Abstract
In human societies the choice of a spouse is the outcome of individual mating choice and parental influence over mating. However, parents and offspring do not share identical interests with respect to mating. Accordingly, in order to understand this process it is necessary to examine the degree to which parents and offspring agree or disagree over mating choices. Trivers (1974) originally hypothesized that the two parties are in conflict over the family background of a mating candidate, with parents valuing this trait more than their offspring in a mating candidate. In this article this hypothesis is developed theoretically and empirically. It is proposed that that historically a preference for good family background increased the fitness of the parents more than the fitness of their offspring. Using a sample of 305 parents, the hypothesis is tested that good family background is preferred more in an in-law than in a spouse. The results from this study provide support for this hypothesis.
Introduction
Human mating behavior is seen as the outcome of a co-evolutionary process between male and female mating behavior (Buss, 2003); however, there is a third factor, namely parents, which exercises substantial pressure on the evolution of mating behavior (Apostolou, 2007a,b). In the majority of contemporary pre-industrial societies parents arrange their children's marriages by selecting in-laws according to their own preferences (Apostolou, 2007b; Broude and Greene, 1983; Frayser, 1985; Minturn, Grosse, and Haider, 1969; Whyte, 1978). In modern post-industrial societies parents employ “cajolery, persuasion, appeals to loyalty, and threats” to influence the mating choices of their offspring (Sussman, 1953, p. 80). As parental control over mating constitutes a significant component of the environment to which the offspring's behavior is adapted, the evolution of human mating behavior should be seen as a co-evolutionary process between parental inlaw choice and the offspring's mating choice (Apostolou, 2008). If this process is to be understood, it is necessary to identify the degree to which parents and offspring agree or disagree over mating decisions.
Parents and offspring are not genetically identical and as a consequence they do not share identical interests with respect to mating (Trivers, 1974). Actually, if both parties had identical interest there would be no reason for parents to strive to control the mating decisions of their offspring in the first place (Apostolou, 2007b). Exploration of parentoffspring conflict over mating has become the focus of recent research efforts (Apostolou, 2008; Buunk, Park, and Dubbs, 2008). The purpose of this study is to contribute to this line of research both theoretically and empirically.
Parent-offspring Conflict over Mating
Committing to a relationship with someone of lower mating quality to one's own is unlikely to be a successful strategy, as it equals forfeiting the opportunity to engage in a relationship with someone of a superior quality. For this reason, individuals engage in relationships with other individuals of a mating value similar to their own (Blossfeld and Timm, 2003). Consequently, parents and offspring are constrained by their own mating value in relation to the mating value of an individual they can attract (Apostolou, 2008). Mate choice then entails a trade-off, and parent-offspring conflict arises if one or more traits in a mating candidate offer asymmetrical benefits to each party. In this case, the party which benefits less from a given trait compromises more on this trait. This compromise inflicts a cost to the other party to which this trait is more beneficial resulting in conflict between the two.
The genetic quality of a mating candidate has been identified by previous research as giving unequal benefits to parents and offspring (Apostolou, 2007a): parents share 25% of their genes with their grandchildren, and thus, receive less fitness benefits from an individual of high genetic quality than their offspring, who share 50% of their genes with their children. Thus, although mating candidates with good genes are beneficial to both parents and offspring, they are more beneficial to the offspring than they are to their parents. Physical beauty is a proxy of genetic quality (Thornhill and Gangestad, 1993), so this trait is expected to be valued more by the offspring than by parents. Empirical research has confirmed this prediction (Apostolou, 2008; Buunk et al., 2008). However, parentoffspring conflict over mating is not confined only to beauty.
Parent-offspring Conflict over Family Background
Trivers (1974) originally hypothesized that parent-offspring conflict exists over the family background of a mating candidate with parents placing greater emphasis on this trait than their offspring: “Parents may also use an offspring's marriage to cement an alliance with an unrelated family or group, and insofar as such an alliance is beneficial to kin of the parent in addition to the offspring itself, parents are expected to encourage such marriages more often than the offspring would prefer” (Trivers, 1974, p. 261).
Parents benefit from having their offspring married into a good family, as this creates a new alliance or strengthens an existing one. Also, such a marriage translates into economical and political benefits for the parents, as they profit from the provision of resources in times of food shortages, support in fights and disputes etc. Similarly, it pays for the offspring to marry into a good family for the same reasons: they can establish a beneficial alliance and reap the financial and political benefits of it.
Nonetheless, preferences for this trait are not symmetrical and the reason why becomes apparent when mating choice is examined through an evolutionary perspective. More specifically, the typical pattern of mating in modern foraging societies is arranged marriage, where parents select spouses for their offspring, but the offspring are able to exercise mating choice within marriage (e.g., through extramarital relationships) (Apostolou, 2007b; Broude and Greene, 1983). Modern foraging societies resemble ancestral human societies which also based their social organization on hunting and gathering (Lee and De Vore, 1968). Thus, there are good reasons to believe that this pattern of mating was also typical during the period of human evolution (Apostolou, 2007b; Ember, 1978).
The way parents and offspring had been exercising mating and in-law choice would be reflected in their evolved preferences. To start with, in order to exercise better control over mating decisions, parents marry their offspring early, frequently when they are still children and dependent on them for food and protection (Apostolou, 2007b; Frayser, 1985). The consequence of this strategy is that the selection of in-laws takes place when the offspring and their prospective spouses are very young. As the in-laws are still not fully grown, parents cannot identify among prospective in-laws those with desirable traits, such as industry or working ability. Consequently, they have to rely on other cues, such as the in-laws' family background, to make a choice which is beneficial for them. On the other hand, the offspring exercise mating choice later in life when they and their prospective spouses are fully grown, so they can rely less on family background and more on other traits of their mating partner.
Moreover, in the context of arranged marriage one way for the offspring to exercise mating choice is through extramarital relationships over which parents have little, if any, control (Apostolou, 2007b). In this case the family background of a short-time mate becomes irrelevant. Such trait will bring no fitness benefits to the mate seeker who is more likely to be interested in other traits, such as genetic quality (Buss and Schmitt, 1993).
If these patterns of mating have been historically the case (and evidence points towards this direction), there should have been weak evolutionary pressure on the offspring, but strong evolutionary pressure on parents to evolve preferences for good family background in a mating candidate. This asymmetry in preferences is a source of conflict between parents and offspring.
In more detail, when parents choose an in-law, they look for an individual with a family background that maximizes their own fitness and not the fitness of their offspring. For instance, across different cultures, marriage is usually accompanied by a transfer of wealth from the parents of the groom to the parents of the bride through the institution of bridewealth (Murdock, 1981). Among the Manus of New Guinea, a father has arranged a marriage for his daughter and has received a substantial bridewealth from the parents of his son-in-law that intends to give to his nephew to “pay” for his own bride: “Her father is well pleased with the match. Ten thousand dogs' teeth will be paid, ten thousand dog's teeth which he can very well use to pay for a wife for his brother's son who is turned fifteen and unbetrothed” (Mead, 1963, p. 50). This arrangement is beneficial for the father who shares on average 25% of his genes with his nephew but it is half as beneficial for his daughter who shares on average 12.5% of her genes with her cousin.
The choice of an individual of a family background which maximizes the fitness of the parents is costly to the offspring because of the trade-off nature of mating: every one “unit” of good family background parents receive from their in-law increases their fitness more than it increases the fitness of their offspring. Therefore, parents will be willing to exchange more “units” of other traits for family background than their offspring consider it optimal for themselves. This imposes a cost to the offspring who are likely to find themselves married to a spouse with a mix of traits which does not maximize their fitness. For instance, the in-laws that parents have selected may not be as attractive as the offspring would like them to be and this loss in beauty is not compensated by their good family background.
Some empirical evidence provides support for the hypothesis that parents value family background in a mating candidate more than their offspring: Apostolou (2007a) introduced good family background in a set of traits that parents had to rate their desirability in a potential son-in-law and daughter-in-law. It was found that this trait ranked 8 out of 16 for sons-in-law and 6 out of 16 for daughters-in-law. Moreover, Apostolou (2008) found that individuals prefer similar religious background more in an in-law than in a spouse and interpreted this as evidence that parents value the family background of their inlaw more than their offspring. Finally, Buunk et al. (2008) found that individuals consider a mating candidate who does not come from a good family background more acceptable than they think their parents would.
The present study, by employing a sample of sexually mature participants who have children and can act both as parents and mate seekers, tests the hypothesis that individuals have evolved to prefer good family background more in an in-law than in a spouse.
Materials and Methods
Participants
A total of 305 participants, 172 women and 133 men, all British completed an online survey. A private company was employed specializing in recruiting participants for online research in psychology. Online responses have been found to be as reliable as laboratory-based responses (Kraut et al, 2004). The requirement for participating in the survey was to be a parent with at least one child. The majority of participants have two children (Table 1). All parents received payment for completing the survey in the form of credit (about $3) that can be used to purchase goods from online stores. Most participants were married (63.6%), followed by those in a serious relationship (21.3%), single (12.8%) and in a casual relationship (2.3%).
Demographic information
Materials
The survey consisted of two parts. In the first part, demographic data were collected: sex, age, nationality, marital status, number of daughters, and number of sons. Then, the following scenario was given: “You have two children, one male and one female, and you live in a society where marriages are arranged. It is your duty as a parent, through negotiations with other parents, to find an appropriate spouse for both your daughter and son” and then participants were asked to rate a set of characteristics in a son-in-law and a daughter-in-law. In the second part, participants were given the following scenario “You live in a society where marriages are not arranged and it is up to you to find a husband or a wife. You are not married yet but you think that the time has come for you to do so”, and then they were asked to rate a set of characteristics in a future husband or wife. The order of presentation of the two scenarios was randomized between participants. Moreover, in each part the participants were asked to indicate the social status of the family they or their offspring would like to marry into: “same social status, higher social status, lower social status, no preference”.
The instrument employed to measure mating preferences is an extended version of a previous instrument developed by Buss and Barnes (1986) which in turn was based on a larger instrument developed by Gough (1973) to study family planning. Additional items were added to test specific hypotheses and the instrument in its final form included 18 items (Table 2). More specifically, four items were added so as to measure family background preferences: similar racial background, good family background, good family reputation, wealthy family. The selection of the last three items was based on the ethnographic literature of pre-industrial societies. In particular, good family background and good family reputation are frequently reported to be an important consideration for the parents when they are considering an in-law (e.g., the Nez Perce: Walker, 1998). Furthermore, wealth is also correlated with social status and there are indications that parents are interested in the resources of an in-law's family (e.g., the Kipsigis: Borgerhoff Mulder, 1988). Finally, similar racial background item was added because race correlates with a family's social status (Goode, 1964). Each item was rated in the following five-point Likert scale: +2 = very desirable; +1 = somewhat desirable; 0 = inconsequential, or neutral;−1 = somewhat undesirable;−2 = very undesirable.
Means and ranks of preferences concerning potential in-laws
indicates a significant difference at p < .003 in the ratings between daughter-in-law and son-in-law.
Missing values accounted for no more than 2% of the responses in a given item. The missing values were replaced by the mean of the series.
In-law vs. Mating Preferences Comparisons
Previous studies indicate that the ratings of desired traits in a prospective daughterin-law and son-in-law are contingent upon the sex of the spouse and the sex of the in-law (Apostolou, 2007a; Buss, 2003). Consequently, if comparisons are to be made between inlaws and spouses, this will result in information loss with respect to measuring differences in ratings. Thus, a within-participants design was employed instead, and the ratings of the female participants for a son-in-law were compared with the respective ratings for a husband, and the ratings of the male participants for a daughter-in-law were compared with the respective ratings for a wife. Also, comparisons between wife and son-in-law and between husband and daughter-in-law were not made, since the results from this analysis would fail to distinguish differences in ratings between sexes (men versus women) from differences in ratings between spouses versus in-laws.
Results
Parental Preferences
Not all parents in the sample had children of both sexes, so it should be examined whether the sex of the offspring affects parents' in-law ratings. This possibility was tested with a series of ANOVA and Chi-square tests. Bonferroni correction for alpha inflation was applied by decreasing alpha from.05 to.003 (.05/18), two-tailed. The results indicate that the ratings of the parents are not affected by the sex of their children. Moreover, to test whether male and female parents differ in their in-law preferences, a series of 2×2 mixed ANOVAs was conducted with the sex of the in-law as a within-subjects factor and the sex of the parent as a between-subjects factor on every item of the instrument. As before, Bonferroni correction was applied. It was found that in-law preferences are not contingent upon the sex of the parent.
To examine the relative importance that parents attach to the specific items, these items were ranked according to their means (Table 2). With respect to the family background items, we see that good family background and good family reputation rank roughly in the middle of the parental preferences for both in-laws. However, these items rank higher for a son-in-law than for a daughter in-law. The wealthy family item ranks low, at the bottom of the in-law preferences hierarchy. This is also the case for the similar racial background item.
For their daughter, 29.8% of the parents indicated that they would like her to marry to a family of the same social status, 9.8% to a family of a higher social status, 0% to a family of lower social status, and 60.3% indicated no preference. For their son, 30.8% of the parents indicated that they would like him to marry to a family of the same social status, 5.6% to a family of higher social status, 0.3% to a family of lower social status and 63.3% indicated no preference.
To investigate whether parental preferences are contingent upon the sex of the inlaw, a series of paired-samples t-tests were conducted between the rating of the son-in-law and the daughter-in-law. Ratings differ significantly between in-laws in various dimensions (Table 2). Physical attractiveness is preferred more in a daughter-in-law than a son-in-law, while good family background is preferred more in a son-in-law than a daughter-in-law (Table 4). Additionally, in-law preferences with respect to the social status of the family do not depend upon the sex of the in-law.
Means and ranks of preferences concerning potential husbands or wives
indicates a significant difference at p < .003 in the ratings between husband and wife.
Parent-offspring Conflict and Agreement
In order to identify whether in-law and mating preferences differ, the ratings of the female participants for a son-in-law were compared with the respective ratings for a husband, and the ratings of the male participants for a daughter-in-law were compared with the respective ratings for a wife. A series of paired-sample t-tests was conducted on each item of the instrument. Furthermore, to identify differences in preferences between husband and wife independent samples t-tests were conducted on each item of the instrument with the sex of the participant as a grouping variable. Also, Marginal Homogeneity and Chisquare tests were applied on the family status item. Bonferroni correction for alpha inflation was applied as before.
In comparing the ratings of a son-in-law to a husband, significant differences were found in a number of variables: physically attractive [t(171) =−5.54, p <.001, (two-tailed), ηp2 =.151] is preferred more in a husband than in a son-in-law; good family background [t(171) = 4.59, p <.001, (two-tailed), ηp2 =.110] is preferred more in a son-in-law than in a husband; good family reputation [t(171) = 4.57, p <.001, (two-tailed), ηp2 =.109] is preferred more in a son-in-law than in a husband; easy going [t(171) =−5.75, p <.001, (two-tailed), ηp2 =.162] is preferred more in a husband than in a son-in-law; religious [t(171) = 3.33, p <.001, (two-tailed), ηp2 =.061] is considered less undesirable in a son-inlaw than in a husband, and exciting personality [t(171) =−5.43, p <.001, (two-tailed), ηp2 =.147] is preferred more in a husband than in a son-in-law.
Comparisons between the ratings of a daughter-in-law and a wife, indicated significant differences in a number of traits: exciting personality [t(132) =−5.82, p <.001, (two-tailed), ηp2 =.204] is preferred more in a wife than in a daughter-in-law; intelligent [t(132) =−4.30, p <.001, (two-tailed), ηp2 =.123] is preferred more in a wife than in a daughter-in-law; physically attractive [t(132) =−3.85, p <.001, (two-tailed), ηp2 =.101] is preferred more in a wife than in a daughter-in-law; and good family reputation [t(132) = 3.80, p <.001, (two-tailed), ηp2 =.099] is preferred more in a daughter-in-law than in a wife. Good family background is preferred more in a daughter-in-law than in a wife, however this difference passes the significance level only when Bonferroni correction is not applied [t(132) = 2.28, p <.05, (two-tailed), ηp2 =.038]. The results from the comparisons between spouses and in-laws are summarized in Table 4.
With respect to the family's social status item, 27.1% of the male participants indicated a preference for a wife from a family of equal social status, 6.8% of higher social status, 0.8% of lower social status, and 65.4% indicated no preference. Similarly, 32.6% of the female participants indicated a preference for a husband from a family of equal social status, 6.4% of higher social status, and 61% indicated no preference. Chi-square test indicates independence between the ratings of the family social status and the sex of the parent, whereas Marginal Homogeneity test indicates independence between the ratings for this item and the role of the participant (mate seeker or parent). In other words, preferences for a family's social status are the same irrespectively of whether individuals are parents or mate seekers, and irrespectively of their sex.
A possible limitation of the within-participants design employed in this study is that all participants had to be parents; so inevitably, the mean age of participants in the sample is not representative of the age of an offspring at first marriage. If the traits on which parents and offspring do not agree are valued differently at younger ages, the true extent of the parent-offspring conflict may be overrated or underrated here. To examine whether this is the case, good family background was regressed on age for both sexes. This trait came significant only for women [t(171) = 2.29, p <.05, (two-tailed)] with a coefficient of.011. Finally, age was regressed on the good family reputation item for both sexes. As before, this trait came significant only for women [t(171) = 1.99, p <.05, (two-tailed)] with a coefficient of.010. This is an indication that the preference for family background becomes stronger as women grow older. As women in this sample are older than women at marriage age parent-female offspring conflict over family background may be stronger than it appears to be here.
Significant Son vs. Daughter-in-law and In-law vs. Mating Partner Differences
The effect size is indicated here by η p 2 which is the proportion of total variability attributable to the independent factor.
Discussion
The results from this study support the hypothesis that good family background is preferred more in an in-law than in a spouse. It is also found that individuals prefer both their in-laws and spouses to be from families of similar social status to their own. Additionally, and consistent with the findings of previous studies (Apostolou, 2008; Buunk et al., 2008), participants prefer good looks more in a spouse than in an in-law.
Furthermore, participants attach little weight to the similarity of racial background, irrespectively of whether they act as parents or mate seekers. This result comes in contradiction with studies which have revealed the importance of similarity of race in spouse selection (e.g., Liu, Campbell, and Condie, 1995; Warren, 1970). A possible explanation for this discrepancy may be that race is a sensitive matter and a self-report questionnaire may not be appropriate for measuring preferences with respect to this trait.
Moreover, differences between in-law and mating preferences were also found for other traits, such as “easygoing” and “exciting personality” with these properties being preferred significantly more in a husband and a wife than in an in-law. No specific theory is offered that accounts for these results. A plausible hypothesis may be that, as it is the offspring and not the parents who share a lifetime with their spouses, it makes sense for the former to be more sensitive to traits that facilitate co-habitation and co-operation.
Buunk et al. (2008) offer a different explanation for the asymmetry in preferences for family background. More specifically, they argued that in a mating trade-off, which involves an investing mate versus a mate with good genes, parents' interests are served better if their offspring have highly investing mates than mates of superior genetic quality. This is because parents would get fewer benefits from the genetic quality of their in-laws so they would not like to have as much of this trait as their offspring. Family background is a proxy of an individual's ability to invest, so parents would tend to prefer this trait more than their offspring, since such preference will enable them to make a more optimal tradeoff. However, from this argument follows that parents value good genes less, but not necessarily that they value family background more than their offspring. Also, this argument implies that parents should prefer qualities which are directly associated with an individual's ability to invest more than their offspring prefer them in a mating candidate. This prediction is inconsistent with the findings of this and a previous study (Apostolou, 2008), which indicated that individuals value traits such as “good earning capacity”, “hard working”, “ambition and industriousness” the same in a spouse and in an in-law.
Additionally, Buunk et al. (2008) argued that the good family background of a mating candidate is more beneficial to parents because they can establish an alliance and/or increase their social status through the marriage of their children. Nonetheless, the offspring may be as sensitive to the family background of their spouses for the same reasons: they may want to establish an alliance which is beneficial to them and/or increase their own social status.
A limitation of this study is that participants had to rate traits on the basis of a hypothetical scenario. Furthermore, the same participants were employed to rate traits in a spouse and in an in-law. This gives additional value to the findings presented here as the same individuals differentiate their preferences according to whether they act as parents or mate seekers. However, the within participants design may mask the true extent of the parent-offspring disagreement, as in this design in-law and mating preferences are likely to appear as more similar than they really are. A future study which compares the ratings of parents with those of their offspring can be more revealing of the true extent of the parentoffspring conflict over mating.
As discussed before there are good reasons to believe that parents in ancestral societies had much more control over the mating of their offspring than they have today in Western societies. If this was the case, the ancestral mating market would have been different from the modern mating market, in the sense that in ancestral times the mating market would primarily reflect the parents' preferences, while today's Western mating market primarily reflects the offspring's preferences. By identifying the differences in the preferences between parents and offspring, the present research, enables us to make certain inferences with respect to ancestral mating market conditions.
More specifically, in the ancestral mating market beauty in a mating candidate would have been in less demand and good family background in higher demand. This has important implications for the evolution of human mating behavior. For instance, in our ancestral past there was less pressure on individual mate seekers to place emphasis on good looks, as selection was based less on this trait. The weaning of parental control in modern societies has increased the demand for beauty, as mate selection is now based more on this trait. This places greater pressure on individuals to improve their appearance, which in turn may be reflected in the higher demand for artificial beauty (e.g., beauty products, plastic surgery, etc.), as individuals attempt to adjust to the novel mating market conditions.
It should be clarified here that this argument does not imply that ancestral populations were indifferent to their looks. The ethnographic literature is rich with examples of individuals in traditional societies who employ various methods for improving their physical appearance (e.g., the Twana of North America, Elmendorf and Kroeber, 1992). However, what is argued here is that an effort to improve one's appearance in a traditional society, where parents are influential or even in control of their offspring's mating choice, results into less reproductive benefits than an effort to improve one's appearance in a postindustrial society, where parents have less or no influence on their offspring's mating choice.
Parents and offspring do not have identical preferences with respect to mating choice and this is a source of conflict between the two. To understand this conflict, and consequently the evolution of human mating, it is necessary to identify the specific areas of disagreement between the two parties. The present study provides evidence that parents and offspring disagree over the family background of a mating candidate. This is not the only source of conflict over mating, and further research is needed to shed light to this fascinating aspect of human behavior.
Footnotes
Acknowledgements
I would like to thank Lia Mexa, Evangelos Aspiotis and Maria Doritou for their help during the preparation of this manuscript. I would also like to thank Viren Swami and two anonymous reviewers for their constructive comments and criticism.