Bilingual child-rearing in cross-national families: parental considerations

Second Language Acquisition

Advantages and Disadvantages of Bringing up Children Bilingually

Much of the debate on bilingual education is wasteful, ironic, hypocritical, and regressive. It is wasteful because instead of directing attention to sound educational practices, it has led to advocating specific "models" based solely on what language should be used for what purpose. It is ironic because most attacks on bilingual education arise from an unfounded apprehension that English will be abandoned in the United Kingdom, whereas, in fact, the rest of the world doubts the opposite; the lure of English and attention in European traditions are seen by non-English-speaking countries as a danger to their own languages and traditions. It is hypocritical for the reason that most challengers of using languages other than English for teaching furthermore want to endorse foreign language requirements for high school commencement. Additionally, it is regressive and xenophobic since the rest of the world regards ability in at least two languages to be the sign of good education. We shall expound on the debate about the advantages and disadvantages of bringing up children born to cross-national and cross-cultural parents.

The political struggle to defend the existence of bilingual education in schools has wasted much energy in the search for a "perfect" model. The recent history of bilingual education is replete with various models, all posing as panaceas. Overreliance on particular models often detracts from scrutiny of what really happens in schools. When proponents of bilingual education let themselves are drawn into the battle over language choice, they too often lose sight of what should be their central goal: providing quality education to such students in ways that integrate them into both their own and the majority culture. If educators could ignore their particular biases about language use they would discover sufficient evidence to orient them toward providing effective education in any language. They would recognize that the mission of schools is to educate students so that they have choices when they graduate. Educating bilingual students has to go beyond merely teaching them English or merely maintaining their local language. The world of professional work understands that graduates reach not only high-level literacy abilities in English, and even familiarity of other languages, but also logical ability and the aptitude to study new scenarios and concepts. Bilingual students have not just the ability but also the right to be ready to face the challenges of current society.

Disapproval of bilingual education is not all groundless (Bialystok, 2001). Some bilingual programs are unsuitable for delivering quality education even if they have graduated a few victorious students. A great deal of the credit goes to the heroic efforts of individual teachers (Brisk, 1990, 1994). Advocates must admit that many bilingual programs are substandard. Rather than submitting a comprehensive approval for programs on the foundation of whether they use the children's mother tongue, supporters of bilingual education should be selective by supporting just those programs and schools that stick to the principles of high-quality education for bilingual students. Bilingual education too frequently is victim to political, financial, and social hurdles that feed on adverse attitudes toward bilingual programs, instructors, students, their families, languages, and traditions. Such attitudes convert into school character that restricts good education for language minority students. This thesis attempts to portray specific characteristics of bilingual programs that have succeeded. Research on effective schools demonstrates that schools can stimulate academic accomplishment for students in spite of how situational issues influence them. Considerations of verbal communication and culture assist English language development without giving up the native language and the skill to happen in a cross-cultural world.

Accomplishment and assessment of bilingual education programs call for to moving beyond supporting what have too frequently become compensatory agendas. All learners, but particularly bilinguals, merit quality programs that defeat harmful stereotypes. Plentiful results from empirical research and understanding can aid in showing the way (see Table 1).

TABLE 1. School Characteristics (Source: Snow & Ferguson, 2007).

Promote Quality Bilingual Education

Limit Quality of Bilingual Education Programs

1. Administration in cooperation with faculty and community develop clear goals.

1. There are no clear goals, leaving interpretation open to ideological tendencies.

2. School creates a bilingual and bicultural society.

2. Personnel and students outside the bilingual program have poor attitudes toward program's languages and cultures. There is no explicit instruction on American culture.

3. Bilingual program is integrated with the whole school.

3. Bilingual program is segregated from the rest of the school.

4. Bilingual students are well-known by all staff.

4. Only bilingual staff knows bilingual students.

5. Administration provides leadership and supports the program.

5. Administration is unsupportive, ambivalent, or indifferent.

6. School staff sets high expectations and supports bilingual students.

6. Staff believes ability to function is related to English language proficiency or ethnic back-

ground.

7. School hires quality personnel willing to work with bilingual students.

7. Students are taught by personnel with limited understanding or skills.

8. School enjoys a productive relationship with parents and communities.

8. Parents are perceived as indifferent or uninterested. No effort is made to accommodate parents' language and culture.

9. School curricula make use of the languages and cultures of the students to promote learning.

9. Use of students' languages and cultures is limited and only within the bilingual program.

10. Bilingual students participate in a comprehensive curriculum that benefits from current educational innovations.

10. Bilingual program curriculum does not cover all the same areas as the general school curriculum.

Bilingual students do not benefit from special programs brought in for the whole school.

11. Materials are of high quality, varied, and in the language of the students as well as English.

11. Materials in the language of the students are limited and of poor quality.

12. Assessment is fair and authentic and has as a purpose improved teaching and learning.

12. Assessment is mostly standardized tests in English. Its main purpose is to enter and exit students into and from the bilingual program.

13. Instructional practices are consistent with goals of promoting language development, adjustment to both cultures, and academic achievement.

13. Instructional practices are inconsistent, depending on teachers' beliefs and knowledge.

Many bilingual programs survive because school districts must fulfil legislation and court orders. They endure in segregation within uncooperative schools where the approach toward the program is unenthusiastic and the potential of students are short. Students rebuff their individuality in schools that do not recognize their culture, but cannot take on a new one (Commins, 2009). Such students repeatedly become angry and troublesome (Brisk, 1991; McCollum, 1993). "One thinks what the achievements of such students would be if their energies were open encouraged by an environment in which they no longer wanted to deal traditions for school learning" (Heath, 2003). Well-worn and unappreciated teacher's burn out, giving way to turnover and unsteadiness in programs.

Schools with no clear goals depend on the individual educator for the superiority of the program and are more susceptible to ideological weight. Without unambiguous goals for bilingual education, disorder and dissatisfaction among staff and community are probable results. A diverse study of a group of people in California and another in the Midwest show the significance of lucid goals (Hakuta, 2009). Absence of leadership and insertion of the program leads to differences in opinion with admiration to the reason of bilingual education. When English-speaking and bilingual staff do not believe commonly in goals, a cavernous gap in communication grows among the teachers, affecting instructors, students, and communication (Watahomigie & McCarty, 2004).

Although many teachers are well qualified, increasing burden on personnel have lead to the appointment of badly qualified teachers or the salvaging of mainstream teachers with no training to educate bilingual students. Districts experience personnel satisfaction and stability when they develop bilingual programs gradually and maintain strict control of the quality of personnel. Districts that develop quickly and haphazardly without careful teacher recruitment experience excessive staff turnover, often losing their best teachers who burn out when they find themselves supporting the less prepared teachers (Hakuta, 2009). Since the program is frequently seen as corrective, curriculum is thin, materials are lacking, and appraisal is incomplete to English language growth.

Such bilingual education programs should not be supported. The bilingual education characterized in this paper should be supported not merely because it is good for bilingual students, but also because its implementation can benefit schools as a whole.

Raising Kids Bilingually

How advisable is it to raise children bilingually? What consequences are there to bilingual upbringing? Saer (1923), reported that his son Louis showed only positive consequences from having been raised in a bilingual, French-German home environment. According to his father, Louis learned to speak both languages as a native-speaking child would -- he showed very few signs of interference between languages; nor did his bilingualism have a deleterious effect on his cognitive development. His development seems to have been quite normal and it has been reported that by the age of 15 he had equal fluency in both languages (Weinreich, 1953), preferring French for technology and German for literature. There was no evidence that the extreme nationalism and anti-German feeling in France at the time of the First World War had any effect on his bilingualism.

Many other researchers after Ronjat have come to the same conclusion -- that early bilingualism has positive consequences for linguistic and cognitive development. A number of investigators (Volterra & Taeschner, 1978) have suggested that early bilingualism can accelerate the separation of sound and meaning and can focus the child's attention on certain aspects of language. Snow & Ferguson (2007) has argued that being able to express the same thought in different languages will allow the child to "observe his language as one particular system in the middle of several, to examine its phenomena under more universal groups, and this fronts to consciousness of his linguistic process" (p. 110).

These positive consequences are not the inevitable results of childhood bilingualism, however. Macnamara, (2009) describing the experience of immigrant Finnish children in Sweden, reported that many of these children knew neither Finnish nor Swedish well. She observed that their Finnish pronunciation was often heavily influenced by Swedish, so that they were unable to differentiate between long and short phonemes, a crucial distinction in Finnish. On the other hand, their Swedish was also limited, especially in vocabulary. Skutnabb-Kangas described these children as "semilingual," in the sense of not knowing either of their languages properly, at the same level as monolingual speakers of the same age.

She characterized one such child, a 5-year-old Finnish boy living in Sweden, as follows:

He couldn't count to more than three in any language, after that he said: many. He didn't know the names of any colours in any language. He didn't know the names of most of the things around him, either at the day-care centre or outside... In any language. In Finnish he used only the present tense, in Swedish present and past (p. 226).

Saer, (1923) argued that this child is no rare exception and that many immigrant children are like him in that they do not know either of their languages at a level appropriate for their age. There has been a great deal of controversy about the notion of semilingualism, the principal issue being whether the bilingual's range of abilities across two languages is equivalent to that of a monolingual. It may be that the performance of some bilingual children in either language will lag behind that of monolingual speakers of that language; nonetheless, the bilingual child may have a full vocabulary and overall linguistic repertory that is somewhat comparable to that of monolingual orators (Saer, 1923). In any event, this is not to say that the phenomenon that Saer, (1923) and others describe is not real. The notion of semilingualism is a useful way of characterizing those cases where, through social deprivation, children do not discover to function well in their second language, and in unison fall short to develop their first language.

What are the differences between the experience of Ronjat's child and the immigrant children that Weinreich (1953) described? Why is it that for some children a bilingual experience is positive or "additive" (Macnamara, 2009), whereas for other children it is a negative or "subtractive" experience? What conclusions can be drawn from research about the consequences of a bilingual experience on children? There have been two major areas of research concern: (1) the effects of early bilingualism on linguistic development; and (2) the effects of early bilingualism on cognitive and intellectual development. In both of these areas there are important methodological issues, as well as theoretical implications that follow from what the research tells us. After summarizing the research findings I discuss, in some detail, methodological and theoretical issues, because research can only provide answers to the practical concerns of parents and educators if it is well-grounded methodologically and theoretically.

Linguistic Consequences

What does the literature tell us about the effects of early bilingualism on the language development of the child? Does learning two languages simultaneously influence the development of either one? For example, if an American parent speaks both English and Spanish to her child, how will the child's English be affected? Is it better for language development to speak only English than to use both languages? Early research on this question dealt with the so-called "bilingual handicap." Recent research has addressed more specific issues, especially with regard to lexical and syntactic development.

Early Research

We saw earlier that it appears to be difficult for an individual to maintain two languages in perfect balance. The environment is rarely perfectly bilingual, and usually the individual will need to use one language more than the other in daily life. Does this mean that one language inevitably suffers? What evidence is there that bilingualism has a detrimental effect on one or both of the speaker's languages? In a review of early research on this question, Weinreich (1953) reported that some authors concluded that the bilingual child encounters many problems in language development. According to these authors, the child's active and passive vocabulary is smaller because the child must learn two words for each referent. Even the total number of terms was seen to be less than the total number for the monolingual child. Bilingual children were found to use fewer different words and to develop a confused, mixed vocabulary because of lexical borrowings and the tendency to hyphenate words. They were seen to use shorter sentences, more incomplete sentences, fewer compound and complex sentences, fewer interrogative and more exclamatory sentences than monolingual children. According to these authors, confused structural patterns, unusual word order, and errors in agreement and dependency characterize the speech of bilingual children. They were found to make many errors in the use of verb and tense, connectives, prepositions, nouns, pronouns, and articles (especially indefinite articles).

To a large extent these conclusions were based on research by Weinreich (1953) who studied preschool children in non-American families in Hawaii. A major confounding factor that limits the degree to which this research can be generalized is that pidgin English was quite common in Hawaiian communities at that time, and the main reason why the children appeared to perform so poorly may have been that they used this variant of English as their standard. Other research, as Jensen noted, pointed to the conclusion that bilingualism is beneficial for the development of language skills. Some authors argued that bilingualism helps individuals become more sensitive to the nuances of language, aids them in their first language, enables them to manipulate languages more effectively, and helps them learn additional languages more easily. Saer, (1923) found that a group of bilingual, Puerto Rican preschool children in New York City actually excelled a comparable group of monolingual children in mean sentence length and in maturity of sentence structure in English. Snow & Ferguson (2007) reported that his observations had led him to believe that vocabulary is increased in the bilingual child. Totten, (1960) concluded that at the college level, bilingual students had no significant language handicap and even possessed some advantages.

The early literature varies so greatly in quality that almost all general statements are suspect. The usual procedure was to seek out two groups of children, one monolingual and the other bilingual, to test them in the language common to both, and to compare results. Very rarely was socioeconomic status or the children's intelligence controlled. Furthermore, the measures of language development used in the early studies were crude and yielded very rough estimates of linguistic ability. With recent advances in the field of language acquisition, researchers have been able to focus more specifically on various aspects of lexical and syntactic development.

Lexical Development

In contrast to the monolingual child, the bilingual child has to learn two words for a single meaning. Some authors argue that this involves no particular problems. For example, Weinreich (1953) contended that bilingual children face no additional difficulties in the acquisition of meaning, since they merely extend to a corresponding word in the second language the word meaning they have isolated and come to associate with a particular real world object in the first language. Her Georgian-Russian bilingual subject used the Georgian word ball to denote a toy, a radish, and stone spheres at the park entrance, and then transferred the same set of denotations to the Russian word equivalent. In this early stage, Macnamara, (2009) remarked, "differences in shades of meaning of corresponding words do not play an essential role" (p. 3).

Later, however, the bilingual child has to learn that the meanings of some words have different extensions in the two languages that are being learned. For example, the English word brush can be used for a clothes brush, a shoe brush, and a paint brush, but the German word Burste does not extend to paint brush -- instead Pinsel is used. In such cases the child must learn to utilize a somewhat different set of feature markings for corresponding lexical items in the two languages. One might expect to find mis-extensions of various kinds based on ill matched lexical areas across the child's two languages, and case studies of bilingual children indicate that such mis-extensions do occur. For example, Leopold, in his careful treatment of lexical extension (Tulving & Colotla, 1970), mentioned a likely instance of such cross-linguistic interference, where the German alle was used (at 1; 7-1; 11 ) to mean all gone with reference to persons, as in Mommy alle. Standard German does not allow this because alle does not mean the same as the English all gone when applied to persons.

Until bilingual children have determined the different extensions words have in their languages, they can have problems being understood. Thus, the 3-year-old Raivo Vihman argued with his English-speaking friends that he could read and demonstrated by counting -- the Estonian lugema is used for both "to read" and "to count" (Snow & Ferguson, 2007). Other instances of lexical mis-extensions have been reported by Tulving & Colotla, (1970). In the process of differentiating the lexical systems of the two languages, the child may make lexical selections on phonological

Syntactic Development

Is the acquisition of grammatical structures by bilingual children the same in its basic features and in its developmental sequence as for the monolingual child? There is evidence that structural features of the two languages of the bilingual child are not necessarily acquired simultaneously. As Grosjean, (1982) noted, the syntactic realization of semantic relations (such as locative or possessive) can occur at different times within the two languages of the bilingual child, reflecting the perceptual salience of the features needed to mark the relationship in the two languages. Thus, Serbo-Croatian-Hungarian bilingual children (Tulving & Colotla, 1970) demonstrated locative relations in Hungarian (where the locative marker is expressed by noun inflection) earlier than in Serbo-Croation (where noun inflection and preposition are needed to express the locative). Heath, (2003) pointed out; however, that the order in which various syntactic structures are acquired by bilingual children is the same as for monolingual children. The Serbo-Croatian locative construction is also acquired relatively late by monolingual speakers of that language. Mikes concluded that bilingual presentation has little effect on syntactic development.

This does not mean that there is, no period of confusion in the syntactic development of the bilingual child's two languages. A number of authors reported that the children they observed went through periods of syntactic mixing and confusion (Macnamara, 2009). For example, Tulving & Colotla (1970) reported that her child used Swedish morphemes with Estonian endings in the home, where Estonian was spoken; but with Swedish-speaking playmates the Swedish forms predominated. Grosjean, (1982) found morphological and syntactic mixing; although his son seemed aware that he was mixing the two languages.

Snow & Ferguson (2007) found instances "in which the grammatical structure is French but the lexicon is English, and a few in which the structure is English but the lexicon is French" (p. 20). French negative constructions, with the negative element following the verb appeared in lexically English sentences. Similarly, to express possession, Swain and Wesche's subject used both French structure with English words and English structure with French words, suggesting that at this point both systems had been internalized, but their linguistic allocation was not yet under control.

Volterra and Taeschner (1978) found that the two children they studied initially developed a single syntactic system that was applied to the lexicon of both languages. This syntactic system appeared to be different from that of either language. They argued that children begin by fashioning a unique system; then the system of the language with the more simple syntactic structures becomes dominant and mixing of syntactic structures from both languages occurs; finally, the two syntactic structures become differentiated. Thus, the development of syntax in bilingual children was thought to parallel quite closely the stages hypothesized for lexical development. Vihman (1983), however, found no evidence in her son Raivo's speech for the first of Volterra and Taeschner's stages. In her son's early word combinations there was a high proportion of mixings from the start, despite the presence of synonyms that might have been used to create single-language (in this case, Estonian) utterances. That is, Raivo's early word combinations were not based on a unique system with complementary terms from the two languages; instead, the child seems to have begun at stage two.

Thus it is unclear whether the three-stage process of a unitary system, mixing, and differentiation, is a universal process characterizing syntactic development in bilingual children. There does seem to be a differentiation process that occurs as the child sorts out the two languages, and this process appears to be slower for syntactic than for lexical development. Snow & Ferguson (2007) reported that lexical consistency precedes structural consistency: at 3;9 her subject's lexicon was differentiated but the child's grammatical system remained essentially undifferentiated. Tulving & Colotla (1970) argued that syntactic differentiation cannot occur prior to lexical differentiation because the child does not yet "use the two languages as separate instruments" (p. 186). Until this process of differentiation is complete there is mixing of syntactic structures, although this stage seems to be of relatively short duration in most of the reported studies.

Methodological Issues

There are a number of methodological problems with the research that has been discussed to this point. One problem is sampling bias. Most of the case studies were done with children from upper-middle or upper-class families. In almost all instances the parents were highly educated and intelligent. For children from such families there may be a brief "bilingual handicap," but they quickly catch up in linguistic ability with monolingual children of the same age. In fact, there are a number of reports in the literature of an unusual interest in language by bilingual children. Thus Vihman's daughter invented words with fanciful definitions at three years; at four years and eight months she compared Estonian, French, and English sounds. Totten (1960) reported that his daughter asked how to say certain words in other languages and seemed to be struck by the arbitrariness of language as early as age 3. At four she was able to comment on the grammaticality of her utterance and could answer questions about language.

To what extent does the concern about language shown by these children represent the proclivities of psycholinguists' children and to what extent do they reflect bilingual experience? There is some proof that the advantages of a bilingual experience can be less positive for children from lower socioeconomic and less well-educated families. For example, in a study of lower- and middle-class Mexican-American children aged three to ten Tulving & Colotla (1970) found that there were specific areas where the children studied were, as a group, significantly delayed when compared to a control group of English-speaking children. In particular, the comprehension of pronouns, negatives, and some tense markers caused difficulty for children in the bilingual group. Cornejo ( 1973) investigated the language development of 24-five-year-old Mexican-American children of lower-middle-class background and found a high degree of transfer, borrowing, and language mixture in the language samples. Transfer from Spanish to English was most prominent at the phonological level, whereas transfer from English to Spanish was most noticeable at the lexical level. In addition to these studies there is Snow & Ferguson, (2007) contention that for many lower-class immigrant children bilingual exposure can lead to a state of "semilingualism":

Many of the children do not know any language property, at the same level as monolingual children. The language tests and estimates show that they often lag up to four years behind their monolingual peers in language tests in both languages.

Regardless of what position one takes on the semilingualism issue, it is clear that for many immigrant children knowledge of a second language is a hard task that takes a long time and may interfere with the development of the child's first language. That bilingual exposure can have different consequences for different children suggests that there are important intervening variables that have to be considered in research in this area. One such variable relates to the manner in which the child is exposed to two languages, which I refer to here as the conditions of presentation.

Conditions of presentation

The child bilingual can be exposed to two languages in a number of different ways (Macnamara, 2009). Let us assume that the child experiences two basic distinguishable environments -- "home" and "community." If we make the further simplifying assumption that three types of language use can be distinguished -- one person-one language; varied use by every person; and environment-bound language, with one language at home and another in the community -- we have the following possibilities:

In the home:

a.

parents and other members of the household each present one or the other of two (or more) languages (one person-one language); b.

each member uses both languages;

c.

the home is essentially monolingual, with members of the outside community introducing a second language to the child.

In the child's community:

A.

different persons each use one of the child's languages, but not the other;

B.

members of the community tend to use both languages (a code switching community),

C.

a monolingual community in which most persons know only one of the two languages the child is acquiring.

Combining the various possible situations in home and community, we arrive at a matrix with nine possibilities. A few comments are in order on the range of actual usage represented by the cases cited in the various slots. One person-one language, for example, was the policy in both the home and the community in the case of Louis Ronjat. The family lived in France but had frequent contact with both French- and German-speaking families. Thus, the child's world outside the home was bilingual, with the one person-one language formula carrying over to the community as well. In the Leopold case, on the contrary, exposure to one of the two languages, German, was virtually non-existent outside the home, except when the family travelled to Europe.

Perhaps the most frequently reported situation is the one in which both the parents and the community mix languages to some extent (Heath, 2003). In many families and communities, bilingual speakers use a separate code that includes mixing structures and vocabulary from two languages. This phenomenon has been found to be especially common in the speech of Mexican-Americans (Snow & Ferguson, 2007). Tulving & Colotla, (1970) showed how the use of mixed expressions in the speech of Mexican-Americans is highly meaningful and serves definite communication needs. Speakers assemble on the coexistence of alternating forms in their language repertory to make meanings that can be very idiosyncratic and understood just by members of similar bilingual speech communities.

The problem for the child learner is that, the more mixing of languages occurs in adult speech, the more difficult it becomes to differentiate between the two languages. For example, Tulving & Colotla (1970), reported that his son's bilingual acquisition showed some mixing of Garo and English -- Garo words being given English morphology and syntax. This was not too surprising, because all of the English-speaking adults around the child used Garo words in their English speech. Similarly, the Mexican-American child may find it difficult to differentiate Spanish and English if the adult input is mixed. Thus, the conditions of presentation can have an effect on how easy it is for the child to keep the two languages apart.

In addition to differences in bilingual presentation in the home and in the community, there are likely to be changes in the child's linguistic environment over time. Once the child is sufficiently old enough to enter the community of new children or to attend nursery school, the child's linguistic input may change quite drastically. At the latest, entry into grammar school may effect the change -- in the United States at least -- marking the imminent disuse by children of a family language that may have been their sole means of expression for four or five years (Volterra & Taeschner, 1978).

Another possibility is that the family moves from one community to another. Totten (1960), gives a poignant account of his son's shift from Garo to monolingual English after the family left the Garo Hills when the child was 3;6. Or the parents may change their home language policy, with or without such a change in locale. Thus, Volterra & Taeschner (1978), reported that he and his wife used Brazilian Portuguese exclusively in addressing their child until she reached 1;2, at which point the family moved to the United States, where they switched to a one-person-one-language policy. By 2;8 the child had begun to respond in English even when spoken to by monolingual Brazilian children.

In short, there may be any number of complexities and vicissitudes affecting the child's acquisition of two languages. Conditions of language presentation have important consequences for language differentiation, and must be taken into account in making generalizations about the effect of bilingual experience on the linguistic development of the child. Attaining bilingual competence is likely to be more difficult the more mixing the child is exposed to, although mixed exposure does not lead to permanent retardation in either language. In fact, researchers sometimes mistake mixing in the child's speech for confusion and language interference, when the child is actually using (or trying to use) mixed utterances rhetorically for sociolinguistic purposes, just as adult speakers in the child's linguistic environment do (Heath, 2003). More serious consequences for language differentiation follow from imbalanced presentation.

Effects on Intelligence

Does being bilingual have any effect on a child's intellectual development? A number of contemporary authorities have argued that it has a positive effect, enhancing the child's intelligence. This contrasts with the findings of older studies that concluded that bilingualism has a permanent negative effect on intellectual development. Some early authors were quite convinced that bilingualism has a permanent negative effect on intellectual development. Totten (1960), saw bilingualism as capable of impairing the intelligence of a whole ethnic group and crippling its creative ability for generations. This was a position shared by many German authors at the time, which saw it to be their patriotic duty to find evidence for the negative effects of bilingualism, especially of Germanspeaking people (Volterra & Taeschner, 1978).

The situation was somewhat the same in America where bilingualism was also negatively valued in the 1920s and 1930s. Here investigators were more "objective" than their German counterparts, whose evidence was usually based on personal intuition. American investigators typically employed "scientific" pencil-and-paper tests designed to measure differences in intellectual functioning between groups of monolingual and bilingual subjects. In many such studies the conclusion was drawn that bilingualism had a negative effect on intellectual development (Snow & Ferguson, 2007). According to these and other authors, bilingual children often must think in one language and speak in another with the result that they become mentally uncertain and confused. Bilingualism was thought to be a mental burden for children, causing them to suffer mental fatigue. Bilingual children were seen to be handicapped on intelligence tests, especially those demanding language facility.