Desktop version

Home arrow Psychology

  • Increase font
  • Decrease font

<<   CONTENTS   >>

Culture and development: childhood across cultures

What this chapter will teach you

  • • What is the relationship between childhood and culture?
  • • Does parenting differ across cultures?
  • • Is attachment theory applicable across cultures?
  • • What are parental ethnotheories?
  • • Does temperament differ across cultures?
  • • What are sex and gender?
  • • Are gender expectations culturally universal?

A global approach to childhood

A lot of people around the world are children. In the Global South every third person is under the age of 15. Globally, the figure is 26% (World Population Bureau, 2007). In some nations, such as Guatemala, under 30s constitute almost 70% of the population (Castanada & Del Pilar Grazioso, 2017). And yet the global population is ageing, with the over 65s share rising from 5% to 9% between 1960 and 2018 (Population Reference Bureau, 2018). Childhood is also experienced differently in different cultures and across different epochs. Being young in medieval Europe was incomparable with contemporary notions of childhood since it lacked the emotional kinship ties that so characterise it today (Cunningham, 2006). It is also suggested that in the modern era childhood in certain cultural settings has unique characteristics. A typically American childhood, for example, may be characterised by a social desire for obedience to parental authority which may not be stressed so much elsewhere (Kessen, 1979). Contrastingly, in many nations across Latin America, the experience of childhood is far more likely to include participation in the labour market, street-connectedness and limited access to regular schooling (Stevenson et al„ 2019b). Arguably, then, where you are brought up has an enormous bearing on the kind of childhood you are in for. Contextual factors such as family size, poverty and access to schooling are all likely to influence a child’s ‘career’, and of course all these factors vary across cultures. We can infer from this that childhood is to some extent a culturally relative phenomenon. How relative, though, is a moot point. The extent of cultural variation - and of any commonalities in childhood experience across cultures-will be explored in the light of psychological research in the following pages.

Culture and childhood in micro and macro

The relationship between culture and childhood can be viewed up close (at a micro level), or from a distance (at a macro level). At close quarters it is commonly portrayed as involving the transmission of values and beliefs between caregivers (usually parents) and children, via socialisation and enculturation. Socialisation (see Figure 8.1) is not a one-way street down which elders’ beliefs pass to a younger generation. Rather, it involves the exchange and formation of ideas between adults, children and peers.

Typically, parents and children engage in socialisation by participating in ongoing transactions that lead to the movement and propagation

Socialisation of ideas in more than one direction

Figure 8.1 Socialisation of ideas in more than one direction (Matsumoto & Juang, 2016). Thanks to this bidirectional exchange, new cultural phenomena (beliefs, behaviours, artefacts) emerge. This bidirectional model of socialisation should strike a chord with any adult who spends time with children, who would surely have to concede that they learn as many things as they teach.

At the macro level the child’s socialisation into its culture is part of a broader process involving biological, environmental and cultural factors. Cole (1998) explores the relative contributions of these factors in the wider context to human development. His cultural mediation model (Figure 8.2) stresses the pivotal part played by culture in the developmental process. Borrowing from Tylor (1874), Cole views cultures as complex entities incorporating beliefs, morals, customs and artefacts (such as tools and language systems). Cultures are shaped by (and shape) external environments. They are also shaped by (and shape) human evolution. Thus, ideas emanating from previous generations influence the way future generations evolve culturally and biologically.

Cultural mediation theory (which is central to the cultural psychology approach (see Chapter 4) casts culture in the role of a filter through which

Macro-models of human development other biological and environmental forces are mediated

Figure 8.2 Macro-models of human development other biological and environmental forces are mediated. The mediating role of culture in the development of humans urges us to understand childhood, adolescence and adulthood as phenomena that should be viewed in their unique cultural and historical contexts.

Micro and macro approaches to human development come together in Bronfenbrenner’s (1979) ecological, layered model, wherein socialisation and its relationship with culture are portrayed in the form of a series of ever-increasing, concentric circles (see Figure 8.3). Here, the developing child is seen as simultaneously transacting with the world as a micro-system, an exo-system and a macro-system. In the microsystem the main agents of socialisation are the immediate, significant others such as families, schools and peers. More distant protagonists (local government agencies, friends of friends, mass media) engage with the developing individual at the level of the exo-system. Finally, larger cultural institutions and historic values constitute the macrosystem: these are perhaps the most deeply rooted ideas to distinguish one culture from another. These three systems are distinct yet intertwined. They provide a context for the development of the individual and for the evolution of the culture itself through meaningful exchanges with its constituents.

However we portray the David and Goliath-like relationship between the individual and his/her developmental context, it seems fair to argue that the elements involved will vary from culture to culture. For example, ideas about how children should be guided through childhood by those who are closest to them (most often parents) are likely to vary

Bronfenbrenner's model of human development from one region to another

Figure 8.3 Bronfenbrenner's model of human development from one region to another. Indeed, you may have noticed your own neighbours disagreeing about how to raise their children, so we should certainly expect these differences of opinion to multiply across cultures. Yet as we shall see in the next section, there is also a high degree of agreement about what constitutes effective childrearing.


  • 1. What is the difference between micro and macro socialisation?
  • 2. What do we mean when we say that the socialisation process is bidirectional?

Culture and parenting practices

One author has written that while in some cultures children grow up, in others they are brought up (Kagitcibasi, 1996). Parenting takes various forms both within and across cultures. Arguably, some cultures traditionally sanction a more interventionist parenting style than others. Such cultural differences in socialisation have a profound effect.

Cultural variations in parenting

Culturally diverse parenting practices manifest themselves in domestic discipline regimes to divisions of labour around the home, to the more general goals that parents set for their children. There are indications that the ways in which parents socialise their children vary considerably cross-culturally. Levine et al. (1996) compared household routines among the agricultural Gusii of Kenya with those observed in a sample of (US) Bostonian suburb-dwellers, and noticed a greater expectation for children to help with chores in the former, coupled with more parental emphasis on intellectual stimulation in the latter.

Domestic sleeping habits also reflect culturally diverse parenting. Anecdotally it is suggested that Spanish children stay up relatively late into the night, that rural African children fall asleep amid a hubbub of outdoor activity and that Japanese children rise relatively early for school. Is there any evidence to support these observations? Cross-cultural studies do reveal cultural differences. For example, US parents encourage their children to sleep alone more often than their rural European and (Central American) Mayan counterparts do (Morelli et al., 1992). Another comparison found Dutch parents to be particularly insistent on an ‘early to bed and plenty of sleep’ regime, with under-fives getting up to two hours’ more sleep than a US comparison group (Super et al., 1996). Interestingly, these findings correlate with what was perceived to be calmer waking behaviour in Dutch infants.

Some of these differences in parenting across cultures reflect material differences in income and space, as well as differences in the desire to allow children more independence. An economic circumstances of need and impoverishment affects parenting undoubtedly. One famous ethnographic study in Brazil found that parents with weak or underfed children regard them as merely temporary visitors to their life-word, such is the rate of infant mortality (Scheper-Hughes, 1993). More recently, it has been shown that parents of street-connected children in Guatemala, through force of economic circumstance, prefer their children to work during the day instead of attend regular schooling (Stevenson et al., 2019b). Perhaps at the other end of the scale in terms of schooling and parenting, the term ‘tiger-mother’ has been used to describe mothers of Asian children whose regimes are more harshly geared towards academic success above all, with Asian-American mothers fitting this description more than European-Americans (Fu & Markus, 2014). It should be pointed out though that tiger-mothering has also been associated with poorer academic outcomes in some cases (Matsumoto & Juang, 2016).

Divisions of parental responsibilities between genders also show cultural variation. Fathers in Mexico, France and Italy indulged in a greater proportion of playtime activities than did those from other national groups, with German fathers most likely to eschew playtime or leave it to mum (Best & Ruther, 1994). Elsewhere, differences have also been observed in parental discipline regimes. As you may remember from your own childhood, several strategies are open to parents seeking to get children to do as they’re told. Compliant behaviour may be requested in the name of authority (because I’m in charge), feelings (jbecause so and so will be upset), consequences (because ‘x’ will happen if you don’t behave), rules (because that’s not allowed), or modelling (because good boys don’t do that). Comparing the use of these strategies across cultures, Conroy et al. (1980) found authority strategies to be common in a US sample, while stressing interpersonal relationships (such as feelings) was significantly more common in Japanese families. In a comparison of disciplinary techniques among cultural groups within the US, European American mothers preferred to use consistency, sensitivity and rule-setting, while Chinese Americans were more likely to yell and administer minor physical punishments, typical of an authority strategy.

It seems evident that from playtime to bedtime, parenting plays by different rules in different places. Arguably, cultural differences in parenting correspond with the transmission of norms, beliefs and behaviours that have an optimum survival value for those who operate in these cultures and for the cultures a whole. However, as we are about to discover, cross-cultural research into parenting has also revealed many common practices.

Cultural commonalities in parenting

Despite their differences, parents the world over operate in many similar ways. Certain common practices, such as baby-carrying to allow handsfree access to other activities, may go back over a million years (Konner, 1972). This may partly explain why the human infant has evolved into a relatively immobile creature, unable to do much for itself in the first months, compared with the young of some other species.

Another global parenting practice with a long history is weaning (the gradual process of accustoming mammalian offspring to do without their mother’s milk). Most non-industrial human societies share a pre-weaning phase of three to four years, which corresponds proportionally to pre- weaning in other primates as it represents one-third to a quarter of the age until female sexual maturity (Berry et al., 2011). There are, however, limits to the cultural universality of this practice, with parents in many industrialised societies typically weaning earlier.


Motherese: Also referred to as infant- directed speech (IDS) or 'baby-talk’, motherese refers to 'the spontaneous way in which mothers, fathers, and caregivers speak with infants and young children' (Saint-Georges et al., 2013).

Language development in children reveals another parenting habit that has been identified as a global phenomenon: motherese, which is also known as ‘parentese’, or ‘infant directed speech’ (Saint-Georges et al., 2013). It appears that many of the world’s parents are fluent in this strange language. This seemingly ubiquitous dialect involves vocal intonation patterns directed towards infants that are characterised by raising the pitch of the voice, exaggerated variations in sound (doing funny voices) and generally talking in nonsense syllables. Papousek and Papousek (1997) listened to motherese worldwide and pronounced it a truly international language, though they found parents throwing themselves into it in some regions more than in others. Japanese mothers, for example, were found to use more ‘singing and nonsense’ than comparison groups.

It appears, then, that while we can identify aspects of parenting that seem to be culturally widespread, the influence of culture is detectible even in these behaviours.

Culturally diverse parental styles and ethnotheories

So far we have presented evidence to suggest that while parents in different cultures may all nurture, protect and feed their offspring, they approach the role of the parent in diverse ways. Could it thus be argued that parenting in different cultures assumes differing styles; some authoritarian, others laissez-faire or easy-going, for example? Baumrind (1971) distinguishes between three parenting styles:

  • Authoritarian: requires high levels of obedience; displays little warmth.
  • Permissive: allows autonomy; displays warmth.
  • Authoritative: firm but fair; displays warmth.

The last of these is widely thought to be most associated with the development of healthy, stable, sociable children (Denham et al., 1997), as well as those with better school performance and more secure attachment with caregivers (Kerr et al., 2012). But how does this model fare when applied cross-culturally? Chen et al. (1997) found a correlation between authoritative parenting and good levels of social adjustment among Chinese children, thus supporting Baumrind’s hypothesis. Authoritative parenting has also compared favourably with authoritarian parenting in relation to levels of adjustment within families in Hong Kong (Chen et al., 2014). However, we should be aware that Baumrind’s typology arises from a Euro-American context, so it may not translate to Asia. We cannot, for instance, assume the connotations of authoritarianism,


Parental ethnothe-

ories. Theories and styles of parenting that originate in diverse locations and cultural belief systems.



Biological predisposition to behave in certain ways.

which may be negative in certain cultural settings, to be similarly negative elsewhere.

Rather than using imported typologies to rate parenting across cultures, it may be more useful to identify theories and styles of parenting that originate in diverse locations, known as parental ethnotheories (Harkness & Super, 2006). These indigenous ethnotheories underwrite the diverse ideas held by parents on matters relating to learning, sleeping, discipline and play. For example, in a study by Tobin et al. (1989), Japanese parents’ valuing of social interdependency (high levels of reciprocal interactions) was reflected in their approval of preschool groups of over 30 pupils (more social interaction is a good thing for my child and for our society as a whole). Such a positive view would not gel with a stereotypically European outlook, emerging from a more individualistic ethnotheory (too many children playing together may hold back my child). Parental ethnotheories also manifest themselves in the expectations parents hold for their children about how old they should be when they acquire various skills. For example, the mastery of various perceptual, cognitive and interpersonal skills has been anticipated at an earlier age by parents from more industrialised, more affluent regions (Flarkness & Super, 2006).

Parenting is a two-way street: the role of temperament

So far we have focused on variations and similarities in parenting practices across cultures. However, it should be obvious from our experiences as parents and children that the development of children is not just a question of how parents treat them. Rather, development is a two-way street along which parental influence and the temperament of the child interact. How a child turns out is due to its own biological predisposition to behave in certain ways, its parents’ and caregivers’ responses, as well as sundry other environmental factors (Berry et al., 2011). Having already discussed various parenting styles, let us now consider the role of an infant’s temperament in this ongoing interaction.

You often hear the phrase, Oh she has a lovely temperament. For psychologists, this term refers to a biologically based style of interacting with the world (Matsumoto & Juang, 2017) which, in turn, affects the parental response. Earlier we considered differing styles of parenting. Distinctions have also been made between various temperamental styles (Thomas & Chess, 1977):

  • Easy: adaptable, consistent in responses and emotions.
  • Difficult: inconsistent, moody, intense.
  • Cool: withdrawn at first, becoming responsive with familiarity.

Together with the parental response, these styles yield a developmental outcome. But do these styles vary cross-culturally? In one cross- cultural study (which you might not want to try at home), a cloth was laid over an infant’s head and face and the reaction described (Freedman,

1974). Chinese infants were more likely to simply let it happen, breathing through the cloth and lying still. A typically more agitated reaction was observed in an American sample. Other researchers have found temperamental differences in similar demonstrations, with Chinese and Navajo infants seemingly less irritable than those from the US and Ireland (Kagan, 1994), yet they have also been found to be more irritable in a comparison with Japanese counterparts (Loo et al., 2005).

What might precipitate differences in biologically predisposed temperament? The answer to this is not immediately obvious, although correlations have been found between average levels of blood pressure in mothers from certain cultural settings and infant irritability (Garcia Coll, 1990). Besides the mother’s health, ecological factors also correlate with temperamental responses in the first 30 days after birth. For example, Peruvian infants born at high altitude were more agitated during these early weeks than were those born around sea level (Saco-Pollit, 1989). Although this suggests that newborns from some cultures may be predisposed towards certain responses, these behavioural differences may equally be explained by identifying very early parental practices such as body contact, massaging and vocalisation in the first 30 days. Clearly, parental responses and temperament are an influential double act whose mutual influence cannot easily be separated.

Temperamental differences, it should be remembered, are no more than a biological predisposition towards certain responses. They may be expressed or inhibited by the parental and environmental influence. Furthermore, we cannot rely on labels such as easy, difficult and cool to be equally meaningful, or indeed equally adaptive, across cultures. A child who is labelled as having a difficult temperament may actually have a positive survival advantage in some situations. Certain environments may require an infant to be fussy, demanding or challenging in order to survive large families or scarce food supplies (DeVries, 1989). Once again this reminds us of the dangers of applying generalised typologies across cultures.

Parenting styles and temperamental predispositions are ingredients that interact to yield a developmental outcome, the growing human being. The cultural manifestations of these ingredients are manifold, but those that have been observed worldwide are the ones that represent the best survival prospects of the cultures themselves, as well as the families who make them up.

Limitations of research on culture and parenting

1 What do we mean by cultural differences in parenting? When drawing conclusions regarding the proposed relation between culture and childhood experience and parenting practices, we should be wary of what is actually meant by the term ‘culture’ in this equation. On closer inspection it may be that cultural influences mask material or economic differences between groups. For example, it has been noted that certain cognitive skills develop at different ages in different cultural settings. Closer inspection of this conclusion reveals that some skills are expected to develop later in less affluent nations, as well as in those where larger family size is the norm, or where mothers are educated to a lower average level (Willemsen, 1996). Findings like these suggest that what appear to be cultural differences in parenting often have a strong educational or economic element. Therefore, rather than labelling these differences as cultural, we may do well to recognise their socio-economic origin.


  • 1. What is the difference between interdependent and individualistic parental ethnotheories?
  • 2. Which of these three concepts is not thought to be biologically based?

a. Socialisation

b. Temperament

c. Maturation

2 Parenting is only part of the socialisation picture. Although research into the effects of diverse parenting practices on the development of children is informative, over-concentration thereon runs the risk of not recognising the role of the agents of socialisation that form Bronfenbrenner’s exo-system (see Figure 8.3). The effects of these exo-system values on the development of the child should not be ignored, especially when we consider that around three-quarters of non-European families in the US see themselves as extended organisations (Fields, 2001). Examples of research into attachments that are formed between children and extended family members will be considered in the next section.

Attachment and strange situations


Attachment patterns. Emotional ties between people.

Monotropy. The

infant's bond with the mother is (biologically) qualitatively different from any other, so any interruption to this bond is necessarily maladaptive.

Developmental psychologists have long recognised that attachment patterns (emotional ties between people, such as parents and infants) formed in early childhood have lasting effects on our sense of who we are (self) and on how we interact (sociability). An influential view of early attachment is the so-called epigenetic model (from epigenesis, meaning to multiply gradually). The emotionally healthy infant is seen as moving from a single caregiver (usually the mother) in the first year towards numerous secondary attachments with extended family and peers (Smith, 1980). Variations on this epigenetic pattern are seen by proponents of this model as detrimental to the child’s development (Bowlby, 1969). Proponents of the classic epigenetic view claim that the infant’s bond with the mother is (biologically) qualitatively different from any other, so any interruption to this bond is necessarily maladaptive. This is also known as monotropy theory (Bowlby, 1951).

Several writers have disputed the unconditional primacy of the maternal bond as an essential building block for healthy development. For example, after extensive naturalistic observations of infant-care-giver interactions, Schaffer and Emerson (1964) concluded that a maternal bond must be of high quality to be primarily beneficial for the infant. Arguably, then, a maternal bond per se is not a prerequisite for healthy attachment.

Ainsworth et al.’s 'Strange Situation' study (1978)

Figure 8.4 Ainsworth et al.’s 'Strange Situation' study (1978)

Mary Ainsworth’s et al.’s (1978) strange situation experiments (see Figure 8.4) sought to investigate further the primary nature of the attachment between mother and infant, and have been used to validate the epigenetic approach. In the original experimental scenario, a 12-month-old infant was observed interacting with its mother, then without her, then in variations on this them. Infants were rated according to the security of their maternal bond as an index of their reactions in each situation. Those with the most secure maternal attachments preferred mother to other carers, though they were not overly upset in the company of strangers. Ainsworth’s research bolstered the epigenetic view of the primacy of the maternal bond as a foundation stone for healthy development.

Culture and strange situations

Various cross-cultural versions of Ainsworth’s research have precipitated a debate about whether the securely attached child should be recognised as a global ideal. Indeed, even before Ainsworth’s studies were conducted, cross-cultural support for the epigenetic model of attachment came from Konner’s (1981) fieldwork with !Kung hunter-gatherers in the Kalahari region of southern Africa (T represents the linguistic clicking sound used by the group). During their first year, !Kung infants were with their mother for 70-80% of the time. This percentage subsequently fell as the social network expanded to include father and mixed-age peers.

Following Ainsworth’s original studies, versions of the strange situation experiment were conducted in many cultural settings in order to investigate the cultural universality of attachment patterns. The classic experimental format was used as a standardised design and then replicated in accordance with a cross-cultural research paradigm.

So, are Ainsworth’s findings replicated across cultures? Well, attachment patterns corresponding to ‘secure’ in Ainsworth’s scenario have shown up to be ideal arrangements in many cultural settings. When Posada et al. (1995) asked mothers from China, Germany, Israel, Colombia and Japan to rate the characteristics of an ideal child, their profiles tallied closely with Ainsworth’s securely attached model. Furthermore, more children invited the ‘secure’ classification than any other classification in a review of 14 attachment studies across four continents (van Ijzendoorn & Sagi, 1999). In another meta-analysis, van Ijzendoorn (1996) also found the securely attached pattern to be the most common across eight nations, inviting the portrayal of the securely attached child as the global ideal or norm.

Yet there is also counter-evidence. Grossman and Grossman (1990), Takahashi (1990) and others have revealed cultural variations in the degree to which infants in different cultures are assigned to Ainsworth’s three categories. In short, infants in some countries appeared to be typically more securely attached, or anxious-resistant, than they were in others. Such regional variations have since invited two alternative explanations:

  • Explanation 1: Some cultures yield more secure infants than others do. The strange situation scenario is a measure of emotional attachment that is equally valid (meaning it measures what it sets out to measure) in all cultures. Variable patterns show secure attachments to be more common in some places (North America) than in others (Germany). Furthermore, the anxious-resistant response seems to be more a feature of Japanese and Israeli society than it does elsewhere.
  • Explanation 2: The strange situation scenario is not meaningful in all cultures. The strange situation scenario is not equally valid across cultures. It has different meanings for participants in different contexts. We cannot assume that categories used in Ainsworth’s original study mean the same thing in all places. This standard scenario cannot provide us with ‘like-for-like’ cross-cultural comparisons of attachment patterns.

Defenders of the first of these explanations may conclude that North American infants are typically more securely attached than their German counterparts, with Japanese culture being associated with more anxious infants. Arguing for the second explanation, Grossman and Grossman (1990) suggest that Ainsworth’s notion of ‘anxious-avoidance’ has a different meaning in Germany, where it is reinterpreted as ‘autonomy’ and considered a virtue. Takahashi (1990) too stresses the special cultural meaning of the strange situation scenario for Japanese one-year-olds, who traditionally rarely stray from their mothers. The strange situation experiment is therefore especially stressful for them, leading to their over-categorisation as ‘anxious-resistant’.

Such diverse cultural interpretations reflect a methodological dilemma for Ainsworth’s paradigm. Explanation 1 invites the replication of her experiment in a standard form across cultures. Yet Cole (1992) expresses concern that this overrides the diverse cultural meanings attached to these experiments. The way forward, he argues, is to develop research scenarios that respond to diverse local meanings and parenting practices. This may involve the use of less standardised, more qualitative or ethnographic methods. Otto and Keller (2014) concur, suggesting that standardisation of research scenarios such as Ainsworth’s leads us to ignore culturally diverse contexts and meanings associated with parenting and attachment. Far better, they argue, to observe and compare parent-infant attachments in the culturally diverse naturalistic home environments (Otto & Keller, 2014). For example, in ethnographic research with the Efe community from the Ituri Forest in the Democratic Republic of Congo (Zaire at the time of the research), Tronick and Morelii (1992) researched multiple attachment patterns in a naturalistic setting. Efe infants of up to four months of age were observed interacting with carers other than their mothers for most of the time. Indeed, they interacted with at least five people per hour. Patterns of multiple caregiving emerged as the norm from these observations. Tronick and Morelii cite soc/'o- ecological explanations for the emergence of these multiple attachment patterns. In other words, elements in the social and physical environment of the Efe were seen as conducive to shared parenting. Efe values of co-operation and sharing, their communal living and community-based work practices, plus the dangers of the Ituri Forest, were all conducive to bringing up children communally. A socio-ecological approach to attachment, unlike a more biological, epigenetic view (monotropy), leads us to view parenting as a culturally diverse practice. Healthy development, for Tronick and Morelii, stems from childrearing practices that are influenced by diverse cultural settings.

Crittenden (2000) also urges us to interpret attachment patterns (multiple, epigenetic or otherwise) as adaptive responses to diverse environments. Granted, the secure attachment pattern as Ainsworth defines it is a globally widespread, effective adaptation. In the Ituri Forest, however, multiple attachments work pretty well too. Elsewhere, where cultural values or material circumstance are conducive, alternative attachment patterns represent other effective survival strategies.

Culture and the development of gender identity

The variable treatment of boys and girls worldwide reminds us that the socialisation of infants into gendered children and adolescents is influenced by culture. To conclude this chapter on childhood and culture, we will look at differing perceptions of gender in different places, beginning with an historical landmark study.

An historical episode in gender identity research

The twentieth century saw many changes in our understanding of gender and culture. Margaret Mead, then a young trainee researcher, was at their vanguard. Sex and Temperament (1935), her ethnographic account of three years spent with the Arapesh, Mundugumor and Tchambuli communities of New Guinea, was inspired by Franz Boas (1858-1942), her teacher at Columbia University. Boas rejected the idea, prevalent at the time, that so-called ‘primitive’ societies aspired towards the status of allegedly more civilised Europeans and North Americans. Rather, Boas sought to investigate culturally diverse, unique practices in order to assess the effect of culture on human development. Furnished with such notions of cultural relativism, Mead embarked on her three-year investigation into gender roles. She wanted to find out whether gender identity manifested itself differently from one cultural setting to another; specifically, in three Pacific communities.

Mead’s fieldwork took place in three sites on the north-eastern coast of the Pacific island of New Guinea, and yielded a detailed, qualitative account of her observations. She focused on aspects of life that related to how gender roles were socialised. She later wrote (1972, p. 196) that her goal was: ‘to study the different ways in which cultures patterned the expected behaviours of males and females’.

What emerged was three societies that apparently socialised gender roles in distinct ways. For theArapesh, Mundugumorand Tchambuli (see Figure 8.5), experiences of femaleness and maleness seemed qualitatively different. The diverse gender roles apparent in the three communities emphasised the importance of cultural factors on their formation.

Subsequent critics have offered Mead a mixed reception. Errington and Gewertz (1987) revisited the Tchambuli region to try to verify the

Gender identity in three Pacific communities original data

Figure 8.5 Gender identity in three Pacific communities original data (Kuper, 1994). While the more contemporary community showed some of the characteristics Mead observed (women were assertive and practical, men vain and decorative), there was no neat reversal of North American gender roles as Mead reported. More complex dynamics were at play. Neither gender neatly fitted the temperamental template of North American men or women.

Other reports of culturally diverse gender roles do support Mead’s view of culturally relative gender identities. In some Madagascan and Alaskan communities males are raised in ways conforming to North American feminine ideals (Gross, 1992). Elsewhere, some American Indian groups incorporate a third gender type of 'man-woman' (or berdache Barfield, 1997), who rejects warrior status to adopt feminine norms and behaviours. The term berdache can be traced to the Arabic ‘bardash’ (male prostitute) and has also been used to describe men dressed as women among North American indigenous Haudenosaunee (Iroquois) communities (Medicine, 2002). Further evidence of culturally constructed gender roles comes from differential manifestations of psychological androgyny, further illustrating the influence of culture on gender identity (see Figure 8.6).

Psychological androgyny across cultures

Figure 8.6 Psychological androgyny across cultures


Gender schemas. Ideas about appropriate behaviours for males and females.

We can reflect that Mead’s albeit simplistic view of culturally constructed gender roles remains a useful point of reference and is supported in principle by other field-workers. Arguably the development of gender roles in diverse settings may reflect differing expectations as to what it means to be male and female in different parts of the world. In other words, it may well be that gender stereotypes vary across cultures. Two researchers, Williams and Best (1982, 1990, 1994), have done much to reveal how much this is the case.

Gender expectations and stereotypes across cultures

Ideas about appropriate behaviours for males and females (or gender schemas) can be present in children at an early age (Westen, 1996). Even five-year-olds can have a good idea about the kinds of activities their culture sanctions for men and women. Global research into gender schemas suggests that there is plenty of agreement across cultures about the nature of these gender-appropriate behaviours. Williams and Best (1982) investigated this by giving a checklist of 300 adjectives (such as caring, dominant, submissive) to students from 30 nations and asking them to rate them as to whether they most typically described female or male behaviour. In most nations active, dominant and aggressive were deemed male-appropriate, while passive, weak and nurturing were seen as female behaviours. Furthermore, relatively stable gender-stereotypical behavioural expectations have been endorsed elsewhere (Rao & Rao, 1985; Trommsdorff & Iwawaki, 1989). These stereotypes, consistent across cultures, are endorsed by research into advertising (Matthes et al., 2016). Analysing television advertisement across 14 nations in Asia, Europe and North America, Matthes et al. found; women to be typically shown advertising domestic products, beauty and cleaning products; men to be associated with

IT and telecommunications products and cars, although this latter effect was not significant in all countries. Interestingly, these authors also found that, in contrast to previous work by Eisend (2010), gender stereotyping in advertising was not associated with indices of gender equality in different societies.

However, in their study Williams and Best showed how some ideas about gendered behaviour can vary across cultures. When asked to identify favourable behaviours, Japanese and South African respondents were more likely to select typically ‘masculine’ behaviours, with Italians and Peruvians preferring ‘feminine’ behaviours. Despite these variations it seems that some degree of global consensus prevails regarding gender-stereotypical behaviour patterns, with males and females across many cultures being expected to fulfil prescribed behavioural characteristics.

Widely accepted ideas about which behaviours are appropriate or desirable for boys and girls are likely to be communicated and generated via socialisation and visual media (Matthes et al., 2016). Gender-appropriate ideals are thus likely to survive through cultural transmission. The development and reinforcement of these gender schemas may be media-generated, with the mass media in the latter half of the twentieth century being especially awash with stereotypi- cally gender-appropriate portrayals of men and women (Fejes, 1992, Eisend, 2010).

We cannot discount the effect of stereotypical or gender-specific portrayals of boys and girls in the media, in relation to subsequent personality and intellectual development (Pacilli et al., 2016). Yet neither should we blame the media for the existence of gender stereotyping. Indeed, media outlets themselves are becoming more diverse and imaginative in their portrayal of gender, with an increasingly globally diverse media arguably contributing to the generation of ideal gender roles that differ from culture to culture, thus challenging the idea of consensual gender schemas.

What's your ideal male and female? Gender-role aspirations

Ideals about what is means to be a boy and a girl can differ between, for example, Latin America and Europe. In one study, Williams and Best (1990) asked participants what they thought males and females ought ideally to be like, rather than what they are like, as in their previous research. Interestingly, these gender-role aspirations varied across cultural groups. Participants from more affluent, traditionally Christian regions (Netherlands, Germany, Finland) were more likely to endorse egalitarian gender aspirations, wherein males and females ideally displayed less behavioural differentiation (both genders actively participating in the work economy, for example). Participants from less affluent, traditionally Muslim regions (of Nigeria, Pakistan, India) were more likely to endorse traditional gender aspirations, consistent with segregated roles and prescribed gender-specific behaviours.

How we idealise male and female gender roles appears to reflect less of a consensus than does a straightforward identification of how males and females are defined. Moreover, these gender-role aspirations may be undergoing some change worldwide. Research in a number of settings not generally associated with a westernised outlook has found many young women, firm in the belief that the woman’s role lies outside the home, in the economy (Gibbons et al., 1991 (Sri Lanka), Mule & Barthel, 1992 (Egypt)). Furthermore, there is, especially in some cultural settings, a greater and increasing level of diversity and flexibility of parenting in relation to historic gender roles (Springer, 2017). The LBGTQ movement too is raising awareness about the importance of parenting practices that rely less on traditional gender schemas (Springer, 2017).

Limitations of research on culture and gender identity research

  • 1 The problem of the imposed etic, again. When assessing gender stereotypes and expectations cross-culturally, some researchers have assessed the responses of participants from diverse cultural backgrounds against concepts of femininity and masculinity that were originally drawn from North America. For example, both Mead’s pioneering research and other studies that used Bern’s (1979) Sex Role Inventory cross-culturally employed this ‘how similar to Americans are they?’ model. Arguably, a less etic approach would involve exploring indigenous notions of masculinity and femininity as a starting point for investigating global gender identity.
  • 2 A call for methodological diversity. The ongoing research of Williams and Best has had an enormous impact on our understanding of the evident cross-cultural consensus in gender stereotypes, as well as variations in gender ideals. Yet most data on this topic have emerged via psychometric, questionnaire-based methods. These might usefully be complemented in future by a more qualitative approach, offering greater detail and insights into the cultural meanings associated with often delicate topics such as the use of child soldiers, or female genital mutilation. Such research might use indigenously generated issues as its starting point.

So, does culture change our experience of childhood?

The research suggests that the short answer to this question is ‘yes, to an extent’. For example, parenting practices and temperament appear to be susceptible to ecological and cultural influences, revealing diverse methods of adaptation. Furthermore, gender roles too, as well as our understanding of the idea that maleness and femaleness are influenced by culture, have undergone a series of changes in the past century.

Margaret Mead alerted us to cultural variations in gender identity, with more recent evidence suggesting a high degree of consensus across cultures about stereotypical gender behaviours. Tellingly, though, aspirations about how males and females ought to behave do seemingly differ from place to place. The fulfilment (or otherwise) of these gender ideals will undoubtedly provide fertile subject matter for future research into the link between culture and gender.


Chapter 8 centres on child development across cultures. The discussion begins by asking how culture, child development and parenting fit into the bigger picture of socialisation. After all, parental influence is only part of the wider context of socialisation. The importance of other agents of socialisation (extended families, for example) may vary from culture to culture. Likewise, parenting styles and attitudes towards parenting practice are culturally relative.

Cultural variations in rule-setting, bedtime routines and assertion of control have all been well researched by global psychologists, whose findings are discussed here. Besides differences and similarities in parenting, we consider the possibility that the child’s temperament (biological predisposition to behave in certain ways) might be influenced by properties of its environment. Research into the formation of secure, insecure and multiple attachments is covered extensively, with particular attention given to the methodological difficulties of applying established attachment theories in diverse contexts. A review of global research on the development of gender identity, gender expectations and psychological androgyny concludes the chapter.



Match up the definitions on the right with the terms on the left (see p. 210 for answers)


  • • Chen, W.W. et al. (2014) The relationship between perceived parenting style, filial piety, and life satisfaction. Hong Kong Journal of Family Psychology, 28(3), 308-314.
  • • Eisend, M. (2019) Gender roles. Journal of Advertising, 48(1), 72-80. doi: 10.1080/00913367.2019.1566103.
  • • Fu, A. & Markus, H. (2014) My mother and me: why tiger mothers motivate Asian Americans but not European Americans. Personality and Social Psychology Bulletin, 40(6), 739-749.
  • • Harkness, S. & Super, C. (eds.) (2006) Parents’ Cultural Belief Systems. New York: Guilford.
  • • Mead, M. (1935) Sex and Temperament in Three Primitive Societies. London: Routledge.
  • • Population Reference Bureau (PRB). (2018) Population Handbook. Available online at: (accessed 13 February 2020).
  • • Saint-Georges, C., Chetouani, M., Cassel, R. et al. (2013) Motherese in interaction: at the cross-road of emotion and cognition? (A systematic review). PLoS One, 8(10), e78103. doi:10.1371/journal.pone.0078103.
  • • Scheper-Hughes, N. (1993) Death Without Weeping: The Violence of Everyday Life in Brazil. California: UCL Press.
  • • Shweder, R. (2003) Why Do Men Barbecue? Recipes for Cultural Psychology. London: Harvard University Press.
  • • Williams, J. & Best, D. (1990) Sex and Psyche: Gender and Self Viewed Cross-culturally. Beverly Hills, CA: Sage.


Culture and psychopathology

<<   CONTENTS   >>

Related topics