Introduction

Jews and Christians of the Eastern Mediterranean and West Asia were involved in the Indian Ocean trade networks since at least the ninth century ce, as attested by at least one famous Old Malayalam inscription, namely the Kollam copper plates, featuring Muslim, Zoroastrian, Christian, and 202 Jewish signatories in Kufic, Pahlavi, and Judeo-Persian.¹ However, and unlike the trade networks operating along the land routes connecting West and East Asia (which are more commonly known as the Silk Road), historical sources after the rise of Islam and before the 1500s witness mainly Arabic-speaking Muslims crossing the Arabian Sea to South and Southeast Asia (Foltz 2010, 13; Wink 1996, 65). It was only the discovery of the Cairo Geniza in the late nineteenth century that added a significant body of sources related to Indian Ocean trade and written by non-Muslim Arabs. As the relevant documents are almost exclusively in Judeo-Arabic, they feature mainly Arabic-speaking Jews and their maritime trade activities. We still lack evidence directly attesting to the premodern history of indigenous Jews and Christians of the period, or evidence of West Asian Christians engaged in Indian Ocean trade after the ninth century and before the sixteenth century. Though there are scattered references to Jews and Christians involved in maritime transregional networks across the Arabian Sea and eastwards, evidence for the extent of their involvement is rather circumstantial. Except for one person, who can be identified as a Christian by his name—ʿAbd al-Massiḥ al-Šammas (“The Deacon”)²—there is no explicit mention of Christians in the Indian Ocean Geniza documents. It stands to reason that Christians did not completely cease their connection with maritime trade eastwards, but the lack of references to Christians in Indian Ocean trade is remarkable even when compared with references to Jewish traders in Muslim sources.³ It is only towards the decline of the 203 Arab trade networks during the early modern period and the rise of the European companies across the Indian Ocean Rim that Jewish and Christian communities were “discovered” along the Malabar Coast, a strategic coastline connecting Southwest and Southeast Asia since Greco-Roman times (Gurukkal 2015).

The emergence of Jewish and Christian communities in the region must have been an outcome of intermarriages between merchants and local women during the heydays of premodern Indian Ocean trade. Kerala Jews vehemently deny intermarriages with non-Jewish women as the source of their origin, which is associated with inferior status and used in intra-communal conflicts as the basis for supremacy claims (Segal 1993, 19; Segal 1983; Schorsch 2008). Christians, too, trace their origins elsewhere, as in conversions of upper-caste Brahmins rather than associating their ancestry with merchants marrying local women (Bayly 1984, 178–9, 184 and 184n13). These approaches stand in sharp contrast to Muslims, who explicitly institutionalized intermarriages with local women by the system of temporary marriage (mutʿa) for the purpose of basing their trade connections across regions (Wink 1996, 71–2; Randathani 2006, 15; Alpers 2014, 58).

The Geniza documents attest conjugal relationships between Jewish merchants and non-Jewish women during their business excursions to the Malabar Coast, though the evidence is rather casual and scarce (Friedman 2010, 171–3). This type of concubinage alliances between merchants and non-Jews is well-known across the Jewish world, as it kept feeding Halakhic debates regarding the legality of such relationships and the religious status of the concubines and their children (Assaf 1965, 230–1). The debates concerned Malabar as well, for in the early sixteenth century, a reponsum 204 from Cochin reports of an inner split in the Jewish community there, where a small group of “pure” Jews accused the majority of Jews in the town of being descendants of intermarriage between local slave girls and Jewish merchants from Turkey, Persia, and Yemen (Qastro 1783, responsum 99; Segal 1993, 24–5; Gamliel 2018, 59). The same responsum also contextualizes the accusations in “jealousy and hatred”, for the “accused” Jews “are learned in the Torah, rich and close to the royal house and the government. They are also the main negotiators for merchants” (Qastro, 1783, Responsum 99).⁴ While surely concubinage with domestic maidservants in overseas market towns was common, there is a rare piece of evidence in the much earlier Geniza suggesting that the closeness to the royal house reported in the sixteenth-century responsum could have been based on conjugal alliances between the West Asian merchants and women of relatively high socioeconomic status. Arguably, evidence in support of this possibility is found in a document attesting the conjugal alliance between a Jewish merchant and an indigenous woman of Malabar in 1132. While the document defines the woman as a slave girl, another document casually refers to the merchant’s brother-in-law in Malabar as Nāyar (נאיר, nʾyr) and as his business associate. The evidence in both documents is thus contradictory, calling for explanation to resolve the discrepancies.

It is important to note right at the outset that the evidence discussed below is not only rare but also too fragmentary and too slim for a comprehensive historical analysis. Nevertheless, it is substantial enough to offer a new perspective for the study of the social history of West Asians 205 in medieval Malabar. Most importantly, the twelfth-century evidence of a Jewish merchant cohabitating a South Indian woman is intricately related to several fields of historical research from South Indian society to Mediterranean society to Jewish and Dravidian inheritance customs and to the history of conversions, slavery, intermarriage, and trade guilds in both South and West Asia. It is difficult, even impossible, to do justice to the various related fields in one single study on the narrow and somewhat arbitrary evidence for a twelfth-century mixed couple. However, the present discussion is the first to incorporate the documentary Geniza in the social history of premodern Jews in Kerala. In closely examining the evidence and its historical context, I aim at indicating the implications of this evidence first and foremost on the history of Jewish networks in the region and, by extension, on the interrelated history of Jews and Christians in the region.

Abraham Ben Yijū and Aśu: The Documents

The evidence at hand is drawn from the Geniza “India trade” letters, dated between the eleventh and the thirteenth centuries. Though mostly dealing with issues related to trade, the documents include occasional references to personal matters, albeit sparingly so. Such references enabled scholars to reconstruct, at least partially, the biographies of several prominent Jewish merchants (Goitein and Friedman 2008, 2009, 2010a, 2010b, 2013). Perhaps the most detailed biography reconstructed based on the documents is that of Abraham Ben Yijū, a Tunisian Jewish merchant who stayed in South India for nearly eighteen years (ca. 1132–1149). Shelomo Dov Goitein was the first scholar to reconstruct the family history of Ben Yijū, which was later followed up by the novelist Amitav Gosh in a study of Ben Yijū’s slave Bama 206 (Goitein 1973; Ghosh 2002, 168–244). The most complete biography of Ben Yijū, based on Goitein’s textual analysis, is published posthumously in the monumental “India Book” completed and supplemented by Mordechai Akiva Friedman (Goitein and Friedman 1999; 2008, 52–89; 2010b). We thus have a considerable amount of information about Ben Yijū; he was born in al-Mahdīya in Tunisia, and his family members—many of them merchants as well—were scattered throughout Tunisia, Sicily, Fustat, Aden, and India. As already mentioned, he settled in South India and established a flourishing business encompassing India, Aden, and Egypt.

The document at the center of the current investigation was found among business letters and accounts written to or by Abraham Ben Yijū. It is a deed of manumission, dated 17/10/1132, attesting the purchase of a slave girl by the name Aśu, her conversion to Judaism, and her manumission by Ben Yijū. Goitein and Friedman assume that the purchase, conversion, and manumission were preplanned by Ben Yijū in order to marry Aśu (2010b, 6–7). Indeed, such practices were common among long-distance trading Jewish merchants, as attested by legal documents of the period (Friedman 1986, 292–6; 2010, 170–1). Ben Yijū, in a move that seems atypical of a Jewish trader in India, decided to return to Aden with his son, Surūr, and daughter, Sitt al-Dār, born to him and to Aśu the convert in India. There is no evidence to the whereabouts of Aśu at the time of return; Aśu might have died prior to Ben Yijū’s return to Aden, or she might have simply stayed behind. Besides the deed of manumission, there is only one other laconic and oblique reference to Ben Yijū’s wife, not in Mangalore but rather in Jurfatan (גרבתן, grbtn), further to the south: “I was told that your wife and 207 children are in Jurfatan, therefore I sent [letters to you] with the above mentioned [merchant].”⁵

Leaving aside the questions regarding Aśu’s background, it is noteworthy that in Aden, Ben Yijū was confronted with allegations that his children were not Jews according to the Halakha, implying that they were not entitled to inherit his enormous wealth. Ben Yijū fiercely resisted these allegations, presumably already before his final departure from Malabar. Evidence for this controversy is found in legal correspondences (responsa) that Ben Yijū wrote regarding the Halakhic status of children born to a Jewish father and a manumitted convert wife. Ben Yijū obliquely refers to himself via the generic character Rәʾuḇen, arguing that Jewish law recognizes children born to a convert wife who had conceived before being manumitted. However, the Halakha forbids intercourse with a slave girl or a non-Jew and marrying her in hindsight. It does, however, rule that in case such a marriage did occur, the husband is not obliged to divorce his wife.⁶ Nevertheless, a son born to a slave girl or a gentile woman is not entitled to become his father’s heir.⁷ More evidence is found in a poem composed by Ben Yijū in honor of his business associate Maḍmūn Ben Ḥasan-Yefet, the head of the Jewish community in Yemen and an influential merchant in his 208 own right, representative of the Jewish merchants (nagid) in Aden. Goitein and Friedman speculate that one of the verses in the poem alludes to the protection provided by Maḍmūn to Ben Yijū in aborting the allegations against him. The responsa documents and the poem are dated to 1140–41, several years before Ben Yijū departs from Malabar with his son and daughter (Goitein and Friedman 2008, 37–47, 73–6; 2010b, 21–4). It is against the backdrop of this controversy that the aforementioned deed of manumission is considered evidence for Aśu being the mother of Ben Yijū’s daughter Sitt al-Dār and son Surūr.

Reading the deed of manumission along with other documents related to Ben Yijū’s life and business in India raises a few questions regarding the socioeconomic status of Aśu in her South Indian social context. There is reason to assume that Aśu was a member of a family of landowners, which makes it highly improbable that she would have been sold as a maidservant. If this assumption can be substantiated, then her conjugal alliance with Ben Yijū can be understood as instrumental in establishing a trade network based on kinship relations and extending over important port towns across the Arabian Sea, especially along the sea route connecting North Malabar with Aden. In what follows, I examine the related documents separately and in juxtaposition to each other and to the historical background of both Jewish and Dravidian societies. 209

The Deed of Manumission Revisited

The most informative document related to Aśu is the deed of manumission dated 17/10/1132, a date closely following Ben Yijū’s arrival in India.⁸ The document plainly states that Aśu is a Tuḷuva (תלויא, tlwyʾ) woman, that is to say, a native of Tulunad, and that she is a convert slave girl.⁹ Tulunad is the coastal region in South Karnataka, including the Kasargod district in nowadays north Kerala, with the Arabian Sea to the west and the Western Ghats to the east. The region is named after its dominant and historical language Tulu and it is part of the region designated as Malabar by Arabic speakers since approximately the ninth century. Tulunad shares with historical Kerala certain geographical features, besides being its immediate neighbor to the north; it, too, is delineated by the Western Ghats to the east and the Arabian Sea to the West. This geographical proximity has shaped a distinctive contact zone along the coast, where maritime communities flourished (Vasanthamadhava 1996; Mailaparambil 2011, 11–16). This shared historical, linguistic, and cultural area fostered transregional connections that are important to consider in the context of premodern Jewish networks, such as the one in which Abraham Ben Yijū operated. Arguably, his conjugal relations with Aśu were instrumental in establishing his business network. 210

Tulu and Malayalam converge into regional dialects at the border lines between South Karnataka and Kerala.¹⁰ While the history of political entities such as royal dynasties is distinctively different, the regions encompassing South Karnataka, or Tulunad, and North Kerala, or Malabar are interconnected and integrated. Along with several other languages such as Kannada and Urdu, Tulu is spoken in the Malayalam-speaking district of Kasargod, the northernmost district of modern Kerala, where some of the most frequented entrepôts of Ben Yijū’s times, like Cannannore (Jurfattan?), Dharmmapaṭṭaṇam (Dahfattan), and Vaḷarpaṭṭaṇam (Budfattan), can be identified as medieval port cities known from South Indian history. Ben Yijū used to travel to these port cities for business when he was living in the region. The local network of this Tunisian Jewish merchant extended along the coastline from Mangalore in the north to Dharmmapaṭṭaṇam in the south.¹¹ It is quite certain that the people with whom the Geniza Arab-speaking merchants came into contact were reasonably versed in both Tulu and Malayalam, for the Geniza documents contain loanwords from both languages (Lambourn 2015). Aśu must have felt at home down south, in the Malayalam-speaking region, or else Maḍmūn Ben Ḥasan Yefet would not have sent letters to Jurfattan, having heard that Ben Yijū’s “wife and children” resided there (see fn 5 above). Aśu, therefore, had the status of a woman free to move between different towns regardless of her husband’s whereabouts (she could have taken his business letters for him in Jurfattan while he was away). This means that she must have had her own network in a port town other than Mangalore, the port town stated at the outset of 211 the deed of manumission. Without her husband, she must have relied on kinship connections to reside with her children and even receive letters on his behalf. Moreover, spatial mobility beyond the boundaries of the village and across regions is typical of upper-class people (Miller 1954, 410, 416).

There are a few hints in Aśu’s deed of manumission that her status in her society was far from that of a slave. According to Goitein and Friedman (2010b, 165n26), the document diverges from typical deeds of manumission from the Geniza, as when stating that Ben Yijū bought Aśu from her mistress for a significant sum of money: “you, whom I bought for the best of my silver from your mistress home.”¹² They also speculate that the deed of manumission was likely intended to certify marriage with the manumitted slave girl, based on several known cases of Jewish India merchants who engaged in conjugal relations with non-Jewish women that were more often than not defined as slave girls or manumitted slave girls (6–7, and note 15). Though Aśu’s deed of manumission is phrased in the conventional manner of divorce certificates, it also contains an allusion to formulaic expressions typical of marriage certificates (kәtuba), according to Goitein and Friedman (164; 166n30). Lastly, the document, untypically of deeds of manumission, explicitly states the Hebrew name given to the newly converted slave girl (7; 166n32).¹³ These remarkable features of the deed of manumission suggest a divergence, at least to a certain extent, from the common practice of medieval Jewish traders to have non-Jewish concubines defined in the Geniza documents as slave girls or manumitted 212 slave girls (Friedman 1986, 291–339). Moreover, one is obliged to wonder why Aśu was left behind in India while Ben Yijū left with his son and daughter to Aden. There is no evidence that indeed Aśu was left behind; she might have died prior to departure or even joined Ben Yijū without leaving traces in the records. Nevertheless, it is quite possible that Ben Yijū followed the common practice of Jewish and Muslim traders to leave their concubines-cum-slave girls behind upon embarking on their journey back home (Friedman 1990, 99–104). What is less common, though, is the fact that Ben Yijū returned with a son and a daughter born in India and was faced with fierce resistance to have them accepted as Jews and as legal heirs of his property. Thus, the peculiar textual features of the deed of manumission as well as the unique biography of Ben Yijū call for a reevaluation of the conclusion that the deed of manumission represents a maneuver typical of other Jewish traders, namely that Ben Yijū bought a slave girl merely to serve him as a concubine or a wife during his prolonged exile (Goitein and Friedman 1999, 263–4; Friedman 1986, 292–4).

It is therefore possible that Ben Yijū’s alliance with Aśu was motivated by concerns exceeding personal needs for domestic help or conjugal delight. Furthermore, it is possible that the alliance was not preceded by the formal agreement attested in the aforementioned document, the deed of manumission, as would have been expected in the context of Eastern Mediterranean arrangements for concubinage (Frenkel 2011, 255–6). Arguably, the alliance with Aśu was primarily intended to establish a kinship-based network through intermarrying into a local family of business partners, as can be gleaned by the reference to Ben Yijū’s in-law discussed below. If this is the case, it is difficult to imagine that a conjugal alliance with a slave girl could have been instrumental in establishing connections with business partners in Malabar. Besides these textual hints, there are also contextual reasons to suspect that the term ‘slave girl’ in Aśu’s deed 213 of manumission is merely a formal term for intra-Jewish legal purposes. Such a claim requires a close reading of the Geniza deed of manumission in juxtaposition to the medieval history of slavery, concubinage, intermarriage, and transregional trade networks in medieval South Asia for viewing Aśu’s relationship with Ben Yijū from the perspective of medieval Malabar society.

Firstly, it is doubtful whether the social history of slave girls and conversions in the early medieval Eastern Mediterranean is applicable to contemporaneous South Indian society. For one, there is hardly any evidence, as far as I am aware of, of institutionalized slave markets, where wealthy people could randomly select male or female slaves for domestic service, as those described in Jewish sources relating to the Eastern Mediterranean (Assaf 1965, 223–41; Goitein 2000, 130–47; Frenkel 2011; Perry 2014). While various forms of human enslavement existed, the concept of and terms for slavery, servitude, and serfdom in Dravidian society cannot be indiscriminately equated with terms used in the Mediterranean context, let alone Jewish society. Thus, a well-known form of slavery in medieval South India is that of agrarian serfdom aṭimai, a term occurring in Old Malayalam and Tamil inscriptions. Agrarian serfs were bound to the land of their masters. Agrarian serfdom (aṭimai) in South Indian society was relegated to land-tilling castes like pulayar and paṟayar, who were situated already at that period on the lowest grade of the “purity-pollution” standard typical of the region. Serfs of these communities were indeed recruited during the early modern period for church servitude and consequently converted to Christianity.¹⁴ If Aśu’s status in Dravidian society 214 was that of a slave, she would have belonged to a community of agrarian serfs or landless people who were attached to a certain territory owned by an upper-class household. Such an assumption makes little sense in the case of a foreign trader who states that he bought a single slave girl from a certain mistress, as aṭimai was provided on a communal basis and along with a grant (or purchase) of land; a landowner would own an extended family of serfs cultivating his lands and seen as an integral and undivided group.¹⁵ None of the documents related to Ben Yijū conveys any hint that Ben Yijū was granted with land ownership soon after arrival in Malabar.

Another possibility is to assign Aśu the status of concubine, which is a well-documented institutionalized form of court and temple female servants, or courtesans and devadāsis, “God’s maidservants”, respectively. However, courtesans and devadāsis were attached to a temple or a royal court (Ali 2006, 45–6; Orr 2001, 211–5). Moreover, due to the rules of purity and pollution that were undoubtedly at play during the time, domestic servitude could not have been performed by women of the so-called slave castes. Under these circumstances, the words of Ben Yijū in the deed of manumission, addressing Aśu as a slave girl (šifḥa) purchased “from the house of your mistress”, suggest that Aśu was of a relatively high status. It is doubtful, therefore, that she would have been designated a “slave” in a sense similar to that of the term in contemporaneous Mediterranean society.

In any case, the purchase of a woman by a foreign merchant from an agrarian community or from a temple is at odds with what is known about the economy and society of the region at that period. There are no documents related to the sale or purchase of slaves in South India 215 during the early medieval period, though that in itself does not rule out the possibility that slave trade was conducted at the time in the region as much as it was in the Mediterranean. Studies on slavery in India reveal a picture different from that emerging from the Mediterranean, where slaves were commodified in institutionalized forms such as slave markets (Goitein 1967, 130–47; Friedman 1986, 291–339; Frenkel 2011; Perry 2014). As already stated, slavery in South India is closely related to agrestic serfdom, on the one hand, and to captives of war, on the other (Ali 2006, 44–6; Subbarayalu 2012, 156–9). While there is evidence for Indian slaves being sold in Western and Central Asian markets (Perry 2014, 39), it is unclear where in India they were purchased and under which “market conditions” precisely. There is no evidence in the Geniza documents, or in medieval South Indian sources, that slave markets operated in the West Coast of South India, where Aśu’s purchase (if indeed it was a purchase) is recorded. Though maidservants, such as the peṇṭātti in the Coḻa court (Ali 2006, 50), for instance, are recorded in historical documents, their status is of domestic personnel rather than a product for sale. It is possible, despite the lack of recorded evidence, that peṇṭāttis were acquired by purchase; nevertheless, such female slaves were dissociated, according to existing records, from any natal or conjugal kinship (56).

It may of course be the case that Aśu’s deed of manumission bears testimony to slave markets operating in twelfth-century Malabar, despite the lack of evidence of such markets in premodern South India. However, in light of the documented history of the economy of human labor and the social status of women in Dravidian society, Aśu’s deed of manumission can be “translated”, so to speak, to the social reality as depicted in relevant studies. Firstly, the “transaction” can be reevaluated based on Ben Yijū’s explicit statement that it is from the house of her mistress that he “bought” 216 Aśu.¹⁶ In twelfth-century Malabar, that meant that Aśu must have belonged to a matrilineal household managed and owned by a woman, whether as a domestic servant or as a family member.¹⁷ During the same time period in the Eastern Mediterranean, the meaning would be that the domestic servant of one lady was transferred to another household. Secondly, an implication of moving from a certain house to Ben Yijū’s ownership could be understood as demarcating the conjugal alliance with Aśu from the Muslim establishment of mutʿa, or temporary marriage, usually with women of fishermen settled along the coast (Wink 1996, 71; 1997, 268; Goitein and Friedman 2010b, 163). Clearly, Ben Yijū intended a long-term relationship with the convert “slave girl”, at least as expressed in the deed of manumission. This intention of his would be in line with the much stricter approach to concubinage and legitimate conjugal alliance with non-Jewish women in Jewish law that is more flexible when the non-Jewish woman is a manumitted slave.¹⁸

The assumption that Aśu was indeed precisely what one would expect of women in the status of slavery led Goitein to read and interpret the name Aśu as derived from the Sanskrit adverb aśu, “quick”, based on a similar name, Ḥidhq (حذق), “dexterity”, given to a Nubian slave girl in the Mediterranean. However, this analogy with Aśu seems awkward, as the Sanskrit adverb aśu, “quickly”, is unlikely to be given as a name for a lady, regardless of her socioeconomic status (Goitein and Friedman 1999, 263 and 263n17; 2008, 55–6). It is more likely that Aśu is an abbreviation of a 217 longer name or, alternately, a “pet name”, typically ending in the vowel /u/.¹⁹ It may also be the case that the letter /ś/ stands for a Dravidian /c/, as sibilants are foreign to the phonemic system of Dravidian languages, and the unvoiced palatal is unrealized in the Hebrew and Arabic scripts. These are, of course, speculations, but they better fit with the reality of names in medieval Malabar than the etymology based on analogy with an utterly different language and culture.

That said, the Geniza documents do provide scanty evidence to justify the assumption that foreign traders would buy slaves in India. The more common practice prevalent in Indian Ocean trade, however, would have been to export slaves to India along the maritime trade routes rather than the other way around (Pouwels 2002). Another piece of evidence is found in documents reporting the sale of Indian slaves in Mediterranean markets. Indian slaves are a minority in this context, and they are not specified for the exact region in India from which they were brought (Goitein 1967, 133, 138; cf. Perry 2014, 39). There is one case attested, apart from Ben Yijū, in which a Jewish trader marries an Indian slave girl, and in that case as well the term slave girl might have been used for legitimizing cohabitation with a local woman (Friedman 1986, 294–6; 2010, 170–2). Naturally, we cannot completely rule out the possibility that Ben Yijū indeed bought a slave girl solely on the basis of lack of evidence for slave markets in twelfth-century Malabar. Nevertheless, the peculiarities of the document pointed out by Goitein and Friedman, the trouble that Ben Yijū went through to legitimize the union with Aśu, and the facts known about slaves and their status 218 in the region at that period altogether warn against taking the deed of manumission at its face value.

Whether Aśu was a domestic servant, a “gift” to a foreign merchant, or a member of a matrilineal household, the “transaction” cannot be taken too lightly as a matter of convenience for the foreign trader; there must have been an additional interest, other than financial profit, on part of the mistress’s household to engage in the transaction. Before turning to more evidence and, consequently, to speculations on the way in which Aśu and her mistress considered the transaction, it is, perhaps, more crucial to understand why Ben Yijū attributed the degrading status of a slave girl to his wife and mother of children even if she was not perceived as such by her own people. One obvious conclusion is that the attribute “slave girl” (šifḥa, שפחה) must have been ascribed by Ben Yijū in the legal deed for validating her conversion rather than for reflecting a reality of women sold as slaves to foreign merchants in a local labor market. Indeed, the period witnessed ample Halakhic discussions and queries on the topic, and, as discussed above, Ben Yijū, too, engaged in responsa on the subject, probably to validate his conjugal relations with Aśu in hindsight. Jewish householders, especially merchants travelling on business overseas, were inclined to enter relationships with slave-concubines, despite the negative light in which such relationships were viewed by Halakhic authorities (Friedman 1985, 11; 1986, 291–339; 2010).

As already noted above, the Muslim legal system allowing for ad-hoc intermarriage (mutʿa) in Malabar did not apply to the legal situation of Jewish merchants. For one, Jewish males were prohibited from having sexual intercourse with female slaves, be they Jewish or not. Prohibition aside, such practices persisted among Jewish men and were even tolerated to a certain extent under the influence of concubinage in Islam that was legitimized as mutʿa marriage (Friedman 1990; Perry 2017, 148). But a 219 Jew wishing to marry a convert would face another obstacle, which was to prove that the conversion was not forced or opportunistic. For that reason, a rabbinic court of law consisting of three witnesses (adult male Jews) was required for passing judgment regarding the validity of the conversion (Segal 2014, 595).²⁰ It is highly unlikely that there were enough Jews (if any at all) in Mangalore in October 1132 to witness and approve the conversion of Aśu. Under these circumstances, the conversion of a woman to Judaism in a distant land, with no Jewish witnesses available for supervising her conversion, could be legitimized only in case she was a slave girl (Goitein and Friedman 1999, 266). In other words, to marry a convert, the convert must have been a freed person prior to the conversion, which, in turn, must have taken place prior to the intention to marry. On the other hand, a Jewish householder had to convert his male and female slaves in order to employ them in his house. In such cases, the process was much simplified, involving immersion in water and teaching of basic Jewish law by the slave owner. It seems, therefore, that Ben Yijū had no other recourse but to claim that Aśu was bought as a slave girl before being converted.²¹

But there may be additional, perhaps more subtle, reasons underlying the formulation of the document as a deed of manumission, conversion, and, by extension, marriage as well. Goitein and Friedman note the remarkable formula of recognition called rašut (רשות) used by Ben Yijū right at the outset of the document as is customary in important documents for subjugating a certain legal deed to a living rabbinic authority. Ben Yijū uses 220 a double rašut invoking two authorities, the Exilarch Daniʾel Ben Ḥisdaʾi of Baghdād and Gaʾon Maṣliaḥ, the head of the Palestinian academy in Cairo. Remarkably, this double rašut signifies the jurisprudent adherence to the Yemenite Jewish legal system. The inclusion of India in the Halakhic jurisprudence of Yemen is reiterated in several other references in medieval rabbinic literature (2008, 633–4; 2010b, 162–4). Ostensibly, Ben Yijū and Aśu formed a nodal point in a Jewish legal network connecting India with Yemen, Baghdād, and Cairo in a shared Jewish legal system. This is an essential and crucial strategy in forming transregional networks, more visibly so in the case of Islamic legal networks across South and Southeast Asia.²² There are a few scattered textual references to a rabbinic court of law in India equating India with Aden as one and the same legislative zone (Goitein and Friedman 2010a, 90–91, and n14). The only evidence for a court of law, possibly of a transient nature, refers to Broach, which is a port town in Gujarat. Though the reference was identified as written by Ben Yijū, the time and place remain unclear, apart from the reported ruling (מעשה, maʿase) being brought before Ben Yijū from Broach (2010b, 281–2). Though this does not substantiate an autonomous authority of Jewish law in India, it shows the attempts by Ben Yijū, and possibly by other Jewish merchants as well, to establish a legal network for Jews in India.

To summarize, the deed of conversion and manumission of Aśu defines her as a slave girl possibly in reaction to allegations against the Jewish pedigree of Ben Yijū’s children born to her. Had the allegations been accepted, the children would have been dispossessed of their father’s inheritance. If the deed of manumission was written in response to the 221 allegations and in hindsight, then the simplest way for Ben Yijū to argue for the proper conversion of his wife would be to ascribe the status of a slave girl to her in the document.

Who Was This Nāyar?

The main reason for speculating that Aśu was not a slave is the reference to one Nāyar whom Ben Yijū refers to as his in-law (ṣihr, צהר, Arabic صِهْر), thus implying that this Nāyar is Aśu’s brother.²³ Nāyar is a caste-name designating a member of the military and ruling clans (Narayanan 2013, 273–4). Since Ben Yijū refers to him as his in-law (ṣihr-i, with the first person possessive suffix -i), it stands to reason that the relationship by marriage between Ben Yijū and that Nāyar were derived through Ben Yijū’s conjugal alliance with Aśu, the mother of his children. That being the case, it is difficult to imagine that the reference to Aśu as a slave girl had any substance in her own society, for her kin—most probably her brother—was a member of an elite group and a business associate of her husband.

The discrepancy between a Nāyar brother-in-law and a wife purchased as a slave was first noted by Shirley Isenberg, an anthropologist studying the Jews of India (1988, 29–30n19). Isenberg notes that Nāyar is a name signifying upper-class people in the Malayalam-speaking region to this day. Ruling out the possibility that the sister of a Nāyar would be a slave, she postulates that Ben Yijū must have married another woman 222 of Nāyar ancestry. She also remarks that the caste system forbids such intermarriages, and that both the brother and the sister would have been obliged to convert to Judaism for escaping excommunication and ostracization by their own community.

Notwithstanding the current attitudes towards inter-caste marriage, there is evidence that caste identity and ethnicity were rather fluid over the ages, with marital relations often defined as acceptable or not based on economic and political interests (Freeman 1999, 282–3; Sharma 1992, 185–6; Durga 2001, 152). Moreover, the caste name Nāyar is generically broad; it refers to a variety of caste identities with various degrees of proximity to the foreign traders who frequented or settled on the Malabar Coast over the centuries (Ayyar 1938, 50; Fuller 1975, 286–7). It is possible that the foreign traders, at least in the coastal regions that depended on overseas trade like the Malabar Coast, were endowed with a status that would be considered equal to that of indigenous trading communities. Especially in the region of North Malabar, which is less favorable to land cultivation, the increased dependence on maritime trade resulted in more favorable sociopolitical terms to foreign traders (Mailaparambil 2011, 11–12). Moreover, had the Nāyar and his sister, Ben Yijū’s wife, been, as Isenberg postulates, excommunicated on the pretext of undesirable inter-caste marriage, it is unlikely that Ben Yijū would have been able to maintain his Indian business network, which embraced quite a few Indian associates. Lastly, the reference to Aśu as Ben Yijū’s wife in the letter mentioned above (fn 5) makes the assumption that Ben Yijū had one slave wife and one Nāyar wife less likely, for the deed of manumission was clearly written in an attempt to legitimize the conjugal alliance with Aśu precisely because she was the mother of his children.

There is at least one more inscriptional reference supporting the possibility that foreign traders of the period would be endowed a caste status 223 eligible for intermarriage with Nāyars. An Old Malayalam inscription dated to the eleventh century mentions a foreign merchant, possibly a Jew, as entitled to honorary rights and caste status which parallels that of the ruling Nāyar castes in Central Kerala. It is especially striking that the inscription ordains the honorary rights as hereditary not only via the foreigner’s male offspring but also via matrilineal descent (marumakkattāyam, “inheritance via sister’s son”). This eleventh-century inscription (also known as the Jewish copper plates) granted to one Joseph Rabban (īsuppu irappāṉ) and to his nephews and sons-in-law (besides his sons and daughters)²⁴ is considered as the fundamental piece of evidence for the origin of Jews in the region, albeit the latter denying intermarriages with non-Jewish women. However, the inscription does not specify the beneficiary as a Jew or as related via marriage to the donor. Nevertheless, complemented by the documents left by Ben Yijū, this inscriptional reference strengthens the plausibility of matrilineal alliances in the early medieval period between Nāyars and foreign West Asian merchants.²⁵ 224

That said, Aśu’s regional identity, defined as Tuḷu in the deed of manumission, further complicates the matter; if her brother was a Nāyar, he must have been a Tuḷu man (for she was a Tuḷu woman), but Nāyars are associated with the Malayalam-speaking region. Furthermore, while there is inscriptional evidence in Old Malayalam since the ninth century for the emergence of the Nāyar class, first as warriors and chieftains and, later on in the post-Cera period (ca. 1100s), as landowners, there is no equally concrete evidence that Nāyars emerged as a landowning caste in Tulunad. Nevertheless and despite the fact that there are currently no Tulu-speaking Nāyars, there is evidence that Nāyars were associated with Tulunad in the past (Narayanan 2013, 273–4). There are several inscriptions found in the coastal region that lies between Kasaragod (in northen Kerala) and Mangalore (in southern Karnataka) that mention Nāyars, Brahmins (Nambis), and merchants (Seṭṭis and others) as Malayali migrants to Tulunad, some as donors to temples or as officials nominated over temples, others as beneficiaries of land grants or as in-laws of Tuḷu people (Vasanthamadhava 1996, 939–44). Interestingly, ethno-historical accounts echo this inscriptional evidence in the textual heritage of the rulers of Calicut integrating the legendary history of Kerala with Tulunad (Logan 2000 [1887], 227–9; Menon 2003, 27, 39; Veluthat 2009, 135). The Judeo-Arabic documents thus further attest to the sociocultural continuity between the Malayalam-speaking region and Tulunad, and the compatibility of Tuḷu identity and Nāyar status in the twelfth century. The document listing Nāyar the brother-in-law and the document ascribing Aśu a Tuḷu identity both portray a group of Tuḷu Nāyar. The connection with Tulunad goes beyond Aśu’s regional affiliation; Ben Yijū established his bronze workshop in Mangalore, an important port town located in Tulunad (Goitein and Friedman 2008, 58–9; 2010b, 8–9, 177–9). It is highly improbable, therefore, that Ben Yijū bought the sister of his business associate as a 225 slave. Under these circumstances, his exceptional choice of words and terms in phrasing of the deed of conversion and manumission betrays the awkwardness of designating Aśu as a manumitted convert slave (Goitein and Friedman 1999, 264, 278–79).

It should be reiterated that the Geniza documents related to Ben Yijū merely represent arbitrary and fragmentary pieces of information about his life and family relations. However, upon examining the references to Nāyar against the backdrop of Indian Ocean maritime trade and the sociopolitical conditions prevalent along the Malabar Coast at the time, the most reasonable assumption would be that Nāyar was indeed Aśu’s brother, and that the conjugal alliance with her was instrumental in forming the business partnership between the two men. As already noted above, the crucial difference between slaves and freed people in medieval South India was land ownership (Gurukkal 2010, 221; Ali 2006, 45). Thus, the socioeconomic status of a slave as equivalent to that in the Mediterranean in the sense of those dispossessed would be translatable to agrestic slavery (aṭimai), as of a land cultivator or tenant. If indeed Aśu was a Nāyar lady, she also had her share of land in her matrilineal ancestral property (Gough 1961, 334, 390–93). Therefore, it is justified to question the meaning in usage of the Jewish term slave girl (שפחה) and its applicability in the socioeconomic context in Malabar at the time.

Two Nāyars and Two Brothers-in-Law

There are two references that complicate the identification of that Nāyar as Aśu’s brother and Ben Yijū’s brother-in-law. Two personalities referred to by Ben Yijū overlap with that Nāyar; one is another in-law called Abū ʿAlī, who is mentioned once in a document listing the donors of oil to a 226 synagogue in Fusṭāṭ in 1153–1156.²⁶ The other overlapping figure is also called Nāyar, but this Nāyar is defined as the brother of the kārdār, “manager” (ʾḵw ʾlkʾrdʾr, אכו אלכארדאר), who was trading in cardamom.²⁷ Since the name Nāyar, like many other names of South Indian origins, even to this day, is derived from the person’s caste affiliation, identifying his relation to Ben Yijū is crucial in determining whether Aśu’s status as a slave girl in the deed of manumission can be understood as merely a formally legal designation. However, to conclude that this is indeed the case, we must first examine the references to that Nāyar and rule out the various speculations brought forward by Goitein and Friedman in order to resolve the seeming contradiction between the low-status wife and the elite brother-in-law and business partner.

Based on Ben Yijū’s reference to Abū ʿAlī as his in-law (ṣihr-i), Goitein and Friedman postulate that the in-law called Nāyar and the in-law called Abū ʿAlī might be one and the same person, and that, consequently, the spelling n-ʾ-y-r might stand for an Indian Jewish name, albeit being unattested elsewhere. Considering that the two may not be the same person, Goitein and Friedman raise another possibility, namely that Ben Yijū might have married another woman, a Jew, in India or in Yemen, whose brother is the aforementioned ʾAbu ʿAlī. Another plausible speculation, as Goitein and Friedman also note, is that the term ṣihr (صِهْر) used in each case denotes different in-law relations, as ṣihr may stand for a brother-in-law, a son-in-law, or a father-in-law. Considering the polysemic nature of 227 the term, it is reasonable to assume that Nāyar and ʾAbu ʿAlī are two different and unrelated people (Goitein and Friedman 2008, 639n17; 2010b, 379n3). This is quite sensible, for the document in which Abū ʿAlī is mentioned belongs to a different phase in Ben Yijū’s life, sometime in 1153–1156, some four to seven years after Ben Yijū left India and sometime after leaving Aden and resettling in Egypt.

As for Nāyar the brother of the kārdār, Goitein and Friedman rule out the possibility that Nāyar the brother-in-law and Nāyar the brother of the kārdār both refer to one and the same person. They assume that Ben Yijū would not have referred to the same person once as his brother-in-law and then again as the brother of a business associate who was blamed in several correspondences for much trouble and great losses (173n26). While it does sound awkward to refer to the same Nāyar once by the attribute brother-in-law (namely, Aśu’s brother) and once by the attribute “brother of the kārdār”, in the context of matrilineal kinship relations it makes sense. In contrast to Goitein and Friedman, Roxani Margariti, following Amitav Ghosh, does not rule out the possibility that the kārdār was indeed related to Ben Yijū through marriage based on the cross-references to Nāyar once as Ben Yijū’s brother-in-law and once as the brother of the kārdār (Ghosh 2002, 214–16; Margariti 2007, 205, 305n130). Considering the family relations in the matrilineal household, the brother-in-law Nāyar and the kārdār’s brother Nāyar could very well be one and the same person. Nāyar could have been Aśu’s brother from her mother’s side, hence both living off the same ancestral land. At the same time, he could have had a half-brother from his father’s side, and consequently from a different Nāyar clan, not directly related to Aśu.

Indeed, the kārdār was a dubious character in Ben Yijū’s life, a business associate who failed to deliver a shipment of cardamom for an advanced 228 payment handed to Ben Yijū by his Jewish business associates in Aden.²⁸ In a letter from Aden sent sometime between 1137 and 1140, Joseph Ben Abraham, a Jewish business associate of Ben Yijū, mentions the kārdār with explicit anger, urging Ben Yijū to pressure the debtor to pay his debt.²⁹ In another letter sent to Ben Yijū in 1146, the merchant Khalaf Ben Isaac refers again to the kārdār’s mischief, but this time the anger is directed against Ben Yijū, for Khalaf demands that Ben Yijū take the responsibility for the undelivered goods.³⁰ Clearly, Ben Yijū did not take any action against the kārdār, as requested by Joseph some six to nine years prior to Khalaf’s letter. He was finally prompted to pay for the loss of cardamom shipment from his own pocket. An oblique kinship relation to the kārdār through his 229 brother-in-law Nāyar might explain both Ben Yijū’s trust and his inability to react against the debtor for so many years.

I am, therefore, inclined to agree with Margariti that the two seemingly contradictory references to Nāyar refer to the same person, precisely because the kārdār was the source of troubles to Ben Yijū. A close reading of the references to the kārdār reveals tolerance on the part of Ben Yijū and even reluctance to act against him despite repeated requests of his Jewish partners in Aden to do so. The reason for this extensive tolerance may very well be the kinship relations through Nāyar the brother-in-law. Ben Yijū also had a problematic brother, Meḇaser, who is referred to in several letters from and to Ben Yijū and who is once blamed by Ben Yijū for being lazy and difficult (Goitein and Friedman 2010b, 20, 158, 223, 229). Tolerance towards unreliable business partners is understandable in the context of family relations. Aśu’s relations through her brother Nāyar to the extensive network of inland merchants like the kārdār simply makes her more desirable for kinship alliance in the eyes of a foreign trader interested in building up a transregional trade network.

The extent of business relations branching out of the alliance between the Nāyar household and Ben Yijū is evident in his draft of accounts mentioning Nāyar “my brother-in-law” (נאיר צהרי, nʾyr ṣhri).³¹ Though there is nothing personal in this list of accounts, it constitutes a remarkable attestation for Ben Yijū’s intricate human connections in India. Like Nāyar, some other names in the document are specified also for their kinship relation with Ben Yijū or, alternately, for their business affiliation with him. 230 Thus, Ben Yijū owes three dirhams and half a fāj (a small Indian coin, Friedman 2016, 685) to one Yosef, the maternal uncle of his workers, who were considered too young to handle their wages. Ben Yijū is also indebted to his maternal aunt’s son, Abū l-Ḵayr Ibn al-Minqār. One Nākhudā Saʿd is referred to as a brother (ʾl-ʾḵ), preceding an honorific “my master” (mwlʾy). The attribute “brother” shows that Ben Yijū considered him a close friend rather than suggesting kin relations between the two (Goitein and Friedman 2010b, 174n34). Still, the insertion of a kinship term underlines the nature of relations associated with the business network in which Ben Yijū was a nodal figure; as much as it is a multi-ethnic and transregional network, it is based—at least partially—on kinship relations.

Another document referencing the unique business connections of Ben Yijū in South India is a letter sent by Maḍmūn Ben Ḥassan from Aden to Mangalore. The letter is a business letter typical of the correspondences between Jewish traders involved in the Indian Ocean trade. It contains a less typical request to convey Maḍmūn’s warm regards to three Indian associates of Ben Yijū in Mangalore, namely Sūs Sītī (סוס סיתי sws syty), Knābtī (כנאבתי, knʾbty), and ʾIsḥāq al-Bānyān (אסחאק אלבאניאן, ʾsḥʾq ʾlbʾnyʾn); the first is thus identified as a Seṭṭi, a term for merchants associated with South India, the second as a citizen of Kambhāt (Cambay) in Gujarat, and the third as a Banian, or a merchant associated with North India, surprisingly bearing a typical Semitic name, Isaac (Goitein and Friedman 2010a, 151n37).³² Seṭṭi merchants are mentioned in several inscriptions related to medieval transregional trade networks with both Nāyars and Muslims connected with West Asia (Vasanthamadhavan 1996; Hall 2010, 128, 131). 231 Kanābtī is probably an appellation derived from Kambhāt (Cambay), the famous and affluent medieval port town in Gujarat (Lamb 1958, 235; Pearson 2003, 94; Ho 2007, 352–3). Banian merchants are related to the Vaiśya, or merchant, caste, which can be traced back to ancient Indian civilization and to the Sanskrit term vaṇij, “merchant”. The earliest known occurrences of the derived term bānyān are in Arabic. Banians are associated mainly with the northwestern parts of India and comprise many sub-castes. They belong either to Jain or to Vaiṣṇava religious groups (Lamb 1958, 235–6; Findly 1997, 289–91). It is therefore surprising to find the designation Banian attached to a Semitic name, a point I shall return to in the concluding section.

Names and appellations of Indian merchants, ship owners, and business associates occur in the Geniza documents time and again. However, the closeness and intimacy projected in this specific letter is uncommon. It demonstrates, I believe, the unique character of Abraham Ben Yijū’s business network in India as a network crossing boundaries of caste, religion, ethnicity, and even language. Such an intricate and closely tied network must have been based on a high degree of social and spatial mobility and on free access to elite groups close to the centers of political power in the various regions along the West Coast. It is for this reason that the relations of Ben Yijū with a Nāyar defined as an in-law cannot be ruled out on the pretext of violating caste or class norms; rather, they can be viewed as a networking strategy shared by both West Asian migrant merchants and local financial and economic agents. It should not come as a surprise that Aśu’s brother, Nāyar, had an interest in an alliance with Ben Yijū and his business associates in India and abroad. It would be surprising, on the other hand, if a man free to own property and to interact in a long-distance maritime trade network were the brother of a slave girl, if Aśu’s designation in the deed of manumission is to be taken at its face value. 232

Whose Property?

If indeed Ben Yijū’s business network in India relied on matrimonial alliance with a Nāyar household, the reference in the deed of manumission to Aśu’s “mistress’ house” (בית גברתיך, byt gbrtyk) denotes, in effect, Aśu’s matrilineal household, or taṟavāṭә, which denotes an impartible house and land unit (Moore 1985). Arguably, Aśu’s taṟavāṭu was instrumental in establishing Ben Yijū’s elaborate and intricate business hub, which consisted of a bronze workshop as well as trade in cardamom, pepper, betel nuts, and raw materials for processing bronze and other metals (Goitein and Friedman 1999, 267; 2010b, 9–10). Another clue for the matrilineal background of the partnership is found in a letter sent many years after the alliance with Aśu took place. In this letter, Ben Yijū’s business associate and coreligionist Maḍmūn Ben Ḥasan urges Ben Yijū to return from the land of India to Aden with his property and children. In the letter, sent in approximately 1145, Ben Ḥasan warns Ben Yijū that he should better return to Aden, for if he dies in India, his property will be lost and his children will be among those accommodated or sheltered (תאויה, tʾwyh) by the land (אלבלאד, ʾlblʾd).³³ What Ben Ḥasan’s concern was is not very clear; was he implying that the children might lose their Jewish identity and become integrated in the local non-Jewish population? While this is possible, it seems to me more likely that the concern was about their inheritance, namely that Ben Ḥasan was concerned that the children would not be entitled to inherit their father in 233 the matrilineal extended household, and that they would depend for their livelihood on their mother’s house, the taṟavāṭә, rather than on their own property.

The term bilād in reference to Ben Yijū’s place of residence is used by Ben Ḥasan twice, once specified as India (בלאד אלהנד, blʾd ʾl-hnd) and once modified by the definite article al- (אלבלאד, ʾl-blʾd). In the first occurrence, the term denotes the country in its widest sense possible, whereas in the second occurrence he refers to the specific town or village in which Ben Yijū lived with his wife and children, namely Aśu’s taṟavāṭә.³⁴ In other words, Ben Ḥasan is aware of the possibility that the members of Aśu’s taṟavāṭә might claim Ben Yijū’s property if he dies and, even worse, they might subject his children to becoming members of the taṟavāṭә, which implies the annulment of their affiliation with the Jewish family of their father. Notably, in a matrilineal system, Nāyar women need not be obliged to a single husband; the father of their children can leave without affecting the social or kinship status of his children. The conjugal relation to a husband is marked by a form of marriage called sambandham, which may or may not be transient. It is, therefore, likely that Ben Yijū was not even married to Aśu in the Jewish sense of marriage; sambandham marriage allows the women to be more or less free to cohabitate with a man of their choice for a certain period as desired by them (Gough 1961, 334–44; Moore 1988). It seems that Ben Ḥasan was aware of these customs and that he warned Ben Yijū against passing on his property to the matrilineal line of the family contrary to the interests of Ben Yijū’s Jewish family back in Egypt. 234

Taking all these considerations into account, there are still many questions left open. Perhaps the most pressing question, as raised by an anonymous reviewer of this paper, is whether upper-caste Hindus would allow a foreign trader to “pollute” their household. The same can be asked about the Jewish traders, namely, how far they would be willing to “bend” the Halakhic regulations against intermarriage with non-Jewish women. To rule out the possibility that traders from both sides of the Arabian Sea would be as liberal (or at least flexible) in such matters is at odds with evidence presented in this paper. The fierce opposition that Ben Yijū encountered in Yemen and the refusal on part of the local Jewish society to recognize his children as his lawful heirs underlines Ben Yijū’s efforts in maneuvering his business and his life between two conflicting socioeconomic systems. The letter by Maḍmūn Ben Ḥasan and the documents regarding Ben Yijū’s appeals to the rabbinical authorities in Yemen show that the transregional trade maneuvers depended to a large extent on kinship relations, which were often fraught with complications. The kinship-based network is one possible and common strategy in the socioeconomic management of production as well as trade (Gurukkal 2010, 307–8; Bhattacharya, Dharampal-Frick and Gommans 2007, 96–7).

Conclusion

That landowners sought alliances with West Asian traders for economic and political reasons is attested in inscriptions from the ninth, tenth and thirteenth centuries (Narayanan 1972, 23–42; Malekandathil 2007). The emergence of monotheistic communities and their continued contacts with West Asia was also due to transregional and intercommunal contacts that hardly, if at all, left traces in history. The records left of Abraham 235 Ben Yijū and his business network, stretching between the Mediterranean and the Malabar Coast, offer a glimpse of processes and patterns of exchange on which official records and historiographies are silent. The conjugal alliance between Ben Yijū and Aśu demonstrates the strategy of intermarriage as instrumental in building up a transregional network based on kinship alliances, notwithstanding the negative attitudes of orthodox Brahmanism and Judaism alike.³⁵ It is possible that during certain periods and in certain regions, Nāyars engaged or interested in overseas trade would consider foreign traders as their equals in socioeconomic status and, hence, eligible to marriage. Such a history of pragmatic and liberal intermarriage may explain the discrepancy between the slave origins and upper-class status of the majority of Cochin Jews as attested in the early sixteenth-century responsum discussed above. Additionally, certain ritual symbols—especially in relation to marriage customs—are attested to for all the Māppiḷa communities (see below), Jewish, Christian, and Muslim alike (Walerstein 1987, 92–113; Bayly 1984, 184; Miller 2015, 44, 179–181).

Thus, besides being a curious life story, the story of Aśu and Ben Yijū contributes to the history of Jewish-Christian relations in the West Coast of South India. That their respective communities evolved based on similar patterns of transregional networking is evident also in the reference to Iṣḥāq al-Bānyān, whose name baffled Goitein and Friedman (2010a, 151n35); how did a man bearing a Semitic name come be termed an Indian merchant? Notably, it is impossible to determine whether Iṣḥāq was a Jew, a Christian, or a Muslim, as the Cairo Geniza letters and documents often mention people whose names do not betray their religion (Margariti 2014, 45–9). However, in the case of Iṣḥāq it is possible to know that while he was 236 of Semitic origin, he was also integrated into the local caste system, for otherwise his Jewish business associates would not have marked his Indian occupational affiliation, al-bānyān, “the merchant”, normally associated with either Jains or Hindus, as mentioned above. The combination of a Semitic proper name and an Indian occupational designation attests to the fact that during Ben Yijū’s time, when the letter with the reference to Iṣḥāq al-Bānyān was written, descendants of intermarriage between West Asian merchants and Indian traders were integrated into the transregional networks. Some other names and people mentioned in the letters seem to refer to indigenous Jews, Christians, or Muslims, like a merchant from Dharmapaṭṭaṇam (drmtn, דרמתן), whose name is Yosef or Yūsuf (ywsf, יוסף) Lanbi (lnby, לנבי), which might also be a combination of a Semitic name with an unidentified South Asian designation.

The origins of Jewish and Christian communities along the Malabar Coast can be safely be attributed to itinerant traders forming trade alliances with local landlords, with intermarriages being one strategy for establishing a transregional network. There could have been various pragmatic reasons for a West Asian trader to cohabitate with a local woman besides merely looking for comfort and domestic service in aligning with concubines or maid-servants. Another pragmatic reason for intermarriage was the begetting of bilingual children, possibly the most efficient way to create a network of translators so essential in conducting business across diverse regions and cultures. The children born to mixed couples carved their own caste status in the social matrix of Malabar known as Māppiḷa, a designation worthy of matrilineal and cross-cousin alliances as it is derived from the words māmaṉ, “maternal uncle”, and piḷḷa, “son”. It should be noted that matrilineal castes were not necessarily Nāyars; there were also castes of fisher folk, artisans, and cultivators of a lower social status (and lesser ritual purity as well) who were following matrilineal lines of descent. The 237 boundaries between those communities that emerged out of intermarriages with upper-caste Nāyars and those whose ancestry is traced to lower-caste communities are still retained, to a certain extent, even today (Bayly 1984, 243–251; Gough 1961, 415). Under these circumstances, interreligious relations between Jews, Christians, and Muslims are shaped by their shared origins in matrilineal kinship relations utilized for establishing transregional business networks, often in collaboration with each other (Margariti 2014).

Naturally, the transregional kinship alliances proved beneficial also for the South Indian traders and landlords. The period in which Ben Yijū resided in India witnessed several historical changes in the state and social formations of the western coast of South India. This is a period in which the Old Malayalam language emerges as the administrative language of the region in inscriptions dated from the ninth to the thirteenth century (Sekhar 1951), with at least two inscriptions, from 849 and 1000, attesting to the alliances between the ruling and landowning classes and West Asian traders. Maritime trade activities witnessed by the Judeo-Arabic Geniza documents contributed, at least to a certain extent, to the socioeconomic development of the Malayalam-speaking region and to its political evolution independently of the historic Tamiḻakam in a period characterized as the early medieval period (Veluthat 2009, 3). The Coḻas to the east posed an ongoing threat to the rulers of Kerala, who became more and more dependent on chieftains and traders for supporting their political power. The alliance with Arab traders is vividly depicted in traditionally attributing the origin of Islam in Kerala to the conversion of a Kerala king in the twelfth century, supported by inscriptional evidence (Narayanan 2013, 129–34). Interestingly, one of the oldest mosques in Kerala was built in 1124 in Māṭāyi, a medieval port town in the same coastal area dotted with port towns inhabited by Ben Yijū and his relatives and frequently visited by his multiethnic business associates. 238

Abraham Ben Yijū settled in an area on the margins of the great empires, the Coḻas to the east and the Cāḷukyas to the north, where a decentralized political system of nāṭuvāḻis, “rural chieftains”, was prevalent (Veluthat 2009, 193–203). It is difficult to imagine a foreign trader like Ben Yijū spreading a network of overseas and hinterland trade without the collaboration or consent of such a nāṭuvāḻi. This alliance between local landowners and chieftains and Arabic-speaking traders left little traces in historical accounts of the period, and it is only much later, after the emergence of full-fledged religious communities, that evidence for Māppiḷa-Nāyar political alliances in both foreign and local accounts begins to emerge. The story of Ben Yijū and Aśu is, therefore, a rare glimpse into the period of formation of transregional networks and of transformation from kinship alliances to religious communities.

Acknowledgments

I am greatly indebted to the European Research Council (ERC) and the European Union’s Horizon 2020 research and innovation program (grant agreement n° 647467 — JewsEast), under whose auspices the research incorporated in this paper was conducted. The paper benefitted greatly from the comments and suggestions of Alexandra Cuffel and of two anonymous reviewers. All remaining mistakes and shortcomings are my own. 239

Reference List