An India Trader’s Partnership in Almería )1139(

∗ This research was supported by the Israel Science Foundation (grant No. 815/02). I benefitted from the use of the late Prof. S. D. Goitein’s “Geniza Laboratory,” the Friedberg Genizah Project (FGP) (http://tinyurl.com/oc2pvww) and the Princeton Geniza Project (http://tinyurl.com/pwfhfqs). Thanks are given to Dr. Ben Outhwaite, Head of the TaylorSchechter Genizah Research Unit, and to the Syndics, Cambridge University Library, for permission to publish the fragments presented in this paper. My former student Dr. Amir Ashur provided valuable assistance. Note the following abbreviations: b. = ben/ bar/ibn (‘son of’); Bodl. = The Bodleian Libraries, University of Oxford; CAJS = Center of Advanced Jewish Studies, University of Pennsylvania; CUL = Cambridge University Library; LG = Bodl./CUL Lewis-Gibson Collection (formerly Westminster College); TS = Taylor-Schechter Collection CUL. The article is based on a presentation given in Madrid as part of the Cartas de Sefarad: Vida y cultura judía en al-Andalus seminar series organized by Dr. María Ángeles Gallego and Dr. Esther-Miriam Wagner in 2012.

This article focuses on newly identified Geniza fragments that document contacts between an India trader, Ḥalfon, and a North African merchant, Ibn al-Naghira. The body of the article is devoted to an analysis and edition of a deed of partnership between the two men drawn up in Almeria, al-Andalus, which supplies pieces of information of interest for various research topics, ranging from the historical record of Almeria to the dating of Judah ha-Levi's Kuzari to Judeo-Arabic. To put the new discoveries in perspective, the following paragraphs introduce the India trade as known from the Geniza documents and review the already published material on the contacts between the two traders. The Introduction also shows how the newly identified fragments make it possible to clarify or emend aspects of earlier research.
nal phenomenon of momentous socio-economic significance. The late S. D. Goitein called attention to the importance of the Cairo Geniza documents, written from the late eleventh century through much of the twelfth century, for understanding this activity. These documents focus on Jewish international traders in Arabic speaking countries, but their activity was inclusive and representative, and the significance of these fragments extends far beyond those geographic and ethnic denominators. The participants in these ventures wrote the so-called 'India Book' letters, and the recent publication of a large portion of this corpus has brought to light significant aspects of this international undertaking. 1 Jewish traders from the Iberian Peninsula, especially from al-Andalus, were active participants in the India trade, and they are mentioned in the Geniza documents as being present in India and along the India route. Most of the Jewish India traders, however, resided in Egypt. Their wholesale import-export business required frequent travel between India and the Far East and North Africa and the Iberian Peninsula. During these trips, they personally purchased and sold goods and arranged for shipping commodities. They also wove trade networks, invested their capital, entered partnerships and engaged the services of agents from one end of the known world to the other.
The Egyptian-Jewish trader and world traveler Ḥalfon b. Nethanel was one of the outstanding participants in the India trade. Geniza fragments that originated in his personal archive document his travels between Egypt, Yemen, India, North Africa, al-Andalus and Syria during the third to fifth decades of the twelfth century. al-Andalus and in other lands he visited and purchased local commodities there for sale in the Eastern Mediterranean and beyond. Ḥalfon was a merchant scholar, and his prolonged stay in the Iberian Peninsula in 1138-1139 was of significance because of not only the commercial activity, which his papers document, but also because of the close relationship that he developed there with the poet laureate Judah ha-Levi and other intellectual luminaries. 3 Ḥalfon's successes in international trade and his resultant extended sojourn in al-Andalus were enabled largely by his contacts with other merchants and the arrangements between them, which he formalized by legal contracts. In this paper, we are concerned with his contacts and contracts with a North African trader named Yûsuf b. Šuayb -a.k.a. Abû Jacob 4 Joseph b. Saul -Ibn al-Naghira. 5 On one of the days between January 5-14 (the last ten days in the month of Tevet), 1138, Ḥalfon and Ibn al-Naghira entered a partnership in Fez, Morocco. Like the other documents discussed here, the deed of partnership (H26) was written in Judeo-Arabic (Middle Arabic in Hebrew characters). 6 This is the only document that mentions the two men in which the entire date formula is intact. Ḥalfon kept the deed in his personal archive, and it eventually found its way to the Geniza chamber in Fustat (Old Cairo), where it was preserved until removed by Solomon Schechter in the late nineteenth century and relocated to the Cambridge University Library.
Ḥalfon invested in the partnership sixty Murâbiṭî mithqâls (gold coins, dinars, each weighing 4.25 grams, minted by the Almoravids), with which Ibn al-Naghira was to purchase goods and conduct transactions on Halfon's behalf, and Ibn al-Naghira added to the principle forty Murâbiṭî mithqâls from his own funds. According to the terms of the agreement Ḥalfon had exclusive rights to determine the duration of the partnership and Ibn al-Naghira was obligated to come, as instructed, to meet him in any locality (in the Maghreb, that is North Africa or al-Andalus, but the word is a restoration in the fragmentary text).
The 1138 partnership and the other contracts discussed below, including the newly identified partnership deed from Almeria, are of interest for the history of Jewish contract law, especially as concerns how they avoid the prohibition of usury. I have discussed these matters elsewhere and do not deal with them here. 7 Ibn al-Naghira appears in letters from Ḥalfon's Andalusian correspondence subsequent to the 1138 contract from Fez. Ḥalfon seems to have been residing in Lucena (in the Geniza papers: al-Yusâna) when Isaac Ibn Bârûk, his business agent in Almeria, wrote to him on July 10, 1138. That letter (H30) deals with funds to be sent to Judah ha-Levi and was written in response to Ḥalfon's letter that Ibn al-Naghira had delivered to Ibn Bârûk. The agent noted in a postscript that he gave Abû Jacob [Ibn al-Naghira] twenty and a half mithqâls, as per Ḥalfon's instructions. 8 In a subsequent letter (H32) to Ḥalfon, probably from the end of August that year, Ibn Bârûk again mentioned this payment and summarized other dealings by Ibn al-Naghira in Almeria. When his anticipated sale of a slave fell through, he was short of funds to pay a silk dyer and asked for an advance of ten mithqâls from Ḥalfon's account. Ibn Bârûk claimed that he had no instructions for such a disbursement of funds. Ibn al-Naghira countered that had Ḥalfon been there, he would have advanced him as much money as he needed and that he would repay the debt immediately upon returning from his business trip. Ibn Bârûk eventually relented and gave Ibn al-Naghira the ten mithqâls. Two additional previously published items from Ḥalfon's archive presumably relate to Ibn al-Naghira. His association with the first (H33), a fragment of a letter that mentions no names, is rather speculative. The hand and language are distinctively Andalusian, and the fragment has been preserved together with other Andalusian correspondence of Ḥalfon. The writer spoke of dealings in Almeria and of his attempt to collect ten mithqâls and requested an advance from the recipient. I have suggested that the recipient was Ḥalfon, the writer Ibn al-Naghira and the ten mithqâls requested as an advance in Almeria the same sum that Ibn Bârûk mentioned in his letter. 10 One of the newly identified fragments discussed below necessitates a reassessment of the suggestion that the writer of H33 was Ibn al-Naghira.
The second item (H29) is a fragmentary partnership deed between Ḥalfon, the investor, and one Joseph/Yûsuf, the active partner. The fragment does not preserve Joseph/Yûsuf's father's name and family name or the sum of Ḥalfon's investment. The partnership was for a limited period, until Passover 1139, and Joseph/Yûsuf undertook to pay a fine of twenty mithqâls to be given "to the poor of Almeria and for [ransoming] the captives," in case of non-fulfillment of terms of the contract. The documents in Ḥalfon's archive contain important information on Jewish captives in the Iberian Peninsula and on the efforts of the Jewish community to ransom them. Judah ha-Levi played a pivotal role in the ransom effort, and Ḥalfon assisted him.

11
Because of the differences between the clauses in the partnership deed issued in Fez (H26) and the fragment of the contract with Joseph/Yûsuf (H29), Gil and Fleischer rejected Goitein's identification of the latter with Ibn al-Naghira. Contrariwise, I have assumed that these differences are not significant and that the fragment probably represents a second partnership between the two men, contracted in al-Andalus, perhaps in Lucena where Ibn al-Naghira traveled to meet with Ḥalfon, after termination of the first partnership. 12 The second of the newly identified fragments studied below supports that general assumption (but not the specific location in Lucena).

NEw DISCOVERIES FROM ḤALFON'S ARCHIVE
(A) Despite the publication of most of Ḥalfon's known archive, this research is still in progress, and recently new relevant items have been identified. Two fragments brought to my attention by Dr. Amir Ashur shed further light on the Ibn al-Naghira-Ḥalfon connection. The first is an extremely damaged and effaced fragment of a letter from the former to the latter. Ibn al-Naghira mentioned a meeting with Abû Ibrâhîm, 13 prices of Oriental commodities, mithqâls from Tlemcen, Algeria, someone who did not listen to him, Almeria and silk.
14 It is tempting to assume that it refers to the same request for an advance mentioned in Ibn Bârûḵ's letter (H32). Nevertheless, the handwriting of this newly identified fragment is clearly different from the handwriting of the anonymous fragmentary letter (H33) that I had assumed Ibn al-Naghira had written concerning the same advance. Accordingly, we must conclude that he was not the author of that fragment -unless a scribe wrote one of the two items for him.
(B) The identification of the second item, a fragmentary deed of partnership between Ḥalfon and Ibn al-Naghira written in Almeria (line 23), was the catalyst for writing this study. This document is of interest for several reasons, and its analysis and edition comprise the remainder of the article. Several Geniza letters were written in Almeria or contain information about that city. Most of them belong to Ḥalfon's correspondence and were published in Goitein & Friedman, India Book IVB: Ḥalfon. To the best of my knowledge, this contract is the only legal document explicitly written there that has been identified in the Geniza papers.
The Geniza documents emanating from or otherwise concerning Almeria moreover provide unique historical data for the research of that important port city. I called the newly identified contract to the attention of Prof. Jorge Lirola of the University of Almeria, and he has confirmed that other than the Geniza fragments there are no known documents from Almeria from the Almoravid period whatsoever and that the earliest 13 Presumably this is Abû Ibrâhîm Iṣḥâq b. Šabbat who wrote H51 (TS 8 J 18.6) to Ḥalfon and is mentioned elsewhere in his correspondence; see FRIEDMAN, India Book IV-A, pp. 207-208 and 403 (index).
14 TS AS 149.119. I hope to publish elsewhere this and other recently identified fragments that belong to Ḥalfon's archive. known legal documentation is from the late fifteenth century. 15 Furthermore, it now seems likely that also the fragmentary contract of partnership that was to last until Passover 1139 (H29) was written in Almeria and not in Lucena, as I had previously assumed. This is the most logical conclusion from the assignment of the fine in case of non-fulfillment "to the poor of Almeria and for [ransoming] the captives." Two pieces of the newly identified contract have been located at present. The larger fragment, which Ashur called to my attention, preserves the full length and most of the right side of the contract. I have matched this fragment with a smaller piece, which comprises most of the lower half of the left side of the original. 16 The contract is dated in the month of Sivan 489[X] (AM). Since there was clearly room for a word conveying the unit after 'nine [ty],' in theory this could be any year between 4891 and 4899, corresponding to 1131-1139 C.E.
Because of his other travels, it would have been extremely difficult or impossible for Ḥalfon to be present in Almeria during most of that period. Furthermore, a number of factors in the court verification and confirmation that were appended to the contract, detailed in the following discussion, provide evidence for the conclusion that the sequence of the three partnership deeds with Ibn al-Naghira was as follows: H26 -Fez, January, 1138; H29 -binding until Passover (beginning March 18) 1139; and this newly identified fragmentary contract, whose date must have been 4899, when the month of Sivan began May 1, 1139. The two partners presumably met in the port city of Almeria after Passover to renew their partnership. As also attested by the verification and confirmation appended to the contract and discussed below, the meeting was undoubtedly occasioned by Ḥalfon's impending departure from al-Andalus for his return trip to Egypt, already known to have taken place in the spring or early summer, 1139.
This conclusion is of significance not only for determining the sequence of partnerships between the two merchants but for a better under-15 A private communication dated December 11, 2013. 16 The two fragments do not appear together in the 'joins suggestions' function in the FGP site. At my request, Dr. Ashur added the join in the 'jigsaw puzzle' function of FGP. As of the time of this writing the image of the join is accompanied on the site by the erroneous comment that "This Join Was Created Automaticaly [!] By The Jigsaw Puzzle Pro." standing of Ḥalfon's biography and the chronology of his extended stay in al-Andalus. This, in turn, has wider implications, especially for the study of Judah ha-Levi's life and works. Ḥalfon's archive is an important source for research on ha-Levi. In one of his letters to Ḥalfon, ha-Levi wrote of having composed his theological magnum opus, the Book of the Kuzari. From their study of that letter and other documents from Ḥalfon's archive Gil and Fleischer concluded that 1127-1129 was in all likelihood the period of the trader's only visit to al-Andalus, that he definitely had not been there during 1138-1139, and that ha-Levi wrote the Kuzari some ten years earlier.
17 I have already provided what I consider incontrovertible evidence that their reconstruction was in error. 18 The lacuna in the date formula of the newly identified partnership contract notwithstanding, the text thus provides additional weight to my earlier conclusion.
Contrariwise, the contract makes it possible to correct my earlier assumption concerning one of the letters from Ḥalfon's archive. I had suggested April 17, 1139 as the date of a letter (H57) that described his arrival in Alexandria from a trip to the West in mid Iyyar (without the year). That dating admittedly caused some tension with other documents when reconstructing Ḥalfon's whereabouts for 1139, and I already speculated that while less likely, the letter could have been associated with an earlier trip to the West. 19 That it refers to another trip follows from the conclusion that the newly identified contract was written in Almeria during the month beginning May 1, 1139.
The formula of the deed is worded as a deposition by the witnesses who quoted and attested Ibn al-Naghira's statement. He confirmed owing fied commodities (see line 10). 20 Unfortunately, the two fragments of the contract do not preserve details concerning the transactions to be undertaken and the division of proceeds. Ibn al-Naghira agreed that Ḥalfon's word concerning repayment of his investment was binding without dispute. 21 The document with addenda consists of thirty-nine lines which can be divided as follows: (1) the body of the legal deed, including both Ibn al-Naghira's statement and concluding legal formulae -twentyfive lines; (2) the signatures of seven witnesses -two lines; (3) verification of the witnesses' signatures by a court of three -five lines; (4) confirmation of this verification by a second court of three -seven lines.
The witnesses (2) clearly signed in Almeria, when the deed was drawn up. As the identity of the signatories to the two court certifications prove, the first certification (3) was executed in Alexandria and the second (4) in Fustat. The contract is written primarily in Judaeo-Arabic with many Aramaic and Hebrew formulas; the main language of the certifications is Aramaic.
The complex triple-tiered attestation -witnesses' signatures (2) and double court certifications (3-4) -enables us to follow the progression of the contract in the hands of Ḥalfon, the great traveling merchant, from Almeria at the western extremity of the Mediterranean to Alexandria at its eastern extremity and on to Fustat, terminus for both the Mediterranean and the Indian Ocean trade. Since each successive entry names the signatories in the preceding one, this procedure also makes it possible to restore some of the names preserved only partially in the original deed or in the respective court certifications. 20 Were it not for the fragmentary reference to purchases or sales in line 10 and the other known data on Ḥalfon and his association with Ibn al-Naghira, one might have thought that this is a simple loan contract. 21 Unlike the partnership contract issued in Fez (H26), where Ḥalfon invested sixty mithqâls and Ibn al-Naghira forty, here presumably Ḥalfon's investment was forty and Ibn al-Naghira's is not preserved. I think it improbable that this contract refers to similar 60:40 investments and an undertaking by Ibn al-Naghira to forfeit all of his capital if he failed to transact Ḥalfon's business as promised. Note that according to the second contract described above (H29) the fine for non-fulfillment was twenty mithqâls, but the amounts of the investment capital are not preserved. Only the discovery of the missing pieces will make it possible to ascertain these matters.
Court verification of the witnesses' signatures was intended to protect against forgeries. Double court certifications, one validating the signatures of the witnesses to the contract and the second confirming the signatures to the first court's verification, are unknown from extant legal formularies. The few examples preserved in the Geniza reflect the highly mobile character of that society. An instance described by Goitein involves another document issued in the Iberian Peninsula and presented in Egypt. The deed was executed and signed by witnesses in Denia about forty years before our partnership contract, its witnesses' signatures verified by a North African court in al-Mahdiyya, and the signatures to the verification confirmed by the court in Alexandria.

22
The procedure involved in the newly identified partnership deed between Ḥalfon and Ibn al-Naghira also reflects special circumstances. The arrangement between them was obviously a long-term affair. This contract must have been drawn up in Almeria shortly before Ḥalfon set sail for Egypt. Four of the seven witnesses (2) evidently planned to accompany him. Accordingly, all four signed the deed first (in line 26) and upon arrival in Alexandria appeared in person before the Jewish court there to authenticate their signatures (3). When he arrived in Alexandria, Ḥalfon was informed of the death of his older brother Eli. He fell into a deep state of depression and was unable to proceed to Fustat and complete his trip for almost a year. Presumably only then, in mid 1140, did he present the contract to the court there for its final confirmation (4).
Ḥârith ('Plowman') is a rare name in the Geniza papers. Abu l-Barakât b. Ḥârith was probably a descendent of Barakât b. Ḥârith who owned property in the Egyptian village Ṣahrajt a century earlier, as reported in a document from 1041, written in Fustat. 24 In none of the documents from Ḥalfon's archive (or other India Book papers) does Abu l-Barakât's Hebrew name appear. It was only logical for Goitein to assume that Abu l-Barakât ha-Levi b. Ḥârith, Elder of the Congregations, the India trader, was the same as Abraham ha-Levi b. Ḥârith, Elder of the Congregations, whose signature appears on a ketubba from Alexandria ca. 1143 and on at least one other document. 25 However, Joseph ha-Levi b. Ḥârith's signature on Ḥalfon's partnership deed in Almeria -where Abu l-Barakât b. Ḥârith was presentcompels us to reconsider Goitein's identification. A letter dated February-March 1140 (H68) shows that by then Abu l-Barakât was back in Egypt with Ḥalfon. 26 Consequently, Abu l-Barakât was almost certainly Joseph's byname rather than Abraham's, and these two men must have been brothers. needless to say, commonplace. Isaac b. Aaron the witness signed with a monumental hand. Comparisons of such styled signatures with the same man's signatures on letters are problematic. Isaac b. Aaron al-Sijilmâsî's signatures on his letters to Ḥalfon 28 bear little resemblance to the witness Isaac b. Aaron's signature, and on that basis, it is difficult to suggest a connection between the two. Nevertheless, since also this signatory traveled back to Egypt with Ḥalfon, he indeed may have been Sijilmâsî.
The fourth signatory, Ḥâvîv/Ḥabîb b. Abraham, is probably the same Shaykh Ḥabîb who appears repeatedly in Ḥalfon's Egyptian correspondence. 29 In his letter to Alexandria from February-March 1140 (H68), Ezekiel wrote to his brother Ḥalfon that Ḥabîb had described details of his experiences in al-Andalus in an audience -presumably with the newly appointed Egyptian Nagid Samuel b. Ḥananya in Cairo. I have already suggested that Ḥabîb traveled from al-Andalus to Egypt with Ḥalfon. 30 Our document substantiates this proposition, since it notes that Ḥâvîv/Ḥabîb b. Abraham appeared before the court in Alexandria to authenticate his signature on the contract from Almeria. Like Ḥalfon, Ḥâvîv/Ḥabîb was probably an Egyptian, who had traveled to al-Andalus on business. It is likely that his father was the Abraham b. Ḥâvîv/Ḥabîb who about ten years earlier wrote a letter from Alexandria to the India trader and representative of the merchants in Fustat Abû Zikrî Kohen. 31 The combination of names of son, father, and grandfather of the fifth signatory, Isaac b. Obadiah b. Isaac, is unique. He was undoubtedly the same merchant scholar Isaac b. Obadiah who appears repeatedly in 28  Michael 14 (1997) pp. 103-108. According to zELDES and FRENKEL, p. 106, n. 2, Ḥalfon's business associate Shaykh Ḥabîb was Spanish, but they did not explain the basis for this assumption.
Ḥalfon's Andalusian letters. 32 He did not appear before the court in Alexandria and presumably remained in al-Andalus. Isaac Ibn Ezra wrote and addressed a letter to Isaac b. Obadiah at Almeria, but it is not clear from all documents that mention him that he was a permanent resident there. 33 One of the remaining three signatures is only partially preserved, and the other two have not yet been identified.
The first signature on the Alexandrian court verification was of Aaron b. Yeŝûâ hâ-rôfê (the Doctor). He was the celebrated judge and poet Aaron Ibn al-Ammânî, who hosted Judah ha-Levi in Alexandria after his arrival there in September 1140. He was well acquainted with Ḥalfon and is mentioned repeatedly in his archive. 34 The second signatory, Isaac b. Joseph, signs as a member of the court of Alexandria after Ibn al-Ammânî also in a document written in January 1143. 35 The minute letters above and below his signature -only partially preserved here -clearly spell ‫ירושלמי‬ 'the Jerusalemite' in that document. The third signatory, Yeŝûâ b. Mevôrak, might be identical with his namesake, 'the Alexandrian residing in Malîj,' from whose hand a fragmentary letter has been preserved. 36 At the present, there is no other known evidence of his having served on a court in Alexandria or having signed as a witness there (see below).
The three judges of Fustat are well known, especially the first and the last. Both of these Nathans, Nathan ha-Kohen b. Solomon 37 and Nathan b. Samuel he-ḥâvêr, 38 were acquainted with Ḥalfon, and the second was a close associate. Both were active between the 1120s and 1150s. 32 See FRIEDMAN, India Book IV-A, p. 420 (index). 33 TS 12.280,ed. GIL and FLEISCHER, Their co-signatory, Yeŝûâ b. Josiah b. Ŝemayâhû Gaon, the scion of an illustrious family, was a perfumer (aṭṭâr) and brother of an India trader. He signed several documents as a witness or a member of the court with one or both of the judges Nathan that validated a contract. Only a few months separated the two known dated documents, which he signed, one is from Iyyar (April-May), the other from Elul (August-September), 1140. 39 In the absence of additional evidence, we can assume that he signed the confirmation of Ḥalfon's partnership contract during the same period more or less, which coincides with the time that Ḥalfon is known to have left Alexandria and returned to Fustat. Accordingly, the data on Yeŝûâ adds further credence to our restoration of the missing units in the date of the contract from Almeria as 4899, corresponding to 1139.
The three judges of Fustat recognized Ibn al-Ammânî's signature, and they validated it accordingly. There was some controversy concerning his title he-ḥâvêr, which as a matter of course he did not affix to his signature. 40 When referring to him, Nathan b. Samuel he-ḥâvêr, who penned the Fustat court confirmation, might have prefixed the title to his name, but the word (in line 35) is fragmentary and effaced and its decipherment and restoration suspect. Two witnesses who appeared before the judges in Fustat confirmed the signature of the second member of the Alexandrian court, Isaac b. Joseph. Confirmation of his signature and Ibn al-Ammânî's sufficed to authenticate the Alexandrian verification, and evidently the unfamiliar signature of the third member of the Alexandrian court remained unendorsed.  49 The symbols over the first three words are apparently only lines, not letters, though the first and third resemble ‫.ש‬ The dot over the resh is superfluous.

Translation
[Ibn al-Naghira's Declaration before Witnesses] (1) Testimony to what transpired in our presence, we the witnesses [whose signatures are below: Our master and lord Joseph b. Saul/Yûsuf b. Ŝuayb] (2) Ibn al-Naghira appeared before us and said: Serve as my witnesses and perform with [me] the symbolic act of commitment 51 [and write and sign that I bind myself with all terms] (3) that are precise and phrases that are absolute and all expressions of deeds attesting to rights 52 that this should be for Ḥalfon] (4) ha-Levi -may the Merciful protect him -b. his honor, the saintly outstanding sage 53 [Nethanel … so that this be in his hand a deed of] (5) proof and evidence, that I declare in your presence in the most decisive [and firm] terms [for declarations, in good health and full capacity, of my free will, under no compulsion] (6) nor force, nor duress, not mistaken or erring, nor with any illness [or anything else that disqualifies testimony] (7-8) and without coercion, that I owe and am responsible for and obligated to pay, from my unencumbered assets, to our lord  (15) not an ancillary [oath] nor a general ban on anything [he says … And any claim against him will be null] (16) and void, completely invalid and considered nil […] (17) has no validity and will not be acted upon […]