Entering full second marriages appears to be a custom peculiar to the community of Old Assyrian expatriates in Anatolia,Footnote 2 designed to accommodate the needs of its travelling men (Michel Reference Michel2006: 163).Footnote 3 The potential role of Anatolian agency in the formation of this new custom, however, is often considered secondarily, despite numerous marriage contracts featuring mixed Assyro-Anatolian couples.Footnote 4 This is partly due to the nature of the textual record, which offers very little information for reconstructing default conditions for Anatolian marriage practices, or gauging the extent to which these may have differed from Assyrian customs. While it is therefore inevitable that discussions of kārum period marriage rely mostly on the Assyrian perspective, it would be a mistake to assume Assyrian interests alone (or primarily) shaped mixed marriages. Anatolian agency should also be taken into account, especially where certain aspects of long-distance bigamy cannot be satisfactorily explained as prioritising Assyrian needs but instead suggest compromise. In other words, in generating a new legal mechanism of second marriages, Assyrians were not simply adapting to the logistics of a long-distance life style, but also to a new set of social expectations. As already noted by Lumsden (Reference Lumsden and Michel2008) and recently reiterated by Larsen and Lassen (Reference Larsen, Lassen, Kozuh, Henkelman, Jones and Woods2014), the nexus of Assyro-Anatolian intermarriage aligns rather nicely with White's (Reference White1991; Reference White2011) model of the “middle ground” as a nexus of cross-cultural compromise.
Middle ground in the kārum period
Scholars are becoming increasingly alert to the need for theoretical nuance in modelling Anatolian-Assyrian interactions, and several such models have been proposed for the kārum period (see Larsen and Lassen Reference Larsen, Lassen, Kozuh, Henkelman, Jones and Woods2014 for a recent overview). Most notable has been Stein's excellent critique (2008, with references), which proposes the trade diaspora as a framework within which to consider the kārum period. A more compelling alternative, however is Lumsden's (Reference Lumsden and Michel2008) suggestion in favour of White's (Reference White2011: xiii) “middle ground”, namely:
[…] the creation, in part through creative misunderstanding, of a set of practices, rituals, offices, and beliefs that although comprised of [sic] elements of the group in contact is as a whole separate from the practices and beliefs of all those groups.
In discussing the usefulness of this term for kārum period contact, Lumsden (Reference Lumsden and Michel2008: 32) has explained the middle ground as follows:
[A] space of social heterogeneity and hybridization, of compromise and accommodation, in which social encounters can lead to new systems of meanings and transformations among all participants. This is not a process of acculturation, of one culture becoming more like the other, but of a more nuanced form of encounter in the “in-between” space of the middle ground, which results in something completely new.
While Lumsden's (Reference Lumsden and Michel2008) recognition of the middle ground as a suitable model for the kārum period holds tremendous potential, it has yet to achieve its deserved scale of impact in the field. Recently, Larsen and Lassen (Reference Larsen, Lassen, Kozuh, Henkelman, Jones and Woods2014) have brought the middle ground and associated notions of cultural hybridity into discussion, focusing specifically on the case of glyptic imagery. We must not, however, lose sight of the fact that hybridity is often the by-product of a wide range of cross-cultural encounters and not exclusive to the idiosyncratic circumstances of the middle ground. What we know of kārum period society from its extensive textual documentation is very much in keeping with middle ground formation in Assyro-Anatolian contact, but how should we distinguish the middle ground from other forms of hybridity, acculturation, imitation, or emulation? According to White (Reference White2006; Reference White2011: xi), a key ingredient is the element of misunderstanding, but this is a notion which is extremely difficult to pinpoint, even within the textual realm (White Reference White2011: xxii; see below). Lumsden (Reference Lumsden and Michel2008: 32, n. 11) also acknowledges this difficulty, noting: “Although I believe that misunderstandings must have played a part in the middle ground process in the Old Assyrian Colony Period, I have not yet been able to find a corresponding phenomenon.”
This article considers whether the elusive element of cross-cultural misunderstanding (perhaps wilful) may be detected in the terminology of bigamous marriage peculiar to the Old Assyrian society in Anatolia.
For the historical setting for which White formulated the middle ground theory, (colonial encounters between the French and Algonquian Indians in the pays d'en haut between the 17th and 19th centuries), “problems in two arenas of contact—sex and violence—seem to have been particularly acute” (White Reference White2011: 60). Sexual relations between the foreign and local populations were particularly significant as the conspicuous absence of women among the former inevitably meant that sexual partners could only be found among the latter (White Reference White2011: 60). A similar situation also applies to the kārum period during which, especially in the early periods, Assyrians travelling to Anatolia were predominantly men, thus determining the character of intermarriage during the earlier part of the kārum period as between foreign men and local women. Michel (Reference Michel and Dercksen2008b: 226) and Larsen (Reference Larsen2015: 250) have attributed the origins of Assyro-Anatolian intermarriage to the emotional and sexual needs of Assyrian men.Footnote 5 Over time, however, texts also attest to intermarriages between Anatolian men and Assyrian women (Michel Reference Michel2006: 173).
While the politics of sex governing French-Indian contact undoubtedly differ from those of the kārum period, a point of inescapable comparison is the creation of mixed families, which constitute a unique context for negotiating cultural differences. As opposed to the public domain of commercial activity in which the main actors are men, mixed families provided a more private setting within which both men and women play a part. Mixed marriages operate not only at the initial level of bringing foreign men and native women (or vice versa) into a context of potential misunderstandings and mutual compromise, but also at the level of children born into mixed marriages, who become cultural mediators themselves (White Reference White2011: 74). Individuals with mixed parents are indeed conspicuous in the textual record of the kārum period, presented sometimes with Assyrian, sometimes with Anatolian names (Michel Reference Michel and Dercksen2008b: 211), pointing to different social identities and with significant implications for middle ground formation over generations (Larsen and Lassen Reference Larsen, Lassen, Kozuh, Henkelman, Jones and Woods2014: 177). The new form of bigamous arrangements peculiar to the expatriate community of Assyrians in Anatolia therefore presents a highly suitable context in which to test the middle ground, and may even reveal that crucial element of misunderstanding as a possibly deliberate misappropriation of the terminology of (mixed) marriage.
Sources
Needless to say, textual sources attesting to kārum period marriage practices present a fair number of challenges of their own, starting with the inevitable Assyrian bias in the documents. Being generated mainly by the literate Assyrian community, texts reflect mainly Assyrian activities.Footnote 6 Information on local customs is therefore quite limited (Veenhof Reference Veenhof2012: 148–50), a considerable difficulty in identifying which points of cultural compatibility may have instigated misunderstanding, misappropriation, and compromise eventually leading to a middle ground.
Even for the textually visible Assyrian community, the available sources are predominantly business-related. This is offset to a certain extent, however, by the high intensity of correspondence with Aššur or within Anatolia. While the primary concern in most letters was again commercial matters, they also contained lively details of family relations (see Michel Reference Michel2001). References to couples’ relationships, mutual expectations, responsibilities towards children, and dealings with in-laws contain a great deal of indirect information which helps outline the basic principles of Assyrian marriage in Anatolia.
As for the legal framework of the conditions for marriage, divorce, ownership of property, and inheritance rights of children, these can be reconstructed from contracts. While there are at least forty such contracts drawn up in Anatolia, these do not necessarily reflect typical Assyrian customs, dealing instead with unusual circumstances (Michel Reference Michel2006: 158–59).Footnote 7 Furthermore, it is particularly frustrating that no Old Assyrian law code has thus far been discovered, although references to “words written on the stela” (Veenhof Reference Veenhof1994–95; Reference Veenhof and Westbrook2003: 431) certainly suggest one existed. This being said, the available records of Old Assyrian legal practice do not contradict contemporary Old Babylonian customs, which, at least in terms of the fundamental principles of marriage, provide reasonable points of reference and comparison.Footnote 8 The second legal corpus from which useful comparisons can be drawn comprise the Middle Assyrian laws, which, although reflecting legal practice of a later period, nonetheless offers a view specifically into the Assyrian cultural sphere.
Anatolian marriage practices
The little evidence available for local Anatolian marriage suggests customs may have been markedly different from, though not wholly incompatible with, Assyrian practices, as intermarriage was not only possible but also fairly common.Footnote 9
Similarities include the exchange of marriage gifts (see below); parents arranging for adopted daughters to marry natural sons (Lewy Reference Lewy1938: 114; Donbaz Reference Donbaz, Mellink, Porada and Özgüç1993),Footnote 10 or rules governing inheritance and adoption (Veenhof Reference Veenhof, Stol and Vleeming1998: 149). Divorce, on the other hand, appears to have been handled differently. In contrast to marriage contracts involving Assyrians, which prescribe a fixed amount of silver to be paid by the initiating party, contracts between Anatolians stipulate the equal division of the couples’ property (AKT I 21; see Michel Reference Michel, Briquel-Chatonnet, Farès, Lion and Michel2009b: 256, n. 11).Footnote 11 If this was the Anatolian socio-legal norm for a divorce settlement, it does not seem to have been adopted by Assyrians, even in mixed marriages.
Standard Old Assyrian marriage and “la bigamie relative”
Texts from Kültepe-Kaneš attest to several stages of a marriage. Although itemized lists of dowry, marriage ceremonies or verba solemnia are not mentioned as in contemporary Old Babylonian documents (Veenhof Reference Veenhof2012: 152–53), gift exchange between the families is attested (Michel Reference Michel2006). Should the marriage end in divorce—a possibility regularly acknowledged in contracts—a monetary compensation was to be paid by the initiating party. Women who were divorced or widowed were free to remarry (Michel Reference Michel2006: 161). At the completion of a marriage ceremony, the husband became responsible for his wife's maintenance (Michel Reference Michel2006: 161).
As for the wife, her principal duty was to produce offspring, i.e. heirs (Michel Reference Michel2008a: 22).Footnote 12 But not all marriages were successful in meeting this goal. Where the wife was unable to bear children on account of (real or perceived) infertility, sickness, or disability, in order for the marriage to fulfil its procreative purpose, the husband was permitted to marry (ahāzum) another woman who, effectively, would serve as a surrogate mother. While children born from such arrangements were treated as the natural children of the married couple and had full rights as heirs, the surrogate retained her slave status (amtum) in relation to the wife (aššatum). Thus the nature of the original marriage contract established between husband and wife (aššatum) was preserved.
In some cases, the burden of providing an amtum as surrogate seems to have been placed specifically on the wife (aššatum), who also retained the right sell the amtum once the latter had fulfilled her childbearing services (ICK I 3, see below). From a strictly “business” point of view, this would have been a practical arrangement whereby the unfulfilled reproductive duties of the wife were effectively “outsourced” without having to dissolve the “partnership” created by the initial marriage arrangement. The investment of material and social capital in a marriage would have been a strong incentive to preserve an already established arrangement rather than dissolve and replace it, particularly if there were specific social and interfamilial interests at stake. Arguably, therefore, the acquisition of an amtum as a surrogate mother does not constitute bigamy in the strictest sense, as it does not replicate the relationship between husband and wife (aššatum), hence Michel's (Reference Michel2006: 168) term, “la bigamie relative.”
Peculiar to the Assyrian community in Anatolia: “la bigamie autorisée”
The Old Assyrian community in Anatolia also practiced another type of bigamy, which allowed men to enter a second marriage irrespective of, and separate from, real or anticipated cases of childlessness. Such cases of what Michel (Reference Michel2006: 161) terms “la bigamie autorisée” were permissible strictly on the condition that one wife remained in Aššur and the other in Anatolia. Thus, while bigamous in theory, such marriages preserved monogamy in practice, as the husband could only cohabit with one wife at a time.
Prima facie, the terminology of bigamie autorisée, in which the first wife is called an aššatum and the second wife an amtum, suggests a similar disequilibrium of status between the two, where the amtum-wife is a slave and/or subservient to the aššatum-wife (see Łyczkowska Reference Łyczkowska1996–98) as indeed would have been the case for bigamie relative.Footnote 13 The evidence shows, however, that this was not so, both wives being of equal rank and in possession of the same rights vis-à-vis their husbands. The terms aššatum and amtum used in this specific context identify the two wives in their chronological order of marriage. In short, Old Assyrian bigamous marriages remain the exception to the rule that: “[Mesopotamian] wives were always of unequal status” (Westbrook Reference Westbrook2003–05: 601). Nor was there an ethnic prerequisite: “The wives could be either Assyrian or Anatolian, and there was no rule that the aššatum had to be the wife in Aššur and the amtum the wife in the colonies” (Veenhof Reference Veenhof, Roth, Farber, Stolper and Bechtolsheim2007: 303). Recently, however, Larsen (Reference Larsen2015: 250) has re-suggested that aššatum-wives were distinct from amtum-wives on grounds of ethnicity and status:
We have no information about the negotiations and deals that must have formed the background for these marriages, but it does seem clear that from the Assyrian point of view the Anatolian wives were not on an equal footing with the Assyrian ones. A different terminology was used, where the proper word for a wife, aššutum, designated Assyrian women, but not the Anatolian wives of Assyrian men. They were referred to as amtum, which means “slave girl”.
As Michel (2006b: 158) observes, most Assyrian men arriving in Anatolia would have been married already in Aššur. Therefore the earlier generations of amtum-wives are more likely to have been Anatolians. There are, however, several instances which show clearly that there was no hard and fast rule confining Anatolian women to amtum-ship and reserving the status of aššatum for Assyrian women.Footnote 14 In fact, Barjamovic et al. (Reference Barjamovic, Hertel and Larsen2012: 63) have noted that Anatolian women initially known as amtum are later attested as aššatum. Assyrian men did not, it would seem, profile the status of their wives on the basis of ethnicity, so neither should we.Footnote 15
One possible difference of status between an aššatum- and an amtum-wife seems to be in terms of the relative rights of their heirs, those of the former taking precedence over the children of the latter (Veenhof and Eidem Reference Veenhof and Eidem2008: 108; Westbrook Reference Westbrook2003–05: 602). A more tangible difference might be that, as Veenhof (Reference Veenhof, Roth, Farber, Stolper and Bechtolsheim2007: 302) observes:
an amtum was more closely tied to her husband when he travelled, since only in contracts with an amtum […] do we find the provision that the husband may take his wife with him on business trips on the condition that he eventually brings her back home to Kaneš.
The rationale for such a condition is not easy to gauge, as it is not immediately clear for whose benefit it would have been formulated. According to Veenhof's interpretation, and certainly in the way it is presented in Prag I 490 (Larsen Reference Larsen2002; see also Michel Reference Michel2006: 162, 166), the provision for an amtum-wife to accompany her travelling husband appears to reflect the interests of the husband, who then has to agree to eventually return his amtum-wife home. In other texts, travelling with one's husband is treated as a duty which a husband could rightfully expect from an amtum-wife,Footnote 16 but which an aššatum-wife was not obliged to fulfil.Footnote 17 It is, however, also possible to take the view that provisions for amtum-wives to travel essentially served the interests of the women themselves. In Kt 94/k 149 (Michel and Garelli Reference Michel and Garelli1996: 298–99), Aššur-malik, who is to be accompanied by his wife Suhkana on his travels, is strictly prohibited from taking another wife whilst in Anatolia (but is free to enter a second marriage in Aššur, in keeping with the principle of bigamie autorisée). Michel (2008: 27–28) sees in this stipulation a conscious attempt to prevent Aššur-malik from acquiring a new wife in every Anatolian city he visits.
Presumably, travelling arrangements between amtum-wives and their husbands depended on the situation and the persons (and personalities) involved.Footnote 18 Although Inna-Sîn has reportedly written to his wife, “If you do not come here, you are not my amtum-wife!” (Kt h/k 73, see n. 11) the failure of his wife's repeated attempts to catch up with him suggests that he was in fact avoiding her. The reverse situation is also attested: in BIN 6 104, we find Puzur-Aššur entreating his evidently reluctant wife Nuhšātum to join him: “On the day you hear my tablet, tur[n] ther[e] to your father (i.e. ask for his consent) and (then) start out with my b[o]ys and do come here.”Footnote 19
Michel (2008: 27) observes that the travelling clause is found in contracts between Assyrian men and Assyrian women with amtum status. If ethnicity was indeed a factor in defining the travelling obligations of an amtum-wife, this potentially has implications for assessing Assyro-Anatolian middle ground in the context of marriage. For the moment, however, it is difficult to offer substantial explanations without being able to determine who would have benefited from such an arrangement, and how. Did Assyrian women make up the majority of travelling amtum-wives because this somehow reconciled their role in bigamie autorisée with traditional Assyrian cultural expectations? Or were Anatolian women largely exempt from the obligations of a travelling amtum-wife because these were somehow at odds with Anatolian customs?
Ultimately, Michel (Reference Michel2006: 162) summarises the current view of amtum-wives as simply having been necessitated by the circumstances in which Assyrian merchants found themselves:
Les marchands assyriens, absents du foyer pendant de très longues périodes et installés dans des comptoirs de commerce loin de chez eux en Asie Mineure, sont autorisés à prendre une deuxième épouse sur place.
Having left their wives tending home and hearth in Aššur and finding themselves de facto bachelors in Anatolia, Assyrian merchants were granted permission to marry a second time. Effectively, then, a legal loophole was created to allow Assyrian men to pursue normal married lives whilst also protecting the fundamental principle of monogamous marriage in practice—an elegant solution to a tricky situation.
Questioning the rationale behind bigamie autorisée
While de facto bachelorhood does certainly seem to be the fundamental problem and legalised bigamy its direct solution, the question remains as to why this particular solution was adopted.Footnote 20
It is easy to see why being deprived of all the benefits of married life may have constituted a problem for Assyrian men. Not only would they have lacked female company, but also faced the more practical concerns of housekeeping or other tasks ordinarily performed by women (cf. Kienast Reference Kienast2008: 37). The complaint of Puzur-Aššur (see above), a lonely husband whose wife appears to be avoiding him, illustrates this basic expectation:
I am alone. There is no one who looks after me (lit. stands at my head), or sets the table for me. If you do not come with my servants then in Wahšušana I will marry a girl from Wahšušana.Footnote 21
Larsen (Reference Larsen2015: 250) has similarly formulated that:
The origin of the use of this seemingly derogatory term must surely lie in the fact that the first generation of Assyrians in Kanesh, men whose wives lived in Assur, would have established sexual and emotional contacts with one of the slave girls in their household in order to make up for the fact that they were separated from their wives for years on end. However, the later amtum wives, or secondary wives of their sons, were clearly not slaves, and several well-known examples of such mixed marriages exist, where we see the Anatolian women functioning in all respects as ordinary spouses.
More importantly, if away from their wives for too long, Assyrian men also faced the potentially much more serious problem which could threaten the most fundamental part of marriage, namely, the production of heirs. It is fair to assume that most Assyrian men needed women—but the question is, did they need wives?
Bigamous amtum-marriages of Assyrian men in Anatolia, insofar as they constitute a socio-legal practice “spécifique à la société marchande assyrienne” (Michel Reference Michel2006: 163), seem too tortuous to have arisen simply in response to the logistics of distance. From a strictly Assyrian perspective, less cumbersome solutions to the problem of de facto bachelorhood could easily have been accommodated within the existing legal stipulations and social expectations guiding marriage, without the necessity to modify the established practice of bigamie relative into bigamie autorisée by redefining amtum-ship into wife-ship.
Existing legal and social practices already allowed Assyrian men to meet all potential domestic and intimate needs by means of an amtum in the ordinary sense of a slave, who could easily have acted in the capacity of housekeeper and/or sexual partner, and could also produce heirs, whose legal status remained at the discretion of the father. As Lumsden (Reference Lumsden and Michel2008: 38, citing Larsen) puts it: “the acquisition of local women's labor [was possible] through marriage or other means” (italics added).
Housekeeping, presumably the most common and pressing need shared by Assyrian men who found themselves in a bachelor's existence in Anatolia, would ordinarily be met by means of female slaves whose typical tasks lay within the domestic realm. Just as female domestic slaves were counted among standard household furnishings in Old Babylonian documents (Seri Reference Seri and Culbertson2011: 50), so were slaves listed alongside various other assets in contemporary inheritance texts citing inheritance matters from Kültepe-Kaneš (see Albayrak Reference Albayrak2000).Footnote 22 Slaves and slave ownership are certainly well attested for kārum period society, both among Assyrians and Anatolians (Veenhof Reference Veenhof and Westbrook2003: 449). While there is no apparent reason why the role of a domestic slave should be transformed into that of a wife, the line between slaves and wives is easily blurred where sexual intimacy is concerned. Regardless of whether a man held sexual rights to his female slaves by default, sexual relations between master and slave did not constitute a social taboo (Westbrook Reference Westbrook1994–95: 1634).
Nor would sexual relations between a married man and his female slave have been considered an act of adultery, which Mesopotamian law codes define “solely in terms of the extra-marital relations of the wife” (Westbrook Reference Westbrook1984: 753). It then follows that lonely Assyrian merchants seeking sexual partners could freely do so without risking their marriages, or indeed without being compelled to marry their slave-women so that they could have (or because they had already had) intercourse with them. More to the point, the option to acquire the services of a slave-woman specifically as a “concubine” (ištariūtu) was certainly available both in Kaneš and in Aššur.Footnote 23
Even in those cases when sexual intimacy between master and slave (or concubine) might lead to the eventuality of offspring, this did not necessarily predicate marriage or indeed a change in the status of the woman. Mesopotamian law codes make frequent references to children born of the sexual union of free men and slave-women outside the context of marriage (e.g. CH §171, Westbrook Reference Westbrook1994–95: 1649). If a man wished his offspring from a slave-woman to be recognised heirs, as long as they were uncontested by the offspring of a wife, he could do so. Marriage was not an absolute condition even for acquiring legitimate heirs and the status of a slave-woman as well as the children she may have borne ultimately remained at the discretion of the man.
Indeed, even within bigamie autorisée, Assyrian men reserved the right to an amtum-slave as surrogate mother in the eventuality that their wives should fail to bear children. The marriage contract between Lāqēpum and Hatala (ICK I 3) is a case in point:
Lāqēpum took Hatala, daughter of Enišru. Lāqēpum shall not take another wife in the Land (Anatolia). In the City (Aššur) he can take a qadištum. If in 2 years she (Hatala) does not acquire for him a child, she herself will buy a slave-woman. Afterwards, once she (the slave-woman) produces (for him) a child she (Hatala) shall sell her wherever she pleases If Lāqēpum himself divorces her (text: him) (Hatala), he shall weigh out 5 minas of silver but if she divorces him, she shall weigh out 5 minas of silver.Footnote 24
What makes this contract particularly interesting is that it not only lays the ground for bigamie relative (i.e. the acquisition of a slave-woman, amtum, against the eventuality of Hatala failing to bear children within the specified time) but it also regulates Lāqēpum's marriage to Hatala in terms of bigamie autorisée (i.e. the permission to enter a second marriage in Aššur). Clearly, bigamie autorisée and bigamie relative were not mutually exclusive. Therefore we must conclude that each arrangement had its distinct purpose. If the reason for the anticipated bigamie relative was childlessness in his first marriage, what was the reason for the anticipated bigamie autorisée permitting Lāqēpum to enter a second marriage in Aššur?Footnote 25
Ordinarily, the practice of introducing a surrogate mother into a marriage (bigamie relative) was justified either by reproductive failure arising from infertility, disability, or sickness; or by reproductive taboo arising from consecrated status, such as in the case of some nadītum-women (Harris Reference Harris1975: 315ff). A third category applicable in our case may be termed reproductive unfeasibility, arising from the sheer physical distance separating wives from their husbands for long periods of time. Instead of extending the well-established practice of bigamie relative into such cases of long-term separation, Assyrian expatriates in Anatolia formulated the auxiliary measure of bigamie autorisée to enter second marriages, whilst retaining the practice of bigamie relative for anticipated childlessness. Why did Assyrian men in Anatolia not simply continue the already established practice of bigamie relative instead of entering the legally bindingFootnote 26 and costly obligation of another marriage? What was it that made them take on the burden of an additional set of wedding gifts to provide; an additional wife to feed, clothe, and shelter; an additional household to maintain; and an additional imposition of in-laws with which to contend,Footnote 27 when it would be far simpler to purchase a slave or acquire a concubine?
The incentive does not appear to be a strictly practical or a purely economic one, as a second marriage would have required greater effort and incurred greater expenses. A moral incentive is also unlikely as Assyrian society did not regulate male sexuality to the extent that seeking sexual partners in slave-women would have been considered adulterous or wrong. A legal incentive is not too convincing either, as the regular course of action for a man needing heirs was to employ a slave-woman as a surrogate mother without having to change the latter's slave status, nor was a man whose casual relations with a slave-woman produced children was obliged to count them among his heirs (e.g. MAL §40, Roth Reference Roth and Chavalas2014: 166). In short, Assyrian men were not bound by any pressing economic reasons, social expectations or legal obligations to confine co-habitation, sexual intimacy, or even the production of heirs strictly to the context of marriage. If a truly compelling incentive for Assyrian men to enter second marriages is not evident within ordinary Assyrian socio-legal norms—which in fact offer far less burdensome means of resolving difficulties that may be associated with de facto bachelorhood—then explaining bigamie autorisée simply as an adaptive response to distance does not fully account for such an elaborate legal loophole. Even if our understanding of the Anatolian perspective on marriage is nebulous, we must nonetheless be explicit in acknowledging its potential role in the emergence of bigamie autorisée as a compromise catering both to Assyrian needs and Anatolian expectations.
Bigamie autorisée as middle ground
White (Reference White2011: xi) stresses that not all cross-cultural compromise qualifies as middle ground, for which mutual misunderstanding is an essential ingredient. The predicament of the historian who is “separated by two centuries or more from a native society that left few or no written records of its own” and must rely on only one side's descriptions (White Reference White2011: xxii–xxiii) is only confounded when reaching as far back in time as the kārum period. How can we identify genuine or wilful misunderstandings in a mixed society that existed over four thousand years ago, and judge the extent to which Assyrians and Anatolians understood the expectations of one another when arranging marriages?
The semantic shift in the term amtum, peculiar to the cross-cultural context of intermarriage between two distinct communities, constitutes a plausible candidate. This shift may simply be a case of retaining a convenient term despite a new meaning, or a deliberate reformulation of an otherwise outlandish practice in familiar terminology so as to make it (more) acceptable. On the whole, as Michel (Reference Michel2006: 164) asserts, the use of amtum as second-wife was part of “un vocabulaire spécifique pour définir le statut des épouses”. However, Michel (Reference Michel and Dercksen2008b: 214) has also observed that the same woman could be referred to both as an amtum and an aššatum in the same text.Footnote 28 This may simply be a case of carelessness. It might, however, also reflect a genuine confusion (or indeed misunderstanding) specifically relation to bigamie autorisée, a practice which has introduced a previously unattested category of “wife” but retained the term amtum, the semantic range of which has now expanded.
While the meaning of amtum as “slave/servant” is an accurate reflection of surrogate mothers in childless marriages, amtum-wives in Anatolia were neither purchased slaves nor subservient to aššatum-wives. It is significant that while the word amtum takes on a new meaning as second-wife in accordance with bigamie autorisée, it also retains its old meaning as slave-surrogate in the context of bigamie relative, which continued alongside and could indeed be merged with bigamie autorisée. In the specific context of bigamie autorisée then, the term becomes a misnomer obscuring the actuality of marriage arrangements involving two non-cohabiting wives. Similarly, the terminology of bigamie autorisée on the one hand masks the equal status between an aššatum- and an amtum-wife (who must be in separate households), while distorting, on the other hand, the otherwise very clear disequilibrium between an amtum-wife and an amtum-slave/surrogate (who are part of the same household). The element of potential misunderstanding in such a context is not only highly likely, possibly even deliberate.
Conclusions
Assyrian merchants arriving in Anatolia found themselves de facto bachelors who, especially in a strange country, would have sought out women as housekeepers, sexual partners, and presumably also in a childbearing capacity. Instead of acquiring the services of ordinary slave-women who could easily fit all these descriptions, we find Assyrian men entering second marriages. A new form of sanctioned bigamy is formulated, peculiar to those travelling between Aššur and Anatolia. Meanwhile the practice of introducing amtum-slaves as surrogate mothers into childless marriages (bigamie relative) continues in Anatolia, from which we can infer that: a) employing slave-women in this capacity did not contradict Anatolian social taboos or legal restrictions; and b) the role and status of an amtum-as-surrogate were distinct from those of an amtum-as-second-wife.
The simplest explanation is that Assyrians – more specifically first-generation Assyrian men – actively sought to enter marriages with local women as a means of establishing and/or strengthening ties with, and to gain acceptance and access into the local community with whom they were conducting business. As Lumsden (Reference Lumsden and Michel2008: 38) points out: “The social repercussions of intermarriage created connections that extended beyond the couple involved” (see also Michel Reference Michel2006: 174). Similarly, mixed marriages would have been of value to the Anatolian community, as a means of creating alternative means of interaction outside of but beneficial to business relations, by making the foreign familiar. Such motivation very much falls within the remit of the middle ground, which “grew according to the need of people to find a means, other than force, to gain cooperation or consent of foreigners” (White Reference White2011: 52).
In demonstrating the significance of preserving financial links between families, the stipulations of the marriage of Ilī-bani and Tataya are a case in point. Although punishable by a heavy fine, neither mistreatment on the part of the husband nor misbehaviour on the part of the wife were considered grounds for divorce. According to Veenhof (Reference Veenhof1997: 368):
[This] was not a normal marriage contract, but rather an agreement reached after or in connection with the marriage, with the aim of preventing by all means a divorce between the partners. Such an arrangement would be understandable in the light of [the related] texts [Kt 91/k 200] and [Kt 91/k 127], which reveal the existance [sic] of financial links between the father of the bride and his son-in-law.
Particularly where mixed marriages are concerned, we could expect a greater incentive for establishing and/or preserving such ties. Both for the Assyrian and Anatolian communities, close family ties would have afforded a competitive advantage not only in terms of increased business opportunities but also by way of consolidating one's interests through the support of allies. We may also speculate an alternative (but not mutually exclusive) explanation that entering marriages with Anatolian women was not simply a commercial expedience for Assyrian men, but perhaps more of a prerequisite. The kind of notional Anatolian identity to be acquired through marriage ties may well have been the crucial factor in creating a commercial common ground, in turn leading to a cultural middle ground.
On the one hand, our image of Assyrian merchants in Anatolia is that of freely intermarrying locals and on the whole adopting local material culture of everyday life; on the other hand we also find Assyrians expressing annoyance with compatriots for “constantly behaving like an Anatolian” (Larsen and Lassen Reference Larsen, Lassen, Kozuh, Henkelman, Jones and Woods2014: 177) or displaying “patriotic” preferences for names with the theophoric component “Aššur” over successive generations (Eidem Reference Eidem and Dercksen2004: 94–95). It is precisely at this nexus of blending in versus maintaining differences that intermarriage would have created a middle ground, simultaneously allowing Assyrians to become a familiar (and therefore non-threatening) part of Anatolian society, within which they could then pursue their business interests; and also to retain their foreign identity, which was closely linked to the monopoly on caravan trade. Maintaining one's position within the Assyrian network was equally important as establishing new ones in Anatolia, if for no other reason than the fact that Assyrian men relied on women in Aššur for the production and shipment of textiles. Stein (Reference Stein2008: 35) calls this a “distinctive diasporic identity” for Assyrians in Anatolia. Through the accommodation of two wives and two households, bigamie autorisée actively created a very particular middle ground for negotiating two identities: an Anatolian identity acquired by marriage, and an expatriate Assyrian identity of mercantile presence.
Individual circumstances presumably took precedence over collective ethnic identity, as we soon find Assyrian women emerging in texts as amtum-wives. Over successive generations of mixed marriages, women with mixed parents, variously bearing Anatolian or Assyrian names, are attested as the wives of Assyrian men (Dercksen Reference Dercksen and Zehnder2014: 199). Also potentially significant is that the (admittedly few) attestations of wives in period Ib texts are all of the aššatum kind, suggesting that perhaps as the two communities mixed, the terminology changed.Footnote 29 All this is perfectly in keeping with middle ground trajectory, according to which what is initially conceived as a compromise between the expectations of two different cultures (but not wholly congruous with the norms of either), eventually becomes part of these norms (White Reference White2011: 52).
Mixed marriages of the kārum period, which facilitated cultural encounters not only at a personal level, but also on one in which the wider community had a stake, certainly stand the test of the middle ground as a theoretical model. Arguably one of the greatest benefits of using the middle ground as a framework within which to think will be to consider Anatolian agency, participation, and interests, even in relationships for which we only have Assyrian accounts. Ultimately, however, the peculiarities of kārum period society itself, especially in terms of how it is manifested in the textual and the material record available to us, impose significant limitations on the extent to which the middle ground lends itself to Assyro-Anatolian contact, and we must remain circumspect in applying it. Testing external models on Assyro-Anatolian contact should not be the end but rather the means to reach an independent formulation with which to explain the idiosyncrasies of kārum period society on in its own terms.
