social aggression and resource conﬂict across the female

AGGRESSIVE BEHAVIORVolume 38, pages 194–207 (2012)

Social Aggression and Resource Conflict Across theFemale Life-Course in the Bolivian AmazonStacey L. Rucas1∗, Michael Gurven2, Jeffrey Winking3, and Hillard Kaplan4

1Department of Social Sciences, California Polytechnic State University, San Luis Obispo, California2 Integrative Anthropological Sciences Program, Department of Anthropology, University of California Santa Barbara, SantaBarbara, California3Department of Anthropology, Texas A&M University, College Station, Texas4Human Evolutionary and Behavior Science, University of New Mexico, Albuquerque, New Mexico

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :

This work explores sources of conflict among forager-horticulturalist women in Amazonian Bolivia, and applies life history theoryas a tool for understanding competitive and cooperative social networking behaviors among women. In this study, 121 Tsimanewomen and girls were interviewed regarding current and past disagreements with others in their community to identify categories ofcontested resources that instigate interpersonal conflicts, often resulting in incidences of social aggression. Analysis of frequency dataon quarrels (N = 334) reveals that women target several diverse categories of resources, with social types appearing as frequentlyas food and mates. It was also found that the focus of women’s competition changes throughout the life-course, consistent with thenotion that current vs. future reproduction and quantity-quality trade-offs might have different influences on competition and socialconflict over resources within women’s social networks across different age groups. Aggr. Behav. 38:194–207, 2012. C© 2012 WileyPeriodicals, Inc.

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :

Keywords: female-female competition; social networks; resource competition; reproductive success; social aggression

INTRODUCTION

The proposal that female reproductive success islimited more by competitive access to resources, whilemale reproductive success is limited more by compet-itive access to females, is a fundamental concept inthe evolutionary biology of mammals. The sexes varyconsiderably in costs of reproduction and potentialfitness benefits to additional matings [Parker, Bakerand Smith, 1972; Trivers, 1972]. The substantial dis-parity in parental investment, with greater contribu-tions by females, creates conflict between and withinthe sexes, contributing to debates surrounding opti-mal investment of resources in quantity vs. qualityof offspring [Hagen, Barrett and Price, 2006], par-enting decisions [Borgerhoff Mulder, 1992; Daly andWilson, 1985], and sex differences in mating strate-gies and relationship conflicts (Clutton-Brock, 1994;Gangestad and Simpson, 2000; Schmitt, Shackelfordand Buss, 2001]. Marital conflicts often reflect dis-parities in the optimal use of resources, where the“optimal” allocation may often differ among spouses[Gurven, Winking, Kaplan, von Rueden and McAl-lister, 2009; Lundberg and Pollak, 1993; Manser andBrown, 1980; Stieglitz, Kaplan, Gurven, Winking and

Vie Tayo, 2011]. Within the evolutionary sciences, theresources that impact human female reproductive suc-cess, which are relevant to these debates, usually fo-cus only on food or mates, with less attention be-ing given to women’s conflicts over social resources.And most work in the social sciences that has fo-cused on women’s social dynamics has not employedevolutionary theory to investigate the complex phe-nomena of resource competition, conflict, and co-operation among women. Additionally, research insmall-scale societies on women’s social networks andconflicts is rather sparse. This work seeks to identifythe resources over which women and girls compete viasocial aggression and how the dynamics and sourcesof competition change over the life-course, and to

Contract grant sponsor: LAII Field Research Grants; Contract grantsponsor: Tinker Foundation; Contract grant sponsor: NSF Grant;Contract grant number: BCS-0136274∗Correspondence to: Stacey L. Rucas, Department of Social Sciences,California Polytechnic State University, San Luis Obispo, CA 93407.E-mail: [email protected]

Received 30 September 2010; Accepted 05 December 2011

Published online in Wiley Online Library (wileyonlinelibrary.com).DOI: 10.1002/ab.21420

C© 2012 Wiley Periodicals, Inc.

Social Aggression and Resource Conflict 195

use evolutionary logic to shed light on this complexsubject. We test several hypotheses derived from evo-lutionary theory, as well as discuss the implications ofthe findings on a diverse range of theoretical areas.

The limited availability of resources instigates com-petition, fueling the evolution of social strategies de-signed to resolve conflicts over assets in one’s favor.Social aggression by quarreling, gossiping or “infor-mational warfare” [Hess, 2006] is a common forumfor competition among women. Quarreling amongwomen is costly in several ways; (1) It takes time awayfrom other activities, (2) Injured parties might retal-iate, and (3) Participation could negatively influenceone’s social standing. Thus, women should only bewilling to quarrel when they stand to gain somethingconcrete (e.g., resources, information) or when it isreputationally advantageous. This current work ad-dresses gaps in evolutionary analysis of the combinedresources that women target for competition and ex-plores the trade-offs of focus on each throughout thelife-course. A woman’s reproductive value, need fora long-term partner, and the number of child depen-dents she has are expected to vary over a woman’s life-course, and hence the type and quantity of resourcesneeded throughout her life should vary accordingly.Thus, we first expect that age-mates will be frequentcompetitors for similar resources. Second, we expectthat women will focus attention on quality long-termmate acquisition when young, because the benefitsof a provisioning long-term partner are cumulative.We also expect that conflict over social resources (in-formal social contracts and friendships) will increasewith age and dependency load due to the increasingdemands of children, which cause women to increas-ingly seek resources outside of the pair bond in orderto satisfy greater reproductive demands. We expect tosee a similar food competition trajectory for similarreasons. And finally, we expect mate-retention con-flicts to be generally focused at younger ages, becauseprevious work in less acculturated Tsimane villagesindicates that men tend to target extramarital rela-tionships when they are younger with fewer depen-dents to risk [Winking, Kaplan, Gurven and Rucas,2007a], and because women’s reproductive value de-creases with age, indicating that younger women willattract more male attention, which in other work hasbeen shown to correlate with significant spousal con-flict and violence [Stieglitz, Gurven and Kaplan, inpress; Stieglitz, Kaplan, et al., 2011].

Resource: Food and Food-Sharing Contracts

Food sharing, a core feature present in all huntingand gathering groups, contributes to the uniqueness

of human social systems and is an essential compo-nent in the evolution of our species [Gurven, Allen-Arave, Hill and Hurtado, 2000; Gurven, Hill andKaplan, 2002; Kaplan and Hill, 1985; Lancaster,1978]. Food sharing lowers the probability of dailyfood shortfalls when success varies within and amongindividuals; however, due to the potential benefits ofwithholding shares (i.e., cheating) or under-producing(i.e., slacking), individuals may not reciprocate inthe expected culturally appropriate manner, result-ing in social aggression via gossiping and quarreling[Gurven, 2006]. The unenforceable nature of infor-mal social contracts creates a demand for honest,trustworthy cooperators. Because of this, we expectthat a significant percentage of female-female quar-rels will concern accusations of food theft, stinginess,and other indications of a failure to meet expectationsgoverning social exchange. The study of quarrels overfood-sharing contract defections can shed light on therelevance of different foods that people acquire andshare. And indeed, were such quarrels to be discov-ered, support would be found for models of reciprocalaltruism; whereas, a lack of such quarrels would bemore consistent with tolerated scrounging or signal-ing [Bliege Bird and Bird, 1997; Blurton Jones, 1987;Gurven, 2004]. Tolerated scrounging predicts shar-ing when the costs of resource defense outweigh thebenefits of hoarding, while signaling predicts sharingto others to advertise information about cooperativeintent or phenotypic quality. These two models donot require contingency between giving and receiving,and therefore are not consistent with complaints orresentments over failed expectations to receive returnshares by specific parties. While it may be interpretedthat a lack of food sharing quarrels could indicate per-fect reciprocity, we believe this is unlikely since foodresources are often scarce, information about others’motives and abilities to produce food is imperfect, andconflicts of interest are likely given the large range ofoptions for sharing resources.

Resource: Material Goods

Human developmental needs are not limited tocalories but include protection, language instruction,good health, mating opportunities, and a varietyof significant social, behavioral, psychological, andphysical skills that require years of teaching and learn-ing accomplished only through instruction, example,exposure, and practice. Food alone cannot supply suf-ficient “resources” in order to maximize female repro-ductive success or quality of life and family. Complexcultural technologies have been utilized by humansfor hundreds of thousands of years to solve problemsof food acquisition, for protection from the elements,

Aggr. Behav.

196 Rucas et al.

predators and microorganisms, and have infiltratednearly every aspect of daily life. Thus, we expect thatmaterial resources, which may greatly enhance condi-tion and reputation, will also be under competitionamong women and result in quarrels. For instance,it may not be obvious that a spoon could increasesurvivorship, but in communal food-eating house-holds, we observed that cooked meals were often-times rapidly consumed from a single pot while stillvery hot. Spoons might assist in getting more foodwithout the risk of getting burned. Additionally, suchimplements are very helpful with food processing andcooking, and have great longevity, saving individualsthe time it would have taken to repeatedly make a lessdurable utensil.

Resource: Men

Male friends, short-term mates, and spouses canprovide numerous resources and protection as wellas serve as long-term cooperative partners. Men areparticularly inclined to such commitment when theyare attempting to win favor or when they share re-productive interests with a woman. Typically, womenattempt to secure several positive traits in male socialand romantic partners, including good genes, ability,and willingness to invest, status, prestige, intelligence,and fidelity—traits that tend to increase the magni-tude, quality, and reliability of investment in currentand future offspring. Courting male friends and ro-mantic partners, assisting in male-male coalitionalconflicts, and harassing other females for their malecompanions are complex social behaviors, intended toincrease the probability of winning high-quality part-ners, thus imparting fitness benefits to self and off-spring [Hooks and Green, 1993; Hrdy, 1999; Mesnick,1997]. Men vary in what they can and are willing to of-fer partners, and competing to secure a higher qualitypartner can often have long-term payoffs, particularlyif it influences a man’s marital choice. For these rea-sons, we predict the acquisition of mates and mate re-tention to cause considerable conflict among women,fueling intense intrasexual arguments. We expect con-flict over mate acquisition to peak in frequency andintensity during a woman’s teenage years, since this iswhen unmarried women in most natural fertility so-cieties first marry. However, marriages of young girlsare often arranged or at least influenced by relatives[Gurven et al., 2009], and courtship periods may bebrief. In the Tsimane context, women may have lesstime to evaluate, select, and contend with competitorsin the game of mate acquisition due to an earlier age ofmarriage and first-birth marriage compared to West-ern populations. We also expect quarrels concerningmate retention to occur early in women’s marriages,

due to previous work demonstrating that most ofTsimane men’s extramarital affairs (a primary, cross-cultural cause of relationship discord) occurred in thefirst five years of marriage, when offspring depen-dency is low [Winking et al., 2007a].

Resource: Social

Parenting assistance may be offered in the form ofallocaregivers who reduce the time and effort requiredby the mother to protect, feed, and socialize depen-dent young, resulting in low interbirth intervals andhigher total fertility [Hrdy, 2005]. Friends may alsooffer help in the form of cooperative tasks, produc-tive activities, and cultural or social information. Ev-idence among baboons indicates that social bonds,measured by grooming and frequency bouts, posi-tively impact offspring survivorship [Silk, Alberts andAltmann, 2003]. Therefore, understanding the impor-tance of friendships among women is of great conse-quence, due to the fact that helpers may significantlyincrease the health and reproductive success of oneselfand one’s allies through a wide variety of pathways.

Humans pay enormous time costs to developingand maintaining relationships with others and secur-ing positions within social communities. Social group-ing provides benefits, however, such as protectionfrom conspecifics and predators, increased food shar-ing, and benefits from economies of scale for variousproductive activities, that must outweigh these costs.The complexity involved with optimizing these bene-fits among equally matched competitors and allies hasfrequently been cited as a major evolutionary pres-sure for increases in brain size, language acquisition,and social intelligence [Barton and Dunbar, 1997;Dunbar, 1992, 1993, 1997]. For example, economicproduction of lactating women is seriously com-promised among hunter-gatherers [Hurtado, Hill,Kaplan and Hurtado, 1992; Marlowe, 2003], and sowomen must rely on spouses, kin support, friends,social ties, and reputation in order to obtain enoughfood. While men reproductively benefit from invest-ment in offspring independent of the need for mater-nal persuasion, social skills, and influence by sociallyintelligent women may also influence the distributionof male-acquired resources. Social intelligence assistswomen in their reproductive struggle by increasing ac-cess to food and other necessities via extended socialnetworks and by decreasing mortality risk throughprotection from predators and other conspecifics. Forthese reasons, we hypothesize that social network sizeand quality, coalitionary allies, and associated socialcontracts with sharing partners (otherwise combinedgenerally into “social resources”) will instigate com-petition and quarreling due to their strong fitness

Aggr. Behav.


potential. Contextual examples of such a quarrelmight involve situations that revolve around distillinginformation about how one’s friend might be with-holding returns of shared foods, or instances wherewomen quarrel with others with whom they share noobvious conflict except in the support of an injuredfriend. And because reproductive resource needs in-crease with the number of dependent children, it wasalso predicted that the number of quarrels over thesesocial resources would increase with dependency loadand age.

General Predictions

We predict that women at similar life stages will becompeting over comparable resources because suchwomen face similar trade-offs, and share analogousdependency loads and relationship statuses. Thus,smaller age differences between women should predictgreater intensity of resource competition, resulting inan overall greater number of within-pair quarrels, andbe represented by a higher probability of engaging inany within-pair resource quarrel.

Previous work indicates that kinship and prox-imity between residences affect friendship, levels ofcompetitiveness, and cooperative coalition formation.Women are often inclined to take resources from closekin and helpers, and they often derogate their femalekin on varying measures [Rucas, Gurven, Kaplan andWinking, 2010; Rucas et al., 2006]. Tsimane womenalso prefer, when possible, to live near helpful femalekin. Therefore, we expect that the probability of quar-reling over resources increases with kinship (proxiedby the coefficient of genetic relatedness) and closerresidential proximity. These two variables togetherimpact the frequency of daily interaction, and all elseequal, quarrels should rise in proportion to interac-tion frequency. Because of this, kinship and distancebetween houses will be used as control variables sothat resource competition can be measured indepen-dently of these other influences.

To our knowledge, no work among women in tra-ditional populations has yet explored the relative re-productive importance of various resource types asmeasured by their potential to cause conflict amongwomen of different ages. However, we recognize thatseveral untested assumptions may be drawn from thisperspective, since many other detailed factors affectthe degree of competition, other than reproductive fit-ness, such as cultural values or effectiveness of compe-tition. Still this work seeks to introduce an evolution-ary lens to examine resource conflict among womenand explore how resource focus may fluctuate in orderto better optimize lifetime fertility.

METHODS

The Tsimane

The Tsimane are an indigenous group of roughly10,000 forager-horticulturalists inhabiting the tropi-cal areas of lowland Bolivia along the Maniqui River[Winking, Kaplan, Gurven and Rucas, 2007b]. Theylive in approximately 90 semi-sedentary villages rang-ing in size from 30–550 persons. They engage in acombination of horticulture, hunting, fishing, andgathering subsistence strategies [Godoy et al., 2004].Horticultural products include plantains, rice, sweetmanioc, and corn. Women spend much of their dayengaged in childcare, cooking, food preparation, andgarden labor. Extended family units typically live inhousehold clusters, with women and girls of varyingages represented at each residential unit, which typi-cally includes a mixture of consanguineous and affinalkinswomen. Whereas some females have many close-blood kin present, others live in clusters as in-lawswith no blood relatives residing in close proximity.

Men typically supply the bulk of meat and fish con-sumed within the household; however, women mayprovide some of these resources themselves when theirhusbands are absent engaging in wage-labor outsideof the community. High-producing women are typi-cally married to high-producing males [Gurven et al.,2009]. Good mothering skills are highly valued, andthis includes attending to children’s needs, keeping aclean house, and providing supplementary foods suchas fish or other staples. Women socialize during workactivities, childcare duties, and in social gatheringssuch as chicha drinking parties and household visits.They share food, friendship, household duties, gardenplots, and sometimes spouses through sororal polyg-yny. Women tend to form coalitions of cooperative(and sometimes competitive) partners among kin andfriends within their villages and engage in varyingamounts of intragroup social aggression. Gossip iscommon and women tend to generally feel that it canhave a negative effect on one’s health and happiness.

Tsimane women tend to have their first child byabout age nineteen and give birth to approximatelynine children over their lifetimes. Marriages, sig-nified by a mating pair sharing the same house,are fairly stable, monogamous unions, althoughpolygynous marriages are not uncommon (<10%).[Winking et al., 2007a]. Postmarital residence rulesare often matrilocal in the first years of marriage,after which partners routinely choose residence op-portunistically to maximize social or food resources,sometimes moving back and forth over the years asconditions and opportunities shift. Parents and otherkin often play a role in the acquisition of marital

Aggr. Behav.

198 Rucas et al.

partners, and are typically asked for courting per-mission by interested males. Cross-cousins are pre-ferred marriage partners and can be close or distantlyrelated.

The Interview

Participants were 121 females, ages 8–70, living infour communities situated along the Maniqui River.The four communities chosen for the study varied inthe extent of their acculturation, proximity, and con-tact with national Bolivian society and local markets.Twenty-nine women from one of the more accultur-ated, 70 from medium acculturated villages, and 22women from a village of very low acculturation (121total females) were asked to participate in a semi-structured interview regarding the state of current andvery recent conflicts or disagreements (quarrels) theywere having with other females in their community, ei-ther social or physical, during any time within the lasttwo months. It should be noted that no physical fightswere reported, thus all data to be discussed only con-cerns instances of social aggression such as rumors,gossip, silent treatment, verbal arguments, or angrydisputes. Within the scope of this research, quarrelsare therefore defined as disagreement or disputes be-tween people.

We coded the contested resource identified to causethe disagreement, participants involved, duration ofconflict, and whether or not the conflict had beenresolved (most had not). Interviewees were queriedabout potential conflicts with all other women inthe community one-by-one using Polaroid photos ofother women as cues to describe current or recentevents concerning those women within the last twomonths in open-ended discussion. It was not requiredthat both parties, during separate interviews, describeor confirm the same quarrel in order to be considered.Evidence of a quarrel having occurred was deemedvalid for coding if at least one woman reported anddescribed the disagreement, and quarrels were notdouble coded when both women reported the sameincidence, which was very frequent, indicating thatdisagreements are often mutually acknowledged. Par-ticipants were allowed to have open interpretationabout what constituted a quarrel worth mentioningand researchers did not prompt women with exam-ples of hypothetical or real quarrels in order that thedata not result in biases toward particular types ofdisagreements. While there exists the possibility thatsome women may not have felt comfortable discussingconflicts openly with an outsider, measures were takento help insure comfort level and interview reliability.Emphasis was placed on informed consent, privacy ofinterviews, and the voluntary nature of participation.

Additionally, it was explained that most data would beaggregate and when not, personal identities would beobscured. Interviews were conducted in a private lo-cation by SLR and a bilingual Tsimane woman. SLRhad spent over six months in two of the three vil-lages, and one month in the third, and had developedtrusting relationships with many of the interviewees.

Resource Categorization

The relative influence of different resources wasevaluated based on quarrel frequency. Quarrels gener-ally fell into eight common themes driven by familiartrends in the topics of women’s complaints: (1) recip-rocal social contracts; (2) non-meat food thefts; (3)adultery; (4) competition for non-attached males; (5)material goods thefts; (6) friendships; (7) meat thefts,and (8) children’s quarrels. Six of the eight categoriescould be further condensed into three broader cate-gories (and are displayed as hatched bars in Fig. 1);all social (contracts + friendships), all men (adultery+ unattached men), and all food theft (meat + non-meat); note these cross-hatched bars exclude materialgoods and children’s quarrels. No quarrels were cross-coded.

The “reciprocal social contract” category com-prised all quarrels in which a woman was accusedof cheating, defecting, or refusing to return her sideof a tacit reciprocal exchange arrangement with an-other woman where the currency of exchange involvedfood, goods, helping with household chores, and otherservices, or social grooming (such as visitations andconversations). The “friendship” category includedconflicts resulting from one’s joining a friend’s quar-rel with a third party. For example, one woman de-cided to join her friend against another person whowas accused of stealing her friend’s pig, even thoughthe pig was not her own. Quarrels concerning foodtheft were coded separately for meat (including fish)and non-meat (such as rice, plantains, or manioc) toexplore differences between the two categories. Non-meat nearly always involved accusations of theft ofstaples from family garden plots when no one wasaround. Quarrels over unattached males resulted fromcompetition to gain the attentions of a potential hus-band. “Adultery” quarrels were the result of marriedwomen who felt that another woman was attempt-ing to gain attention, services, or investment fromher husband. “Material goods” resource quarrels re-sulted from theft of a wide variety of items such asspoons, knives, clothes, etc. “Children’s” quarrels oc-curred when mothers conflicted as an extension oftheir respective offspring’s arguments. It is importantto note that resource labels were largely driven bywomen’s reports, who are quick to identify the source

Aggr. Behav.


Fig. 1. Number of quarrels by resource type. Cross-hatched bars: all social = reciprocal contract defections + friendships; all men = adultery +nonattached males; all food = food theft (nonmeat) + meat thefts.

outright by making such statements as; “she stolerice from my garden,” “ . . . our daughters are fight-ing . . . ,” “ . . . my friend is mad at her so I am too,”“she stopped sharing food with me,” “she won’t visitme anymore,” etc.

RESULTS

The frequencies of the different types of quarrels(N = 334) are shown in Figure 1. In descending order,reciprocal social contracts were the most common(19.5%), followed by non-meat food thefts (13.5%),adultery (12.9%), competition for non-attached males(12.6%), material goods thefts (10.2%), friendships(9.0%), meat thefts (5.1%), and extensions of chil-dren’s quarrels (4.2%). A significant number, 44 quar-rels (13.2%) could not be assigned to any resourcecategory. These involved cases in which women iden-tified a persistent aggressive state with a longstanding“enemy,” but were unable to remember the originalcauses of the disagreement or subsequent deteriora-tion of the relationship. Such fights were typically be-tween older women who had more or less permanentsocial aversion to each other. It cannot be discountedthough that at least some of these women perhaps didnot want to provide details about the origin of theirdisagreement. The other remaining unassigned casesinvolved very young girls who were unable to articu-

late exactly why they were upset with each other andwhat exactly had instigated the disruption.

The frequency data displayed in Figure 1 indicatethat of the eight resource categories, reciprocal socialcontracts account for the greatest number of conflicts.Even when the eight categories are collapsed into thethree cross-hatched bars (all social, all males, and allfood), the greatest number of conflicts still revolvearound the social resources of friendship loyalties andsocial contract infringements. Such arguments are of-ten more about the fidelity of the sharing relationshipthan about the individual disputed items themselves.

Figure 2 provides a further breakdown of the recip-rocal social contract category (N = 65). Of these, quar-rels over food (including both meat and non-meat) ac-count for 38% of the total, with meat-related quarrelscontributing the majority, two-thirds. Comparatively,as illustrated in Figure 1, meat theft quarrels, wherethere existed no social contract of expected exchange,were only half as frequent as non-meat theft quarrels.Socializing exclusions were also very common (28%),which are characterized by situations such as, for ex-ample, when one woman invites a friend to a beerdrinking party, but the friend does not later recipro-cate with an invitation. Defections on household dutysharing accounted for 18% of the quarrels, and finally,disputes over money-goods exchanges, such as whenone woman receives some good and fails to provide apromised payment, accounted for 15%.

Aggr. Behav.

200 Rucas et al.

Fig. 2. Frequency of reciprocal contract-defections by type.

Fig. 3. Mean age by quarrel category (standard error bars).

Age Effects

Figure 3 illustrates the average age of participantsfor each type of quarrel. Average ages vary from16–40, illustrating the changing foci of competitionacross the life-course. Discriminant analysis indicatesa significant effect of conflict type on average ages(P < .001 and F = 10.02, N = 290). Figures 4 and 5show the average number of quarrels per personacross the life-course for each decade of age, wherebyFigure 4 collapses the groups into three major cate-gories (Male, Social and Food), and Figure 5 illus-trates the same data expandedin to finer categories.These data indicate that conflicts over men occur rel-atively more frequently at younger ages, and thoseover social resources peak in the thirties, while con-flicts over food resources increase only very slightlywith age. A significant amount of variation in conflict

Fig. 4. Average number of resource quarrels per person across the life-course.

over resources may be due to age and its corollary in-creases in dependent young. Further contextual andqualitative description is provided in a section belowillustrating what is perhaps a common life history tra-jectory for Tsimane women with respect to resourcecompetition at different ages.

Figure 6 illustrates the age difference averages foreach resource and provides average absolute age dif-ference between women who are vs. those who arenot quarreling over different resources. Multiple lin-ear regression indicates that women closer in age arestatistically significantly more likely to be quarrelingabout men, food, and friends (P < .01 for all three),but not for children, materials, or social contract de-fections. Women quarreling over unattached maleswere the closest in age, on average being only approx-imately 1.6 years apart. There was very little variationin the absolute age difference of women not quarrelingover any of the resource types. Indeed, it appears thatwomen not quarreling over the different resource cat-egories exhibited the exact same age difference, butthis is not the case, it is simply that the observablevariation is negligible.

Using multiple linear regression, the number of con-current quarrels reported between pairs of women wasmodeled as a function of their absolute age differ-ence after controling for other potential predictors ofquarreling, such as the coefficient of genetic related-ness between women and the geographical distancebetween houses. A greater absolute age differencebetween women predicts fewer concurrent quarrels(P <. 001) (Table I). The mean age differences for

Aggr. Behav.


Fig. 5. Average number of resource quarrels per person within each age group for each type of resource.

pairs of women engaged in varying numbers of con-current quarrels, with the maximum reported beingthree between any one dyad, were calculated. Pairsof women experiencing three quarrels were on aver-age 6.92 years apart; women with two quarrels wereon average 8.13 years apart; women with one quar-rel were on average 9.69 years apart, and the averageage difference for those reporting no quarrels wasapproximately 15.02 years. The differences were con-

siderable when comparing intense competitors withnoncompetitors, whereby women not quarreling weremore than twice the age difference of women experi-encing three concurrent quarrels.

DISCUSSION

Data on social aggression, social networking, andresource competition among women in small-scale

Fig. 6. Absolute age differences between pairs of women quarreling and not quarreling over different resource types.*sig. < .1; **sig. < .05; ***sig. < .01. Controls: coefficient of relatedness and distance between residences.

Aggr. Behav.

202 Rucas et al.

TABLE I. Multiple Regression Analysis of Absolute Age Difference Effects on the Number of Concurrent Quarrels Reported BetweenPairs of Women Controlling for Kinship and the Distance Between Their Residence Locations

Unstandard coefficient Standard Coefficient

Predictor variables Beta Standard error Beta T Sig.

(Constant) 0.321 0.037 8.688 0.000Absolute age difference − 0.003 0.001 − 0.116 − 5.644 0.000Coefficient of relatedness 0.136 0.062 0.054 2.203 0.028Distance between houses − 0.058 0.010 − 0.143 − 5.870 0.000

traditional societies are very limited. Because net-working and social aggression are costly, it shouldbe expected that women are more willing to pay thecost when the potential gains are high. It was also pre-dicted that women compete over different resourcesat different times throughout their lives in accor-dance with fitness-maximizing goals, making womenof similar ages more likely to be resource and socialaggression competitors. Our data generally supportthese ideas. We would caution, however, that othermeasures of intensity of conflict, beyond frequencydata, were insufficient in our dataset for use in analy-sis. That being said, it is important to note that somequarrels may have far greater intensity than othersand reporting of a disagreement may reflect some levelof salience to the participant. Indeed, while womenwere asked to report recent or current disagreements,it was found that women sometimes reported long-standing quarrels with another woman that had lastedfor several years, to the point that their relationshiphad evolved into a general state of constant discordcharacterized by continuous dissatisfactions with oneanother. Beyond quarrel frequency, we did not collectdata to provide a suitable proximate measure of inten-sity, other than the duration of conflict, which was notusable in this circumstance since many quarrels werestill occurring. But we do believe that issues such asthe duration of the conflict or perhaps the exact na-ture of the aggressive medium might be importantindicators of relative resource importance and futurestudies should attempt to parcel such factors.

Based on the fact that social conflict is costly, itwas assumed that women would actively quarrel morefrequently over those resources that have a greaterimpact on reproductive success. Our results indicatethat food, men, and social resources are the most im-portant fitness-related factors that women discern areworth the cost of female-female competition via so-cial aggression, and are most effectively dealt withvia quarreling. Indeed, these three broad categoriesencompassed 87% of all quarrels. It is also worth not-ing that all recorded quarrels were social in nature;no physical fights were reported. When women were

asked about occurrences of physical fighting, the gen-eral consensus was that such events do happen, butonly rarely and more typically only among youngergirls.

Social resources (34% of quarrels), made up of so-cial exchange networks, friendships, and social con-tracts, are clearly vital to women and appear fre-quently in events of social aggression. Many quarrelsin this category involved failed expectations concern-ing food or money exchanges; however, 48 of them(51%) were specific to friendship loyalties, situationsin which women supported the social conflicts of oth-ers, and breakdowns in social contracts about visi-tation and conversation exchange. Given the heavyresearch focus on mate competition in evolutionarypsychology, and on feeding competition in primatol-ogy, it was illuminating and instructive to find thatsocial resources in this population are at least as im-portant as both food and men. Fighting over a singlepiece of stolen fruit may not be as profitable as quar-reling over a friend or social contract fidelity becausethe latter two may provide long-term benefits such asinsurance against future food shortages, coalitionalsupport, or household task assistance. Also, it couldfree one from free-loader or “fair weather” friendsand advertise to others the unlikelihood that anyonecould take advantage of you.

Since additional caretakers should improve off-spring quality, we expected significant conflict overallocaretaking to occur, but, there were no reportedquarrels specifically related to allocaretaking eitherin circumstances of shared babysitting duties or ofwomen reporting that others commonly saddled themwith unattended offspring. However, in one interview,a married woman admitted to persuading a youngersister who was vying to become a second wife to herhusband to help babysit her children. This did notactually cause conflict, however, because the youngersister was eagerly taking on the duties in an attemptto persuade her sister concerning the courtship at-tentions of the husband; this suggests that there isbargaining going on with respect to allocaretaking,but not so much fighting or social aggression. This

Aggr. Behav.


example should not be taken to mean that co-wives donot quarrel, because indeed they do, and such quarrelsmay be frequent and intense; however, it is simply thatthe nature of co-wife quarreling very rarely involvesallocaretaking.

As expected, there was a high frequency of quar-rels regarding males, likely due to the fact that Tsi-mane men provide substantial caloric contributionto the diet, and other forms of support and pro-tection [Winking, 2006]. A good mate can mean thedifference between a lifetime of support vs. a lifetimeof fending for one’s self and one’s children. Somewomen competed in order to get mates while oth-ers competed in order to retain them. We found thatcompetition to obtain mates would be concentratedat younger ages, around 16 years. This is likely muchyounger than the average age of mate competition inmodern cultures because the latter exhibit later agesof marriage and first birth resulting in extended yearsof stressful mate (and status) competition for qual-ity marriage partners (for example, according to theCenters for Disease Control, the age of first marriageand first birth in the US is 25.6 and 25 respectively,and medians in some other developed nations are evenhigher). Our data support the suggestion that girls in anatural fertility population engage in short term, buthighly concentrated competition for life-mate part-ners. So intense is this competition that groups ofTsimane girls were often observed to be competingover several boys simultaneously. Thus, it seems pos-sible that high quality males may sometimes be ableto choose among several spousal options; whereas,others may have difficulty attracting girls’ attentionsas some boys were rarely or never mentioned in thediscourse of conflicts.

The Tsimane divorce rate is lower than that ob-served in other forager-farmer groups such as theAche, particularly once a family has a child or two[Hill and Hurtado, 1996; Winking et al., 2007a]. Mar-riages more often end due to the death of a spouse asopposed to electing for alternative mating opportu-nities. The average age of women making adulteryaccusations was 29, with several accusers over 40 anda majority in the early-mid twenties (Fig. 3). The av-erage age of the accused woman was 18, and she wasnearly always unmarried. Younger, unattached girlsare likely feared and distrusted more because they arehighly desired by men because they exhibit greater re-productive value due to their age, and they lack thedeterring presence of an existing spouse.

Previous data indicate a negative association be-tween extramarital affairs and number of dependentoffspring; thus, we expected to find that women sus-pecting husband infidelity in this dataset to be in their

early twenties; yet, the average reported age was 29with several accusers over 40. Contrarily, men of ourprevious study were more likely to have affairs whenthey were younger with fewer children to support[Winking et al., 2007a]. There are a few explanationsfor the difference in these male vs. female reportedfindings. First, Winking’s data only included two ofthe four villages used in this study, so perhaps culturaleffects account for the difference. Second, these data,unlike the male reports are unsolicited accusations ofaffairs, as opposed to Winking et al., infidelity data,which was direct male incident report. By contrast,the women’s data included only a few cases offeredduring interviews by the offending women; whereas,the majority of the cases were simply described as un-confirmed suspicions by women. Finally, there seemsto be an age gap in charges whereby women in theirtwenties experience a high degree of conflict in thisarea, which drops strongly in the thirties and thenincreases slightly again in the forties. Perhaps, too,women become more vocal and competitive aboutsuspicions of adultery as they become much olderand perhaps more independent and socially power-ful where they were less so in their thirties (thoughthis does not explain the high frequency in the twen-ties). Finally, some of the adultery charges in thisdata set were “ancient history” that had fueled a gen-eral adversarial status between the women, negativelycharacterizing their interactions for several years. Anevent that may have occurred several years in the pastis still credited sometimes as being the reason for latersocial conflicts. Whatever be the case, wives are force-ful in competing with younger girls over accusationsof affairs, and may play a part in the common Tsi-mane belief that philandering fathers may cause childillness.

Nearly 30% of all quarrels involved food, whetherthrough theft or failed expectations concerning reci-procity. While it may appear that conflict over meatis less important given its lower frequency in theftquarrels, non-meat food items show up far more fre-quently in the daily course of life, allowing for moreopportunity for conflict. Additionally, staples are eas-ier to steal, since agricultural fields are often leftunattended. Nonetheless, even though meat materi-alizes more rarely in daily life, it was still involved in17 theft quarrels and 18 reciprocity defections un-derscoring its extreme importance to the diet andwomen. And, non-meat quarrels due to sharing de-fections were less frequent than those involving meat.This is perhaps due to the fact that staples, fruits,or other vegetables may be less frequently sharedthrough social contracting, or it could be that they aremore likely to be reciprocated than meat items. While

Aggr. Behav.

204 Rucas et al.

non-meat quarrels were far more likely to result fromtheft than failed food-sharing contracts, meat quarrelswere nearly equally represented by both thefts and re-ciprocal contract defections. These findings are moreconsistent with food-sharing models involving reci-procity than those emphasizing tolerated scroungingor signaling, particularly for meat items. If toleratedscrounging were responsible for food distributions,frequent quarreling over failed food-sharing contractswould not be prevalent at so high a rate.

We expected that items that can improve the qual-ity of offspring, such as material goods, to be un-der competition, but compared to men, food, andsocial resources (which also improve offspring qual-ity), few quarrels involved material goods. Addition-ally, their probability of appearing in quarrels wasstrongly influenced by increasing age, likely reflectingthe accumulation of material items over the lifespan.Also, while goods are often used within a householdby all women, ownership tends to be attributed tothe oldest matriarch. Plus, such thefts are more diffi-cult to hide from others because, unlike food, whichis quickly consumed, items such as clothes are re-peatedly worn in front of others, and knives, spoons,and plates are continually used at communal meal-times. Nearly all of the material goods quarrels in-volved these items, with clothing appearing the mostfrequently.

Women quarreling as an extension of their “chil-dren quarreling” occurred less frequently than allother types of quarrels (N = 14). These extendedfights typically concerned the marital success of theirolder children and more rarely about access to food orother resources. For example, two women were quar-reling with each other because they both wanted theirrespective daughters to be the wife of the same high-status male. With such a low frequency of quarrelsregarding children, it might be concluded that Tsi-mane women do not substantially micromanage thesocial lives of their younger children, but this can-not be known without further cross-cultural compar-ative studies. These data indicate though that Tsimanemothers actively become involved when they feel itnecessary, they do not seem to concern themselves inevery minor quarrel. Also, the mean age of conflictover children’s quarrels was greater than all other cat-egories, a direct reflection of the fact that nearly 80%of these quarrels concern the marital success of theirchildren; thus, one must be of an age to have a mar-riageable aged child in order to have such a quarrelabout their competitiveness in the mating market.

Several different analyses provide support for thehypothesis that resource competition is age-specific.Intensity of social aggression (frequency of quarrel-

ing) is strongly predicted by closeness in age, evenwhen controling for features such as relatedness andresidential location. Women closer in age are morelikely to be competing for similar resources (mates,friends, food, etc.) and are thus more likely to engagein social aggression via quarreling.

The following integrates qualitative informationfrom our interviews to complement the quantitativedata (i.e., Fig. 5, which details the average numberof quarrels per person at each decade of life for eachresource type) in order to better characterize the lifeof a typical Tsimane woman. Identities have been ob-scured with substituted character names.

Less than 20 age group. Compared to otherwomen, this age group is not strongly characterizedby food-sharing defections, but more so instead byfood and meat theft accusations. While theft is stillprobably a minor way to get food, it may be thatthis age group is relying more upon theft as a food-getting strategy (beyond what is gifted to them bykin) over food sharing through reciprocal social con-tracts. By comparison, young girls are overwhelminglyconcerned with getting mates and defending them-selves from adultery accusations. In one example ofthis type of quarrel, Sofı́a explained to us that she wasmad at another young girl, Dulce, who used to be afriend but is not any longer due to the fact that Dulcehad started showing strong attentions to their mutuallove interest, Domingo. Conflicts in this age group arealso characterized by scenarios in which girls reportedquarreling with a third party in support of a friendwho was actively engaged in mate competition withthe third party girl. In the mid-teens to early twenties,girls report strong levels of aggression from marriedwomen due to fears of adultery with their husbands.One girl in particular, Georgiana, was accused by noless than six married women of such a behavior, eventsthat the girl denied. Still, this did not deter the accusa-tions and Georgiana was continually plagued by thesocial aggression. Evidence also indicates that younggirls are concerned with friendship making and loy-alties. For example, Sylvania, explained that anothergirl, Juanita, had stolen her best friend Felicia awayand supplanted her as Felicia’s seating companionduring the occasional school lessons. Thus, girls ofthis age group appear to invest in acquiring two highpayoff resources; a long-term mate and a social net-work of good loyal friends.

20s. Women in their twenties are usually marriedand have their first children. Men target extramaritalrelationships in this age range, and our conflict datasupports that women in their 20’s make frequent accu-sations of adultery toward younger girls. In one primeexample, Lolita, a 24-year old with two children and

Aggr. Behav.


a husband, informed us that she was upset with a 16-year-old girl of her community, Consuela, who flirtedtoo much with her husband Jorge. Food thefts in-crease in the 20’s probably because women need morefood due to dependent offspring, but also becausethey begin to develop their own fields, which is thelocation of nearly all non-meat food thefts. In one ex-ample, Cynthia, a 28-year-old woman with three chil-dren accused Analuz, a 24-year-old woman with onechild of sneaking into her field and stealing rice, whichshe proceeded to eat in secret with her kinswoman.Interestingly, meat thefts decline in the 20’s and re-main relatively low throughout the rest of the life-course until they ultimately disappear around age 50.However, meat-sharing conflicts begin to rise, reflect-ing that women develop sharing relationships that be-gin to take precedence over thieving activities at leastfor meat. Friendship loyalties have a relatively smallinfluence on women in their twenties who seem to bemore concerned with mate retention and strugglingto acquire enough food to feed their newly growingfamilies.

30s. The 30’s are characterized by high invest-ment by women in friendship loyalties and resourcesharing between these friends. For example, Alejan-dra described quarreling with Gabriela, simply be-cause Gabriela was the stated enemy of her bestfriend Maribel. The two women reported no othercomplaint aside from the state of their relationshipwith the third-party friend/enemy, Maribel. Anotherwoman, Rosita, reported that she had invited Susanato a chicha (fermented manioc beverage) party at herhouse, but Susana subsequently later made chicha anddid not invite Rosita back to drink. In another situa-tion, Teresa said she had stopped talking to Ysabel,because the latter had not shared fish with her fromthe a group fishing event, even though Teresa hadgiven her some coati meat the day before. It is duringthese ages that conflict over men seems to be at its low-est, partially reflecting that women have succeeded infinding a mate and their increasing dependency loadcombined with still present reproductive value im-proves the chances that their mate is diverting signifi-cant energy away from extramarital relationships andtoward paternal investment. Thus, their competitiveenergy is instead taken up with creating and maintain-ing cooperative same-sex coalitions and improving so-cial contract making skills and alliances. Extensionsof their children’s quarrels have begun to increasenearer the end of the decade and remain constantthroughout their 40’s before increasing significantlybeyond their 50’s. For example Vanessa, a 37-year-old woman reported arguing with Lola, a 68-year-oldwoman, because the elder woman’s adult daughter

accused the younger woman’s nearly-adult daughterof infidelity with her son-in-law.

40s. Adultery accusations (by wives of their hus-bands) which were very low in the 30’s appear to riseslightly in the 40’s, possibly as a result of the factthat women in their 40’s exhibit rapidly diminishingreproductive value but still a relatively high depen-dency load of children of different ages (though wefound in other interviews that women in their fortiesare sometimes given high attractiveness rankings as aresult of their greater power and social status, Rucaset al. [2006]). Quarrels over informal social contractsare more numerous than those over friendship loyal-ties among women in this age group, as women arefocused on maintaining equity within their specificcooperative relationships that they developed earlierin life. Women are less invested in outward displaysof loyalty to friends and are more interested in so-cial contract fidelities. Dalia, a 44-year-old woman,described a quarrel she was having with Antonia, be-cause Dalia had given Antonia some fish when shewas hungry one day, but Antonia had not given herany meat in return when her husband came homewith deer a week later. Non-meat food theft accusa-tions by women in their 40’s also jump from the 30’sand are likely a result of the cumulative investmenthusbands have been making with the increasing num-ber of children in the size and variety of staple foodsgrown in their fields. The result of all of this labor isproducing a surplus that might be attractive to others.Women’s oldest children are getting mates and start-ing their own families, which are reflected in the slightincrease in children’s quarrels. Aside from some at-tempts to protect the mating future of their children,and keep husbands invested, they also find it neces-sary to protect material goods that have been skillfullyacquired.

Greater than 50s. Paulina, a 52-year-oldwoman, reported to us that she was upset with her sis-ter, Petrona (age 69), because their respective daugh-ters wanted to marry the same boy and thus her sisterhad started rumors about Paulina’s daughter’s inabil-ity to cook and properly take care of children. Matingsuccess of children becomes paramount as well as so-cial strategizing to improve the social prospects oftheir children. The 50’s also see a large jump in thenumber of quarrels over material goods. Modern con-sumer items such as spoons, mosquito nets, and potsare rare and more often claimed in ownership by olderwomen (mothers) within households who have morepower to control them. Thus, older women, begin-ning in their 40’s and continuing throughout life, aremore likely to report such conflicts and thefts. Socialcontract infringement and friendship quarrels are still

Aggr. Behav.

206 Rucas et al.

present but not greater than other resources quarrels,probably reflecting the fact that older women are rela-tively self-sufficient, powerful, and lack many depen-dent young. Adultery accusations, which increased inthe 40’s, begin to drop slightly, probably reflecting thedecreasing mate value of their often even older spousewhose hunting productivity peaked in the previousdecade and continues to decline.

In final conclusion, this study demonstrates theimportance of social network resources to womenand highlights the current lack of knowledgeconcerning women’s social and resource competitivelives in small-scale societies. While no other such stud-ies exist for comparison, we expect similar generalresults in other traditional societies where coopera-tion and food sharing are critical for survival andsuccess. However, in wealthy modern societies, wherefood sharing is largely symbolic and cultural andwhere women have direct access to resources (throughjobs as opposed to food-sharing, hunting husbands),we would expect quarrels to more frequently revolvearound mate acquisition, social status, jobs, and re-sources directly. Questions for the future include: Inwhat other ways might intensity of conflict be mea-sured? How do women influence the size and com-position of their social networks? To what extentdoes social capital and social aggression impact fit-ness and by what pathways, e.g., increased offspringsurvival, shorter birth intervals, etc? How does invest-ment in social capital trade off against other kinds ofresources, and is this different for men than women oracross the life-course? Additional work among tradi-tional populations is needed in order to better under-stand the costs, benefits, and trade-offs of women’ssocial networks.

ACKNOWLEDGMENTS

We would like to thank the women of our studycommunities who so generously agreed to participatein our study. Acknowledgement and thanks are madeto Maria Crespo for her assistance with data collec-tion. For their insightful comments and criticisms,we would also like to thank Steve Gangestad, RandyThornhill, and Jane Lancaster of the Human Evolu-tionary and Behavior Sciences program at the Uni-versity of New Mexico.

REFERENCES

Barton RA, Dunbar RIM. 1997. Evolution of the social brain. In:Whiten A, Bryne RW (eds), Machiavellian Intelligence II. Cam-bridge: Cambridge University Press, pp 240–263.

Bliege Bird RL, Bird DW. 1997. Delayed reciprocity and toleratedtheft: The behavioral ecology of food-sharing strategies. Curr An-thropol 38:49–77.

Blurton Jones NG. 1987. Tolerated theft, suggestions about the ecol-ogy and evolution of sharing, hoarding, and scrounging. Soc SciInform 26:31–54.

Borgerhoff Mulder M. 1992. Reproductive decisions. In: Smith EA,Winterhalder B (eds), Evolutionary Ecology and Human Behavior.New York: Aldine de Gruyter, pp 339–374.

Clutton-Brock TH. 1994. The evolution of sex differences and theconsequences of polygyny in mammals. In: Bateson P (eds), TheDevelopment and Integration of Behaviour: Essays in Honour ofRobert Hinde. New York: Cambridge University Press, pp 229–253.

Daly M, Wilson M. 1985. Sex, Evolution and Behavior. Boston, MA:Willard Grant.

Dunbar RIM. 1992. Neorcortex size as a constraint on group size inprimates. J Hum Evol 20:469–493.

Dunbar RIM. 1993. Coevolution of neocortical size, group-size, andlanguage in humans. Behav Brain Sci 16:681–694.

Dunbar RIM. 1997. Human conversational behavior. Hum Nat 8:231–246.

Gangestad SW, Simpson JA. 2000. The evolution of human mating:Trade-offs and strategic pluralism. Behav Brian Sci 23:573–644.

Godoy R, Gurven M, Byron E, Reyes-Garcia V, Keough J, Valdez V,et al. 2004. Do markets worsen economic inequalities? Kuznets inthe bush. Hum Ecol 32:339–364.

Gurven M. 2004. Reciprocal altruism and food sharing decisionsamong Hiwi and Ache hunter-gatherers. Behav Ecol Sociobiol56:366–380.

Gurven M. 2006. The evolution of contingent cooperation. Curr An-thropol 47:185–192.

Gurven M, Allen-Arave W, Hill K, Hurtado AM. 2000. “It’s a won-derful life”: Signaling generosity among the Ache of Paraguay.Evol Hum Behav 21:263–282.

Gurven M, Hill K, Kaplan H. 2002. From forest to reservation: Tran-sitions in food-sharing behavior among the Ache of Paraguay. JAnthropol Res 58:93–120.

Gurven M, Winking J, Kaplan H, von Rueden C, McAllister L. 2009.A bioeconomic approach to marriage and the sexual division oflabor. Hum Nat 20:151–183.

Hagen EH, Barrett HC, Price ME. 2006. Do human parents face aquantity-quality tradeoff? Evidence from a Shuar community. AmJ Phys Anthropol 130:405–418.

Hess NC. 2006. Informational Warfare: The Evolution of FemaleCoalitions and Gossip. Santa Barbara, CA: University of Cali-fornia.

Hill K, Hurtado AM. 1996. Ache Life History: The Ecology andDemography of a Foraging People. New York: Aldine De Gruyter.

Hooks BL, Green PA. 1993. Cultivating male allies—A focus on pri-mate females, including Homo-sapiens. Hum Nat 4:81–107.

Hrdy SB. 1999. Mother Nature: Maternal Instincts and How TheyShape the Human Species. Toronto: Ballantine Books.

Hrdy SB. 2005. Comes the child before the man: How cooperativebreeding and prolonged post-weaning dependence shaped humanpotentials. In: Hewlett BS, Lamb ME (eds), Hunter-GathererChildhoods. Piscataway, NJ: Aldine Transactions, pp 65–91.

Hurtado AM, Hill K, Kaplan H, Hurtado I. 1992. Trade-offs betweenfemale food acquisition and child care among Hiwi and Ache for-agers. Hum Nat 3:185–216.

Kaplan H, Hill K. 1985. Food sharing among Ache foragers-tests ofexplanatory hypotheses. Curr Anthropol 26:223–246.

Lancaster J. 1978. Carrying and sharing in human evolution. HumNat 1:83–89.

Lundberg S, Pollak RA. 1993. Separate spheres bargaining and themarriage market. J Polit Econ 101:988–1010.

Manser M, Brown M. 1980. Marriage and household decision-making: A bargaining analysis. Int Econ Rev 21:31–44.

Aggr. Behav.


Marlowe FW. 2003. A critical period for provisioning by Hadzamen: Implications for pair bonding. Evol Hum Behav 24:217–229.

Mesnick SL. 1997. Sexual alliances: Evidence and evolutionary impli-cations. In: Gowaty PA (ed), Feminism and Evolutionary Biology.New York: Chapman and Hall, pp 207–260.

Parker GA, Baker RR, Smith FCF. 1972. The origin and evolutionof gamete dimorphism and the male-female phenomenon. J TheorBiol 36:529–553.

Rucas SL, Gurven M, Kaplan H, Winking J. 2010. The social strategygame: Resource competition within female social networks amongsmall-scale forager-horticulturalists. Hum Nat 21:1–18.

Rucas SL, Gurven M, Kaplan H, Winking J, Gangestad SW, CrespoM. 2006. Female intrasexual competition and reputational effectson attractiveness among the Tsimane of Bolivia. Evol Hum Behav27:40–52.

Schmitt DP, Shackelford TK, Buss DM. 2001. Are men really more‘oriented’ toward short-term mating than women: A critical reviewof theory and research. Psychol Evol Gender 3:211–239.

Silk JB, Alberts SC, Altmann J. 2003. Social bonds of female baboonsenhance infant survival. Science 302:1231–1234.

Stieglitz J, Gurven M, Kaplan H, Winking J. in press. Infidelity, jeal-ousy, and wife abuse among Tsimane forager-farmers: testing evo-lutionary hypotheses of marital conflict. Evol Hum Behav.

Stieglitz J, Kaplan H, Gurven M, Winking J, Vie Tayo B. 2011. Spousalviolence and paternal disinvestment among Tsimane’ forager-horticulturalists. Am J Hum Biol 23:445–457.

Trivers R. 1972. Parental investment and sexual selection. In: CampbellB (ed), Sexual Selection and the Descent of Man. Chicago: Aldine,pp 136–179.

Winking J. 2006. Are men that bad as fathers? The role of men’sinvestments. Soc Biol 53:100–115.

Winking J, Kaplan H, Gurven M, Rucas S. 2007a. Why do men marryand why do they stray? Proceedings of the Royal Society of LondonSeries B-Biol Sci 274:1643–1649.

Winking J, Kaplan H, Gurven M, Rucas S. 2007b. Why do men marry,and why do they stray? Proceedings of the Royal Society of LondonSeries B-Biol Sci 274:1643–1649.

Aggr. Behav.

social aggression and resource conﬂict across the female

Documents