¹⁰⁵ Gellner, , p. 124.Google Scholar

¹⁰⁶ The super-tribe is Tibi's expression (p. 134).

¹⁰⁷ Lapidus, , p. 30.Google Scholar

¹⁰⁸ Lapidus, , p. 28.Google Scholar

¹⁰⁹ Though the tribes of Iran and the Fertile Crescent can well be described as rural populations, I cannot with the best of my will bring myself to describe the Arabian desert or the Sahara as countryside.

¹¹⁰ Cf. Crone, , “The tribe and the state”, p. 76.Google Scholar Compare Crone, P. and Cook, M., Hagarism (Cambridge, 1977), P. 118,Google Scholar on “the richness and variety of the Berber presence in Muslim North Africa as contrasted with their barbarian anonymity in the days of the Romans”; Lapidus (p. 29) makes much the same point.

¹¹¹ Barfield, , p. 153.Google Scholar