Skip to main content Accessibility help

The Sunnah Jāmi'ah, pacts with the Yathrib Jews, and the Taḥrīm of Yathrib: analysis and translation of the documents comprised in the so-called ‘Constitution of Medina’

  • R. B. Serjeant


For almost a decade I have taught the ‘Constitution of Medina’ at Cambridge as the eight distinct documents it comprises, issued on various occasions over the first seven years or so, of Muḥammad's Medinan period, using a typescript text of these documents, designated A to H. My attention was first attracted to it when reviewing the late Professor A. Guillaume's translation of the Sīrah of Ibn Isḥāq/Ibn Hishām, because of the astonishingly close parallels with ḥawṭah treaties from Ḥaḍramawt upon which I was working about that time. Originally I had intended to publish a typical ḥawṭah agreement together with an analysis of the ‘Constitution’ which of course is not really a constitution at all. I still intend to publish my ḥawṭah corpus and demonstrate the similarity, but it has long been time to publish the analysis and identification of the separate documents contained in the ‘Constitution’, more particularly as it has important methodological and historical implications if anything approaching a definitive modern account of Muhammad's career is to be written.



Hide All

1 Guillaume, A. (tr.), The. life of Muḥammad, London, 1955, 231 seq.

2 of. my ‘Ḥaram and ḥawṭah, the sacred enclave in Arabia’, in Badawi, 'Abd al-Raḥmān (ed.), Mélanges Ṭaha Ḥusain, Le Caire, 1962, 4158.

3 ‘The “Constitution of Medina”’, Islamic Quarterly, VIII, 12, 1964, 316.

4 cf. introd. to Arberry, A. J. (ed.), Religion in the Middle East, Cambridge, 1969, II, 321, especially p. 10, but further researches have since brought new developments in my views.

5 cf. The Saiyids of Ḥaḍramawt, London, 1957, p. 6, n. 1; al-Wāqidī, Kitāb al-maghāzī, ed. Jones, Marsden, London, 1966, II, 547; al-Bukhārī, , ṣaḥīḥ, ed. Krehl, M. L.. Leyde, 18621908, II, 342, 348, 350, III, 322 (Anbiyā', 8, 14, 19, and tafsīr of Sūrah XII).

6 op. cit., I, 184. A da'wah can mean a confederation to aid or assist (Lane), but to assign it this sense at the early period is possibly a little previous.

7 cf. my ‘Constitution’, 67.

8 al-Iṣfahānī, Abū 'l-Faraj, Kitāb al-Aghānī, Būlāq, 1285/18681869, XIX, 94.

9 A pointer toward confirming the Jewish tradition of emigration from Palestine to the Ḥijāz lies in the name Fiṭyawn, of the Banū Tha'labah b. al-Fiṭyawn (Guillaume, , op. cit., 239).Al-Rawḍ al-unuf, II, 24, says, ‘Al-Fiṭyawn is a Hebrew word meaning anyone who is in charge of governing the Jews and their king (waliya amra 'l- Yahūd wa-maliku-hum)’. The late Dr. M. Gertner considered he had found the origin of this word in post-Biblical Hebrew and we had intended to write a note on it from the Hebrew and Arabic sources. Professor John Emerton has kindly undertaken to try and identify Dr. Gertner's word, and has supplied me with the following references: Levy, J., Wōrterbuch über die Talmudim und Midraschim, Berlin, 1924, I, 69, = Greek ὑπατεα, properly ‘consulate’, then ‘every era marked by an epoch-making event’. There is also = Greek ὑπατικς'. Cf. (ibid, IV, 26) , an abbreviated form of = Greek ὑπατεα. Cf. Krauss, S., Griechische und lateinische Lehnwörter im Talmud Midrasch und Targum, Berlin, 1899, n, 3940, = Greek ὑπατικς, consularis, a governor (who has been consul) appointed as an imperial governor. Cf. Jastrow, M., Dictionary of the Targumim, repr., New York, 1950, I, 58 = II, 1155. The word is also found in Syriac. I seem to recall that Gertner said that in the usage of the period it meant something like a village headman, but this requires confirmation.

10 Al-Aghānī, XIX, 95.

11 ibid, XIX, 97.

12 For al-YahŪd the text requires that al-Yahūdi be read.

13 The situation is very close to that which I have described in ‘A Judeo-Arab house-deed from Ḥabbān’, JRAS, pts. 3–4, 1953, 119.

14 Laja'a ilā = istanada ilā, i'taḍada bi.

15 Dīwān, ed. Kowalski, T., Leipzig, 1914, 35–6.

16 Al-Samhūdī, Wafā' al-wafā', Cairo, 1326/1908–9, I, 125.

17 Aḥmad Zakiyy Ṣafwat, Jamharat khuṭab al-'Arab, Cairo, 1352/1933, I, 5, in a passage on boasting against the Maqāwil Ḥimyar in the pre-Islamic era.

18 Al-Ṭabarī, Tafsīr, ed. Maḥjūd and Aḥmad Muḥammad Shākir, Cairo, 1374/1954–5–, v, 407 seq., in comment on Lā ikrāha fi 'l-dīni. Cf. al-Suhaylī, , al-Rawḍ al-unuf, Cairo, 1332/1914, II, 24. Some of these converts refused to abandon Naḍir when they left Yathrib. For a somewhat parallel practice, vowing a child to a saint, see my ‘Sex, birth, circumcision’, in Leidlmair, A. (ed.), Hermann von Wissmann-Festschrift, Tübingen, 1962, 203.

19 Ibn al-Athir, al-Nihāyah fī gharīb al-ḥadīth wa-'l-athar, Cairo, 1311/18931894, I, 43.

20 Bell, Richard (tr.), The Qur'ān, Edinburgh, 19371939, I, 232.

21 Tafsīr, Cairo, 1374/19541955–, XVII, 61.

22 Al-Ṭabari, , Tafsīr, Cairo, 1374/1954/1955/, VII, 61 seq.: Bell, , op. cit., I, 55, using Western numbering of the verses, those below commencing as no. 98.

23 ibid., I, 43.

24 Tafsīr, , Cairo, 1374/19541955–, VII, 59.

25 Reading so for the jimā' of the editors.

26 Tafsīr, VII, 70. Professor Beeston has supplied me with an invaluable note on ḥabl in ancient South Arabian. ‘Ḥbl is used in a stereotyped formula when the mkrb (mukarrib = mujammi') “gave juridical status to every community of (dhū, presumably here implying ‘united by’) god and divine patron, and/or of ḥbl and ḥmr”. It cannot be said even now whether the second pair of terms has a religious or purely secular association.’ The significance of ḥabl in this context is that it parallels in sense and circumstance, the ḥabl Allāh of Qur'ān, , III, 101. Cf. Beeston, 's Warfare in ancient South Arabia (2nd–3rd centuries a.d.) (Qahtan, Fasc., 3), London, 1976, 34, 38, and particularly 36, for mḥkm b-ḥbl, a settlement by treaty/negotiated peace, sought by Shammar of Raydān from the ting of Saba’ and Raydan.

27 The ancient and modern phrase, Kalimatu-hum wāḥidah, means that they are of one accord, united in their attitude, etc.

28 Tafsīr, VII, 71.

29 ibid, VII, 74.

30 ibid., VII, 91 seq.

31 Bell, , op. cit., I, 55, places this passage from Sūrah III, 100 (his no. 98) before Uḥud.

32 ‘Awn al-Sharīf Qāsim, Diblūmāsiyyat Muḥammad, Khartoum, n.d. [1971], 241–4, has printed the ‘Constitution’ adopting the division into separate documents made up till the time he wrote his London M.A. thesis. Muḥ. Ḥamīdullāh, Majmū'at al-wathā'iq al-siyāsiyydh, third ed., Beirut, 1389/1969, has noted my first article but simply reprinted the text uncritically without divisions, though he has given a long list of texts in which pieces of the eight documents are quoted. This could be valuable for further study, but I have not used all his sources. Qāṭī Muḥ. b. ‘Alī al-Akwa' al-Ḥiwālī, al-Wathā'iq al-siyāsiyyah al-Yamaniyyah, Baghdad, 1396/1976, 51 seq., appears only to have reproduced Ḥamidullāh's text and numbering. Miss O. H. Rahman has drawn my attention to Gil, M., The constitution of Medina: a reconsideration, Jerusalem, 1974, which adds one or two references to my accumulation (not all utilized here) of sources, but it takes no account of the division of the text into eight well-defined documents.

33 See Naṣr b. Muzāḥim al-Minqarī, Waq'at ṣiffīn, Cairo, 1382/19621963, 505, for a Shi'ah commentary on the teḥkīm document and, 510, for the document in one of its recensions. It is not, of course, suggested that this was the original title of documents A and B.

34 Qāsim b. Sallām Abū ‘Ubayd, Kitāb al-amwāl, Cairo, n.d., 202 seq.

35 Al-Bidāyah wa-'l-nihāyah, Cairo, 1351/1932, III, 224 seq.

36 For al-Zuhri see Duri, A. A., ‘Al-Zuhrī: a study on the beginnings of history writing in Islam’, BSOAS, XIX, I, 1957, 112.

Gil, M., op. cit., 47, cites another isnād from Sayyid al-Nās, Ibn (first half of the eighth/fourteenth century), ‘Uyūn al-athar, Cairo, 1356/19371938, but it seems to have no special evidential value and is not to be regarded as an isnād missing from Ibn Isḥāq who, in fact is quoted by Ibn Sayyid al-Nās as an authority in his own right.

37 Al-Zuhrī's text has been collated with that of Ibn Isḥāq; see pp. 40–2. Al-Zuhrī narrated Tradition on the authority of the Prophet's servant Anas b. Mālik (ob. c. 90/709) (Usd al-ghābah, Cairo, 1280/1863–4, I, 127), in whose house the two ḥilfs were contracted.

For al-Zuhrī's wording kataba bi… cf. Dozy, Supplément, kataba ilay-hi bi-dhālika, il lui écrivit pour lui ordonner faire telle chose.

38 Muḥ. b. Ḥabīb, Kitāb al-munammaq, Haydarābād, 1384/1964, 88.

39 Al-Wāqidī, , op. cit., II, 612. The Prophet sometimes sealed a document with his fingernail (ibid., III, 1025)—here, I suppose in the sense of closing it. A paper was sometimes endorsed on the back (Muḥ. b. Yaḥyā al-Ṣūlī, Adah al-kuttāb, Cairo, 1341/1922, 149).

40 cf. infra, p. 11.

41 Tafsīr, Cairo, 1374/19541955-, x, 110. See also al-Jāḥiẓ, , Rasā'il, Cairo, 1384/1964, 1, 14; Ibn al-Athīr, al-Nihāyah, IV, 167, the naqīb is the ‘arīf over the tribe (qdwm), set over them (almuqaddam ‘alay-him), who becomes acquainted with their news, and examines their conditions, i.e. inspects (yufattish). Ibid., III, 86, defines the ‘arīf as al-qayyim bi-umūr al-qabīlah, the person who manages the affairs of the tribe, or the collective body (jamā'ah). The functions of the naqībs are well set forth in al-Wāqidī, , op. cit., 11, 547. Ḥabīb, Muḥ. b., Kitāb al-muḥabbar, Ḥaydarābād, 1361/1942, 38, lists the naqībs of the Banū ‘l-'Abbās b. ‘Abd al-Muṭṭalib, of Khuzā'ah, Tamīm, and others—perhaps indicating that the family had a control over this group of tribes associated with the Meccan ḥaram. Al-Bukhārī, Ṣaḥīh, IV, 395, notes that ‘arīfs refer a case to Muḥammad.

42 In the north of the Yemen a naqīb is a tribal chief lesser than a shaykh al-mashāyikh.

43 op. cit., 11, 19.

44 ‘Alī b. Ibrāhīm Nūr al-Dīn al-Ḥalabī, al-Sīrat al-Ḥalabiyyah, 1, 479, 11, 119.

45 Al-Suhaylī, , op. cit., 1, 276, avers that the naqībs are like qawm Mūsā. Cf. al-Balādhurī, Ansāb al-a shrāf, ed. Muḥ, Ḥamīdullāh, Cairo, 1959-, 1, 253–54, quoting the Prophet as saying, Inna-kum kufaiā' ‘alā, qawmi-kum … wa-anā kafīl ‘alā qawmī ‘You are guarantors/responsible for your tribe and I for mine’.

46 Islamic Quarterly, loc. cit.

47 Qur'ān, , LXIII, 7. Ibn al-Athīr, op. cit., 11, 190, quotes a Tradition, ‘Do not call the Munāfiq lord (sayyid), for even if he be your sayyid he is still a Munäfiq’.

48 Anwār al-tanzīl, ed. Fleischer, H. O., Leipzig, 18461848, 11, 335.

49 Tafsīr, 1374/19541995-, XIV, 297–8.

50 op. cit., 11, 566. Cf. al-Bayḍāwī, , 11, 320, for their false oaths.

51 op. cit., 1, 186. Cf. Qur'ān, IX, 55.

52 Tafsīr, ed. cit., XIV, 431. Cf. ‘Alī b. Aḥmad al-Wāḥidī, Asbāb al-nuzūl, Cairo, 1388/1968, 287: ‘Abdullāh b. Ubayy says, Lā tunfiqū ‘alā man ‘inda Rasūli ‘llāhi, referring to the A'rāb auxiliaries paid by Muḥammad.

53 op. cit., 1, 399.

54 op. cit., XIV, 433.

55 Islamic Quarterly, loc. cit., 11.

56 Bravmann, M. M., The spiritual background of early Islam: studies in ancient Arab concepts, Leiden, 1972, 26 ff. This is a most useful study to be read along with the ‘Constitution’.

Cf. Sattar, M. A. (Muḥ. ‘Abd al-Sattār), Al-Hīrī's life and works with an edition of his Wujūh al-Qur'ān, Cambridge Ph.D. thesis, 1974 (Ph.D. 9008), 5, where iymān means al-amn and al-thubūt with a reference to Qur'ān LIX, 23, wa-āmana-hum min al-khawf. (Al-Ḥīrī, fl. 361431/971–1039.)

57 op. cit., 11, 276. Al-Bayḍāwī quotes the incident which gave rise to the revelation of this verse. A party of Banū Asad came to Medina in a droughty year, making a show of the two shahādahs. They used to say to the Apostle of Allāh, ‘We have brought you our families and dependants (al-athqāl wa-'l-'iyāl) and we did not fight with you as the Banū So-and-so did’. They wanted ṣadaqah and were obliging the Prophet to do them a favour. Cf. al-Tha'ālibī, Thimār al-qulūb, Cairo, 1385/1964, 116–17, for verses in similar vein. The rhyming cliché, which I Burmise in its original form may also have included the word qitāl, is characteristic of tribal society. It may even be a part of a formula in this case for the action of forcing a donation by bringing one's womenfolk to a powerful or wealthy protector as a gesture of seeking for his aid. This used, at any rate np till recent years, to be done in some tribal areas of the former Western Aden Protectorate.

It looks as if, under the circumstances, their use of the term āmannā showed impudence.

58 op. cit., 7 seq.

59 Ed. Fayḍi, Ḥusayn b. ‘llāh al-Hamdānī, Cairo, 1375–7/19561958, 11, 70 seq. Unfortunately ths editor did not live to complete the edition of this little-known work.

60 Tafsīr, ed. Shākir, XI, 492504. Al-Ṭabarī quotes one authority who identifies ‘those who have guaranteed security’ as I have rendered alladhana āmanū here, with the Muhājirūn. This, in my view, suits the situation well, but in this case the verse is perhaps to be considered a Medinan verse in a Meccan sūrah, which does in fact have some other verses considered Medinan by al-Bayḍawī in it. However, if Meccan, it may be better to render alladhīna āmanū in its more conventional senses.

61 Baneth, D. Z. H., ‘What did Muḥammad mean when he called his religion “Islam” ?’, Israel Oriental Studies, 1, 1971, 183 seq. He cites in particular the concept of the individual who aslama wajha-hu li-'llāhī. Wajh has among its meanings that of ‘honour’.

Cf. al-Ḥīrī, , ed. Sattar, Islām, for similar data, especially the last phrase.

62 The many sources and citations supporting senses assigned to individual words in the documents cannot be quoted in full here.

63 cf. Ibn al-Athīr, Usd al-ghāhak, Cairo, 1280/18631864, 1, 127 seq.

64 Guillaume, , Life of Muhammad, 59; Wüstenfeld, , text, 88.

65 cf. Islamic Quarterly, loc. cit., 7. Ibn Kathir, al-Bidāyah, 111, 224, words this—between Qnraysh and the Anṣār, or the Muhājirūn and the Anṣār.

66 Defined by Ibn al-Athir, al-Nihāyah, IV, 201, in the words, Wāda'a Banī Fulān-in ay ṣālaḥa-hum wa-sālama-hum ‘alā, tarki ‘l-ḥarbi wa-'l-adha’.

67 cf. p. 9, n. 33. I intend to deal with this subject at greater length in the Cambridge history of Arabic literature at present in preparation.

68 Wafā' al-wafā', 1, 162, in a tantalizing note, he says that a group of tribes forming Aws Allāh did not accept Islam until after Khandaq under the influence of their chief Abū Qays b. Ṣayfī al-Aslat who would seem to be a leader and poet of prominence according to notices in the Aghani but does not seem to figure in the Sīrah unless under some other name.

69 op. cit., 1, 176.

70 Tāri, kh, ed. de Goeje and others, Leiden, 18791901, 11, 111, 1395. Cf. al-Wāqidī, , op. cit., 11, 454.

71 Dahima-hum is the term used in document C/2. Perhaps both al-Wāqidī and al-Ṭabarī have in mind the agreement with Qurayẓah rather than this earlier agreement, and the ‘Constitution’ as presented by Ibn Isḥaq in one document would presumably have seemed to them to belong to Year 1 of the hijrah.

72 Al-Nihāyah. 1, 43.

73 cf. the Tradition, Al-Muslimūna tajma'u-hum da'watu ‘l-Islāmi.

74 Gil quotes Ḥusayn b. Muḥ…. al-Diyārbakrī Tārīkh al-khamis, 1, 398, not available to me. As seen above, however, it is only document C that is to be considered as such.

75 Guillaume, , op. cit., 235; Wüstenfeld, , 346.

76 Al-Samhūdī, , op. cit., 1, 151.

77 op. cit., XIX, 96; al-Samhūdī, , op. cit., 1, 152.

78 The dating of the murder is discussed by Jones, J. M. B., ‘The chronology of the maghāzī— a textual survey’, BSOAS, XIX, 2, 1957, 262 seq. The agreement at the house of Ramlat bint al-Ḥārith is likely to have followed the murder of Ka'b and of the Ḥārithī tribesman's ḥalīf fairly soon.

79 Al-Wāqidī, , op. oit., 1, 191.

80 op. cit., XIX, 106.

81 op. cit., 1, 76–7.

82 cf. Jones, J. M. B., art. cit., 254, where various dates at the end of Year 6 or beginning of Year 7 are given, and one date as late as Jumada I, Year 7.

83 It is attractive to consider Abū Bakr's remark (al-Wāqidi, , op. cit., 11, 460) that one of the things by which Allāh turned back Qurayẓah from what they intended was that Medina was guarded (tuḥras), embodies the idea of divine protection—but the passage itself and another (p. 467) seem to indicate simply protection by patrols.

84 cf. Guillanme, , op. cit., 509 seq.; Wüstenfeld, , 754–6. Most sources place Ḥudaybiyah in Dhū ‘l-Qa'dah, Year 6, and Khaybar about a month later or in Muḥarram Year 7.

85 op. cit., 11, 440. Cf. Jones, J. M. B., art. cit., 251.

86 op. cit., 11, 454, and for Qur'ān quotations relating to the event ibid., 11, 494. Al-Wāqidī adds, Wa-yuqīmū ‘alā ma'āqili-him al-ūlā bayna ‘l-Aws wa-'l-Khftzraj, which in fact does not belong to this document G, but would doubtless be implicit for the Jews in the earlier documents A, etc.

87 op. cit., 11, 455.

88 Al-Ṭabarī, , Tafsīr, 1374/19541955-, X, 258 seq.; al-Wāqidī, , op. cit., 11, 458; Guillaume, , op. cit., 267; Wüstenfeld, , 395. Al-Wāqidī, loc. cit., says that Sa'd b. Mu'adh was reviled by Ka'b b. Asad, but the retort was made to him that you are no equal (kuf’) for him. Naḍīr were nobler (a'azz min-kum) than you, and your blood-wit (diyah) half theirs.

89 Al-Wāqidī, , op. cit., 11, 455.

90 ibid, 11, 456.

91 Al-Bidāyah, IV, 103.

92 This might be Tha'labah of documents C/3 and D/l.

93 cf. my ‘Ḥaram and ḥawṭah’, p. 1, n. 2, supra. Makdisi, G. (ed.), The notebooks of Ibn ‘Aqīl: Kitāb al-funūn, Beirut, 1970, 1, 349, discusses the question of ‘the murderer when he takes refuge in the ḥaram (al-qātilu idhā ‘ltaja'a ilā ‘l-ḥarami)’.

94 op. cit., 11, 485.

95 Mu'jam al-buldān, ed. Wüstenfeld, F., Leipzig, 18661871, IV, 11, 619.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

Bulletin of the School of Oriental and African Studies
  • ISSN: 0041-977X
  • EISSN: 1474-0699
  • URL: /core/journals/bulletin-of-the-school-of-oriental-and-african-studies
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed