Version classiqueVersion mobile

The rules of Barat. Tribal documents from Yemen

 | 
Paul Dresch

Part 3. Texts and translations

List of Guarantors (text B)

اسماء الضمناء

Texte intégral

  • 1 For guaranty in ʿurf see al-Sudumī Naẓarīyat al-ʿuqūbah pp. 149-74. His inclusion of wide a range o (...)

1The role of guarantor or ḍamīn (pl. ḍumanā’) has been mentioned several times, and the term itself occurs prominently in our major texts: in the market agreement and in both versions of the broader pact called qāʿidat al-malāzim. Attention has already been drawn (part three) to the large number of guarantors who signed the latter. Their names are listed, below, in the present section. A brief note is in order, first, however, on the general phenomenon of guaranty. Although a range of terms is found in Arabic (ḍamān and kafālah are perhaps the best known of these), and the instances to which one term applies may differ, there seems always to be in evidence a simple concept of taking responsibility for others.1

  • 2 Dresch Tribes pp. 89-91, 96. The view given there of Dhū Muḥammad needs qualifying. It seemed in th (...)

2Certain figures may be spoken of as shaykhs of guaranty, and Imams sometimes recognized men by that title as responsible for collecting tax or providing hostages. In tribal terms, however, “shaykh of guaranty” (shaykh al-ḍamān) is hardly an administrative rank. People may disagree as to who deserves the name, and as to how many such figures or families there are in a given tribe or section.2 The present documents (texts A and B) suggest that prominent shaykhs at Baraṭ had no kind of monopoly on speaking or signing for elements within their section: one could not, I think, safely guess from the lists of guarantors, as opposed to the names of witnesses, even which were the most prominent. Everyone equally who signed, for however many men or however few, qaddama wa-kafala, as the formula usually runs: he “put forward” (his face or wajh: in other words his honour) and “gave surety” (for his men). In doing so he made himself answerable to others.

  • 3 Plainly there are limits to these expectations. As we mentioned in the introduction (part one) many (...)

3The guarantors of the market or of the Code are meant all to act together, but each is responsible for his own people. If someone refuses to pay his share of a debt incurred by the whole tribe, for instance, his specific guarantor or guarantors must pursue him (text A section 9): they are responsible for seeing the payment made, indeed, even if the man himself is “cut off” from their tribal section. If one of Dhū Muḥammad wrongs his fellow at market, his guarantor has to pay in the culprit’s name four-fold, then recover the debt (text 1 section 3). If someone takes refuge with another section and his own section refuse amends for a wrong that he suffered, their guarantor is “vilified” (text B section 38). Penalties seem to fall mainly, and at first solely, on the guarantor. In the case of shaykhs, questions about the apparent “cost” of the role –about the risks attendant on default by those for whom one “gave one’s face”– are met with what seems incomprehension. Providing guaranty is part of being a shaykh.3 The relation itself is markedly asymmetrical: guaranty by the shaykh requires no formal loyalty from the tribesman for whose actions he provides surety.

  • 4 For the equation of ḍamīn and lazīm, text 1 section 12, text B section 40; for lazīm and kafīl, tex (...)

4Guarantors for qawāʿid al-malāzim, or indeed the market-guarantors (ḍumanā’ al‑sūq), might seem to act as such almost ex officio, as shaykhs of their minor sections. Certain other ḍumanā’, or guarantors, do not. Those at arbitration (text A section 21, text B sections 15, 39, 40), for instance, could be anyone agreed by the parties at odds. The grouping in our texts of amends for breach of guaranty with those concerning truce and sharing food (text A section 13, text B section 13) suggests that guaranty might involve any tribesman and might arise in a range of circumstances. Certain terms overlap here (ḍamīn, kafīl, lazīm) and usually in fact are virtual synonyms.4 Different circumstances, however, can be dealt with separately.

5The most prominent cases of formal guaranty are those that attend arbitration, truce, or the settlement of debt. Each principal party, whether single or collective, chooses guarantors for the other’s actions. Party A thus have a guarantor (or several guarantors) li-A ʿalā B, responsible to A for B’s conduct; part B have a guarantor li-B ʿalā A. The guarantors for each side may agree among themselves to act together (shabk), but each principal has claims only on their own guarantor, that is, their ḍamīn or kafīl. Although the choice involves all manner of practical considerations, such guarantors may in theory be drawn from anywhere. The guarantor is answerable for a debt, no doubt, but the emphasis is all on forcing the principal to pay, or otherwise to act as required, and one might often choose a guarantor who for instance lives on a route by which the principal must reach the main road, so he can easily bring pressure to bear if need be.

  • 5 See the introductory essay (part three). Al-Sudumī (Naẓarīyah p. 151) says the debtor can “cover” h (...)

6A rather different case arises at the start of disputes. Person A wrongs a person B from another family or section, and commonly someone else from A’s own section steps forward to provide B with surety that the debt will be paid, providing “cover” (ḥajb or ḥujbah) for the debt and debtor.5 The two whole sections are therefore not placed at daggers drawn or at rifles raised (part three, above). In this case, too, one can speak of a ḍamīn or kafīl. All these guarantors, whether acting for their own section or ensuring a truce for others, are “responsible” to someone else, and in our documents each is referred to on occasion as lazīm. The person to whom a guarantor of any kind is answerable is his mulazzim.

  • 6 Cf. al-Sudumī Naẓarīyah pp. 153-4. Usually the formal claim should be waived.

7A lazīm paying a debt on behalf of the original culprit is said to get half the total amount due, and half goes to the original creditor on whose behalf he acts (text B section 37). Shaykhs at Baraṭ these days think that excessive: a third seems to them a more proper maximum, and in practice only the costs of mediation (farāʿah, of the coming and going required) are usually sought.6 But even half the debt, if one ever received it, appears a poor return on what can be a large risk. In fact guarantors for arbitration and the settlement of debts can be hard to find. People stepping forward to “cover” their fellows in immediate dispute with outsiders is less problematic, for obvious reasons: one does not usually want a fight between whole tribes or sections. But relations of guaranty, some for a specified period and some permanent, all of them specific, are extremely numerous.

  • 7 D.E. Walters, The general features of archaic European suretyship in T.M. Charles-­Edwards et al (e (...)

8Concern with personal commitments to ensure payment or truce recurs in many “archaic” or non-state systems. It recurs in early Wales or Scotland, for instance, in the Frankish world, and in much of mediaeval Europe. Security for an undertaking was provided in the form of claims to persons or property; and persons other than the principals bound by that undertaking acted as sureties, or bondsmen, or guarantors. Modern security is “essentially collateral or ancillary” to some other obligation. In so‑called archaic law, by contrast, “the participation of sureties [was] central to the formation of agreements which gave rise to ‘debt’... as well as to the performance of other acts”.7 The duties of these sureties or guarantors could be quite specific.

  • 8 Kelly Irish Law pp. 167-73 and passim. The material is rehearsed in Walters ‘European suretyship’ p (...)

9Early Irish law presents a rich set of distinctions.8 The ráth was a surety who pledged the value of a debt, and if the principal defaulted twice then the ráth’s own property was distained by the creditor. The aitire, by contrast, was a surety who pledged his freedom, agreeing to be held for a time by the creditor if the debtor failed to pay. Both types of bondsman or guarantor seem to act with a naidm, an “enforcing surety”, whose honour was pledged to enforcing payment and who could rightfully attack, if need be, the debtor who refused payment. The concepts themselves are not crystal-clear. We are told that a ráth (who guaranteed the amount owed) should be someone who would pay meekly; yet he should not only possess ten cows and himself be free of debt, but apparently should be a large man, with a shield and a sword, who had won three fights. One would have thought that more what was needed of a naidm. What does seem clear is that the naidm (who enforced compliance) had no liability for the debt itself, but a ráth did.

  • 9 Cf. the discussion in note 3 to text B section 13/2 (part six, above).
  • 10 For further discussions and examples see P. Dresch, Episodes in a dispute between Yemeni tribes, De (...)

10Unlike tribesmen further west and south, people at Baraṭ sometimes distinguish between a kafīl, responsible for monetary debt, and the ḍamīn, responsible for persons’ conduct, of which paying a debt might be an instance. But there seems not to be a clear distinction between persons on the Irish model. More to the point, perhaps, is that merely financial debts present one set of problems: they are likely to arise from commerce, there are several ways one might address them, and financial surety might therefore perhaps be given by a non-tribesman.9 The guaranty of others’ actions presents quite other problems, which can only be taken on by men of tribal standing. But in either case one could usually say equally ḍamīn or kafīl. A terminological distinction which does arise often, however, is that between primary and secondary guaranty, the latter being called radm or ṣawn. Secondary guarantors back up the kufalā' or ḍumanā', and in effect provide guaranty for the primary relation between guarantor and principal.10

  • 11 Stewart ‘Contract’ pp. 176-7, 254 ff. Aligning terms from other systems with those of English or Ro (...)

11In all these cases, the parties to the original dispute (the antagonists, or ghuramā', creditor and debtor) cannot act directly against each other, nor can a guarantor act against one of these principals without a demand from the other or, in the case of arbitration, from the arbitrator (cf. text B section 15). The guarantor for person A must act against B if person A demands it (al-lazīm ʿalā ra'yi mulazzim-hu); the guarantor for A’s conduct must act against A if B demands it. In a sense, though not quite that of common or of civil law, each of the disputing parties is bound to the other by a unilateral contract.11 Meanwhile, action by one of the original principals or antagonists, no matter how sound his claim, directly against the other insults the guarantors, who are due from him amends.

  • 12 Stewart ‘Contract’ pp. 223-28. For the quotation below, Walters ‘European suretyship’ p. 101. Stewa (...)
  • 13 The complexities of “modern” law are considerable. Anglo-American arguments about liability seem us (...)

12Stewart invokes here a Germanic distinction between duty and liability (Schuld and Haftung), such that the original creditor may owe a duty, or otherwise a debt, and only the guarantor be liable.12 “The distinction between the ideas of debt and liability is useful”, as Walters remarks, “even if their primitive location in separate persons is unproved”, and their practical identity in much state-centred law, one might think, is indicative of the way that states impose duties uniformly, at all times, on citizens or subjects: a duty recognized in law, as distinct from simply morals, will usually entail liability. An easier solution to the (tribal) Arab cases, however, may be to distinguish simply between separate duties, and to distinguish rights from means of restitution. After all, the original debtor is still liable for the original debt and has a duty to pay, though he is not rightly vulnerable to retaliation; the guarantor has a duty to enforce payment, and is separately liable, for the creditor can summons him (indeed, “blacken” him) if he fails to act.13

13Whatever juristic approach is taken, the practicalities are clear. In all but the simplest debts, restitution is made by way of third parties, and our disparate cases all illustrate the general idea of covering (ḥajb). As we saw in the introduction (part three) a person from the tribe, but unrelated to the antagonists, can simply step in (darak) and take responsibility; or a protector (rabīʿ) can secure to his fellows the promise of payment from a stranger (text B section 7); or an existing guarantor can pay others on the culprit’s behalf (text 1 section 4). The debt or duty is everywhere displaced. Person A owes person B either payment or some other duty, and person C provides B with surety.

  • 14 Stewart ‘Contract’ pp. 176-7, 186-7, 229 and passim.

14Stewart offers two causal explanations of what he calls the three-party contract (the case where A agrees to pay B or attend arbitration and B selects a guarantor for A’s conduct).14 First, payment is arranged by way of guarantors to keep the antagonists apart, for pursuing the original debt directly would simply exacerbate the original dispute. That is certainly how people in Yemen speak of guaranty and settlement. Secondly, the guarantor is liable to “blackening” (i.e. loss of honour), not just liable for the debt, because otherwise one faces an infinite regression: person A fails to pay, for instance, and his antagonist B therefore calls on C, who is guarantor for A’s behaviour; B now has a claim against C and to ensure that C pays this second debt he finds a guarantor for C’s behaviour, called D; and if D fails in turn to act, then one needs an E or, if that fails, an F.

  • 15 See Dresch ‘Episodes in a dispute’. For something of the broader context, idem. Tribes pp. 379‑83. (...)

15An instrumental approach may serve to describe single actions, perhaps, but hardly to explain a social fact. The system, if we choose to call it such, does not always work. Just a few decades ago, for example, men from ʿIdhar fell out with each other, and guarantors from al‑Ahnūm, nearby, were arranged to oversee settlement. One of the original antagonists seems not to have paid; the guarantors summonsed the defaulting principal, then the secondary guarantors (kufalā' al‑radm), and tried to enforce the judgement; the principal again defaulted and lodged claims against the guarantors; these claims and counter-claims went more than once to arbitration, which each time involved further (quite separate) guaranty, and for years the trouble simply spread with people being shot at and property seized on every hand.15 Even the supposedly more definitive practice of “blackening” (part three, above) can unravel. A jidhn or malām placed to “blame” a defaulter may be met by a jidhn elsewhere in reply.

  • 16 Elsewhere clarity and certainty in this sense may be related more closely. Where rank is admitted, (...)
  • 17 Sections or tribes may figure in disputes as plaintiffs and respondents, but a collective identity (...)
  • 18 al-Sudumī Naẓarīyah pp. 175 ff. Someone who wrongs outsiders may even give a rifle to one of his ki (...)

16Pacts such as those examined here try to fix responsibility, or at least to fix the course of procedure: men contract to recognize each other’s covering of others, or not to do so, in specific circumstances. More generally, however, the prominence of surety implies not certainty of outcome so much as clarity of definition.16 Any disagreement implicates the two largest tribal sets to whom the principals respectively belong, and potentially also their respective friends, acquaintances, neighbours, and in-laws. As we said in the introduction, arbitration specifies, among other things, the disputing parties. But even before that, and again afterwards, guaranty pins down answerability. Although sections or tribes as such may figure as plaintiffs and respondents, they fragment indefinitely into potentially conflicting views from this family-group or that (quite who is answerable for what is labile), while guarantors, in strong contrast, are named personally in a dispute-document or otherwise have been named already in signing some variety of pact.17 In the former case pledges are given, usually in the form of rifles.18 In the latter case one gives one “face” in writing, and responsibility is assumed to be passed on within the family.

  • 19 al-Ḥajrī Buldān al-yaman, al-Maqḥafī Muʿjam al-buldān. As noted earlier, spelling and vowelling are (...)

17Text B carries about 150 such names. It seems worth reproducing them in the hope that one day more documents will come to light from the same place and period and we can start on the detail of local history. Obviously, caution is needed. The texts that we have show the problems one has to expect in copying out mere lists, whether saut de même au même or repetition, and no “redundancy” allows firm corrections. I have tried to reconcile the personal names as they appear in the different copies of texts A and B, showing major alternatives and ambiguities in parentheses. Most of the family names can be found in gazetteers.19 Some names that are prominent nowadays are contained within the (agnatic) sets that names in the documents define, as Bayt Thawābah, for instance, probably form part of Āl Ṭashshān in Dhū Zayd. Personal names often recur between generations within a family, but attention is drawn to how close those of Dhū Mūsā are here to those in the market-guaranty.

List of guarantors, text B

List of guarantors, text B

Witnesses

18يعلم ذالك وإذا سدوا اللحمتين بينهم بغير قصا وبغير حضور قاعدة فلا هي تمحق القاعدة وهذا بحضور الشهود وهم النقيب منصور بن مهدي والنقيب حسن بن ناصر جزيلان والشيخ أحمد صالح ابو رأس والشيخ مرشد بن حسين طشّان والشيخ حسين بن قاسم دمينه والشيخ صالح البحر والشيخ هادي بن علي فَرَج وعلي بن راشد عاطف وشهود المعاطرة ناصر بن حسن الضُويني وناصر بن جابر وحسن بن أحمد بن محمد والولد القاضي علي بن محمد العَنْسي وعِدة من الناس بتاريخه شهر محرم سنة 1211

Transcriptions

19وكتب محمد بن أحمد علي نقلها القاضي محمد بن يحي عافاه الله من القواعد المرقومة بخط القاضي محمد بن أحمد المتناظرة وكتب محمد بن يحي وفقه الله ونُقلت عن خط الوالد القاضى العلامة بدر الدين محمد بن علي بن محمد رضوان الله عليه من القواعد المذكورة بخط مَن ذُكر ومِن النقل الذي بخط حسين بن مفلح ابو رأس حسبما استأمر به مِن المعاطرة وذو عَمْر نقل قاعدة للمعاطرة وقاعدة لذو عَمْر من تلك القواعد وشواهد لهم يُعلم ذالك بتاريخه شهر جماد الاول سنة 1214 وكتب القاضى محمد بن علي لطف الله به وهذه منقولة بخطي من الام مِن خط المذكور نقلتها لآل الضويني أهل حجان اللفظة باللفظة لا زيادة ولا نقصان تاريخ النقل الذي بخطي شهر محرم الحرام سنة 1238 وأنا الحقير إلى الله صالح بن أحمد (بن صالح بن أحمد) العَنْسي هذا النقل مِن الاصل لفظاً وكان الجار ثقة اذا ادعا بمدة دون ماله وكان مَن عاب في وجيههم فهو بالمحدش على شروع اهل الملازم في سوق العتب وغيره تم نقل هذه القاعدة على الصورة المنقولة مِن خط النقيب محسن بن اسماعيل أحمر الشعر كما وجدت بدون زيادة ولا نقصان كان نقلها للأخ الشيخ أحمد عبده بن ناجي ثوابه يُعلم ذاك وتاريخ هذا النقل 29 شهر شوال سنة 1414 الف عام وأربعمائة عام وأربعة عشر عاماً هجرية

20كتبه الفقير إلى الله

21يحي بن أحمد عتيق البرطي وفقاه الله

Notes

1 For guaranty in ʿurf see al-Sudumī Naẓarīyat al-ʿuqūbah pp. 149-74. His inclusion of wide a range of cases under the same general heading of ḍamānāt seems to me right. He begins, very reasonably, with “topical” or “substantive” cases of ḥajb or ḥujbah, and goes on the types of pledge or guaranty that accompany “procedure”.

2 Dresch Tribes pp. 89-91, 96. The view given there of Dhū Muḥammad needs qualifying. It seemed in the early 1980s that there was one family in each fifth who were spoken of as shaykhs of guaranty. Obviously families rise and fall, so Bayt Dāris, for instance, had become prominent in Āl Dumaynah after the death of Fayṣal ʿAwfān in the 1960s. But there seems in fact a more basic disagreement over whether there need even be a single family called mashāyikh ḍamān for each fifth.

3 Plainly there are limits to these expectations. As we mentioned in the introduction (part one) many shaykhs recently have “declared themselves quit” of the market guaranty; to avoid a specific guaranty being imposed on them by people presenting slaughter-beats etc. shaykhs will sometimes flee the house. On the other hand, standing forward as guarantor in transactions with outsiders is an obvious part of maintaining more general influence for politically ambitious shaykhs. But it is hard to avoid wondering at the sheer trouble involved in responsibility for people over whom one has no formal authority, and often no real sanction. Managing relations among them occupies one’s whole life.

4 For the equation of ḍamīn and lazīm, text 1 section 12, text B section 40; for lazīm and kafīl, text A section 13, text B section 13. Plainly there are contexts where one term would be more natural to use than another, but I have not found at Baraṭ any basic differences of meaning among them. Examples are discussed below.

5 See the introductory essay (part three). Al-Sudumī (Naẓarīyah p. 151) says the debtor can “cover” himself rather than being covered by a third party. I have not heard this usage among friends from Baraṭ, but it deserves noting in discussions of duty and liability, below. My own understanding is that covering applies primarily to persons, rather than the debt.

6 Cf. al-Sudumī Naẓarīyah pp. 153-4. Usually the formal claim should be waived.

7 D.E. Walters, The general features of archaic European suretyship in T.M. Charles-­Edwards et al (eds.) Lawyers and Laymen (1986), p. 93. This provides a convenient set of references to a range of early European systems. A good many more are provided in F.H. Stewart, The contract with surety in Bedouin customary law, UCLA Journal of Islamic and Near Eastern Law vol. 2/2 (2003).

8 Kelly Irish Law pp. 167-73 and passim. The material is rehearsed in Walters ‘European suretyship’ pp. 99-101. Both the ráth and aitire, below, had a separate claim against the defaulter whose debt they had guaranteed. The naidm, it seems, was by contrast meant to act against the principal without further or later claim at law. Both the naidm and the ráth, however, stood to lose their honour-price if they failed to keep faith with the person to whom they had given surety.

9 Cf. the discussion in note 3 to text B section 13/2 (part six, above).

10 For further discussions and examples see P. Dresch, Episodes in a dispute between Yemeni tribes, Der Islam vol. 64/1 (1987), ‘Ḥūth’ (1990) p. 73-4,81. I have pursued with friends from Baraṭ the possibility of a distinction between a “paying surety” and an “enforcing surety” but found no clear instances. In discussion we have usually ended up saying kafīl for the first possibility and radīm or kafīl al-radm for the second, which may result from my own rather Ḥāshidī way of putting things.

11 Stewart ‘Contract’ pp. 176-7, 254 ff. Aligning terms from other systems with those of English or Roman law, although hard to avoid, produces problems. Revealing though Stewart’s analysis is here, one cannot help doubting whether “contract” is quite the crux. A “unilateral contract” might arise from an offer to all the world, as in such splendidly named English cases as Carlill v. Carbolic Smokeball Co.; Civil Law, as in France, deals with contract quite differently. But the ethnography is clear enough without the jurisprudence.

12 Stewart ‘Contract’ pp. 223-28. For the quotation below, Walters ‘European suretyship’ p. 101. Stewart has uncovered here a set of questions to which anthropologists have given sadly little thought since Davy’s La Foi jurée.

13 The complexities of “modern” law are considerable. Anglo-American arguments about liability seem usually, if I understand them, to begin from a breach of duty. The Roman tradition argues instead from a wrong (culpa, in modern French law une faute). Both traditions, however, distinguish liability or obligation from restitution. “Natural obligations” in French law thus cannot be legally enforced, but nor can a payment which discharges such an obligation be recovered. The same applies in English law, oddly, to gaming debts.

14 Stewart ‘Contract’ pp. 176-7, 186-7, 229 and passim.

15 See Dresch ‘Episodes in a dispute’. For something of the broader context, idem. Tribes pp. 379‑83. For the jidhns, below, Dresch ‘Placing the blame’ and Glaser Ma’rib pp. 76-7, 86, 103. As we shall see in part seven, functionalist argument works poorly.

16 Elsewhere clarity and certainty in this sense may be related more closely. Where rank is admitted, a guarantor may be admitted like the Germanic Leistenbürger whose promise might explicitly overawe the debtor: Walters ‘European suretyship’ p. 104. That happens in Yemeni practice (where Irish warnings against choosing too powerful a surety have also their echo), but it does not appear formally in law.

17 Sections or tribes may figure in disputes as plaintiffs and respondents, but a collective identity has no face (wajh), cf. Dresch Tribes p. 79. To form an agreement one needs individual “faces” (wajīh), a common classical usage, of course, for notables.

18 al-Sudumī Naẓarīyah pp. 175 ff. Someone who wrongs outsiders may even give a rifle to one of his kinsmen or fellow tribesmen, securing “cover”, and the second man then offer rifles to the wronged outsiders. In large-scale confrontation and truce such pledges come in bundles. Although compensation “forty-four-fold” is in my experience mere rhetoric, 44 rifles may often be pledged by one tribe to another. It used to be, even two decades ago, that the ground floor of houses of noted shaykhs was often stacked with bundles of rifles as surety. The quality of such weapons varied from excellent to simply rubbish.

19 al-Ḥajrī Buldān al-yaman, al-Maqḥafī Muʿjam al-buldān. As noted earlier, spelling and vowelling are generally more reliable in the former work.

Table des illustrations

Titre List of guarantors, text B
URL http://books.openedition.org/cefas/docannexe/image/907/img-1.jpg
Fichier image/jpeg, 887k

© Centre français de recherche de la péninsule Arabique, 2006

Conditions d’utilisation : http://www.openedition.org/6540

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search