Cultural and Social Differences

Approximately 2 million individuals are homeless in the United States (National Law Center on Homelessness and Poverty, 1999). Of these about 40 percent are families (National Coalition for the Homeless, 1999). Language, learning, and cognitive delays are common in preschool children.

Data from mothers and children in homeless shelters is complicated by factors such as poverty, health issues, and race and ethnicity. Nonetheless, we find that both children and mothers in homeless shelters exhibit deficits or delays in at least one of the following: auditory comprehension, verbal expression, reading, and writing (O'Neil-Pirozzi, 2003).

In the Deaf culture, among parents and children who are both deaf and for whom American Sign Language is the primary means of communication, motherese is conveyed by sign and facial expression. Use of sign can present a potential problem because facial expression marks both affect and grammatical structures, such as questions. With only limited use of paralinguistic cues, such as higher pitch and exaggerated intonation and stress, a mother's nonvocal facial expression takes on added importance as a conveyer of her intentions and as a device to hold a child's interest. Prior to a child's second birthday mothers of children with deafness use facial expression primarily for affect. There is a shift to more grammatical uses after that point . (Reilly & Bellugi, 1996).

Cultural differences are evident in the maternal behavior of Japanese and North American middle-class mothers. While American mothers talk more with their children and encourage them to respond, Japanese mothers engage in more rocking, carrying, and "lulling:' In responding to their infants, American mothers use more facial and vocal behaviors, while Japanese mothers are more nonverbal, responding with touch (Fogel et al., 1988). With toddlers, Japanese mothers employ more vocalizations similar to the American English uh-huh, which is not surprising given the importance of omoiyari, maintenance of harmony, in that culture (Maynard, 1986; White, 1989).

The intentions of American mothers are providing information and directing. In contrast, the Japanese mother exhibits fewer of these behaviors, preferring to use nonsense words, sound play, and emphatic routines, such as discussing feelings (Morikawa et al., 1988). Her productions are usually very easy for her child to imitate.

In general, Japanese mothers are less likely to talk about objects; when they do, it is often without the use of the object's name, used more frequently in the United States. Although both American and Japanese mothers use questions frequently, American mothers use them more in the context of labeling. It is not surprising, therefore, that American toddlers have larger noun vocabularies while Japanese toddlers have more social expressions (Fernald & Morikawa, 1993; Hess et aI., 1980).

Still, similarities exist across languages. Both American and Japanese mothers use linguistically simple forms when addressing young language-learning children, repeat frequently, and use intonation to engage the infant (Fernald & Morikawa, 1993). The common motivation for these changes seems to be an intuitive sense of the developmental level of the child.

Cultural differences may reflect three related factors (Schieffelin & Eisenberg, 1984):

1. The role or status of children.
2. The social organization of caregiving.
3. Folk beliefs about how children learn language.

We must also be careful not to assume that the way middle-class mothers in the United States interact with their children is the only way or the most correct way. In general, interactive patterns between children and their caregivers have evolved to fulfill the special needs of the populations and cultures in which they occur.

In the middle-class American family, the child is held in relatively high regard. This is also true among the Kaluli people of New Guinea. In contrast, the relatively lower standing of children reported in western Samoa and among some African Americans in rural Louisiana results in an expectation that children are to speak only when invited to do so (Ochs, 1982). It is important to remember that low status does not mean a lack of affection for children. Within these same rural southern African American communities, a child is not expected to initiate conversation but to respond to adult questions in the shortest possible form. A child is not expected to perform for adults, and most of a child's requests for information are ignored. What expansion exists is an expansion by adults of their own utterances, not those of the child. It is believed in this culture that children learn by observation, not interaction.

Middle-class American mothers talk with their children, not at them. Many maternal utterances consist of comments on topics established by a child through word or action. This tendency to follow a child's conversational lead is evidenced in maternal expansion and extension of the child's utterances. Although these semantically related maternal utterances can enhance language acquisition, it has not been proven that they are crucial to the process. While Chinese and Western mothers both interpret babbling as meaningful, talk about what their children are doing, do not overtly correct, and recognize that their infants understand some words prior to speaking, Chinese mothers use less expansion and conversational prompting and more direct teaching of language (Johnson & Wong, 2002).

Not all cultures value verbal precocity in children or demonstrate the adult modifications seen in motherese. Among the Kipsigis of Kenya and rural African Americans in Louisiana, for example, comprehension is more important than verbal production in young children; many of the utterances directed to them consist of directives and explanations. Kaluli parents and Samoan parents rarely follow their children's conversational leads. In all the cases, language acquisition does not seem to be slowed or delayed in any way.

Mothers may use other strategies that seem equally effective to those described in this chapter. For example, Kaluli mothers mentioned previously and some Mexican American mothers provide models of appropriate language for specific situations and direct their young children to imitate these models. In situations with other adults, children are directed by their mothers in the appropriate responses. This recycling of appropriate utterances for recurring situations is a language-learning device. Like semantically related adult utterances found in middle-class American homes, these predictable situational responses may be highly comprehensible to a child without complete grammatical knowledge (Snow, 1986).

The expectation of a quiet child does not necessarily reflect children's low status. Within the Apache nation, it is a societal norm to value silence from all people. In general, Japanese parents also encourage less talking by their children, although children are held in very high regard. Nonverbal behavior is more important in Japan than in the United States, and Japanese parents anticipate their children's needs more often, so children have fewer reasons to communicate.

The second factor, social organization of caregiving, also varies widely and reflects economic organization and kinship groupings. In some cultures, such as that of western Samoa, older siblings are more responsible for caregiving than in middle-class American homes. This arrangement is also characteristic of many inner-city households in the United States. There is no evidence, however, that children raised by older siblings learn language more slowly than those raised by adults.

Finally, folk "wisdom" on language acquisition affects the language addressed to the child. The Kipsigis of Kenya believe that a child will learn by himself or herself. Thus, there is no baby talk or motherese. A child is encouraged to participate in conversation through imitation of its mother's model of adult speech. The Kaluli of New Guinea also require imitation from a child in certain social rituals, even though the child may not understand what he or she is saying.

Among both middle- and working-class African American families, a general belief exists that children learn language by listening and watching, thus there is little need to adapt adult behaviors for a child (Scheffner Hammer & Weiss, 2000). Even so, middle-class mothers seem to have a "teaching agenda" that emphasizes production of language by their children. In general, middle-class mothers include more language in their child play and use a wider range of words with their children than working-class mothers (Scheffner Hammer & Weiss, 1999). As a consequence, middle-class African American infants initiate more verbal play and produce twice as many vocalizations as working-class infants. In contrast, working-class mothers have a very limited teaching agenda and interact less with their children.

Children are not limited to direct language input and can acquire language-based knowledge by drawing upon a range of experiences. They can also learn language by indirect means, such as conversational exchanges between other individuals. Children can learn language from speech that is not addressed to them (Oshima - Takane, 1988). This may be especially true of some pronouns, which may be best learned by observing their use in various contexts.

Television can also provide some limited input (Lemish & Rice, 1986). Unlike conversations, television is passive and does not require a response. In addition, the language provided by television is not related to ongoing events within a child's interactive context. Although having adults read to a child positively affects the size of the expressive vocabulary of English- and Spanish -speaking preschoolers, watching television does not have this beneficial effect (Patterson, 2002).

Even with all this variation, children still learn their native languages at about the same rate as middle-class American children. In general, in the United States, most adults treat a child as a communication partner. The language-learning American child is raised primarily by his or her parent(s) or paid professionals or paraprofessionals who model and elicit language. Even within the United States, however, there is no definitive pattern.

Of most importance among children in the United States are maternal stimulation and the overall quality of the home. For example, among African American families, a strong correlation exists between maternal sensitivity, responsiveness, stimulation, and elaborativeness and a child's cognitive and communicative skills at age 1 (Wallace, Roberts, & Lodder, 1998). Although socioeconomic differences exist within the African American community, there is strong evidence of these maternal behaviors among all African American mothers.