Friday, November 09, 2007

The evolution & economics of human mate choice   posted by Razib @ 11/09/2007 11:58:00 PM

Tyler Cowen points and interest working paper, What's Love Got To Do With It?
Parental Involvement and Spouse Choice in Urban India
...when parents are involved in mate choice, sons are significantly less likely to marry college-educated women and women engaged in the labor force, after controlling for individual and family characteristics. I show that these effects are driven, at least in part, by parental preferences and cannot be entirely attributed to correlation between arranged marriages and unobserved characteristics or preferences.....

To cut the chase the paper suggests that parents prefer that their sons marry someone who is in a relatively weak bargaining position in terms of allocation of economic resources in the future. The rationale from a parental perspective is that as they age they will depend upon their son's household to provide financial support when their own earning power diminishes (i.e., familial social security). Though a daughter-in-law who is an economically productive professional increases the resources which might be allocated toward her in-laws, the concomitant independence is judged to more likely result in resistance to intergenerational transfers of wealth.

The author focuses upon upper-middle class families in Bombay in her survey, as presumably they are are the cultural "cutting edge" and more likely to manifest the noticeable shift toward "love matches" within Indian society. She notes that arranged marriages are ubiquitous amongst the lower classes and less economically advanced. But I have to wonder, how recent is the ubiquity of this particular cultural practice within South Asia? A survey of the ethnographic literature will show that within the past few centuries (even within the past century) "elite emulation" has been a noticeable dynamic which has transformed Indian society. Within the Hindu context there has also been a great deal of "Sanskritization," an attempt to shift away from local customs and tradition and conform to upper caste practices which often have a North Indian provenance. For example, in South India the Nair caste shifted away from their customary matrilineal & matrifocal systems of inheritance and residence only within the past few centuries (the literature makes clear that other Hindu groups found Nair practices abhorrent). The custom of dowry is often considered part and parcel of Indian culture, but there is a good deal of evidence that many non-upper caste groups have only begun to practice it in the 20th century, prior to which they adhered to brideprice if they engaged in wealth transfer at all. Similar shifts have occurred in terms of the consumption of beef and meat-eating in general.

In the data above arranged marriage is associated with multi-generational extended family households. These marriages are seen as an alliance between families and a way to advance material interests through connections with other lineages of similar power and status, at least on the conscious level (note that the data above implies that the parents are also behaving as rational actors maximizing their own security even if it reduces the greater glory of the family). From a Western perspective the dynamics operant upon European elites until recently serve as a good analogy. Napoleon married the daughter of the Hapsburg emperor because he perceived that the best way for a parvenu soldier to secure his status among the royal families of Europe was to marry a member of one of the oldest aristocratic lineages of note. In return the Austro-Hungarian empire had hope that they'd won a measure of peace and security by cementing an alliance with the conqueror of much of the continent. Though such world-historical implications are not operative in the typical arranged marriage, quite often attention to material wealth, status and the positives & negatives of the possible inter-familial association are considered in great detail. These are the explicit raison de'tre of the marriage.

With that in mind I think that when looking at arranged marriages in the long-view it seems that the relatively rigid form with little input from the principals must have emerged relatively recently in human evolutionary history; likely within the last 10,000 years due to the rise of "civilization" and the accumulation of power, status and wealth among specific lineages within a locale. In many societies, such as early modern Europe and Japan, an arranged marriage was a sign that the familial context was elite. For the typical peasant the pool of potential partners was likely small because of constraints of geography, a man and woman were economic partners in maintaining a lifestyle above the margins of subsistence. Norms and values promoted by elites are certainly extant within the literature which captures the Zeitgeist of any particular age, but until humanity broke out of the Malthusian Trap these values were irrelevant for the overwhelming majority of the population. The current situation in Indian society, where there are competing cultural ideals and the economic wherewithal to engage in a range of choices, is atypical and characteristic of a culture in transition. A wave of elite emulation by the lower orders within Indian society within the past few decades is manifest evidence of some level of economic prosperity, after all dowries are a major burden, while curtailing consumption of potential sources of protein due to concerns about pollution and loss of status also suggest a robust enough diet where such choices are an option (there's actually a good deal of evidence that in some regions of India Hindus converted to Islam simply because they had already lost caste by the consumption of polluting foods, either inadvertently or under duress). The particular dynamics fleshed out above might also not be the norm in the historical scale, as mass adoption of the practice below the highest elites might be relatively recent (social security would have been less of a concern for the extremely elite in most cultures who were safeguarded by rents).

Prior to the Reformation the Catholic Church had generally increased the choice that potential spouses had in regards to their partners and also banned moderately consanguineous relationships (e.g., first cousins). After the Reformation elite lineages began to engage in cousin marriage, often as a way of consolidating property, concentrating wealth and cementing familial bonds. There was also marked tendency toward increased parental control over the marriages of their offspring. These dynamics show that the pent up tendency was there, only constrained by the fiat of the Catholic Church (marriage was a sacrament under religious control, and marriages which produced illegitimate issue could be useless in terms of perpetuating the family legacy). Yet neither of these are major issues for most of Protestant Europe as cultural norms have evolved and shifted naturally. In the case of India as the upper and upper-middle classes continue to grow in wealth it seems likely that norms will naturally evolve away from parental control toward individual choice. As the elites abandon practices such as arranged marriage and dowry once their group-level utility decreases one assumes that the lower orders will emulate them quickly, in particular since the gains from some of the practices (e.g., consolidation of wealth via arranged marriage) are minimal and the losses from others are more significant (e.g., debts incurred due to dowries or more marginal nutrition because of constrainment of dietary sources of protein).

An interesting final point is that the paper above notes that there is some positive correlation between arranged marriage and caste, religious and ethnic endogamy vis-a-vis a love marriage, while there is some disassortative mating along the axis of educational attainment (i.e., sons with a college education are less likely to marry a woman of the same educational status). I have often wondered as to the effect of the changes in the cultural environment in regards to assortative mating and the heritability of particular traits (assortative mating often increases within population heritability). The classic example is the fact that most female physicists are married to other physicists. This is a function of the lopsided sex ratio, but in the early 20th century obviously this wouldn't have been a nearly irrelevant fact because there were hardly any female physics Ph.D.s. A pooling of extremely intelligent individuals at elite universities or graduate schools from which they select their partners is a radical change; after all, higher education for women is a relatively new phenomenon. One assumes that the likelihood is that most men with higher educations would find a wife from the same social order in the past, but because of within family variation in propensities I assume that the assortative mating was less pronounced (fathers & brothers are imperfect proxies). An economist may see in love marriages a more efficient allocation of human capital, but I can't but help wonder as to the possibility that the enormous sample space of mates of like mind and character is a more efficient across population genetic architecture (at least for the purposes of increasing the variance of trait value).

Note: The popularity of tales of tragic love attest to the recent origin and shallow extent of arranged marriage where the offspring have little choice. The lack of evolution of the underlying psychology suggests little selective pressure.