Abstract
Child rearing is a universal task, yet there are differing solutions according to the dynamics of socio-cultural milieu in which children are raised. Cultural models of what is considered good or bad parenting become explicit in everyday routine practices. Focusing on early mother–infant interactions in this article we examine the discursive practices and strategies that foster cultural values such as autonomy and relatedness. Drawing on micro-analysis of videotaped mother–infant interactions from middle-class families in Muenster, Germany and farming Nso families in Kikaikelaki, Cameroon, we aim at illustrating how diverse discursive strategies construct alternative versions of the child’s experience of self and self-in-relation-to-others. In each case, mothers draw on discursive practices that convey cultural norms and values that fit the relevant cultural context.
Keywords
Introduction
‘What is the single most important thing that one could do to influence the development of that infant?’ is a question raised by Tom Weisner (2002: 276) who then argues that the most important would not be physical protection, good nutrition, providing love and trust, insuring good attachment, or stimulating the infant although these are all important for development. The most important thing, he then goes on, is ‘to decide where on earth – in what human community – that infant is going to grow up’ (p. 276). The central message of this statement is that cultural every day activities in which children participate will have a crucial impact on the developing self and that children who grow up in distinct socio-cultural communities will follow different developmental pathways. While caregivers’ task in any culture is to raise their children to competent members of the society, there are culture-specific solutions for shaping these developmental processes (Keller, 2007; Quinn, 2005). Different socio-cultural environments offer different alternatives well suited to the specific physical, cultural, social, and economic opportunities and restrictions of these environments (Bronfenbrenner, 1979; LeVine, 1977; LeVine et al., 1994; Whiting, 1977, 1981).
Parents’ socialization practices are informed by cultural models (Holland and Quinn, 1987) of good parenting that have evolved in an adaptive process to these different historical, ecological, and cosmological dimensions. There is obviously a variety of cultural ideas of what it means to raise a child to be a good person just like there is a variety of cultural environments. Nevertheless, it has been argued that there are specifically two dimension that are important everywhere: the dimension of autonomy, and the dimension of relatedness. These dimensions can be regarded as universal human needs that co-exist in any given culture (Bakan, 1966). There are, however, culturally distinct ways in which both dimensions may be particularized in daily activities, which can be systematically related to particular socio-cultural and sociodemographic profiles (Kagitçibasi, 2005; Keller, 2007).
From birth on, infants habitually and tacitly internalize cultural knowledge through repeated participation in culturally shaped routine practices through which the child learns how to participate in the larger cultural community (e.g. Miller, 1996; Quinn, 2005; Rogoff, 2003). Communication plays a major role in this process (e.g. Ochs and Schieffelin, 1984; Ochs et al., 2005; Schieffelin and Ochs, 1986). Discursive practices, comprising both nonverbal and verbal communication express cultural and social values as well as cultural notions of self and modes of interpreting social reality (Keating and Egbert, 2004). Studying caregivers’ discursive practices with infants therefore offers access to cultural beliefs and expectations regarding the nature of children’s development. Analyses of caregiver–child interactions have demonstrated how particular cultural discursive practices lead to the construction or limitation of children’s autonomy (Fasulo et al., 2007), how children gain access to cultural understandings of norms and transgressions (Pontecorvo et al., 2001), acquire an understanding of their roles of participants in interactions (Fatigante et al., 1998), or of family time and family togetherness (Kremer-Sadlik et al., 2008), as well as culturally appropriate senses of self-in-relation-to-others (Miller et al., 1990, 1996). Drawing on a discursive practice approach, we therefore want to investigate how cultural orientations are communicatively co-constructed in mother–infant interactions. Special emphasis is given to how autonomy and relatedness are mediated in these interactions. We focus on two distinct cultural communities: the Nso in the village in Kikaikelaki in the Western Grassfields of Cameroon and middle-class families in the north German city of Muenster. Muenster was chosen because it reflects the cultural background in which the two first authors (who both are German) live and work. Kikaikelaki was chosen because it is the cultural context in which the third author (who is a Nso) grew up and is familiar with. Moreover, previous studies comparing North German middle-class families with Nso farming families could show systematic differences with regard to autonomy and relatedness (see Keller, 2007, for an overview) as we will outline in the following.
The study sites: Kikaikelaki farmers and Muenster middle-class families
Muenster is a university city of about 281,000 citizens and a well-developed administrative and educational infrastructure. The city is marked by a strong economy and a low unemployment rate compared to other parts of Germany (Jahresstatistik, 2008). Average income amounts to 33,498 Euro compared to the national average of 31,192 Euro. Forty-three percent of the population hold a high school degree, 22 percent hold a university degree. There is a high percentage of single households and only about 18 percent of the households in Münster include children. Families live in their own condominium or house, often with a separate room for the child from early on.
North German middle-class families have been found to value self-realization, self-confidence, and the ability to judge and to assert oneself (Pross, 1982), as well as children’s psychological independence and the development of individual competencies (Keller et al., 2005a; Keller et al., 2005b). They encourage the child to sleep alone at an early age (LeVine and Norman, 2001; Norman, 1991), and to play by themselves (Grossmann et al., 1985) thus fostering autonomous self-regulation and self-reliance (Selbständigkeit). Children’s uniqueness is emphasized for example in the choice of their first names (Keller et al., 2005b) or in socialization goals like ‘learning to be different from others’, ‘expressing own ideas’, or ‘being assertive’ over goals like ‘learning to share with others, obeying the parents’, or ‘maintaining social harmony’ (Keller et al., 2006). Mother–infant interactions have been found to be dominated by face-to-face contact (Keller et al., 2005c) with exclusive attention, and unconditional child centeredness with responsivity to a child’s personal wishes and preferences. Infancy research suggests that such contingent responses toward infants’ positive communicative signals support the perception of autonomy (Keller, 2007). These infant care practices correspond to the general Western conception of maternal sensitivity (Ainsworth et al., 1978): a good mother is expected to be aware of an infant’s signals, accurately interpret them, and promptly respond by showing empathy and insight into the baby’s wishes and moods.
Kikaikelaki village is part of a locally ruled chiefdom (fondom) of the Nso who live in the northeast corner of Cameroon’s North West Province. The total Nso population has been estimated to comprise 217,000 people (Goheen, 1996). Most Nso are Christians but many are still deeply rooted in traditional beliefs and continue to practise indigenous beliefs leading to a mixture of the new religion and traditional practices (Mbaku, 2005). Families earn their living primarily through subsistence agriculture (Nsamenang and Lamb, 1993) and the average amount of formal schooling is very low (Keller, 2007). The village is composed of patrilineal compounds, consisting of houses that are usually unfenced exposing children to the wider village life from early on. On average 6.7 persons live in one household (Yovsi, 2003). Large families with many children are favoured since children form the basis of symbolic and material wealth (Nsamenang, 1992; Verhoef, 2005), meant to be shared throughout a community. Children grow up in a dense social network including parents, siblings, relatives, grandparents, and neighbours (Yovsi, 2003). Older children are expected to help on the farm, do household chores, and take care of younger siblings (Mbaku, 2005).
The Nso community has been characterized by norms of collective responsibility, sharing, and exchange (Goheen, 1996; Nsamenang and Lamb, 1994) as well as harmonious and hierarchically organized relationships between family members and the wider social reference group (Mbaku, 2005; Nsamenang and Lamb, 1993; Verhoef, 2005). Previous studies found that Nso mothers’ conceptions of good parenting centre around obedience, respect for elders, conformity, and compliance to rules as well as social responsibility and subordinating individual interests to those of the group in favour of a strong community spirit (Nsamenang and Lamb, 1994; Yovsi, 2003). For instance, they value socialization goals like ‘obedience to parents’, ‘learning to share with others’, and ‘maintaining social harmony’ above goals like ‘learning to be different from others’, ‘expressing own ideas’, or ‘being assertive’ (Keller et al., 2006). The goal is to socialize children towards acquiring a ‘good character’. It is to this end control and regulation is often used in child care. It serves to prevent the child from developing a sense of pride about his own achievement which would be regarded as ‘showing off’ and a ‘bad character’ (Nsamenang, 1992; Nsamenang and Lamb, 1994; Tchombe, 1997). Infant care primarily focuses on survival, physical growth, and motor development. Mother–infant interactions have been characterized by a frequent body contact and motoric stimulation like moving the infant up and down (Keller et al., 2005a). Such stimulation is believed to accelerate motor development and make the child strong so that the child can help in the household from early age on (Keller et al., 2002). Physical stimulation also allows parents to check on possible health problems. This corresponds to the Nso conception of maternal sensitivity defined in terms of closeness, monitoring, instructing, training, directing, and controlling infant activities (Yovsi et al., 2009).
The study
Data corpus
The data corpus is part of a larger video-archive collected from October 2002 to November 2003 in Kikaikelaki and from July 1995 to June 1996 in Muenster. The families were recruited following local practices. In Kikaikelaki the hierarchical system required permission from the chief before getting in touch with families through a health centre. The Muenster mothers were contacted individually in hospitals and preparation classes. Twenty native German mothers living in Muenster and 20 native Nso mothers from Kikaikelaki consented to participate. The Muenster mothers were between 26 and 40 years at the time of infant birth (average 30.7 years) and had an average school education of 14.4 years. All infants were firstborn. The Nso mothers were between 17 and 47 years at the time of infant birth (average 27.8 years) and had an average school education of 8.5 years. They all lived from subsistence farming. Four infants were firstborn, 16 were later born. 1 They were video recorded for 10 minutes on a weekly basis over a total period of 16 weeks. The present account draws on the recordings from the 12th week session, when infants were three months of age.
Procedure
The families were visited at home and videotaped in settings that reflect typical social encounters with infants. The Nso mothers were mostly sitting in front of their house, in an open space with people passing by. 2 The Muenster mothers were inside the house, usually in the living room or the kitchen, alone with their child. The mothers were told that we would like to learn more about parenting and childcare in different cultures. For this purpose, we would like to videotape mother–infant free-play interactions. 3 The interactions where videotaped for 10 minutes by local research assistants.
Analysis
We transcribed interactions following the conventions common in conversation analysis (Jefferson, 1984). Some notations were added to include specific features of infant communication (see Appendix). The Nso interactions where directly translated into English by a native Nso research assistant since the written form of the local language, Lamnso, has only recently become available. Other modalities of communication such as body movement, gaze, mimic and gesture were also included in the analysis. Transcripts are presented in a format suggested by Ochs (1979) allowing for parallel vertical lines arranging nonverbal behaviour co-occurring with verbal/vocal utterances for mother and infant, respectively. Analysis was based on conversation analysis (Heritage, 1984; Sacks, 1992; Sacks et al., 1974) and discourse analysis (Potter, 2007; Potter and Wetherell, 1987). The approaches have in common that they take a practice perspective on language and aim at analysing what is socially accomplished in an ongoing interaction on a micro-analytic turn-by-turn basis. While conversation analysis aims at inferring how speakers take on certain roles or identities through their talk and hence produce social order, discourse analysis aims at deriving recurrent discursive patterns (‘interpretative repertoires’) that reveal how members of a social group construct versions of the world.
By systematically comparing the two data sets, groups of sequences were derived that show similar and contrasting patterns of communicative strategies that point to shared understandings and interpretation of interaction in each case. Analysis revealed salient different patterns between the two groups which will be discussed in the following. The excerpts discussed in the following represent typical patterns found repeatedly within each group.
1.Dyadic turn-taking and rhythmic-synchronous structuring
While features like turn-taking, repetition, and musicality were found in both groups, there were salient differences in the way and extent to which turn-taking is established and in the extent and structure of repetitive and rhythmic features of communication.
Within the Muenster group, the prevailing pattern was a diachronic narrative-eliciting structuring along dyadic turn-taking with a question-answer pattern, interpreting infant’s vocalizations as communicative acts, and providing missing parts of the conversation on the child’s behalf. Mothers were found to establish a narrative structure by 1) explicitly prompting the child to narrate, 2) referring to the child’s past experience and announcing upcoming events. In doing this, mothers would often gradually unfold a narrative as the interaction developed.
The following is an excerpt of an interaction of a 29-year-old mother of the Muenster group with her three-month-old daughter Tina. 4 The interaction takes place in the living room with the mother sitting on the sofa and her daughter lying next to her. The excerpt starts shortly after beginning of the recording:
Excerpt 1: 5 time: 00:01:07
Several features are noteworthy in this excerpt. First, the mother follows the child’s lead by bending down and re-establishing face-to-face contact. She then takes up on the child’s signals by mirroring the child’s (assumed) internal state (being tired) and further elaborating on it by successively adding information and unfolding a narrative. By relating the child’s recent experience to the present, she brings the child’s experience in a timely autobiographical structure of a past and a present. Second, by reformulating her statement as a question (line 16–17) and answering on behalf of the child (line 18–19), she initiates a turn-taking pattern similar to adult conversation positioning the child as a quasi-equal communicating partner.
We find a similar pattern a few turns later:
Excerpt 2: time: 00:01:20
The mother uses several strategies to establish dyadic turn-taking: she calls the child’s attention and leaves a pause for the child to take up the next turn (lines 21–24). She takes up the child’s behaviour and elaborates it by first mirroring what the child is doing and adding a tag question (‘aren’t you?’) (lines 24–25). By using such ‘facilitative tag’ (Holmes, 1984) the mother invites the infant to take a conversational turn and establishes a pattern of protoconversation. In both excerpts, the mother produces repetitions of partial sentences or paraphrases a previous utterance. On a meta-communicative level, this discursive strategy functions to express participation in the ongoing communication (Tannen, 1987), to ratify B’s communicative signals and to demonstrate focus and attention (Johnstone, 1994) on the child. It can hence be seen as a means to keep protoconversation going in the interaction.
A very different pattern was found in the Nso group: a clearly prevailing feature was the highly rhythmic and repetitive character comprising musical chorusing as well as repeated addressing of the infant by name. The infants were greeted by a variety of names such as proper names, gender role (‘boy/girl’), term of endearment (e.g. ‘little child’), often in combination with a term of respect (‘grandma/grandpa’), or with a linguistic appendix (o:: as in ‘girl=o::’). In synchrony with these utterances, mothers would typically rhythmically bounce or shake the child.
The following is an excerpt of an interaction of a 32-year-old Nso mother with her three-month-old son, Francis. The interaction takes place in front of the house with the mother sitting on a chair holding the child in front of her. The excerpt starts shortly after beginning of recording.
Excerpt 3: time: 00:00:48
This sequence shows a large amount of onomatopoeic rhythmic vocalization that gives the impression of imitating an instrument and parallel rhythmic body stimulation accompanied by smiling and laughing (lines 14–15). This has an inviting character as if the mother tries to say ‘join in’. The child is addressed by name in a very repetitive way with hardly any pauses but simultaneous rhythmic body stimulation (lines 16–19).
This pattern of rhythmic onomatopoeic vocalization with parallel body stimulation continues as the interaction proceeds. Through chiming of the repeated vocalization and simultaneous kinaesthetic stimulation, the mother brings the child’s bodily and auditory experience in unison with her own behavioural and vocal expression. In terms of musical aspects, the repetitions in this interaction result in a rhythmic pattern that creates ensemble. The importance here seems to lie not primarily on the content of what is being said, but on establishing and maintaining a connection through a synchronous rhythmic pattern. The metamessage on the relational level conveyed here is close intimacy through the construction of mutual rapport and experience.
2. Co-operative and normative-hierarchical strategies
A second major difference in the way mothers structured the interactions can be described in terms of granting autonomy to the child and of varying degrees of directness to which mothers exert control over the flow of the interaction. These patterns became particularly evident when negative affect was expressed by the child. Table 1 gives an overview of the different types of reactions in the two groups.
Reaction to baby’s expression of negative affect
In the Muenster group, the prevailing pattern in handling a child’s expression of negative affect was the use of co-operative and negotiating strategies. Mothers try to find out what is wrong with the child by asking whether the child is hungry or tired, making tentative suggestions, and asking for confirmation (e.g. ‘I think you are tired, can that be?’), and encouraging the child to express the reason for the discomfort (e.g. ‘What are you complaining about?’). The emerging pattern is one of adapting to the child’s presumed preferences and treating the child as quasi-equal partner.
Excerpt 4: Muenster mother-infant dyad (same as in excerpt 1). After having playfully interacted for almost 10 minutes, the child starts to whine at the end of the recording session. Time: 00:09:01
The mother reacts to B’s whining by expressing non-acceptance in form of a rhetorical question (line 233). Rhetorical questions (RQ) are not information-seeking questions but function to challenge propositions and actions in order to redirect or terminate the actions of an addressee (Schieffelin, 1986). She continues by a repeated negative imperative. The repetition intensifies her response, however, the playful intonation softens the strictness of the negation. She uses the inclusive first-person plural pronoun form ‘we’ (line 235) which functions to invoke a sense of collaboration and shared goals by utilizing ‘point of view operations’. This is a strategy used in adult-adult exchanges for politeness and extended here in a context intended to alter the child’s behaviour (Brown and Levinson, 1987). This move also softens the tone of the mother’s request and appeals to imagined reason of the child (Sirota, 2006). It is interesting, in this context, that the mother continues by referring to her own personal feeling and opinion as justification for the restriction against crying (line 242). This appeal indirectly requests B to respect her individual needs and positions B as quasi-equal negotiating partner. However, she also refers to herself in the third person as ‘mommy’. By using this membership category she positions herself as someone who is entitled to give instructions to the child. She conveys a social norm (crying is a disapproved behaviour) but simultaneously grants the child autonomy and freedom in other domains by telling her that she may do what she wishes except crying (line 242).
Another, however less prominent, pattern was to use downplaying and indirectly controlling discourse strategies, for example, irony or mitigating devices as the following example illustrates:
Excerpt 5:
This sequence differs from the previous example in that the mother reacts with a vocal expression with an exaggerated intonation contour followed by playful teasing in form of an indirect reproach that is keyed as a question (line 1–2). This formulation suggests that M does not really see a severe problem in B’s crying. This is even further emphasized by the prosody of the vocal expression in line 6 followed by a comment phrased in an exaggerated fashion and formulated in the third-person singular past tense thus resembling a dramatic quotation from a novel. This linguistic device makes the expression clearly ironical. Ironic expressions intend to draw attention to some discrepancy between the world as someone else is describing it and the way things actually are (Wilson, 2006). That is, from B’s crying an observer might infer that she is suffering but M calls this into question. B calms down and briefly smiles. The mother takes this up by teasing her about not knowing whether to laugh or to cry (lines 9–10) and ‘pities’ her in an exaggerated way that underscores the ironic tone in her message (lines 15–17). This interpretation of teasing the child is confirmed in line 18 in which M mirrors that her reaction (from B’s perspective) is really ‘mean’ and starts laughing, as well as in line 20, where she repeatedly encourages B verbally and nonverbally to move on and ‘rant’. Teasing here can be interpreted as pointing to B’s complaining as being done in an overdone manner. It may also be considered ‘a mild and indirect form of reproof of a mild kind of transgression’ (Drew, 1987: 250). M thus constructs B’s problems as seemingly ‘unspectacular’ and in no relation to the expressed behaviour and as mildly socially unacceptable behaviour.
Finally, some Muenster mothers used straightforward directive strategies to promote time orientation and self-regulation.
Excerpt 6:
The mother uses the negative imperative ‘don’t say’ (line 1) which indicates that what follows (‘you are hungry yet again’) carries a socio-moral message with it. She constructs the interaction in a way that 1) indicates that she knows what the reason for B’s fussiness is (being hungry), 2) there is a certain feeding schedule to stick to. This is confirmed in line 3 where the mother claims that the child has been given a certain measurable amount of food indicating that food will be provided not according to the child’s (assumed) desire but according to a certain schedule that the child has to stick to. Since the time for the next portion has obviously not yet arrived, there is no ‘objective’ reason for B to cry which leads her to prompt B to regulate himself by sucking the thumb and to instruct him physically how to do so (lines 5–8).
In the reactions of the Kikaikelaki mothers to their child’s expression of negative affect, there was less variation and there were overall fewer instances of situations in which babies started whining or crying. The Nso mothers commonly used discursive devices to exert direct control and positioned the child as having to obey and comply in a hierarchical setting. Although reactions to whining varied in their intention and intonality across dyads, the reactions followed a very similar pattern: vocal distraction and rhythmic animation, expressing disappointment and scolding the child, conveying a social norm, commanding the child vigorously to stop, asking rhetorical questions indicating that there is no reason to cry, and assuring relational accord after the child complied. This pattern is illustrated in the following excerpt.
Excerpt 7: Nso mother-infant dyad. The mother has been rhythmically bouncing her child for about one minute. The child has been frowning and staring into the camera and now starts to whimper.
In this excerpt, the mother abruptly shakes the infant when he starts whining. She poses a question without providing an answer or leaving any pause indicating that hers is a rhetorical question. As mentioned earlier already, RQ’s function to challenge propositions and actions in order to redirect or terminate the actions of an addressee. RQ’s can thus be conceived of as verbal forms of social control used in order to change B’s behaviour. There is an underlying proposition to the question ‘Has someone touched you’, namely ‘there is no good reason for you to be crying, therefore, you should not cry’ (if someone would have ‘touched’, i.e. hit the child, he would have had reason to cry, but since this is not the case, there is no reason). The mother then continues by a vigorous repeated negative imperative (line 37) which intensifies her response and then tries to gain the baby’s attention by vocalizing (lines 39–40) and directing his attention to a stimulating object (lines 41–42). The mother uses directives as she prompts the child repeatedly to look at and take the object. Eventually she ‘makes’ the child do what she wants him to do (line 43) by pressing the leaf into his hand in an imposing gesture. We argue that she confirms in this way that she does not accept whining behaviour and that she expects immediate compliance and obedience from the child.
The child gradually calms down and mother and child engage in a playful sequence of alternating vocalizations. The child starts to whimper a few times, however, and eventually starts crying again:
The mother uses a shaming expression (‘terrible’) indicating that the child’s crying behaviour is socially unacceptable. She verbally and physically prompts the child to redirect his attention (by shaking him and prompting him to look at and take the leaf), followed by a repeated negative imperative. The reason she mentions is that it is not common in this area to cry. Note also that she first uses a cooperative expression, ‘we’ (line 130) and then switches to ‘they’ (lines 131–139) indicating a normative discourse. In a stern voice she commands the child to stop and again poses a rhetorical question (133) similar to the first excerpt. Note that after the baby stops crying (line 134), the mother smiles and the tone of her voice becomes softer. In this strategy we argue the mother moderates the initial severity of her reaction and re-establishes social harmony in her interaction with the child.
Conclusion
In this article we have argued that discursive practices are a rich source for the study of how a cultural understanding is produced and mediated in everyday interactions. The aim of this articel was to illustrate how cultural expectations are mediated already in early infancy. For this purpose we compared early mother–infant interactions in two distinct cultural groups: middle-class families from Muenster, Germany and farming families in Kikaikelaki, Cameroon. Microanalysis of these interactions allowed to lay open some of the interactional processes that mediate cultural knowledge about the child’s role of self and self in relation to others.
Our analysis revealed that, while there seem to be similar communicative devices that the Muenster and Nso mothers of our study use in play interactions with their infant (such as focusing attention, establishing rhythm, or taking up on what the child is doing), they augment these devices with strikingly different strategies for socialization. We have further argued that mother–infant communicative practices need to be understood against the backdrop of the historically evolved cultural logic relevant to a given community.
In the Muenster group, the child is positioned (Davies and Harré, 1990) as quasi-equal partner who responds and engages in turn-taking. The child receives exclusive attention by the mother through repeated attention seeking on her part and her repeated uptake and elaboration of behaviours initiated by her child. Protoconversation is established and sustained along a dyadic, narrative turn-taking model. Interaction is facilitated by mirroring the child’s inner state, imputing meaning to the child’s bodily and vocal expressions, and by taking the perspective of the child. The focus is on the individual and agentic experience of the child. On the relational level, we argue the child has the opportunity to learn to perceive him- or herself as an individual who is encouraged to express own thoughts and preferences but who also needs to negotiate own preferences with those of others. We contend that such strategies are consonant with and likely support the development of a sense of uniqueness, individuality, personal choice, and autonomy. Including the child in a negotiated solution finding process may encourage a child even from the earliest months of life to express personal preferences. These are important lessons to be learned in order to be successful in a society marked by self-determination and self-reliance. More straightforward directive strategies are used with regard to a feeding schedule. We argue that these strategies are likely to promote time orientation and self-regulation which prepares the child for the broader cultural practice of having scheduled daily activities. These strategies not only convey that there are certain time schedules within the family to be observed but also promotes self-regulation independent of others (e.g. by sucking one’s finger).
In the Nso group, the child is positioned as a subordinate novice whose activities are actively and lopsidedly structured by the mother. Signs of negative affect by the child are responded to by strategies requiring compliance. Social interaction is facilitated by jointly rhythmic co-participation and simultaneous kinaesthetic stimulation. Intimate contact is established by bodily proximity as the mother stays in constant skin-to-skin touch with the child, but also and mostly through chiming rhythmic interaction. Sharing in the fluidity of rhythmic patterning, by moving the child in the same rhythm as mother speaks and moves creates a pattern of symmetrical co-participation rather than a negotiable dyad. This communicative pattern provides a context in which the child can learn emotional symbiosis as well as proper demeanour and avoidance of expressing negative affect. We infer that these strategies are consonant with Nso emphases on social responsibility and will support the development of compliant behaviour as well as close emotional enmeshment. In an environment that is characterized by a highly hierarchical social structure and high appreciation of interrelatedness and harmony in relationships, especially among the family but also among the wider social reference group, these strategies are functional as they prepare the child to maintain social harmony and good relationships with other people of the community. In line with other authors, we argue that rhythmic speech patterns provide a means of connecting individuals in close relationships and becoming members of a social group such as families and larger cultural communities (Cowley, 1994; Merritt, 1994). The different communicative strategies prevalent in the each group can hence be regarded as fostering values that will prepare the child for the relevant cultural organization of life.
The different communicative practices correspond to the prevailing cultural models of good parenting. In Germany, a good mother takes up the signals of the child by interpreting the child’s intentions, feelings and thoughts. Mind-mindedness (the tendency to focus on the mind of infants) in the first year of life is believed to contribute to subsequent child’s mental abilities (Meins et al., 2001). She is expected to treat the child as ‘quasi-equal’ partner and to teach the child self-reliance. In the Nso view, a good mother monitors, directs and controls the child’s activities by taking the lead of the interaction. She is expected to primarily teach the child proper demeanour and respect (Keller, 2007; Keller et al., 2008). The mother‘s priority is to keep the child calm, that is, prevent the child from crying. This is also related to the belief that if the child does not show any negative signals he or she is happy and in good health. Scolding a child is not conceived of as unloving expression but as an expression of responsibility toward the child. Moreover, by taking back some of the initial harsh reaction and smiling at the child after he obeys, the mother creates a close emotional bond with her child. The rhythmic stimulation is considered by the Nso as a useful strategy to keep the child in a happy mood by distracting the child from any negative moods (Yovsi, personal communication).
Similar differences have also been found between British mothers and mothers in a kwaZulu community in South Africa (Cowley et al., 2004). Cowley and colleagues found highly rhythmic and lopsided patterns in mother–infant interactions similar to our findings among the Nso which leads us to suggest that there might be similar cultural orientations with regard to good infant care in these African communities. The protoconversational pattern found in the British group was similar to that of our Muenster mothers and might reflect similar orientations among North European middle-class mothers. We caution however against overgeneralizing our findings. As pointed out elsewhere (Keller, 2007; Keller et al., 2008) there exists for instance considerable variability within the Nso culture regarding mothers’ ideas on good child care depending on sociodemographic differences such as formal schooling and age. While some cultural orientations might be shared by the broader given society, we would, for instance, expect systematic differences when comparing urban middle-class Nso mothers with farming Nso mothers, or when comparing working-class north German mothers with little formal schooling with the Muenster mothers.
Concluding, we want to stress that socialization is a lifelong process and that culture is not a static entity. The process of acquiring culture is not a process of passively internalizing a given cultural model but a continuing dialogical process between self and society that primarily takes place through communicative interaction. In order to take into account both individual and cultural change future research needs to draw on longitudinal studies to follow up on the developmental outcomes. For comparability purposes, the present study used video recordings of mother–infant play interactions. Future research on children’s participation in diverse everyday practices that also include interactions with other caregivers may provide a more holistic view of infants’ experience in cultural context.
Footnotes
Appendix
Explanation of transcription conventions used in the above excerpts (Jefferson, 1984; Ochs, 1979):
↑↓ pitch movement
CAPITALS speech that is obviously louder than surrounding speech
°I know it,° obviously quieter speech
(4) pauses in seconds
(.) micropause, hearable but too short to measure.
: elongation of the prior sound
> < speeded-up talk
= immediate ‘latching’ of successive talk
hhh Aspiration (out-breaths); proportionally as for colons
.hhh Inspiration (in-breaths); proportionally as for colons
(text) Additional comments
Voc vocalization
(WN) whining
(GR) grunt
(CR) cry
> indicating direction of gaze
Acknowledgements
The project was funded by the German Research Foundation (DFG). We thank Prof. Dr. Arnold Lohaus, University of Bielefeld, for cooperation in assessing the Muenster data. We would like to express our gratitude to Prof. Janet Dixon Keller, University of Illinois for her very helpful comments and reflections on a previous version of this article.