(A shorter version of this paper has been published as
``Losing the Battle for Balanced Bilingualism: The German-American Case."
in J. Pool (ed.), Linguistic Inequality. Special Issue of Language Problems and Language Planning
(Vol 11, No. 1, Spring, 1987). pp. 66-81.)
Some of these questions have not been raised until quite recently, while others have never been explicitly dealt with in the literature on bilingualism or bilingual education. I will draw my conclusions from the evidence available, using the schools of the German-American churches as an illustrative case, while also comparing them to other ethnically-affiliated religious-supported language maintenance schools in North America. In the discussion of linguistic assimilation in the German-American church, I will also challenge the notion that assimilation proceeded at a similar rate throughout all German-American denominations and sects, since it is important to compare differential language policies of different sects and to see how these varying policies had dissimilar results.
Linguistic assimilation in the German-American church has been studied by Hofman (1966:139-50, 1972) and tangentially by Schneider (1939), Beck (1939), Dietz (1949), Stellhorn (1963), Graebner (1965), and Kloss (1966). Except for Hofman and Kloss, most of these studies have accepted the notion that linguistic assimilation was a given, and that the process only needed to be documented. Most also have studied the ``official'' change in language policy (from German to English) in the German-American churches by examining the statistics on number of churches offering English-language services, confirmation classes in English, and number of church publications sold in English vs. German. Hofman concludes, for instance, that
``the most crucial variable in the relatively greater persistence of German [as opposed to Scandinavian or other ethnic languages] may well be the greater numerical concentration of Germans. ... [C]onservatism is an unnecessary assumption in order to account for the somewhat greater retentiveness of German''. (1972:623).
Kloss (1966) goes beyond this to offer a useful taxonomy of factors affecting language maintenance, but admits the contradictory effect of some of them (in some cases a given factor will aid maintenance, while in others the same factor will hasten assimilation).
While these studies are important for our understanding of the linguistic assimilation of the largest1 body of non-English-speaking immigrants that the U.S. has ever had to deal with, none of them discuss any differential rates of assimilation within the rubric of the so-called ``German-American church''. None address the reasons for the failure of the large school system(s) operated by many of these denominations (including the Missouri Lutherans, the Evangelical (and Reformed), the Roman Catholics, and other bodies, to say nothing of schools operated by `non-church' Germans such as Freidenker and others), to produce bilingual individuals. Balanced bilingualism has not been the natural outcome of these schools, such that stable and balanced bilingualism never became an accepted part of German-Americans' linguistic make-up, and thus a natural and accepted part of the American linguistic personality. That some bilingual individuals resulted from some of these schools is an accepted fact, but that the members of the next generation were monolinguals or English-dominant bilinguals also seems to be accepted as natural. No one has asked why the ``bilingual'' schools operated by German-American and other denominations during the 19th century did not produce stable community bilingualism among these groups. The literature seems to assume that the official language policy of these German-American churches is representative of the actual de facto language preference of their members, and then attributes the quick change-over to English after World War I to the harsh regulations of that era; or, these groups are acknowledged to be already English-preferential, but their English dominance is attributed to the ``enormous assimilative power of American civilization'' (Glazer 1966:360). The overwhelming evidence from internal documents of these churches, and particularly their schools, however, indicates that the German-American school was a bilingual one much (perhaps a whole generation or more) earlier than 1917, and that the majority of the pupils may have been English-dominant bilinguals from the early 1880's on.
Since at least some immigrant societies in the world are not monolingual, and linguistic minorities in some of them have remained distinct (for example Canada and South Africa) it seems incumbent on us to ask, not why immigration in some societies results in bilingualism, but why immigration in American society did not result in widespread bilingualism. Why did bilingual schools, whose goal was to maintain German and other languages (with concessions to English for ``practical'' subjects such as arithmetic and geography), produce in effect assimilationist English-dominant bilinguals who melted into the American mainstream? While the ``enormous assimilative power of American civilization'' may have been in the end an irresistible force that, combined with factors such as those enumerated by Kloss (1966:206-12), simply overwhelmed any language maintenance efforts to the contrary, I think we should also look at the schools themselves. The sociological arguments are strong arguments, but they are lacking in an understanding of the nature of bilingualism that parallels a similar failure correctly to evaluate this phenomenon both by scholars and policy makers within the German-American church and by those studying it in retrospect. Naturally much of what we know now about successful methods of teaching bilingual children was not known a century ago, so it is perhaps easy to make these judgements with the benefit of hindsight. Yet it is useful and instructive to do so, since many bilingual programs even today do not apply the best knowledge we have available, and may be doomed to failure even with the best of intentions. Kloss in his taxonomy of linguistic assimilation factors (1966) notes the paradoxical nature of factors that in some areas or with some groups have favored assimilation and in other cases have favored retention of the immigrant language. What has not been examined is the language policy and particularly the curricula of 19th-century bilingual schools in the light of what we now know about bilingualism. That is, if we know what kinds of schools (if any) produce truly balanced bilingual individuals who are capable of using both their languages in all situations in their lives, and have a positive self-image about themselves as bilinguals, even in the face of negative feedback from the society at large, we could compare these schools with those historically operated by various American ethnic groups. We could then perhaps determine whether those schools played an effective role in language maintenance among the group members they served.
The work by Lambert and his colleagues at McGill University (Peal and Lambert 1962, Lambert and Tucker 1972, Lambert 1977) indicates that there exist effective methods of producing such bilinguals, but these methods were of course not known or not applied in the German-American schools during their heyday in the 19th century, so that the linguistic policy of these schools may have been not only ineffective, it may have in fact been unwittingly contributory to eventual linguistic assimilation in America. This is also a strong claim and needs careful justification but, if it can be substantiated, it could have far-reaching implications, both for our understanding of linguistic assimilation in the history of the United States, and for future policy planning in the area of bilingual education.
To begin with, we need to examine two areas: definitions and examples of bilingualism and in particular the methods known to be most effective in producing bilingual individuals, and the language policies of the German-American schools, both in fact and in fiction. By `fiction' I refer to the official policies of the various churches, which I contend were often not in line with the actual day-to-day policies in schools and churches. The later de facto policies are often not explicitly reflected in the official historical records, although it is possible to discover them in other historical materials, or to extrapolate them from other evidence. In particular I want to examine the policies of two important German-American denominations, the so-called Missouri Lutheran Church (Evangelical Lutheran Synod of Missouri, Ohio and Other States) and the Evangelical and Reformed Church (especially one of its forerunners, the Evangelical Synod of North America). These two denominations differed in important ways in their theological bases, and it is my contention that these differences in theology not only had a differential effect on language policy of the two churches, but contributed to differential rates of assimilation that has been largely ignored by the attention paid only to the official change in policy of the two around the time of World War I.
Before we examine the language policies of the two churches and how these were reflected in their schools, we should first discuss three uses of the term ``bilingual'': ``bilingual'' schools, ``bilingual'' states/communities/countries, and ``bilingual'' individuals. To begin with bilingual schools, there are several typologies available that can be used to characterize schools that use more than one language for instructional purposes. Fishman (1974:40-46, 118-23) gives two different typologies of bilingual education programs (which is not the same as bilingual schools ). One distinguishes between (1) transitional bilingualism, (2) monoliterate bilingualism (3) partial bilingualism and (4) full bilingualism. The first has as its goal the use of a (usually minority) language as a stepping stone to a monolingual program; the second has as its goal literacy in only one language, although another language is used as an emotional and community link; the third seeks full bilingualism but only in certain subject matters, ``most generally related to the ethnic group and its cultural heritage. In such a program, reading and writing skills in the mother tongue are commonly developed in relation to the social sciences, literature and the arts, but not in science and mathematics.'' (Fishman 1974:43-44). The fourth kind of program has as its goal developing all skills in both languages in all domains.
``Typically, both languages are used as media of instruction for all subjects (except in teaching the languages themselves). Clearly this program is directed at language maintenance and development of the minority language. From the viewpoint of much of the linguistically and psychologically oriented literature this is the ideal type of program, since ... it results in `balanced, coordinate bilinguals--children capable of thinking and feeling in either of two languages independently.'' (Fishman 1974:45).
Fishman's other (``sociological'') taxonomy of bilingual education classifies programs according to variables such as the language of primary emphasis (LPE), language of secondary emphasis (LSE) and language of the major institution. Languages are classified according to whether they are the student's own or other language, and whether the language is a ``major'' or ``minor'' language. This taxonomy, based on actual cases, results in sixteen different types of schools and school systems; it assumes, however, that there is always inequality between the two languages used, i.e., one is always the LPE and another is always the LSE. Mackey's typology of bilingual education (1970) goes into exhaustive detail, allowing for different types of learner (unilingual to bilingual), differences in medium (single, dual), differences over time (transfer pattern vs. maintenance pattern), direction (toward acculturation or irredentism), and distribution of subjects (gradual vs. complete change), not to mention differences between languages of home, school, wider culture, and nation. When all these differences are taken into account, a typology of at least 90 different kinds of bilingual schooling emerges:
``In fact, the entire typology may be viewed as a series of patterns of distribution of two or more languages in the education of the learner, within the home, the school, the area, and the nation.'' (Mackey 1970:603)
In actuality, there are even some other kinds of programs that have been called ``bilingual'' that do not fit under any of these typologies, such as
Bilingual states, countries, and communities can differ greatly as well. There are a number of useful taxonomies of multilingual countries that may be referred to (Mackey 1972, Kloss 1966, Shapiro and Schiffman 1981) for an idea of the many different conditions that may obtain in so-called bilingual or multilingual countries. Suffice it to say that some bilingual states, like Belgium, contain very few bilingual individuals, while a country such as Paraguay (Rubin 1967) has a population purported to be as high as 65 to 90% bilingual in Guaraní and Spanish (Fishman 1974:89) but with very little official2 recognition of Guaraní.
India, the [former] Soviet Union, the Republic of South Africa, and the United States are all countries characterized by significant populations of bilingual individuals, but with different societal conditions surrounding their status. There is no space here to enumerate all the kinds of multilingual nations and communities in the world and in fact no way to be sure just what degree of multilingualism exists in them given the different ways language and dialect are defined there (see for example Kloss and McConnell 1975). As with definitions of bilingual schools, we are faced with many different kinds of situations that have been or might be called ``bi- or multilingual'' but that differ from each other in significant respects. It is not my goal here to clear up the definitional problem created by the many uses and misuses of the term ``bilingual'' but merely to indicate which of those many uses I am referring to in describing the ``bilingual'' schools of the German-American churches and the kinds of ``bilingualism'' that was the product of these schools.
Bilingualism in individuals is perhaps the most crucial to define clearly, and in fact the most difficult to do. Most scholars begin with the distinction made by Weinreich (1952, repr. 1964:9) between `coordinate' and `compound' bilingualism, although we shall see that there are problems with this distinction. According to this formulation, `compound' bilinguals have one semantic system (one Weltanschauung or world view) but more than one code that this semantic system can be linguistically realized as. `Coordinate' bilinguals, on the other hand, are thought to have two semantic systems, with a code associated with each. Typically, coordinate bilinguals have learned these languages separately, and do not make rapid mental associations between the two systems--often they have trouble translating between the two codes. Some researchers have posited that compound bilingualism is often associated with linguistic codes that are relatively close, such as two cognate languages like Italian and Spanish, while coordinate bilingualism would be more likely with unrelated languages like Japanese and English. Compound bilinguals are often thought to have one `dominant' language and one subordinate language, while coordinate bilingualism is thought of as `true' or `pure' bilingualism or bilingualism `par excellence' (Weinreich 1966, Haugen 1956). Since the distinction between the two types is theoretical, in actuality speakers range themselves along a continuum, if not outside the system altogether.
One factor that has been ignored until recently is the importance of biculturality as a factor in bilingualism; a bilingual who has learned both languages within the dominant culture of only one of them is a different kind of bilingual (if a bilingual at all) from one who has learned both languages in different cultural milieux, perhaps even in different geographical surroundings.
Defining bilingualism (and biculturalism) implies that there are ways of assessing bilingual proficiency, and various researchers have addressed themselves to the problem of devising accurate and comprehensive methods of such assessment. Out of these attempts has come the realization that different kinds of bilingual situations, including bilingual school programs, tend to produce different kinds of bilingual individuals. One of the most typical characteristics of bilingual individuals who have not had equal exposure to two languages in all social situations is that of ``compartmentalization'', where vocabulary for different activities or subjects is kept separate, and the individual finds that s/he can function effectively only in the language that is appropriate to the context. The ``home/ school'' bilingual is typically compartmentalized, using one language at home and another at school, or using one language for certain school subjects and another for others. Often this compartmentalization is reinforced by the society, so that it is not only inconvenient or difficult to discuss the same subject with both languages, it is socially inappropriate. Such situations have been referred to as ``diglossic'' (following Ferguson 1959) and are typical in colonial and post-colonial countries where a colonial language is used for many subjects for which the native language does not have the vocabulary. We have also not mentioned the problem of ``interference'' where one language exerts pressure on the other such that deviations from the norm of the second language occur (Weinreich 1953); this can happen in the lexicon, grammar, syntax, or semantics of either language. When there is widespread community bilingualism existing for centuries or even millenia, as in India, interference can be massive and can cause widespread change in the grammars of the languages in contact (Shapiro and Schiffman 1981). In such a situation there can be little `bilingualism par excellence' but only compound bilingualism, because the entire society shares one macro-semantic system, no matter how many different codes there are. In such situations difference in codes seems to be used to maintain group boundaries and preserve the illusion of differences in subculture; but since the semantic systems of the various languages have already long since converged, speakers are not obliged to make an effort to keep their semantic systems separate. Such an effort would consume a great deal of energy that could more profitably be used elsewhere. As Fishman has pointed out,
``a fully balanced bilingual speech community seems to be a theoretical impossibility because balanced competence implies languages that are functionally equivalent and no society can be motivated to maintain two languages if they are really functionally redundant.'' (Fishman 1974:45-6).
Here it is obviously necessary to distinguish between language and code since many citizens of India are obviously able to control two codes, and this control of more than one code is widespread and of long-standing duration in the subcontinent. Many segments of Indian society obviously are motivated to maintain more than one code because these codes have social significance that is far more important to the society than the functional redundancy that Fishman posits (Shapiro and Schiffman 1981).
From the above it should be clear that there are many definitions of bilingualism and of bilingual competence, and that not all of them are congruent or exhaustive of the possibilities. Different kinds of bilingual situations produce different kinds of bilingualism, and many social factors seem to be involved in fairly complicated ways. What is important for my argument here is to ascertain whether there are any bilingual situations, including bilingual schools, that are successful at turning out balanced bilingual individuals who avoid interference and compartmentalization, are truly bicultural and are able to use both their languages on an equal footing. Among the various types of bilingual schools that have been discussed above only one, Fishman's type (4), seems to have the production of this kind of bilingual as a goal.
``Type IV. Full Bilingualism.'' In this kind of program, students are to develop all skills in both languages in all domains. Typically both languages are used as media of instruction for all subjects (except in teaching the languages themselves.) Clearly this program is directed at language maintenance and development of the minority language. From the viewpoint of much of the linguistically and psychologically oriented literature, this is the ideal type of program, as illustrated by these comments:
Since one of our purposes is as nearly as possible to form and educate balanced, coordinate bilinguals--children capable of thinking and feeling in either of two languages independently--instruction should, we believe, be given in both languages ... (Michel 1967).
An education, both in and out of school, which respects these basic principles [to gain ``progressive control of both languages'' and ``a sympathetic understanding of both cultures''] should hopefully produce after us a generation of bilinguals who really are fully bilingual as well as bicultural (Andersson, 1967).
Programs such as these enable us to examine the difference between developing balanced competence in individuals and producing a balanced bilingual society Though bilingual societies might find individuals with highly developed competence in all skills and domains very useful in a variety of interlocutor roles (teachers, translators, business representatives), a fully balanced bilingual speech community seems to be a theoretical impossibility because balanced competence implies languages that are functionally equivalent and no society can be motivated to maintain two languages if they are really functionally redundant. Thus, this type of program does not seem to have a clearly articulated goal with respect to societal reality (Fishman 1970:88-9).
Because such programs are expensive and difficult to set up and maintain (and for other reasons that Fishman claims to be unrealistic, but which need to be examined further), there are few of them in existence. Where they do exist, however, researchers report that fully-balanced bilingualism is in fact the result (Michel 1967, Gaarder 1967, Barik and Swain 1974), or at least, no measurable difference between performance in the two languages is discernible, especially if all other factors are kept equal. Since this means that the two languages must be equally valued in the society, the pupils must be of equal socio-economic rank irrespective of mother tongue, the teachers of the two languages must be equally well-trained, the materials used must be of equal quality, etc., in fact it is not possible to keep all social factors equal, so naturally the more social variables there are, the more likelihood there is that measurable differences in performance on various kinds of tests of intelligence, reading, bilingual competence, etc. will be discernible. Certainly the school that attempts such programs (called ``dual immersion'' or ``Dual Medium Equal Maintenance'' programs) is not in a position to control the social factors that exist outside the school, although some kinds of control of the choice of pupils (in certain pilot programs) or choice of socio-economic background of certain neighborhoods can be made by researchers. The point is that such DMEM programs, while they represent the ideal, can probably never be realized in actuality. Fishman claims (above) that societies will never maintain (and by this he must mean `maintain school systems that maintain') two languages if they are really functionally redundant. That may be true, but since in the Indian subcontinent separate codes are regularly maintained, apparently because they are socially or functionally different, Fishman's dictum must be qualified somewhat. Bi- and tri-lingual communities are regularly reported in South Asia (Shapiro and Schiffman 1981) and this seems to have been part of the linguistic scene for millenia; the maintenance of these separate codes must have some purpose in the society, or this would not be the case.
In any event, if DMEM programs are in fact so rare as to be almost impossible to find in their ideal state, we must ask whether any other kinds of school setup might be expected to produce balanced bilinguals, and conversely, what kinds of bilinguals are produced by other kinds of schools. Remember that we are looking for schools that do not produce compartmentalization and do not result in replacive bilingualism, which by definition cannot be a kind of language maintenance. If language maintenance is the goal of a bilingual system, and balanced bilingualism is the desired product what other kinds of schools, if any, lead to this goal and these products?
Lambert (1977) reports on immersion schools (where English-Canadian children are taught exclusively in French) that work well for English children being taught in English. Here English is defined as the ``majority'' language and French as the ``minority'' even though in Quebec English speakers are in a minority and French speakers in the majority, because for Canada (and North America) as a whole, French is the minority language, and the English the dominant majority language. According to Lambert and his associates, majority language children (here, Anglophones) do well in immersion schools because their home language receives massive support from the rest of the environment; minority language children (here, Francophones) require more support for their language than they get from the society at large, because they also see only the majority language (English) reinforced by the environment at large. Majority language children immersed in a minority language program do as well in all subjects including their mother tongue as, or better than monolingual children of either the majority of a minority schooled in their own language. Minority children (e.g. Francophone children in Canada or Hispanophone children in the U.S.) seem to require a kind of program not given in Fishman's typology, one where instruction in the mother tongue predominates, and the amount of exposure to the other, majority language is more carefully controlled. That is, for a Francophone child in Canada to achieve bilingualism in French and English, very little school exposure to English is necessary, and much more instruction in French is required to maintain French, given the onslaught of English in the North American environment. For an Anglophone child in Canada to achieve bilingualism, however, the child can actually receive no instruction through English except for English as a subject, and still perform as well as or better than monolingual Anglophone children by the end of the sixth grade. Much of the dynamics of this situation has to do with developing cultural pride in the minority-language child, such that the child perceives that the minority language is as important as the majority language and is as adequate a vehicle as the majority language.
Aside from the DMEM school and the immersion or partial immersion school, no other type of bilingual program seems to lead to balanced bilingualism; in fact, most do not explicitly have it as a goal.
In what follows, I will examine language policy in the German-American church-supported schools of the Lutheran Church-Missouri Synod (LC-MS) and the Evangelical Synod (ES) to see how they compare in goals and actual outcome with other schools that have balanced bilingualism and language maintenance as both a goal and an end product.
Language policy in the LC-MS has been examined primarily with regard to the period of rapid change before and after World War I (Beck 1939, Dietz 1949, Hanley 1964, Gräbner 1965, Stellhorn 1963), since this is the period during which the ``official'' policy of the churches changed and the use of German in ``official'' contexts decreased while the use of English increased. Most of the comparisons have been based on statistics on the use of English in services, confirmation classes, Sunday school literature use, and the like. These are, of course, important indices of use, but if taken at face value, only serve to illustrate the change in official language policy, and furthermore these studies tend to view the change only as a function of the pressures of World War I and its aftermath. That is, even if we assume that German predominated in ``official'' use in the LC-MS until WWI, there is also evidence that English predominated in ``unofficial'' use beginning much earlier, and that the two languages existed in what is sometimes called a diglossic relationship (Ferguson 1959, Fishman 1967), or at least in a kind of complementary distribution as far as the domains of use of the two languages are concerned. In other words, if the growing unofficial use of English is ignored and attention is paid only to official German use, then the changeover to English after World War I seems abrupt and precipitous, and furthermore leads one to conclude that even such an edifice as the huge German church and school system could be brought to its knees by a few edicts of the State Councils of Defence. But if English can be shown to be in actual though unofficial use much earlier, the quick changeover to English in 1918 is not so astounding.
The evidence for the claim that English was making inroads in unofficial contexts is ample. The very fact that it had to be repeatedly banned from any official use in the LC-MS is proof of the challenge it was constantly making to German. For example, it was stated time and time again in various statutes 3 and constitutions of the Synod and its educational bodies that German and German alone was to be used as the language of reports, of instruction, of examination, etc.
Yet at the same time that the Fort Wayne Seminary was deeded to the LC-MS under condition that German alone be used as medium of instruction, the Seminary in Altenburg, Perry County, Missouri, also supposed to be German in medium, agreed to hire a Norwegian professor in order to help out the Norwegian Lutherans by teaching their Seminarians in Norwegian, and it was also agreed4 to establish an English-medium academy (i.e. secondary school) in Fort Wayne.
Similarly, at the same time that the Missouri Synod was refusing to accept into membership English-language congregations that were otherwise doctrinally acceptable (forcing them to form an independent ``English Synod'' that was not accepted into membership in the LC-MS until 1911 [Dietz 1949:99]), the Missouri Synod was also sending out missionaries to work among both English-speaking groups and speakers of other (immigrant) languages (Sommer 1922).
As early as 1865, Fölinger, writing in Lehre und Wehre (11, 8, pp. 236-42) admitted that the children of the church were becoming English speakers, and that this development could not be hindered, but should be taken cognizance of so that the church would be prepared for this eventuality instead of fighting it; that pastors should learn English in order to serve the descendents (Nachkömmlinge) and that the LC-MS should establish a periodical in English alongside translations of Lutheran writings, both dogmatic/catechetical and devotional, including hymns. Föhlinger's advice went unheeded for many years.
In fact the use of English among Missouri Lutheran Germans is amply documented in what must be now seen as a ``revisionist'' history, Stellhorn's 1963 volume entitled Schools of the Lutheran Church-Missouri Synod. In it he devotes all of Chapter 9 (``The Fathers and the English Language'', pp. 100-111) to showing that the LC-MS was not anti-English at all, but used English in its schools and seminaries to an increasing degree as time went on. He quotes Stach (l942) as saying that `In the Missouri Synod Schools ``the teaching of the English language was in many instances obligatory from the very beginning.''' and that for instance the Immanuel School in Grand Rapids, Michigan ``was reported teaching only English in the afternoon'' in 1867 (Stellhorn 1963:107). Various articles in the Evangelisch-Lutherisches Schulblatt, a journal for Lutheran school teachers published by the LC-MS are quoted by Stellhorn to indicate that the purpose of the schools was to maintain doctrinal purity, not to maintain language and that English was necessary if children were not to be ``estranged from the church'' or from the society in which they lived. From early beginnings when English was taught only as a subject, there was an increase in the use of English for other subjects as well. In an article published in the E.-L. Schulblatt in 1901 the author indicates that although 50 years before ( i.e. in 1850) English was banned from some schools, by the turn of the century ``in practically all the parochial schools not only reading and writing but also arithmetic and geography are being taught by means of the English language.'' (quoted in Meier and Mayer, 1964)
Meier and Mayer also quote an article by the important churchman Ludwig Fürbringer in Der Lutheraner of September 1906, who claimed that since the purpose of parochial schools is to train children in God's word,
``we should and will bear in mind particularly in those places where the English language prevails, ... that we make the proper preparations to change the parochial school, when the right time has come, from the German into the English.'' (p. 327, my translation, hfs).
I think it should be clear from the above, that in spite of an official policy of German only in the Lutheran Church, Missouri Synod (until 1921), English was in use in many areas of the church where not explicitly banned and where it was felt not only not to be a threat but to be necessary to meet the needs of many of the members. Since German and English existed in what has been termed and what is in fact an extension of the notion of a ``diglossic'' relationship (Fishman 1967), it is instructive to note that in classical diglossic situations, the speakers of the official (``formal, high'') language are content with the situation as long as the official language alone receives recognition and is used in formal situations. The other (``low, unofficial, spoken'') language in such situations is treated as if it doesn't exist; policy makers in such situations think that they are in control of language usage if they can control the ``high'' language, whereas teachers and others who have to deal on a day-to-day basis with the situation develop strategies that lead to fulfillment of their goals by whatever means are necessary, such as by paying lip service to the official policy but using the other language (in this case English) whenever and wherever necessary as long as the policy makers are unaware of it. Thus the LC-MS could be satisfied that officially only German existed; the fact that unofficial use of English was actually undermining the official policy was either not noticed or blindly ignored.
The language policy of the Evangelical Synod5 shows a parallel development to that of the LC-MS, but at a faster pace. It, too, began with an exclusive use of German although in the early records there was no mention of either German or English since the latter language was furthest from the minds of the founding fathers. However, many of the early pastors found a knowledge of English useful. The Rev. Henry Tölke, pastor of a flock at Evansville, Indiana, wrote in l849 to the Home Missionary Society6 that he was trying to use English and was writing some of his sermons in that language.
If the early Evangelische Kirchenverein did not have an explicit language policy, its seminary at Marthasville, Missouri (later moved to St. Louis) had evolved a policy7 by 1849:
#3. The German language is and will be forever the language in which instruction in this seminary will be provided, with the exception of the condition mentioned in #11.
#11 It should be explicitly noted that instruction in English should be provided through the medium of English, as well as instruction in arithmetic and in geography.
The Evangelische Synode, like the LC-MS, was not to concern itself much with the language question (known in both churches as ``die Englische Frage'') for another 40 years, since German immigration was increasing every year, reaching its peak in 1882 at approximately 250,000. It was all that the Evangelische Synode could do to keep up with the demand for pastors being trained in its seminary in St. Louis. The children of these immigrants who were becoming more comfortable with English than German were not noticed during this period. The Evangelische Synode, of course, like the LC-MS, recognized the need to found parochial schools, and the number of schools run by affiliated congregations increased steadily. Records date from 1854, with 20 schools at first, increasing to 88 in 1872.8 I lack figures for succeeding years, but there were apparently enough schools and teachers being trained that a need for a school journal Pädagogische Zeitschrift, which was founded in 1893 and ran until 1898) was felt.
Almost simultaneous with the decline in German immigration from 1882 onward came the first demands for help in ministering to English-speaking members. The President of the Synod in his report to the Districts in 1888 took the bull by the horns. He admitted that there was a need to do something about English, but recommended that the Districts not admit English-speaking congregations without the matter being decided by the whole Synod, since the Synod (or some of its members) had a plan to take in English congregations until a total of 12 was reached, then to require these twelve congregations to form themselves into a separate ``English'' synod. An alternative, the formation of bilingual congregations, with services in English at one hour and German at another, did not seem to appeal to the Evangelische Synode, just as it did not appeal to the LC-MS. Nor did it seem possible to train pastors to be theologically bilingual in both English and German, although the need for a knowledge of some English was recognized. The debate over whether to Anglicize gradually or whether to hold out, meanwhile increasing efforts to found schools and provide an opportunity for children to learn German was to continue for several more decades, until World War I and anti-German feeling made the decision to switch to English inescapable. But the Evangelische Synode did not have a firm united policy--some districts were very pro-English (e.g. Indiana) while others were luke-warm or anti-English. Since the Evangelische Synode was organized more democratically than the LC-MS, demands that came up from the ranks were more clearly heeded, or were at least discussed, even if no decision was taken. Gradually, however, the camel was allowed into the tent, with translation of catechism books and other liturgical writings, the employment of an English-speaking professor of theology at the Seminary, and various other measures.
The language policy of the Evangelische Synode parochial schools during this period is also indicative of an increase in bilingualism, although again the ``compartmentalized'' bilingualism that we have already seen in the schools of the LC-MS. A copy of a proposed Stundenplan published in the PZ (1893:70) indicates (see Appendix 1) that about half the school day may have been devoted to German-medium instruction, and the other half to English medium. But the subjects taught in German are such things as Bible, penmanship, reading, catechism, language, singing, and spelling, while subjects taught in English are the practical subjects of arithmetic; reading, writing and spelling in English; and geography. German-medium subjects are taught in the morning hours (except for music) and English-medium subjects, except for arithmetic, are taught in the afternoon hours. The day is long--from 9 to 4--although the proposed schedule is only a suggestion, and some subjects might be given less time, or, in the case of one-room schools, older children might work with younger ones while the teacher works with an intermediate group. But the picture is clear--German is for the primarily religous or abstract subjects, and English for the more practical subjects. We have another type-3 bilingual school, where compartmentalized bilinguals are produced. Since at the same time that this school plan was proposed the Evangelische Synode was moving toward gradual introduction of English services, English confirmation classes, and training of pastors in English (albeit with the objection still of many of its churches and older pastors), it seems clear that, whatever the ``official'' policy of the Evangelische Synode was at this time, English was gaining a foothold earlier than in the LC-MS.
Theological conservatism in the LC-MS has already been mentioned as one reason for greater antipathy to English in that body than in the E. Church. Another interesting reason has to do with the background of the trainees in their seminaries. The E. church (and the LC-MS as well) had trouble finding enough trainees who could handle the German language adequately, since many of the Nachkömmlinge were becoming Anglicized too rapidly, so they arranged to recruit trainees from Germany. The LC-MS had also tried this route, but stopped it since the German-born trainees, recently arrived from Germany, had ideas that were antithetical to conservative (i.e. LC, MS) Lutheranism and tried to import their ``Prussian'' attitudes into American lutheranism.9 But the E. church was itself an offshoot of this very church merger that the LC-MS people had left Germany to avoid union with, so the Evangelische Synode had no theological disagreement with German-born recruits. The problem that they found with those born in Germany was that they lacked enough English to deal with any matters outside the church, and they lacked a sensitivity to the American way of life. Hiring an English-speaking professor of theology was thought to be a way to prepare these German-born recruits for better service to their congregations. In 1894 the Indiana District made special mention of this problem and this need in their report to the Director (Indiana District, 1894, Protokoll) and two years later (1896) a suitable person was found to undertake this training.
A comparison of the two churches' differing responses to the threat of English seems to reveal that, while the LC-MS objected to English on doctrinal grounds as well as on emotional grounds, the Evangelische Synode's response (although earlier doctrinal), 10 was in my opinion primarily sentimental. By `sentimental' I mean a response that we might characterize today as loyal to the language for no reasons other than that the language has a familial or emotional (perhaps even cultural or patriotic) value for its speakers, whereas in previous centuries (and indeed for some people still today) such a reason for maintaining a language could not be given much credence. I must therefore disagree with Schneider (1939:420) who claims no early doctrinal loyalty to German, because of the `non-confessional' 11 character of the Kirchenverein generation but finds a rise in doctrinal concerns later, with a corresponding rise in language loyalty. I find rather the opposite--in the District reports, especially those opposed to the introduction of English, the reasons given are not doctrinal ones, but for the most part historical or emotional, when indeed any reasons are given at all. The later, ``revisionist'' analysis of early LC-MS opposition to English tries to tone down the emotional and/or sentimental attachment to German, which often blatantly equated German with truth and righteousness, and emphasizes the doctrinal purity stance. It seems to be the case that latterday historians of this period, particularly church historians, do not want to dignify language loyalty by examining it in its own right, or else they do not understand it to be a phenomenon that can exist independent of other factors, but try instead to cosmeticize it by attaching it to a doctrinal issue of some sort. For sociolinguists and sociologists of language there need be no justification given for the existence of language loyalty, but for church historians seeking to place the policies of their founding fathers in an understandable context, language loyalty per se must be explained away.
Let us now consider the question of what kind of bilingual children the de facto use of English alongside German in schools of this period was producing. From the reports of English usage already documented, and especially from the distribution of subjects shown by the model Stundenplan (appendix 1), it is clear that these German-American schools are clearly of type (3) in Fishman's taxonomy (``Partial Bilingualism''), since certain subjects are taught in German and other subjects in English. In this case religion, bible, catechism, and German are taught in German, and arithmetic, geography and English are taught in English only. According to the E. Church model Stundenplan, music (Singen/singing) may have been taught in both languages. As already mentioned, this kind of school tends to produce `compartmentalized' bilinguals, people who can function in some subjects better in one language and in other subjects better in another language. The problem with this kind of setup for German-American children (or any North American children) is that there was and is great reinforcement for English from the society at large, especially in areas where children were surrounded by English speakers outside the home and school, while German received reinforcement only in the church and perhaps at home. (There is often an implication in the policy makers' discussions of the ``English Question'' that even in some homes the parents did not insist on German being spoken). From what we now know of the importance of societal reinforcement of the equal value of both languages lest children draw their own conclusions about one language being more valuable than another (Lambert and Tucker 1972), we could infer that pupils in these schools could easily conclude that English was a more useful and practical language than German. This is especially so since the subjects that were taught in English were the useful and practical subjects (arithmetic, geography, science), while German was used only for religious subjects (Bible, catechism). In the pragmatic, increasingly secular American milieu, English could easily be seen as the useful and practical language while German could be identified as an abstract language with only an increasingly restricted domain for its use.
I conclude that compartmentalized bilingual schools of type (3), whose goal is only partial bilingualism, will not only fail to produce balanced bilinguals (especially since that is not their goal), but will actually contribute to the production of (in this case) English-dominant bilinguals. Lacking the reinforcement of a contiguous homeland like the French and Spanish-language minorities in this continent (Québec and Latin America, respectively), German-Americans relied on the establishment and maintenance of a large number of ``German'' schools to perpetuate their mother tongue. In fact these schools constituted a transitional system to aid children through a period of compartmentalized bilingualism to eventual monolingualism (or functional monolingualism) in English, while perpetuating the belief that, because religious subjects were being taught in German (to maintain doctrinal purity or for whatever other reason), both the spiritual and the linguistic goals of the community were being met.
Before this case can be laid to rest, there is some additional neurolinguistic evidence on the subject of bilingualism that is pertinent to this case and ought to be examined here. In the now increasingly extensive literature on brain lateralization and its relationship to bilingualism, there seems to be some evidence 12 that bilinguals process information differently from monolinguals. It seems to be fairly established that the left hemisphere of the brain is dominant for verbal, analytic, logical, sequential, and mathematical kinds of mental processes, while the right brain is the hemisphere in which non-verbal, synthetic, intuitive, ``artistic'', musical, and ideational kinds of cognition takes place. This is at least true for monolingual right-handed people; left-handed and right-handed monolinguals with familial histories of sinistrality may have a different pattern of lateralization. But bilinguals, especially bilinguals who learned a second language later than the first, show a pattern of right-hemisphere involvement in second language learning, and even cases of the first language shifting to the right hemisphere as the second is learned (Albert and Obler 1978). It is impossible to determine in retrospect whether the German-American pupils of the schools we have been examining were English-dominant or German-dominant bilinguals at the time that they entered school 13 or first encountered the second language, be it German or English. But given the fact that the subjects that were taught in German were those that tend to involve the right brain (religion, Bible), while the subjects taught in English tend to involve the left brain (arithmetic, science, geography) it is possible to conclude that this is a causative factor in the development of compartmentalized bilingualism. Since the left brain is in most cases the dominant hemisphere, especially as education and school-favored cognitive skills increase, it is no accident that the language used for left-hemisphere cognitive functioning will tend to be or eventually become the dominant language of the child. In the case we are discussing, and in most bilingual schools of type (3) in North America, this language happens to be English. Since English is the dominant societal language as well, type (3) bilingual education simply has no chance of producing stable bilinguals in this context--they will all be English-dominant bilinguals.
Add to this the very strongly assimilative character of ethnically-affiliated all-day schools, and there is an even weaker case for type (3) bilingual programs to succeed in their language-maintenance goals. Fishman and Nahirny (1966) have shown that the ethnically-affiliated All-Day School (which is of course the type of school we are considering, rather than afternoon or Saturday schools) is today the most Americanized of all ethnic schools, offering the least amount of language or any other ethnic instruction. ``By every available index the All-Day School is far less embedded in ethnicity, and, therefore, far less concerned with language maintenance, than any other type of ethnically affiliated school.'' (Fishman and Nahirny, 1966:95) It is not hard to see how such schools became so de-ethnicized, since the policy of gradually introducing compartmentalized bilingualism is, as I hope to have shown, a covertly (or perhaps I should say unwittingly) assimilationist policy contributory to eventual monolingualism. That this would be the case was probably furthest from the minds of the policy makers of the German-American church, but it is my inevitable conclusion. One can easily see why certain groups (some German-American groups among them) insisted on monolingual ethnic-language schools (e.g. the clergy in Québec insisting on French only) as a bulwark against both anglicization and Anglo-American mainstream religious thought and practice. The implication is that, unless more attention is paid to what ``partial'' bilingual programs actually do, the battle for language maintenance may be lost in the very arena in which it is appearing to succeed. This is the tragic irony of the failure of bilingual German-English schools to preserve the German language, an irony that the church policy makers never comprehended, even if the tragedy of it was clear to them.
1 Kloss (1966) estimates German immigration to total 6 million by 1920. The official language of these church bodies has almost always been High German, irrespective of what dialect (often mutually unintelligible with High German) was spoken in the home, but very little information is available as to what these dialects actually were like. Therefore, to classify all ``Germans'' (who may have come from Germany, Switzerland, Austria, Belgium, the Ukraine, Alsace, or various other territories) as speakers of High German is erroneous and extremely misleading.
2 This is reported to have occurred in 1967 (Rhodes 1980:4)
3 For example, see Synodal-Bericht der deutschen Evangelischen-Lutherischen Synode, 1848, where the statutes of the Fort Wayne (Indiana) Seminary are stated: ``Daß als alleiniges Lehrmittel in dem Seminar die deutsche Sprache angenommen sei und unverbrüchlich bleibe;'' (p. 46, section 1,2).
4 Synodal-Bericht 1860, p. 3.
5 The Evangelical Synod, known at first as the Deutsch- Evangelische Kirchenverein des Westens, was organized in May of 1841 in rural Missouri.
6 Transcript of American Home Missionary Society Correspondence, Dealing with Work among the Germans in Indiana, p. 142 (translation).
7 Kirchen-protokoll I: Protokollbuch des Deutsch-Evangelischen Kirchenvereins des Westens, 1849:51, extraordinary meeting of 12th and 13th February, 1849. (My translation, hfs).
8 According to Lowell Zuck, letter of March 1977. He refers to Muecke's history of the E. Synod, dated 1915, a source that was unavailable to me.
9 The Lutherans had left Germany in the 1840's when the Prussian state had forced a union of all Protestants in the territories of Germany that were then part of the German Reich, into one united Evangelische Kirche. To then bring recruits from this `oppressive' Prussian union church and retrain them to be true Lutherans was more than the LC-MS could stomach.
10 See Kirchliche Mittheilungen, 1845-6, for anti-Methodist sentiments.
11 The Kirchenverein, or `church union' generation compromised on the issue of whether the Heidelberg confession or the Augsburg confession would be adopted, and allowed both, or either.
12 See, for example, the bibliography in Albert and Obler 1978.
13 Anecdotal evidence from various unrelated sources indicates that some children were monolingual in English when first enrolled in German schools similar to these, or had at the most only a passive knowledge of German.