This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
[k] [f] in | | [X]in[Xt] and [X]#
[z] Table 1: Spelling norms for fifteen phonemes under five different spelling systems
WIM VANDENBUSSCHE
32
3. Spelling reality 3.1 Lower class writers None of the preceding models was ever adopted consistently by any of the lower class writers in our corpus. It is clear that the strict normative views of the successive spelling designers were either unknown or ignored, but this does not necessarily mean that the spelling of the Flemish lower class writers was normless or unsystematic. I contest the traditionally held opinion that the relative uniformity of written Dutch in Holland contrasted with absolute spelling chaos in Flanders (Suffeleers 1979:19), and I do so with respect to the upper, middle and lower classes. A systematic analysis of the variants used for the cited phonemes shows that each of the lower class authors had developed their own spelling system (Vandenbussche 1999). The use of spelling variants for the same phoneme within the same text was a core element of these spelling systems. Variability should, in other words, neither be seen as chaos, nor as a deviation from existing official norms, but as an essential characteristic of the distinct spelling systems of all lower class writers. It was perfectly normal for lower class writers to write the same word in two different ways in the same text or sentence. We can thus find gemeensaemheijd ("association") next to gemeenzaamheyd, without any correction or attempt to make consistent spelling choices. A striking example of this spelling tolerance was found in the following text, which contains three successive identical formulae on the same page which were beyond any doubt written at the same sitting. Although the author could have copied the first example twice, the second and third versions contain divergent spelling forms; one also notes that certain words were suddenly capitalized or linked up with other words. Dem zelve dag wierd Geresloveerd En vast Gestel by deze Gemeenzaemheyd Den Zelve dag wierd Geresolveert Een vast Gestel by deze Gemeenzameheiid Den zelve dag wierd als dat zal gegeven worden aen Agt Druck Een wessel als dat zal gegeven worden aen Joseph Goorens, op wessel, de somme gegeven aen Eugenius annoij, op wessel desomme van 700 francs Capital tegen den penning sesen ten honderd van 700 frfrancs capital, tegen den penning sesen ten hondert van 700 f francs, Capital, tegen den penning sesen ten hondert,
DUTCH ORTHOGRAPHY IN 19 -CENTURY FLANDERS
33
voor den tyd van twee Jaeren begenende met den twee-en-twintigsten novembre 1814 voor den tyd van twee jaeren Beginnende met den twee en twintigsten Novembre 1814 voor den tyd van twee Jaeren beginnende met den twee en twintigsten Novembre 1814 En Vervallende met den 17 Novenbre 1816 memori Endevervallede met den 17 No ve 1816. Memori En de vervallede met dem 17 Novembre 1816 Memori [The same day it was decided and stipulated by this association/ that will be given to Agt Druck/Joseph Goorens/Eugenius annoij a bill of exchange worth the sum/ of 700 francs capital at an interest rate of six percent/ for the time of two years starting on the 22nd November 1814/ and ending on the 17th November 1816 Memori.] (In the third example the introductory phrase was shortened to [The same day was/ given to...]) (Meeting minutes tailor apprentices, November 21 st , 1814; our translation)
There were clear constraints on spelling variation, however. Analysis of the corpus made it possible to predict where variants would most probably occur (the fifteen cases cited above), but also to predict which allographs could be used. An overview of all possible allographs which were actually used in the whole corpus throughout the 19th century is presented in Table 2, which should be interpreted as a maximal system capturing the greater part of all tolerable spelling variation. All writers applied their own restrictions within these limits; the frequencies with which the variants were used also differed from writer to writer. It is not surprising to see that most variants also occurred in one of the different official spelling norms, which once more indicates that lower class spelling was not inspired by chaos but rather by a certain tradition. In this context one can refer to the remarks made by Milroy (1992:133-134) about the spelling of Early Middle English authors: "[I]f the scribes really had used variants 'at will', we would actually be unable to read the texts [...] there must always be some order in any spelling system that we can read, even if it is a variable system [...] it is our task to attempt to specify the constraints on spelling under which they were working, always admitting that after we have done this, there may well be residues of apparent randomness that we cannot explain."
34
WIM VANDENBUSSCHE
[a:] closed syllable [a:] open syllable [e:] open syllable [o:] open syllable [i:]
[u] [0] [ei]
<e> <ee> <eu> <ey> <ei> <eij>
[œy]
[cw] [p] in [t] in [t]#
[k]
[f] in [X] in [Xt] and [X]# [z]
<s>
Table 2: Allographs used in the corpus throughout the 19th century Many of these 'residues' can either be explained as attempts to use formal or foreign terms with which the lower class writers were probably not familiar — and which, accordingly, were reconstructed in the best possible way, rendering e.g. refuest instead of refus ("refusal") and gedisgecrutiers instead of gedisgratieerd ("disgraced") — or as dialectal interferences (krus for kruis ("cross") and alk for elk ("each") reflecting the Bruges dialectal pronunciations [krys] and [alk]) (Vandenbussche 1999). It should be clear, however, that the writers did not attempt to write dialect. On the formal level these regional interferences were very modest (compared to, for example, the (fully intended) dialect transcriptions in Lootens 1868 and the description of the Bruges dialect in De Wolf n.d.). As far as functional aspects are concerned, their written variety may be defined as an 'intended standard language', a term used by Mihm 1998 ("intendiertes Hochdeutsch") to refer to a variety which does not meet the formal requirements of a standard language (e.g. consistent spelling and grammatical correctness) but which is nevertheless intended by the writer
DUTCH ORTHOGRAPHY IN 19 -CENTURY FLANDERS
35
to fulfil the functions attributed to a standard variety (e.g. supraregional communication, prestige variety) (Vandenbussche forthcoming). How did this concept of 'consistent spelling variability' evolve throughout the 19th century? There is no evidence that writers changed their spelling habits when a new official norm was adopted, nor of an overall partial move towards the official norm around 1900. Some writers from the second half of the century even displayed more variation than their colleagues fifty years earlier. In the writings of certain authors we can, however, distinguish a growing personal standardization. This should not be understood as the gradual approach towards an existing official norm, but rather as the ongoing limitation of the personal tolerance of variability. In other words: the basic philosophy of consistency ('one character for one phoneme') became ever more apparent in certain lower class texts, too, but this need not result in a system consistent with the official spelling guideline. Variable spelling was combined throughout the 19th century with the socalled zusammengebrochener Stil or Stilzusammenbruch (Mattheier 1986). This concept (lit. "stylistic breakdown") from German sociolinguistics refers to authors' inability to control the stylistic and grammatical properties of the texttype used — be it letters or official meeting minutes — which results (among other things) in syntactical anomalies, the omission of constituents and, in some cases, the loss of text structure. The co-occurrence of variable spelling and Stilzusammenbruch is believed by certain German authors to be typical of Arbeitersprachen (Klenk 1997, Mattheier 1990), exclusive lower class varieties in the 19th century. At first sight, the simultaneous presence of these linguistic features in Flemish and also in English (Fairman 2000) lower class texts seems to confirm this universalist conception. Then again, the everyday life of a Flemish trade apprentice had so little in common with that of a German mineworker or an English pauper that one can justly question the existence of a uniform shared lower class identity from a sociological point of view; the shared characteristics in their written output should, in any case, not be attributed to the factor 'class' as such. 3.2 Middle class writers Since the middle class writers displayed the same amount and identical patterns of spelling variation and since no official guideline was ever fully adopted by any of them either (Vandenbussche 1999), there is no evidence at this stage to support the claim that middle class writers spelled better or more consistently than their lower class colleagues, either at the beginning or at the end of the 19th century. Once again, the influence of the spelling
36
WIM VANDENBUSSCHE
standardization movement seems not to have reached this part of the population. There is more to this claim, however, than the simple observation that lower and middle class spelling were virtually identical. Despite the strong correspondences between lower and middle class texts until 1850, a striking qualitative divergence was noted towards the end of the century on the level of text and sentence structure (Vandenbussche 1999). The Stilzusammenbrüche which were also present in the middle class meeting minutes during the first decades of the century — albeit in a more idiosyncratic way — disappeared almost completely after 1850, rendering a text image which came very close to the texts produced by upper class associations. The implications of this are at least threefold: • until 1850, the combination of variable spelling and zusammengebrochener Stil was found in both lower and middle class texts. This contradicts the traditionally held views of an exclusive Arbeitersprache, at least as far as Flanders is concerned. Further research is needed to confirm whether similar patterns also occurred in middle class English and German. • the stylistic and grammatical standardization movement had little impact on the lower classes during the 19th century, but did successfully reach the middle classes from 1850 on. This may indicate a phased diffusion from the upper towards the lower classes. • standardized spelling may have been considered less important or seen as less of a hindrance for 'getting one's meaning across' than stylistic and grammatical correctness. 3.3 Upper class writers Research on the upper class material is not yet complete, so that it is not possible to pronounce a full judgement on the spelling behaviour of the higher social echelons in Bruges. Texts from the first quarter of the 19th century indicate, however, that there was also a considerable degree of spelling variation in the formal meeting reports of one of the most prestigious upper class associations in Bruges (the Sebastian archers' guild mentioned above). These writers used exactly the same spelling variants as their socially inferior colleagues, for vowels, diphthongs and consonants alike. For some phonemes, though, the stage of consistent variable choice had already been reached (see Table 3). Apart from a few syntactical anomalies, neither massive loss of style control nor incoherent text structures occurred. The persistence or evolution of this text pattern during the rest of the century is the subject of ongoing research.
DUTCH ORTHOGRAPHY IN 19TH-CENTURY FLANDERS
[a:] open syllable |[e:] open (syllable [εi]
[œy]
[z] [X] [k]
'ambt'
'ontvangen'
Lower/Middle Class variants until 1900 <e> <ee> <eij> <ey> <ei> <s>
Upper Class variants around 1820 |
Examples from UC texts
<e> <ee> <eij> <ey>
contenteren gouverneeren qualiteijt societeyt
huijs uijt huys uyt
<s>
sal deese gesegt zal deze gezegt
ampt amt
verpacht gebragt trecken trekken contracte contrackte boek boeck
ampt amt ambt ontfangen ontvangen
37
vergaderinge vergaederinge
|
ontfangen
Table 3: Comparison of spelling variants used by Lower/Middle Class writers until 1900, and by Upper Class writers around 1820. 4. Discussion How should we interpret the corresponding spelling problems of lower, middle and upper class writers? What is the key to the apparent phased social stratification of increasing writing quality during the 19th century? I believe that the answer to both questions is to be found in the history of writing education in Flanders, and in the patterns of literacy acquisition in 19th-century Bruges.
WIM VANDENBUSSCHE
38
Although there was no direct link between writers' social class and the quality of their written language, there most certainly was a connection between relative wealth and the availability of quality writing education. The richer the parents, the better the schools they could afford for their children. Expensive daytime schools offered a full language curriculum with composition and grammar exercises, whereas Sunday schools for the poor hardly rose above the level of alphabet learning (Michiels 1978, de Clerck et al. 1984). One effect of this layered system is reflected in Figure 1 (based on Callewaert 1963), which convincingly shows that literacy in Bruges gradually spread from the highest towards the lowest social classes in the second half of the 18th and the whole of the 19th century. The graph illustrates the percentage of illiterate people for four different social classes during the first half of the 19th century in Bruges and indicates that in Bruges the general literacy level — and, presumably, the quality of writing education — rose first among the higher social classes. Whereas the large majority of the middle classes (here further divided into 'middle class' and 'skilled workers') was fully literate around 1850, the lowest social classes (roughly corresponding to 'unskilled workers') only reached the level of full literacy at the beginning of the 20 th century.
Illiteracy in Bruges according to social class (from Callewaert 1963) D 1797-1815 ■ 1815-1830
90
□ 1830-1840
80 70 60
50 40 30 20 10
0
0
upper class
middle class
skilled workers
unskilled workers
Fig. 1: Illiteracy in Bruges according to social class (from Callewaert 1963)
DUTCH ORTHOGRAPHY IN 19TH-CENTURY FLANDERS
39
As indicated earlier, I have been able to show (Vandenbussche 1999) that the spelling and style problems of the lowest social classes during the last phase of the acquisition of literacy (second half of the 19th century) were identical to those of the middle classes during their last phase of the acquisition of literacy, half a century earlier (first half of the 19th century). There is reason to suggest, therefore, that the lower and lower middle class texts may provide examples of transitional varieties, which should be located on a sliding scale between literacy and illiteracy. The lower and middle class language in the research corpora could then be considered as reflecting two phases of this literacy acquisition, which started at different points in time for each class. If our hypothesis about the transitional character of literacy-linked varieties is right, one of the future results of our ongoing research may be the discovery of consistent examples of these frequent spelling and style problems in the writings of the Flemish upper classes before 1800. Literacy percentages may reveal the educational profile of different social classes, but they also convey information about the importance attached to the mastery of reading and writing in these groups. Next to the need for a detailed analysis of Flemish pauper education, it is therefore necessary to clarify the sociological impact of literacy in the lower, middle and upper classes. I assume that upper class life in Bruges contained a substantial written core by the beginning of the 19th century. Lower and middle class writers, on the contrary, may still have lived in cultures which were to a large extent orally based; many of their writing problems are indeed typical of writers in the transition from an oral to a written culture (Fairman 2000). This is reminiscent of the German distinction between writers whose professions were either handarbeitorientiert or schriftarbeitorientiert (Mihm 1998); an insight into the evolution of the lower and middle class professions in Bruges on the continuum between both poles would most certainly contribute to a better understanding of the research data. The professional obligation to become literate probably did not imply immediate 'perfection' on the levels of spelling, grammar and style. During the first years of this transition not all class members may even have felt the need to be literate, as long as someone was able to write and perform the expected written tasks. The relative importance of standardization in different socio-professional contexts (once a literate culture had been adopted) is a third factor which may partly explain the social stratification of writing quality. By the end of the 19th century many associations may have felt the need to perform certain written tasks, but depending on the amount of external communication and the intended reading public, a high degree of stylistic and grammatical standardization was not always required. Our data show this standardization
40
WIM VANDENBUSSCHE
should, in any case, not be understood in the traditional present day sense of the word. The process did not involve the acceptance of a single norm for spelling and grammar, but rather a growing concern with intelligibility. Consistent spelling was of little importance in this respect, but a sound text structure was. 5.
Conclusion In sum, spelling standardization (in the present day sense of minimal variation in form) was unimportant to the majority of the 19th-century writers in Bruges. Whereas lower and middle class writers continued to use a variable spelling system up until 1900, there are indications that upper class writers, too, may have shared this custom for a long period into the 19th century. In the light of these findings the virulent controversy among contemporary spelling designers in 19th century Flanders is reduced to a sterile theoretical discussion which took place above the heads of the actual writers. To these language users, the 'war on spelling' was a non-issue. There are indications, however, that — due to the growing impact of supraregional communication — a certain need for standardization was felt in the realm of style and grammar, first among the higher and, soon afterwards, among the lower social classes.
References Barton, David & Nigel Hall, eds. 2000. Letter Writing as a Social Practice. Amsterdam & Philadelphia: John Benjamins. Betten, Anne, ed. 1990. Neuere Forschungen zur historischen Syntax des Deutschen. Tübingen: Max Niemeyer. Callewaert, Herwig. 1963. Bijdrage tot de studie van het analfabetisme en het lager onderwijs te Brugge. Licentiate thesis, Katholieke Universiteit Leuven. Cherubim, Dieter, Siegfried Grosse & Klaus J. Mattheier, eds. 1998. Sprache und Bürgerliche Nation. Berlin & New York: Walter de Gruyter. Couvreur, Walter & Roland Willemyns. 1998. Spellingoorlog. NEVB 1998. 2802-2805. De Clerck, Karel, Bie De Graeve, Frank Simon, Boudewijn Delaere, Kristien Depoorter. 1984. Dag Meester, Goeiemorgen zuster, Goeiemiddag Juffrouw. Facetten van het volksonderwijs in Vlaanderen (1830-1940). Tielt-Weesp: Lannoo. De Groof, Jetje. Two Hundred Years of Language Planning in Belgium. This volume. De Vries, Jan, Roland Willemyns & Peter Burger. 1994. Het verhaal van een taal. Negen eeuwen Nederlands. Amsterdam: Prometheus.
DUTCH ORTHOGRAPHY IN 19TH-CENTURY FLANDERS
41
De Wolf, Karel. n.d. (c. 1930-1940?). Het Brugsch, lijk het nog leeft, en vooraleer het nog meer verdwijnt. Ms., Brugge. Fairman, Tony. 2000. English Pauper Letters 1830-34, and the English Language. Barton & Hall 2000. 63-82. Godar, Henri. 1947. Histoire de la Gilde des Archers de Saint Sébastien de la Ville de Bruges. Bruges: A.G. Stainforth. Görlach, Manfred. 1999. English in Nineteenth-Century England. An Introduction. Cambridge: Cambridge University Press. Klenk, Marion. 1997. Sprache im Kontext sozialer Lebenswelt. Eine Untersuchung zur Arbeiterschriftsprache im 19. Jahrhundert. Tübingen: Max Niemeyer. Lootens, Adolf. 1868. Oude kindervertelsels in den Brugschen tongval. Brussel: J. Nys. Mattheier, Klaus. J. 1986. 'Lauter Borke um den Kopp.' Überlegungen zur Sprache der Arbeiter im 19. Jahrhundert. Rheinische Vierteljahrsblätter 50. 222-252. Mattheier, Klaus. J. 1990. Formale und funktionale Aspekte der Syntax von Arbeiterschriftsprache im 19. Jahrhundert. Betten 1990. 286-299. Mattheier, Klaus. J. 1998. Kommunikationsgeschichte des 19. Jahrhunderts. Überlegungen zum Forschungsstand und zu Perspektiven der Forschungsentwicklung. Cherubim et al. 1998. 1-45. Michiels, Guillaume. 1978. Uit de wereld der Brugse mensen. De fotografie en het leven te Brugge 1839-1918. Brugge: Westvlaamse Gidsenkring. Mihm, Arend. 1998. Arbeitersprache und gesprochene Sprache im 19. Jahrhundert. Cherubim et al. 1998. 282-316. Milroy, James. 1992. Linguistic Variation and Change. On the Historical Sociolinguistics of English. Oxford, UK & Cambridge, Mass.: Blackwell. Molewijk, G.C. 1992. Spellingverandering: van zin naar onzin (1200-heden). 's-Gravenhage: SDU. NEVB. 1998. Nieuwe Encyclopedie van de Vlaamse Beweging. Tielt: Lannoo. Suffeleers, Tony. 1979. Taaiverzorging in Vlaanderen. Brugge & Nijmegen: Orion & B. Gottmer. Van de Craen, Piet & Roland Willemyns. 1988. The Standardization of Dutch in Flanders. The International Journal of the Sociology of Language 73. 45-64. Vandenbussche, Wim. 1999. Arbeitersprache, a fiction? Belgian Journal of Linguistics 13. 87-103. Vandenbussche, Wim. Forthcoming. Triglossia and Pragmatic Variety Choice in 19th-century Bruges: A Case Study in Historical Sociolinguistics. Journal of Historical Pragmatics.
42
WIM VANDENBUSSCHE
Vandenbussche, Wim & Roland Willemyns 2000. Historische sociolinguïstiek: het Brugge-project. Taal en Tongval 52. 258-276. Van den Toorn, Maurits.C, Willy J J . Pijnenburg, J.Arjan van Leuvensteijn, & Joop M. Van der Horst, eds. 1997. Geschiedenis van de Nederlandse taal. Amsterdam: Amsterdam University Press. Van Eenoo, Romain. 1959. Een bijdrage tot de geschiedenis der arbeidersbeweging te Brugge 1846-1914. Mémoire IV. Leuven: Interuniversitair Centrum Hedendaagse Geschiedenis. Witte Els, Jan Craeybeckx & Alain Meynen. 2000. Political history of Belgium from 1830. Brussels: VUB Press. Witte Els & Harry Van Velthoven. 1999. Language and Politics: the Belgian Case Study in a Historical Perspective. Brussels: VUB Press.
STANDARD GERMAN IN THE19THCENTURY? (COUNTER-) EVIDENCE FROM THE PRIVATE CORRESPONDENCE OF 'ORDINARY PEOPLE'
STEPHAN ELSPAß University of Münster, Germany 1. Introduction: German standard(s) With the emergence of Historical Sociolinguistics in the last two decades, varieties between standard and the traditional basal dialects have come into the focus of attention. The study of these new varieties has cast a new light on language standards and led to a reconsideration of standardization processes. A notable tendency in standardization within the German sociolinguistic tradition may be considered here. According to Besch (1983), Mattheier (1988:5f.) and other German scholars, a standardization process can only be regarded as complete when the predominant part of the population have access to the written and the spoken standard.1 If we accept the existence of a spoken standard of German, the majority of people in the German speaking countries certainly could not make use of Standard German in the 19th century. It was no earlier than the mid-20th century that a form of Standard German became the mother tongue of greater parts of the German population (cf. Durrell 1999:302). British linguists like Milroy & Milroy (1985:24), however, maintain that "absolute standardization of a spoken language is never achieved (the only fully standardized language is a dead language)". That this viewpoint also holds for the German situation may be supported by König's (1989) findings in his Atlas zur Aussprache des Schriftdeutschen ("Pronunciation Atlas of Written German"). König has impressively shown that the pronunciation of Standard German speakers is still a far cry from what German Orthoepic dictionaries prescribe. One may add that this has always been the case since the first edition of the oldest of these dictionaries, the Siebs, was 1 On the differences between the German and British sociolinguists' view on language standard(s) cf. Durrell (1999:292).
44
STEPHAN ELSPAß
published in 1898. From this perspective, even the standardization process of German is still under way. In view of these terminological and conceptual disparities, the following analysis will concentrate on developments in the written standard. The German language has often been described as being standardized by the end of the 18th century. This certainly reflects a general agreement in traditional linguistic handbooks. To what extent does it reflect the reality of written language production, though? Two aspects may be pointed out here which leave this statement at least debatable: The first aspect is whether a written standard had been achieved on all linguistic levels. For the morphological and syntactic level, the answer may be a Yes. For orthography, the answer is a clear No, as the first official orthographic dictionaries (Preußische Schulorthographie etc.) were only published after the new German Empire was founded in 1871, and official spelling guidelines for all German-speaking countries did not come into effect until 1903.2 And the lexicon? The Wortatlas der deutschen Umgangssprachen (Eichhoff 1977ff.) does not just reflect sub-standard but clearly standard variation in German. Consider notable variants in spoken and written German like Sonnabend/ Samstag ("Saturday"), Jungel Bub ("boy"), viertel nach 6/ viertel 7/ viertel über 6/ viertel ab 6 ("quarter past six"). One part of the lexicon which has been largely overlooked is idiomatics, or rather 'phraseology' (as it includes the notion of collocation). It has been maintained that the set of German idioms and collocations had reached its present norm by the end of the 18 th century (Burger 1987:69). Data collected from professional speakers and writers (Elspaß 1998:217ff.) cast doubt on this view, and I argued in a recent study that even today, a phraseological standard hardly exists in German (Eispaß 2002). A second objection to the 'general agreement' on the standardization of German must be raised with respect to access to the standard. Most older studies have focussed on the language of literature (e.g. Blackall 1959) or the theoretical debates of grammarians and other language specialists (e.g. Naumann 1986). How little this affected the linguistic struggle of the majority of people from the lower and lower middle classes — for whom the written language was not an everyday need — has concerned only few linguists to date.3 2
Even printed media did not set a good example here (Wells 1985:349f.). Some of the most notable German studies on written Alltagssprache ("everyday speech") of 'ordinary people', mostly from the end of the 19th and beginning of the 20th century are: Mattheier (1986); Grosse et al. (1989); Schikorsky (1990); Klenk (1997); Mihm (1998). In a report on their 'Bruges Project' about workmen's written language in 19th-century Belgium, 3
STANDARDIZATION IN 19TH-CENTURY GERMAN?
45
The fact that in the middle of the 18 century a written variety of German emerged that may more aptly be called a standard variety should not stop us from looking at standardization processes since then.4 In my view, the most interesting aspects of standardization only begin at the time when the majority of people in a country are able to read and write and are therefore able to participate in the use of a written variety. In German speaking countries, this did not happen before the first half of the 19th century when the majority of German people became literate. Some of the questions that arise from this mass literacy are: In what way did people learn the written standard? Which grammars did they use? Did they use grammars at all? If not, which 'norm authorities' could they rely on? Did teachers master the standard variety? To what extent, and for how long, did regional influences prevail in their actual written language production and their teaching? Which language levels were mostly affected by regional variation? These are the main issues that will be addressed in this paper. First, a brief outline of the literacy campaigns in the German speaking countries in the 19th century will be given (2.). Following that, some new text sources will be introduced which shed new light on the standardization process of German (3.). I will then present results of my own research (4.) and will lastly discuss the implications of these findings (5.). 2. Literacy in the German speaking countries in the 19* century One major reason why language historians in the past neglected the written language of 'ordinary people' in the 19th century is that these people were largely considered to be illiterate. Moreover, even leading German sociolinguists assumed that written language in the private domain was Willemyns & Vandenbussche make it very clear that the language norm debate in late 19 century Flanders did not affect lower class writers' basic problems with writing texts and coping with orthography, grammar and style. In view of their everyday linguistic difficulties, the norm debate appeared to be no more than a "sociaal impactloze discussie met byzantinistische trekjes." (Willemyns & Vandenbussche 2000:270; cf. also Vandenbussche 1999 and in this volume). 4 The term 'high variety', which sociolinguists would use in this context, is somewhat misleading, as the verticalization of linguistic varieties in the 18th- and 19th-century German speaking countries did not necessarily correspond to the social structure of society. There is plenty of evidence that many members of the nobility, for example, did not master written Standard German (see text samples in von Polenz 1994:208ff.; 1999:238). Tony Fairman (2000:65) in his analysis of English pauper letters from the beginning of the 19th century considers the terms 'Standard' and 'dialect' inappropriate for the time. He prefers instead to refer to 'Schooled' vs. 'Open' language and to designate respective realizations of these on a scale between 'Refined' and 'Vulgar' (corresponding to 'Standard' and 'Nonstandard' in today's terms).
46
STEPHAN ELSPAß
confined to the (noble) upper class and the bourgeois middle class, especially the ambitious Bildungsbürger of the 19th century. The lower and lower middle classes were thought to be limited to spoken language in their private communication (e.g. Mattheier 1989:105). The wealth of private letters, chronicles and personal diaries from farmers, artisans, maids etc. that have come to light in recent years have produced sound evidence that these assumptions are simply no longer tenable.5 We may contend today that in the first half of the 19th century the majority of people in the German speaking countries were able to read and write (von Polenz 1999:51). By the mid-19th century the leading German state of Prussia had achieved a literacy rate of 85% — compared to only 52% in England, for example (Gawthrop 1987:29).6 It is imperative, though, to bear in mind significant regional differences. Regions in central Prussia could always boast of relatively high literacy rates, whereas Catholic and rural districts in the east, the far west and the south — particularly in Austria — had to cope with widespread illiteracy (François 1983; von Polenz 1999:52f.). 3. Standardization processes as reflected in private correspondence In recent years, much research has been carried out on 'informal sources' (Russ 1992) in private domains. Whereas much of this work in the English language area was carried out explicitly within the framework of historical sociolinguistics (e.g. Palander-Collin 2000 and the Corpus of Early English Correspondence), some scattered studies in German have so far concentrated on the private correspondence of VIPs, e.g. the letters of the Mozart and Schiller families (Reiffenstein 1993; Russ 1998). Much data, particularly in the context of language variation and change, can be gained from private letters, diaries etc. of 'ordinary people'. Letters from 19th-century emigrants and their relatives and friends, in particular, supply valuable text material for linguistic analysis. The prime asset of this text type is the sheer wealth and the regional distribution of letters which were produced and which have come down to us. Between 1830 and 1914, over 100 million private letters were written from the U.S. to Germany alone (Helbich, Kamphoefner & Sommer 1988:31f.); some nine to ten thousand letters have been collected and archived by various institutions and individuals. This abundance of private documents can be 5
It may be conceded that many text sources were simply not available in print. The first people who started to collect and edit private texts from the lower classes in Germany were social historians and folklorists — not German historiolinguists. 6 Gawthrop has demonstrated that the high literacy rates in 19th-century Germany were due to early 'literacy drives' which go back to the Reformation movement in the 16th century.
STANDARDIZATION IN 19TH-CENTURY GERMAN?
47
attributed to the new socio-communicative situation caused by emigration. In order to stay in contact, emigrants and their families and friends at home were compelled to take up pen and paper. I know of no other instance which illustrates this new situation better than the last few lines of a letter written by 37-year old Catharina Mannott, a farmer's wife from the north German region of Ostfriesland: (1) Hir mit mus ich schließen dies iß mein eßß erß Brief zu schreiben Catharina. mus gut Studiren M. C. Man [With this I must finish. This is my first letter that I am writing. Catharina must study hard. M.C. Man.] (Margaretha Catharina Mannott, born 1831 in Wiesede near Friedburg [Ostfriesl.], letter from late 1868)
Apparently, as can be seen from her quite clumsy handwriting (cf. excerpt), this was the first time since Catharina Mannott had left school, i.e. the first time in nearly 25 years, that she had written a longer text, and it is her first private letter ever. Another advantage of these private letters is that they not only represent written language but also (and probably best of all text types) reflect elements of orality. They even allow for reconstructing historical dialect variants. García-Bermejo Giner & Montgomery (1997) have demonstrated this with evidence from letters written by English emigrants to the United States.7 It should be noted, however, that in spite of the dominance of the traditional dialects in 19th-century spoken language, neither in the German speaking countries, nor e.g. in England (cf. Fairman 2000:75) would people write in their local dialect, but would rather use a written language that was orientated towards the standard variety.8
7
For text samples from the English language area, cf. also Fitzpatrick's (1994) anthology of letters from Irish emigrants to Australia. 8 Dialect would only be employed very rarely, either humorously or as quotations (10 instances in 610 letters — almost exclusively by male writers from the North and Northwest!).
48
STEPHAN ELSPAß
Excerpt from an original letter by Margaretha Catharina Mannott A clear advantage of private letters over more 'official' letters by 'ordinary people', e.g. applications and petitions,9 is their degree of authenticity. The main problem with applications and petitions is that "we can't always be sure whether the people whose names appear at the end of these letters were the actual writers" (Fairman 2000:64), whereas even people with only elementary writing competence were able to write private letters and would prefer to do so, particularly since — as a Northern Irish school inspector in 1867 observed in view of the general appreciation of reading and writing abilities among the local peasantry — [...] he who employs another to write his letters, puts himself in a very humiliating position, as he thereby makes the village teacher, or perhaps some doubtful friend, the depository of little family secrets. (cited after Fitzpatrick 1994: 500)
The aim of my approach is to reconstruct ordinary people's route to the written standard variety. I have analysed 610 letters by 290 male and female writers. The following three text samples may serve as a first illustration of the range of language variation in these letters: (2) An die zweitausend Zeitungen, rückhaltslos geschrieben, unterrichten das Volk von dem, was geschehen ist und was zu hoffen und zu befürchten steht für die nächste Zukunft. Du findest bis zum äußersten Pioneer eine Zeitung, also politisch gebildete Menschen. Jeder Man gehört einer Partei an, er hat eine durchdachte Ansicht von unseren politischen Zuständen und weiß Dir seine Meinung so klar darzulegen, wie der erste Senatas in Waschingston. Von diesen Leuten nehme ich wieder aus die Catholiken und die Alt-Lutheraner, meine Nachbarn. Diese Leute lesen keine Zeitung, sie überlassen das Denken ihren Pfäfflein. Durch die Presse und durch die freisinnigsten Institutionen in unserer county und town Verfassung ist das republicanische Leben in das Volk selbst gedrungen und kein Vergleich, den man
9
Mattheier (e.g. Mattheier 1986) and Klenk (1997), in particular, focussed on petition letters.
STANDARDIZATION IN 19TH-CENTURY GERMAN?
49
mit Frankreich Rom oder einer griechischen Republick machen möchte, paßt auf die vereinigten Staaten [...] (Dietrich Sigismund Gerstein, born 1828 in Rheda [Westphalia], letter from 15.02.1860)10 (3) [...] ich muß euch benachrichtigen das hir die Kattuffelen Zimlich gefault Sint, Sojahr das wir nich wohl mehr zu flantzen haben Aber es ist noch keine deurigkeiten hir Von lebensMitelen ich muß euch benachrichigen, das wir dieses Jahr eine Neue Kirch bauen, und auch wie ville Vamillgen hir Sint 75 Sint Jetz hir und giebt noch Jetes Jahr Mehr das dauert nich manches Jahr den Sint es Über 100 Welches Alle Katollische Sint, und wie es heisch das in Zeit 2 Jahren die Eisen bann hir Verbej kumen Solte Welche uns Großen nutzen Machte Ubrigens kan ich euch nicht viele Neuichkeihten schreiben Als das ich 6 Söhne haben und Ein Methgan [...] (Matthias Simon, born ca. 1810-1815 in Staffel [Eifel], letter from 21.04.1847) (4) Liebe Eltern! Euern Brief habe ich erhalten und gesehn, das ihr alles voll fürt habet was ich euch zuletzt in dem kleinen Briefgen mittheilte. [...] meine Schwiege Mutter will mier 100 Thl. lehnen, welche ich auch warscheinlich gegen Frühjahr schicken werde [...] Liebe Schwiegereltern Es ware mir gewiß auch eine große Freude, wenn ich mich mit Euch vielmals so unterhalten könnte, als wie ich mich mit meiner Mutter unterhalte, allein dieses kann nicht geschehen. [...] Ich weiß das Ihr gewiß gerne hören, daß ich auch etliche Zeilen von Wilhelm schreibe, Er wohnt bei mir im Hause, und sind des Abends immer beisammen und beten dan miteinander den Rosenkranz. ich bin mit ihm sehr gut zufrieden wie ich von Wilhelm gehört, sind sie sehr bekümmert gewesen wegen dem heirathen allein ich denke nicht, daß er sich reut, das Er geheiratet habe [...] (Wilhelm Meis, born 1844 in Uckerath [Sieg] and Theresia Meis, 03.12.1871)
Only sample (2) is close to what we would regard as modern 'Standard German'. It is taken from a letter written by one of the so-called '48ers' who fled Germany after the revolutionary upheavals in the 1840s. The writer was born and brought up in Westphalia with a bourgeois family background and enjoyed a secondary education. Sample (3) is a text written by a farmer from the Rhineland, sample (4) is from a letter written by a locksmith from the same area and his wife. The writers of letters (3) and (4) did not have more than an elementary education. The question is whether samples (3) and (4) can be shrugged off as just 'not standard' and 'incorrect'. This is a way of looking at deviant language use exclusively from today's perspective. Children may get a decent instruction in 10
19th-century letters were almost exclusively written in old German script. Single morphemes, words (esp. names, dates and loan words) appear in Latin script; in this text sample, Latin script is marked by italics.
50
STEPHAN ELSPAß
written Standard in their elementary school education today — though not necessarily in a spoken standard (cf. Davies 2000) — and there are plenty of reference dictionaries, grammars etc. that a writer can consult today when in doubt. This perspective does not apply to the 19th century. Whereas written language instruction at the secondary school level (esp. Gymnasien) and the grammars used there have been extensively researched, there is still no clear picture about writing instruction in elementary schools (Ludwig 1998). Many elementary school teachers — or rather part-time teachers — in the first decades of the 19th century had received no formal teacher training and could hardly afford decent teaching material. Messerli (2000) reports that elementary reading instruction in German speaking Switzerland up to ca. 1830 was often based on handwritten text material from the 18th century — which clearly followed grammatical and orthographical rules that were up to 100 years old. It is not certain what knowledge elementary school teachers had of the standard variety. It would be more accurate to say that they taught a form of German that they considered correct. Variation in writing could then be attributed not only to plain mistakes and direct interference from the spoken dialects, but also and at least partly to regional or even network traditions and patterns of writing that were passed on via basic language instruction.11 In other words, when in the first half of the 19th century, for the first time in the history of German, the majority of people were able to participate in written communication (Grosse 1989:12), there were more "non-standard norms of usage" (Milroy & Milroy 1985:25) rather than just the norm of the standard variety. These norms of usage cannot, of course, be revealed just by comparing grammars and scanning metalinguistic remarks by prominent writers of the time. This has been the traditional approach to 18th- and 19th-century language. It is rather necessary to look at the actual language production of those who received the kind of basic instruction that was taught in elementary schools. The aim of the following analysis is to bring to light the fact that grammatical variation, rather than one monolithic standardized system, was characteristic of 19th-century German. I will present three examples from the two different linguistic levels which were supposed to have been standardized in the mid18th century: morphology and syntax. The analysis of these features will, I believe, paint a somewhat differentiated picture of the standardization 11
Mattheier (1986:243) considers Arbeiterschriftsprache ("workers' written language") a reflex of "gesprochene Sprache der Arbeiter in ihren verschiedenen Schichtungen" ("workers' spoken language on their different [linguistic] levels"), but pleads for an investigation of Arbeiterschriftsprache "in erster Linie als Schriftsprache" ("primarily as written language"). For a critical discussion of the concept of Arbeitersprache "working class [dia-]lect" as a classrelated language variety cf. Mihm (1998:282ff.) and Vandenbussche (1999:57ff.).
STANDARDIZATION IN 19TH-CENTURY GERMAN?
51
processes and perhaps also of traditional views on the development of the German language in the 19th century. 4. Analysis of selected grammatical
features
4.1 Diminutive The first example is a morphological feature: the diminutive form in German. A brief outline of the German diminutive system in the 18th century is necessary to explain the linguistic situation in the subsequent century: Blackall in his book The Emergence of German as a Literary Language 17001775 (1959:130) mentions only two competing variants: -chen and -lein (as in Mädchen "girl" and Mägdlein "little maid"). -chen appears to be the dominant form at the beginning of the 18th century. It has been maintained that the -lein variant in German was revived after 1770 (Öhmann 1972).12 This claim, however, is solely based on the observation that the writers of literary movements like the Sturm und Drang and the Göttinger Hain groups made excessive use of obsolete forms like the -lein diminutive.13 (This may serve as a good example of how misleading a focus on literary language can be in the study of standardization processes.) The actual distribution of diminutive forms in the 18th century can be seen from Figures 1 and 2.14 In the first half of the 18th century, the old West Central German diminutive form -gen (in fact pronounced like -chen) spread out over other 'German writing' regions.15 It eventually became the leading variant south of the Benrath line.
12 Influential 18th-century grammarians like Johann Christoph Gottsched apparently favoured the -lein form for stylistic reasons. According to Blackall (1959:130), Gottsched regarded -lein as "more refined and elevated" than -chen and dismissed as wrong other variants, e.g. the southern German -/ (as in Madl "girl"). 13 These movements are known for their revival of medieval poetry and folksongs. 14 The data, taken from a study by Schebben-Schmidt (1990:314) and Wegera (2000:50, 52), do not include isolated variants like -el, -le or -i. 15 Wegera (2000:46) considers and in Early New High German as allographs of /x/.
52
STEPHAN ELSPAß
Low German -chen (-gen) West Central German
-gen (-lein)
f-el] [-le]
[-el[
East -gen West Upper German
-gen (-lein)
[-chen] [-el]
-chen (-lein) (-gen)
East Central German
-l (-lein)
East Upper German
Franconian (-chen)
He] [-i]
Fig. 1: Diminutive variants in written German, first half of the 18th century It was not until the second half of the 18 century, that it gave way to the -chen diminutive, which has remained dominant until today.16 Even eminent authors from the West Central German dialect areas, however, like the young Johann Wolfgang Goethe, continued to use it in their writing (Eckardt, Stiiber & Trumpp 1999:136f.) until Johann Christoph Adelung (1781:102) in his Deutsche Sprachlehre, which is considered to have been the leading German school grammar until the mid 19th century, prescribed "chen, (nicht gen)". This dominance of the -chen diminutive is also apparent in private letters from the mid 19th century (Figure 3). 75 out of 104 writers use the -chen form exclusively, another 11 writers use it along with other diminutive forms. The -lein form is more or less negligible: only two writers from the south-west employed it frequently, a few others used it in clearly marked contexts. Thus, the writer of text (2) clearly makes ironic use of -lein in the word Pfäfflein (la): (2a) (...) die Alt-Lutheraner, meine Nachbarn. Diese Leute lesen keine Zeitung, sie überlassen das Denken ihren Pfafflein. [The old-Lutherans don't read newspapers, they leave the thinking to their little parsons.] (Dietrich Sigismund Gerstein, 15.02.1860)
16
-lein is still frequent in certain text types (like fairy tales) and is employed for euphonic reasons, particularly after and (e. g. Zwerglein, Bächlein), cf. Duden (1998:504f.).
STANDARDIZATION IN 19TH-CENTURY GERMAN?
53
Low German -chen (-lein ) West Central German
-chen/ -gen
-chen
East -chen West Upper German
-chen
[-le] [-el] [-chen]
East Central German
Franconian (-gen) [-lein] [-el]
-chen
East Upper German
Fig. 2: Diminutive variants in written German, second half of the 18 th century The writer of text (4) uses this variant as part of a poetic cliché (4a): (4a) Ja kleines Brieffelein gehe geschwint zu dem Eiteren hin und bringe ihnen schöne bortschaft da hinn [...] [Yes, little letter, go swiftly to our parents and bring them good tidings.] (Matthias Simon, 21.04.1847)
Some of the other exceptions to the -chen rule are more noteworthy: -le (as in Mädle) by seven Swabian writers, -li by a Swiss writer, -/ by an Austrian writer, -el {Mädel) by a Silesian writer, and finally -gen (Mädgen, Metgen, even Methgan) by 14 writers from the West German region. The -/-variants (-le/ -l/ -li/ -el) can quite easily be attributed to the corresponding forms in the spoken dialects and regiolects (cf. Seebold 1983:1252; König 1994:157; Eichhoff, vol. II, 1977:121), also the two instances of the West Central German plural diminutive -cher (as in Mädcher). As for the notable use of the -gen variant by many West German writers, however, an interference of spoken dialect does not seem very plausible. In West Central German areas, for instance, the dialect form would bc [-çe(n)], if not the coronal version [-se(n)], which appeared in the second half of the 19th century (Elspaß 2000:266f.). More likely is that the writers in the West employed a variant that had 'officially' disappeared by the end of the 18th century. The fact that some grammarians at the beginning of the 19th century took the trouble to mention the -gen variant as incorrect indicates that it was still in use. From the regional distribution it can be concluded that in the West,
54
STEPHAN ELSPAß
it still had the status of a norm of written usage that was passed on to new generations of writers who considered it 'correct'.17
(One symbol stands for one writer. Merged symbols designate different variants used by a single writer, with symbols on the left showing the more frequent variants)
Fig. 3: Diminutive forms in 19th-century letters
17
This interpretation may be supported by the fact that the -gen suffix was and still is apparent in family and place names from the Rhineland area, e. g. Schnütgen, Päffgen, Stiissgen; Röttgen, Heidgen (both near Bonn) etc., pronounced [-çe(n)] throughout the old dialects.
STANDARDIZATION IN19TH-CENTURY GERMAN?
55
One possible domain where such older variants could have been transmitted well into the 19th century was elementary school. We have, in fact, evidence that elementary school teachers used such forms in their own writing. Text (5) is an excerpt from a diary by a teacher from the Rhineland area: (5) Das hier auf der Nebenseite hat der Matthias Huppertz, mein lieber Vater noch geschrieben, und ist das letzte so er diesem Bücheigen einverleibt hat. Er ware seit anfangs July kräncklich, so besonders von einem Bruchschaaden herrührte, das ihme die heftigste schmertzen verursachte. [...] Er war 9 Jahre Kirchenmeister der hiesigen Pfarr Contzen, und zwarn zur zeit des französischen Krieges von 1792 bis 1801, auch hate er ein schönes Stückgen Brod, obwohl mit vieler Mühe zwischen Cölin und Monjoye mit seinem Fuhrwerk verdienet [...] (Johannes Huppertz, born 1782 in Konzen [Eifel], diary entry 21.12.1807)
As was pointed out earlier, many school teachers at the beginning of the 19 century had not received any formal training when they officially or unofficially started their job. Few of them could live on a teaching salary; many of them were artisans or farmers by profession, like Johannes Huppertz, who worked as a farmer as well as a schoolteacher and sexton in a small village of the Eifel region.18 Noteworthy in this text sample are the two words in bold print, Bücheigen ("little book") and Stückgen ("little piece").19 Huppertz's writing appears to be quite elaborate, but it is also somewhat conservative, as can be seen from other morphological features like the -e in ware.20 Thus we can assume that this teacher used -gen according to what he thought was the norm, and we have no reason to doubt that he taught in school what he considered to be correct. th
4.2 Comparison particles The second example, this time on the syntactic level, is the use of comparative markers in German. According to Adelung and the grammarians of the 19th century, two different instances and basically two different forms may be distinguished here: An adjective in the comparative form + the particle als "than" is used to mark differencc: süßer, als Honig ("sweeter than honey") (Adelung 1782:479) 18 Huppertz started teaching in a nearby village in 1808. In 1815, he writes about his first official teacher training under the Prussian authority. 19 Bücheigen actually consists of a double diminutive: Umlaut + ~el- + -gen. 20 The epenthetic -e in the 1st and 3rd person singular of the preterite is originally an Upper German form which was widespread in the West Central German area in the 18th century (Elspaß 2000:252, 271).
STEPHAN ELSPAß
56
An adjective in the positive form + the particle wie (or als) "as [...] as" is used to mark correspondence/ agreement: so reich wie du ("as rich as you") (Adelung 1782:479) so roth, als eine Rose ("as red as a rose") (Adelung 1781:224) Whereas Adelung in the second instance preferred the use of als for stylistic reasons, the grammarians of the 19th century strove for a functional distribution, i.e. one conjunction was to stand for one grammatical form, thus als after comparative and wie as the standard unmarked form after positive (although als was still allowed in certain stylistic contexts). als after positive after comparative
6 39
wie 31.6%
13
68.4%
90.7%
4
9.3%
Table 1: Use of comparison particles in letters of writers with secondary education With respect to the use of comparison particles in private correspondence, it can be quite revealing to take separate looks at letters written by people who had received a secondary education (Table 1) and letters by those with a basic, primary education (Table 2). Writers with secondary education seem to have adopted the codified norm quite well (Table 1). Strictly speaking, only 3 writers (in 4 instances) used an "incorrect" form (wie after comparative). wie
als after positive after comparative
denn
als wie
38
15.1%
181
72.1%
32
12.7%
-
317
62.0%
150
29.4%
40
7.8%
4
0.8%
Table 2: Use of comparison particles in letters of writers with primary education With regard to the actual language use of writers with only primary education, the standards seem to get blurred (Table 2). Especially after comparative, 38% of the writers do not use the prescribed form als. The wie form seems to be very 'normal'. Surprisingly, in a few instances the old denn form was employed and a fourth variant was in use, which the grammars
STANDARDIZATION IN 19TH-CENTURY GERMAN?
57
mostly ignored, namely als wie. This form was common usage in the 18 century, cf. the famous lines from Goethe's Faust: Da steh' ich nun, ich armer Tor! und bin so klug als wie zuvor. Let us have a look at the regional distribution: For the regional distribution of comparison particles (Figure 4), we don't have as clear a picture as with the use of diminutive forms (Figure 3). It is still discernible, however, that writers from the West and North show greater inconsistency and variation in their use of particles after the comparative, with a particular fondness for the wie form. It might be added that in these areas wie is still commonly used in colloquial spoken language today. Here again, the results cannot be attributed directly to the corresponding dialect forms (Figure 5). The dominant particle after comparative in western and northern dialects is als (or as). According to Schikorsky (1990:268), the frequent use of the wie form can be regarded as hypercorrection. Whether we follow this interpretation or not, we are confronted with the empirical evidence that writers with minimal instruction in formal grammar in some regions had two or more variants at their disposal and clearly made use of them — regardless of the grammarians' prescriptions. Yet again, we have data which indicate that the non-standard forms were also passed on in school. In his letters, Friedrich Martens, a young emigrant from the northern region of Holstein, frequently employs the different variants. Here are just some samples (6) from his letters: (6) daß doch des Vaters Auge weiter sieth, gewöhnlich, als wie das des Sohnes [...] er würde vielleicht beßer ausmachen denn ich [...] er wird ein guter americaner abgeben, beßer wie ich, weil er, trotz seines guten benehmen gegen mich, beßer zu sich nehmen kann wie ich [...] (Friedrich Martens, born 1838 in Delve [Dithmarschen], letter from 18.04.1858) (7) Ihr schreibt mir in Eurem Briefe, ich sollte diesen Winter Schullehrer wieder werden, no Sir ich liebe die Veränderung, und demzufolge werde ich diesen Winter mal wieder Kellner werden [...] es bringt wenigsten mehr ein als das Schullehrerwesen [You write in your letter that I should become a school teacher again this winter, no sir, I like the change, and therefore I am going to be a waiter again this winter [...] it pays better than the school teacher job] (Friedrich Martens, letter from 24.09.1858)
As can be seen from the last extract, Martens worked at least temporarily as a (primary) schoolteacher. His use of comparison particles does not follow a
58
STEPHAN ELSPAß
clear rule, let alone Adelung's norm. Rather, he seems to make free use of the three variants mentioned above plus even the archaic particle denn.
Fig. 4: Particles after comparative in 19th-century letters
STANDARDIZATION IN 19TH-CENTURY GERMAN?
59
Fig. 5: Particles after comparative in German dialects (from Lipoid (1983:1237) 4.3 wegen + CASE A third and final example is a prominent grammatical feature in German grammar, namely the use of prepositions with the genitive, in this case the preposition wegen ("because of') + GENITIVE. Again, Adelung's dictate seems to have set the official standard for the following two hundred years: wegen is to be used with the genitive, it is incorrect to use it with the dative (Adelung 1781:349). It is well-known that people in spoken Modern German tend — or rather have always tended21 — to use the dative after prepositions like während, wegen, trotz.22 21
"Wir haben Grund zu der Annahme, daß der Genitiv in der Mundart des Volkes schon zu Anfang des Neuhochdeutschen der Hauptsache nach untergegangen war; in unserer heutigen Schriftsprache steht er noch da ohne jedes Zeichen des Verfalls" ("We have reason to
60
STEPHAN ELSPAß
The 125 contexts with wegen in the letters render only 68 instances where the grammatical case can be clearly identified. However, in only 8 of these 68 instances does the correct genitive case appear. In 42 instances, writers used the 'incorrect' dative, in another 18 instances even the accusative. Again, this result is not surprising in view of the fact that some of these writers received their language instruction from teachers like Huppertz. Sample (8) is one of several instances in his diary in which Huppertz uses the preposition wegen: (8) Das Brod, welches im Herbst wegen dem bevorstehenden Kriege nicht wohlfeil war fängt im Januar 1833 an abzuschlagen. Es gilt itzt 4 Sgr. 4 Pf. [Bread, which was dear last autumn because of the impending war, has started to become cheaper since January 1833. It is now 4 Silbergroschen 4 Pfennige.] (Diary Johannes Huppertz, *1782 Konzen, 05.01.1833)
Once more, this teacher uses a variant in his own writing that does not comply with the school grammars of the time. (A clear indication that he employs this form very consciously is the -e ending in Kriege which Adelung had declared the compulsory dative ending.) Here, we have yet another instance of a norm of usage that deviates clearly from what has so far seemed to be the only grammatical norm in the 19th century. 5. Conclusion: Standard German in the 19* century — a myth? In this paper, three different grammatical features have been discussed which according to German grammar books of the late 18th century and the 19th century were standardized and allowed only one standard form. It has been demonstrated, however, that in the written production of 'ordinary people' from the mid 19th century, different non-occasional variants were in use. These variants could only partly be attributed to interference from regional spoken dialects, while some were rather to be traced back to written language variation in the 18th century. Especially among writers who had received only elementary education, certain regional variants seem to have prevailed until the second half of the 19th century. One reason for the existence of unofficially 'non-standard' forms was, as I have tried to show, the persistence of regional norms of usage. These norms of usage in written language seem to have been at least partly reinforced believe that the genitive had already virtually disappeared from the spoken language of the people at the beginning of the New High German period, whereas it still stands firm in written language without the least sign of decline") (Behaghel 1900:219). 22 The Duden grammar notes the varying use, but still insists on the genitive as the standard form (Duden 1998:392).
STANDARDIZATION IN 19 -CENTURY GERMAN?
61
by teachers in elementary schools who were not aware of the official standard variety or felt insecure about its correct use themselves. Einar Haugen considers a vernacular standardized when it has been "given a uniform and consistent norm of writing that is widely accepted by its speakers" (Haugen 1994:4340, emphasis added). If we follow Haugen's definition of standardization, I contend that in the German speaking countries, written standardization had not been reached by the end of the 18th century, but from the perspective of the majority of speakers and writers was still in full swing in the mid 19th century. The notion of a continuous standard language since the Middle High German period has been revealed as one of the national myths of 19th-century Germany (Durrell 2000) — the idea of the existence of 'Standard German' in 19th-century Germany may prove to be yet another one.
References Adelung, Johann Christoph. 1781. Deutsche Sprachlehre. Zum Gebrauche der Schulen in den Königl. Preuß. Landen. Berlin: Voß. Adelung, Johann Christoph. 1782. Umständliches Lehrgebäude der deutschen Sprache, zur Erläuterung der Deutschen Sprachlehre für Schulen. 2 vols. Leipzig: Breitkopf. Behaghel, Otto. 1900. Geschriebenes Deutsch und gesprochenes Deutsch. Festvortrag, gehalten auf der Hauptversammlung des Allgemeinen Deutschen Sprachvereins zu Zittau am 1. April 1899. Wissenschaftliche Beihefte zur Zeitschrift des Allgemeinen Deutschen Sprachvereins 17/18. 213-232. Besch, Werner. 1983. Dialekt, Schreibdialekt, Schriftsprache, Standardsprache. Exemplarische Skizze ihrer historischen Ausprägung im Deutschen. HSK 1:2.961-990. Besch, Werner. 1988. Standardisierungsprozesse im deutschen Sprachraum. Sociolinguistica. International Yearbook of European Sociolinguistics 2. 186-208. Blackall, Eric A. 1959. The Emergence of German as a Literary Language 1700-1775. Cambridge: Cambridge University Press. Burger, Harald. 1987. Normative Aspekte der Phraseologie. Beiträge zur allgemeinen und germanistischen Phraseologieforschung ed. by Jarmo Korhonen, 65-89. (= Veröffentlichungen des Germanistischen Instituts, 7.) Oulu: Oulun Yliopisto. Davies, Winifred V. 2000. Linguistic norms at school: a survey of secondaryschool teachers in a Central German dialect area. Zeitschrift für Dialektologie und Linguistik 67. 129-147.
62
STEPHAN ELSPAß
Duden. 1998. Duden. Grammatik der deutschen Gegenwartssprache. 6 rev. ed. by Peter Eisenberg et al. Mannheim, Leipzig, Wien & Zürich: Dudenverlag. Durrell, Martin. 1999. Standardsprache in England und Deutschland. Zeit schriftfür germanistische Linguistik 27. 285-308. Durrell, Martin. 2000. Standard Language and the Creation of National Myths in Nineteenth-Century Germany. Das schwierige neunzehnte Jahrhundert. Germanistische Tagung zum 65. Geburtstag von Eda Sagarra im August 1998 ed. by Jürgen Barkhoff, Gilbert Carr & Roger Paulin, 15-26. (= Studien und Texte zur Sozialgeschichte der Literatur, 11). Tübingen: Niemeyer. Eckardt, Hans Wilhelm, Gabriele Stüber & Thomas Trumpp. 1999. 'Thun kund und zu wissen jedermänniglich. ' Paläographie — archivalische Textsorten — Aktenkunde. (= Archivhefte/ Landschaftsverband Rheinland, Archiv beratungsstelle Rheinland, 32.) Köln: Rheinland-Verlag, Bonn: Habelt. Eichhoff, Jürgen. 1977ff. Wortatlas der deutschen Umgangssprachen. 4 vols. Bern & München: Francke. Elspaß, Stephan. 1998. Phraseologie in der politischen Rede. Untersuchungen zur Verwendung von Phraseologismen, phraseologischen Modifikationen und Verstößen gegen die phraseologische Norm in ausgewählten Bundestagsdebatten. Opladen & Wiesbaden: Westdeutscher Verlag. Elspaß, Stephan. 2000. Rheinische Sprachgeschichte von 1700 bis 1900. Rheinisch-Westfälische Sprachgeschichte ed. by Jürgen Macha, Elmar Neuß & Robert Peters (unter Mitarbeit von Stephan Elspaß), 247-276. (= Niederdeutsche Studien, 46). Köln, Weimar & Wien: Böhlau. Elspaß, Stephan. 2002. "Alles unter den Tisch gekehrt". Phraseologische Gebrauchsauffälligkeiten im Urteil von Sprachbenutzern. Wer A sägt, muss auch B sägen ed. by Dietrich Hartmann & Jan Wirrer, 127-160. Hohengehren: Schneider. Fairman, Tony. 2000. English Pauper Letters 1800-1834 and the English Language. Letter Writing as a Social Practice ed. by David Barton & Nigel Hall, 63-82 (= Studies in Written Language and Literacy, 9). Amsterdam & Philadelphia: Benjamins. Fitzpatrick, David. 1994. Oceans of Consolation. Personal Accounts of Irish Migration to Australia. Cork: Cork University Press. François, Etienne. 1983. Alphabetisierung in Frankreich und Deutschland während des 19. Jahrhunderts. Zeitschrift für Pädagogik 30:5. 755-768. Gawthrop, Richard L. 1987. Literacy Drives in Preindustrial Germany. National Literacy Campaigns. Historical and Comparative Perspectives ed. by Richard F. Amove & Harvey J. Graff, 29-48. New York & London: Plenum Press.
STANDARDIZATION IN 19TH-CENTURY GERMAN?
63
García-Bermejo Giner, Maria F. & Michael Montgomery. 1997. British Regional English in the Nineteenth Century: The Evidence from Emigrant Letters. Issues and Methods in Dialectology ed. by Alan R. Thomas, 167183. Bangor: University of Wales Bangor, Department of Linguistics. Grosse, Siegfried, Martin Grimberg, Thomas Hölscher&Jörg Karweickh. 1989. "Denn das Schreiben gehört nicht zu meiner täglichen Beschäftigung". Der Alltag kleiner Leute in Bittschriften, Briefen und Berichten aus dem 19. Jahrhundert. Ein Lesebuch. Bonn: Dietz. Haugen, Einar. 1994. Standardization. The Encyclopedia of Language and Linguistics. 12 vols., ed by Ronald E. Asher, vol. 8, 4340-4342. Oxford, New York, Seoul & Tokyo: Pergamon Press. Helbich, Wolfgang J., Walter D. Kamphoefner & Ulrike Sommer, eds. 1988. Briefe aus Amerika. Deutsche Auswanderer schreiben aus der Neuen Welt 1830-1930. München: Beck. HSK 1:2 = Dialektologie. Ein Handbuch zur deutschen und allgemeinen Dialektforschung. 2 vols. (= Handbücher zur Sprach- und Kommunika tionswissenschaft, 1) ed. by Werner Besch et al. Vol. 2. Berlin & New York: de Gruyter. Klenk, Marion. 1997. Sprache im Kontext sozialer Lebenswelt. Eine Untersuchung zu Arbeiterschriftsprache im 19. Jahrhundert (= Reihe Germanistische Linguistik, 181.) Tübingen: Niemeyer. König, Werner. 1989. Atlas zur Aussprache des Schriftdeutschen in der Bundesrepublik Deutschland. Ismaning: Hueber. König, Werner. 1994. dtv-Atlas zur deutschen Sprache. Tafeln und Texte. Mit Mundart-Karten. 10th ed. München: dtv. Lipoid, Günter. 1983. Möglichkeiten der Komparation in den deutschen Dialekten. HSK 1:2. 1232-1241. Ludwig, Otto. 1998. Alphabetisierung und Volksschulunterricht im 19. Jahrhundert. Der Beitrag der Schreib- und Stilübungen. Sprache und bürgerliche Nation. Beiträge zur deutschen und europäischen Sprachgeschichte des 19. Jahrhunderts ed. by Dieter Cherubim, Siegfried Grosse & Klaus J. Mattheier, 148-166. Berlin & New York: de Gruyter. Mattheier, Klaus J. 1986. 'Lauter Borke um den Kopp'. Überlegungen zur Sprache der Arbeiter im 19. Jahrhundert. Rheinische Vierteljahrsblätter 50. 222-252. Mattheier, Klaus J. 1988. Nationalsprachenentwicklung, Sprachenstandardisierung und Historische Soziolinguistik. Sociolinguistica. International Yearbook of European Sociolinguistics 2. 1-9. Mattheier, Klaus J. 1989. Die soziokommunikative Situation der Arbeiter im 19. Jahrhundert. Voraussetzungen und Grundlagen der Gegenwarts sprache. Sprach- und sozialgeschichtliche Untersuchungen zum 19.
64
STEPHAN ELSPAß
Jahrhundert ed. by Dieter Cherubim & Klaus J. Mattheier, 93-107. Berlin & New York: de Gruyter. Messerli, Alfred. 2000. Das Lesen von Gedrucktem und das Lesen von Handschriften — zwei verschiedene Kulturtechniken? Lesen und Schreiben in Europa 1500-1900. Vergleichende Perspektiven ed. by Alfred Messerli & Roger Chartier, 235-246. Basel: Schwabe. Mihm, Arend. 1998. Arbeitersprache und gesprochene Sprache im 19. Jahrhundert. Sprache und bürgerliche Nation. Beiträge zur deutschen und europäischen Sprachgeschichte des 19. Jahrhunderts ed. by Dieter Cherubim, Siegfried Grosse & Klaus J. Mattheier, 282-316. Berlin, New York: de Gruyter. Milroy, James & Lesley Milroy. 1985. Authority in Language. Investigating Language Prescription and Standardization. London & New York: Routledge & Kegan Paul. Naumann, Bernd. 1986. Grammatik der deutschen Sprache zwischen 1781 und 1856 (= Philologische Studien und Quellen, 114.) Berlin: Schmidt. Öhmann, Emil. 1972. Suffixstudien VIII Die deutschen Diminutivsuffixe -lein und -chen. Neuphilologische Mitteilungen 73. 555-567. Palander-Collin, Minna. 2000. Historical Sociolinguistics and Linguistics. Neuphilologische Mitteilungen 101. 466-469. von Polenz, Peter. 1994/1999. Deutsche Sprachgeschichte vom Spätmittelalter bis zur Gegenwart. Vol. II: 17. und 18. Jahrhundert. Vol. III: 19. und 20. Jahrhundert. Berlin & New York: de Gruyter. Reiffenstein, Ingo. 1993. Sprachvariation in den Briefen der Familien Mozart. Vielfalt des Deutschen. Festschrift für Werner Besch ed. by Klaus J. Mattheier et al., 361-381. Frankfurt am Main, Berlin, Bern, New York, Paris & Wien: Lang. Russ, Charles V.J. 1992. Informal Sources in the History of English and German. Who Climbs the Grammar-Tree ed. by Rosemarie Tracy, 107116. Tübingen: Niemeyer. Russ, Charles V.J. 1998. Die Sprache von Schillers Mutter in ihren Briefen. Ein Beitrag zur Sprachgeschichte im Südwesten Deutschlands im späten 18. Jahrhundert. Deutsche Sprache in Raum und Zeit. Festschrift für Peter Wiesinger zum 60. Geburtstag ed. by Peter Ernst & Franz Patocka, 643650. Vienna: Praesens. Schebben-Schmidt, Marietheres. 1990. Studien zur Diminution in der deutschen Schriftsprache des 18. Jahrhundert. Deutsche Sprachgeschichte. Grundlagen, Methoden, Perspektiven. Festschrift für Johannes Erben zum 65. Geburtstag ed. by Werner Besch, 313-321. Frankfurt am Main, Bern, New York & Paris: Lang.
STANDARDIZATION IN 19TH-CENTURY GERMAN?
65
Schikorsky, Isa. 1990. Private Schriftlichkeit im 19. Jahrhundert: Unter suchungen zur Geschichte des alltäglichen Sprachverhaltens "kleiner Leute" (= Reihe Germanistische Linguistik, 107). Tübingen: Niemeyer. Seebold, Elmar. 1983. Diminutivformen in den deutschen Dialekten. HSK 1:2. 1250-1225. Vandenbussche, Wim. 1999. 'Arbeitersprache' in Bruges during the 19th century. Beiträge zur Historischen Stadtsprachenforschung ed. by Helga Bister-Broosen, 49-65 (= Schriften zur diachronen Sprachwissenschaft, 8). Vienna: Praesens. Wegera, Klaus-Peter. 2000. 'Gen, oder wie Herr Gottsched will, chen'. Zur Geschichte eines Diminutivsuffixes. Wortschatz und Orthographie in Geschichte und Gegenwart. Festschrift für Horst Haider Munske zum 65. Geburtstag ed. by Mechthild Habermann, Peter O. Müller & Bernd Naumann, 43-58. Tübingen: Niemeyer. Wells, Christopher J. 1985. German: a linguistic history to 1945. Oxford: Clarendon Press. Willemyns, Roland & Wim Vandenbussche. 2000. Historische Sociolinguïstik: het 'Brugge-project'. Taal en Tongval 12. 258-276.
ON THE IMPORTANCE OF FOREIGN LANGUAGE GRAMMARS FOR A HISTORY OF STANDARD GERMAN
NELS LANGER University of Bristol, England
1.
Introduction Standard German today is used as a means of supraregional communication, and functions as the norm variety in the typical sociolinguistic domains, such as written language, regional and national media and education.1 The question arises: how did it get there? and even despite intensive research over the last few decades, the puzzle of the origin and development of standard German consists of many little pieces rather than a coherent picture. Whilst it may well be the case that the history of standard German might never be drawn up completely because of the long period of time and the many, partially disputed factors involved, it appears nonetheless a worthwhile venture to explore this sociolinguistic phenomenon. Apart from the purely academic merits, research into the reasons and developmental processes of the prestige variety Standard German might also have some educational justification, namely to disperse the belief in 'good' and 'bad' languages, or of standard German as being a form of perfect or ancient German — views that were very common in the 17th century amongst scholars, but which are still prevalent amongst folk linguists today (cf. Langer 2001a for similar views in relation to the Rechtschreibreform).2 1 But cf. Davies (1995 and 1999) for differences between perception of use and actual use of standard German by secondary school teachers. Clearly, what speakers believe themselves to be speaking (i.e. standard German) does not always correspond to what other speakers or even the normative grammars believe to be standard German. 2 Any attempt to stop folk linguists from classifying people (positively or negatively) according to their language variety will not succeed, since this is one of the primary sociolinguistic functions of language, as countless studies have shown (cf. Preston & Niedzielski 2000). But it may nonetheless be worth our while to point out to folk linguists that linguistically a synthetic genitive (e.g. meines Bruders Kleid) is not in any way superior to an analytic genitive {das Kleid von meinem Bruder).
68
NILS LANGER
The research community is agreed that in the process of standardizing German, many factors were involved; but there is little agreement about which factors were involved and what role these factors played. However, early explanations have by now been rejected, including the views that standard German had its origin in Charlemagne's attempts to promote the vernacular or that Martin Luther somehow created rather than promoted a supraregional German (cf. Langer 2000b for a summary). On the other hand, it is comparatively clear that supraregional language varieties did exist as early as the 14th and 15th centuries (Besch 1967) and that the development took a new turn in the 17th century, when grammarians and other scholars interested in linguistic matters (writers, poets, members of language societies) promoted the idea that if one wanted to replace Latin as the primary language of discourse and get rid of the increasing influence of French from German writing, one would have to offer an alternative: an elegant, an ancient, an original — a perfect German. Since this ideal form of German did not exist, it needed to be created, either by way of forming something new or by way of taking an existing variety and 'cleansing, purifying', it. Konopka (1996) and Takada (1998) recently showed the influence of prescriptive grammarians as regards the mclusion of certain linguistic elements in the emerging standard German, while I hope to have shown their influence in the exclusion of certain constructions (Langer 2000a, Langer 2001b). However, an area as yet underresearched is the process of 'knowledge transfer' — it is all very well to show that particular grammatical prescriptions in the 17th century can be found both in contemporary texts and in the standard language today and that it is therefore reasonable to assume that the grammarians' prescriptions triggered or at least influenced the inclusion in such a prestige variety. Yet it is not entirely clear how these prescriptions were actually transferred so that a language user would comply with them. Studies have been conducted to measure the success of a grammar by number of copies published or sold. However, to measure the success of a book in this way, e.g. the Spelling-Duden today, would ignore the fact that whilst this book can be found in virtually every household in Germany, it is probably rarely touched or read. What makes (made) the Duden so influential is the fact that it is used by the media and in education — not its circulation rate per se. Returning to the 17th century, the piece in the standardization jigsaw I wish to examine in this article is the correlation between language prescription by 'scholarly' grammarians on the one hand and language use and prescription of 'practical' grammarians, namely foreign language (L2) teachers on the other hand. The views of L2 teachers appear particularly useful since in teaching a foreign language, one will (traditionally) pay special attention to the variety of language used and taught, namely the
EARLY FOREIGN LANGUAGE GRAMMARS
69
accepted, the norm, the prestige variety. Comparing the type of German used and taught in early L2 grammars with contemporary academic discussions about an emerging or existing prestige German will shed light on the question of to what extent language users were aware of these discussions and were prepared to accept the prescriptions in their own teaching. 2. The history of standard German As indicated in the introduction, the standardization of German was influenced by many factors. This is partially explained by the lack of a political or academic centre in Germany, in contrast to France and England, for example, where the respective political centres were also the cradles of the standard varieties of their respective national languages. In Germany, the standard language was not formed on the basis of one particular regional dialect or speaker community (e.g. the language of the Court). Tendencies towards supraregional varieties existed in the form of the medieval poets' language3 or trade languages (Besch's 1967 Schreiblandschaften) but crucially, as late as in the 16th century we find references which indicate that no variety had a greater prestige than another {Schryfftspiegel 1527 in Müller 1882). The varieties of German were ordered horizontally, not vertically, as regards their sociolinguistic status (Reichmann 1988). This verticalization, however, did take place in the 17th century, with the emergence of an intense debate, motivated in part by the rejection of the increasing influence and presence of the French language (cf. Jones 1999) and a striving towards some kind of a national identity despite political heterogeneity. In the absence of a central language academy (cf. the Academie Française), the debate over the model and the properties of a prestige German focussed on two factions: the anomalia viewpoint, endorsing the promotion of an existing dialect (namely Meißnisch, the dialect spoken by the proponents of this position) versus the postulation of analogia, i.e. the claim that all of the existing dialects are to some extent corrupt and that the original, perfect German can only be found by (re-)creating a new grammar (cf. Gardt 1999, Hundt 2000, Takada 1998). In sum, standard German was influenced by many factors, such as an increase in the use of written documents, the invention of the printing press, the prestige of the language of Martin Luther, the fame of particular cultural centres, the expansion of schooling, and the influence of prescriptive grammarians (Takada 1998:1). Crucially, however, the standardization of 3
This is currently being investigated by Klaus-Peter Wegera et al. in their project to write a new Mittelhochdeutsche Grammatik based on a corpus which is much less biased to poetic language and high literature than were earlier accounts of Middle High German.
70
NILS LANGER
German since the Middle Ages was not a linear process but rather can be divided into two periods, before and after c.1550. Prior to 1550, varieties existed which constituted supraregional compromises. Features which were regionally or locally marked would be avoided as much as possible. These varieties were not ordered vertically in a sociolinguistic sense and standardization was used for practical purposes to facilitate communication across dialect boundaries. The view that a supraregional or even national variety is better in a philosophical and communicative way only emerged after 1550 and in particular after c.1640, when a standard German was viewed as a prestige variety, to be used intraregionally as well as supraregionally, in order to advance the cause of the German language but also to distinguish oneself from the dialect-speaking lower classes {Pöbel). In order to disentangle the web of factors involved in this process, researchers have attempted to concentrate on individual influences — as regards prescriptive grammarians, Konopka (1996) and Takada (1998) focussed on the question to what extent grammatical features recommended by grammarians are attested in contemporary language use. In Langer (2000a, 2001b), I investigated three morphosyntactic constructions which are excluded from Modern Standard German in order to find out to what extent their modern stigma as bad German can already be found in the 17th century, both in the views of prescriptive grammarians and in contemporary language use. The results showed that the three constructions were indeed stigmatized by prescriptive grammarians but, crucially, at vastly different times and in clearly distinguishable stages (based on the changing reasons given for the stigma). Also, the constructions were still present in language use despite their mention as undesirable by prescriptive grammars, showing that prescriptive influence, albeit ultimately successful (all three constructions are clearly ungrammatical today despite their presence in all non-standard varieties), does not straightforwardly and immediately change language use. 3. The importance of Early L2 Grammars A simple comparison between language prescription and language use will tell us to what extent certain recommendations did catch on — but crucially it does not tell us how they caught on, since this method does not allow us to view the motivations of a language user which made him or her accept and use a certain word, a way of spelling or a grammatical construction in place of one that s/he used before. For obvious reasons (the mortality of 17th-century speakers), these motivations may never be known for certain, but we may extrapolate metalinguistic views by a careful choice of data. Traditionally, studies that wish to investigate the influence of grammarians on actual
EARLY FOREIGN LANGUAGE GRAMMARS
71
language use will compile a corpus that contains a large number of literary texts (Langer 2001b is no exception), i.e. texts from newspapers, pamphlets but also plays and travel descriptions.4 However, these texts, whilst probably as close to spoken and everyday language as one could get with historical data, do not give metalinguistic clues. It is for this reason that the study of early L2 grammars and phrase books appears an ideal source of data in an investigation of acceptance and awareness of the contemporarily emerging standard language, since L2 teachers will pay particular attention to teaching the (in their view) best German used. The most important foreign language in Germany was, unsurprisingly, Latin, which was the primary language of most official discourses, e.g. law, church, medicine, and which was taught to children in schools. It is somewhat symptomatic that only in the beginning of the 17th century was the case made (by Wolfgang Ratke) for the study of German in German schools — up until then, formal education in language was more or less restricted to the teaching of Latin. However, despite the vast exposure of all educated speakers to Latin, we have some evidence that Latin would not always be used between Germans as a means of supraregional communication. This is somewhat surprising since the differences in the German of speakers from very different regions may have been greater than the differences between their (learned) varieties of Vulgar Latin. Today's complaints by schoolchildren about Latin as a difficult language were, however, already known in the 17th century, and Caspar Seidel (1647:10) complains that the Germans had no problems with learning modern languages (Italian, French, Polish, Bohemian) but could not speak Latin, despite having learned it for "zwelff vnd wol mehr Jahr". Latin was often considered too academic and this may well have had an influence in the 17th century on parents' choice to send their children to Ritterakademien where they learnt French (as well as fencing, dancing and other courtly delights) rather than Latin. The influence of all things French increased due to the envied and copied model of the French court and had repercussions for the promotion of both French loanwords and the learning of French as a foreign language. Whilst 38 grammars of French were published in the 16th century, this number increased to 173 in the 17th century and 400 in the 18th century (von Polenz 1994:65f.). Other foreign languages were much less important, although loanwords entered the German language not only from French (military, society, dancing, 4
However, as shown in Langer (2000a), the use of stigmatized constructions such as the auxiliary tun is not necessarily more frequent in literary texts than e.g. in theological or scientific ones.
72
NILS LANGER
diplomacy), but also Italian (banking, music), Spanish (diplomacy, overseas colonies), and Dutch (shipping, also a catalyst for French words). My corpus of foreign language grammars and dictionaries contains some 60 works from 1534 to 1794 with a quantitative emphasis (40 texts) on the 17th century.5 Core of this corpus are the collections of the Herzog AugustBibliothek in Wolfenbüttel, Germany, which, in the 17th century, was one of the largest libraries in the world and thus may be considered to provide a representative sample of the range of L2 grammars and dictionaries that were published and bought at the time.6 The 16th century (10 works) is represented by multilingual dictionaries and phrasebooks (Latin, German, French, Dutch, Spanish, Italian in various combinations with at least three languages in any given book) on the one hand and Latin grammars of German on the other hand. It is striking to note that L2 grammars of German were written in Latin, showing very clearly the status of Latin as the omnipresent lingua franca. In the 17th century, there are also grammars and phrasebooks which include English, but also more exotic languages such as Rotwelsch, Lettish, Czech (böhmisch) and Anglo-Saxon. The 18th century sees a substantial increase in grammars of English, as well as the inclusion of a Yiddish grammar. Whilst this corpus cannot claim absolutely to present a balanced cross-section, it is hoped that it provides a sufficiently broad sample to assess the status of standard German in this type of grammar. Crucially, not all grammars aim to teach German — a substantial portion uses German as the L1 and teaches a foreign language to Germans. This, however, is no obstacle to our investigation of the promotion of certain German constructions as good German. Foreign language teachers will be more aware of their language than 'normal' language users, since it is their profession to reflect on language and to be able to explain both L1 and L2 to their pupils in order to make any differences or similarities clear. In this sense, we can safely assume that a foreign language grammar written in German will contain a language variety that is more likely to be reflected on or filtered than a text written by a non-linguist. Therefore, both L2 grammars of German and from German provide us with exciting data, able to provide us with clues to what extent the recommendations of prescriptive grammarians were known and accepted by (linguistically aware) language users.
5
For full details of the corpus, cf. Langer (2001b). One of the advantages of the Herzog August-Bibliothek is that most of the books were actually bought in the 17th century and therefore may provide clues as to the status of a given book in its time rather than today, i.e. whether it was considered important enough to be bought for the collection. 6
EARLY FOREIGN LANGUAGE GRAMMARS
73
Despite the claim above that the 17 century provides the stage for the emergence of Modern Standard German, it is of course very difficult to date this development. In relation to L2 grammars, the question is thus: what kind of German was taught in L2 grammars in the 17th century, at a time when there was no prestige variety (before 1640), when the debate about standard German intensified (1640—1700) and when the existence (although not the actual grammatical properties) of a standard German was generally accepted amongst prescriptive grammarians (after 1700)?7 At least four possibilities present themselves: • the regional dialect of the author or place of publication • a supraregional standard German (although this did not exist in a generally accepted form) • a compromise, perhaps with foreign, i.e. Latin or French influences • a faulty German, when authors were not native speakers of German 4. The data Having presented the general objectives of the study, I will now provide some tentative results based on an investigation of two morphosyntactic constructions, namely polynegation and auxiliary tun. Both constructions are ungrammatical in Modern Standard German but are well attested in the language use of the Early New High German period (1350-1650). Polynegation, i.e. the presence of at least two negative elements in a clause with overall negative reading, is a common property of the medieval varieties of all West Germanic languages. In German, its use had already decreased, however, by the early 16th century to about 2% (Pensel 1976), and in this sense it is rather surprising that its stigmatization by prescriptive grammarians as 'bad' German did not take place before 1750. Today, it is still found in German dialects, albeit infrequently. Auxiliary tun emerged in the Middle Ages in the West Germanic languages, as a grammatical offspring of causative tun. It was a common property of all Early New High German dialect areas and text types and can still be found in all modern German dialect areas (cf. Langer 2000b). In contrast to polynegation, its stigmatization began much earlier, with the first objection voiced by Erasmus Alberus as early as 1550. Importantly, its stigmatization as poor German occurred in stages from being seen as bad practice in poetry (from 1550) to bad written German (1640/1680) to bad, 7
In this context, it is important to point out that the year dates given above are based on a more or less accepted consensus of the research literature. I believe I have shown in my own research (Langer 2000a, Langer 2001b) that these groupings are still very much to be disputed.
NILS LANGER
74
lower class German (1741). (Note in this context that the construction took a very different route in English, where it is fully grammaticalized to enable questions and negation in the standard language.) The data are presented in four sections, looking at the usage and comments on both constructions separately. The most important finding is the surprising scarcity of evidence, given that both constructions are completely ungrammatical in standard German. 4.1 tun: usage There are six grammars where tun is used as an auxiliary, either in the body text as part of an explanation or in the preface (Anchinandrius 1616, Klatowsky 1603, Sumaran 1623), or as part of a dialogue in the phrasebook section which was often appended to grammars (Anon 1 1607, de la Faye 1615, Güntzel 1648, Sumaran 1623). In both cases, we can safely conclude that the use of tun indicates that the author did not consider it 'bad' German. Klatowsky (1603:XCVI)
[Czech grammar]
Thun sie dennoch bey uns nicht schaden / halt ich doch es künd nun nicht grösserer schaden vnd verderben sein /[...]
Anon 1. (1607:no page)
[French grammar]
Beaucoup promettre & rien donner, c'est aluy qui chasse & rien ne prend Vil verheissen vnd nichts geben / thut den vnseligen trösten.
de la Faye (c. 1615:19)
[French phrase book]
Aber lieber thut mann nicht ewer Hoheit groß unrecht das mans derselben sorgen thut
Anchinoandrius (1616:Vorrede, last page)
[Italian grammar]
Hiermit thue ich E.E.G. in den Schutz des Höchsten befehlen / vnd wünsche euch sampt vnd sonders ein glückseligs anstehendes Newes Jahr /[...]
Sumaran (1623:14)
[quadrilingual phrasebook and grammar]
Die Frantzosen vnd Welschen haben kein Genus neutrum, vnnd do sie es schon haben / thun sie dem generi masculino zuaignen.
Sumaran (1623:459)
[quadrilingual phrasebook and grammar]
K: Wer auff bit verkaufft / thut vil weggeben / vil verlieren / vñ nie kein Gelt haben Qui vende à credit, vende beaucoup, l'ais perde, & argent n'a iamais. Chi vende à credenza, spacia robba assai, l'amico perde, denari non ha mai.
Güntzel (1648:319)
[Italian grammar]
Heyraten. Heyrathe[n]. N. freyen. L. in den Ehestand treten. Ehelich werden. [...] in die Freundschaft zu freyen. L dauon sagt man, in die Freundschaft zu nahe freyen,
EARLY FOREIGN LANGUAGE GRAMMARS
75
thut vielen gerewen. Si dice nel Pron. Non bisogna entrar frà Parente e parente item i parenti escono fuor di casa. Item il sangue non fi mai acqua.
Note, however, that all of these instances are comparatively early, dating from a period when the use of tun as an auxiliary was only stigmatized as poor poetry. In this sense, this data corresponds to the recommendations from prescriptive grammarians, namely that tun is used in written German when no stigma argued against it and that no examples were found at a later period, say after 1680, when the use of tun was frowned upon. 4.2 tun: rule The apparent avoidance of auxiliary tun after 1682 is partially challenged by a quotation from a French grammar of German where auxiliary tun is given as an optional way to realize the present indicative form of any verb. Intriguingly, at a time when it was more or less unanimously agreed by prescriptive grammarians that the use of tun as an auxiliary is undesirable, especially in written German, Anon 2 (1682) suggests that the auxiliary is not only acceptable but can be used fort élegamment when conjugating German verbs. This recommendation of the use of the auxiliary to the foreign learner is repeated in J.J.M.'s universal grammar (Allgemeine Sprachkunst, 1763). Similarly Adelung (21801), who strongly objects to the use of the auxiliary in German, nonetheless argues that it is very convenient when speaking English since it allows the learner to avoid difficult ablaut patterns by simply using tun with an infinitive.8 Anon 2 (1682:100)
[French grammar]
lieben, aimer - Indicatif Present Ich liebe, ou Ich thue lieben, j'aime Du liebst, ou du thuest lieben, tu aimes Er liebt, ou er thuet lieben, il aime
Anon 2 (1682:105) Tous les verbes se conjugent fort élegamment par le verbe thuen, faire: avec leur infinitif qui devient substantif, & cela seulement au temps présent & à imparfait de l'indicatif, à l'imperatif, & au present & imparfait du subjonctif: comme vous voyez dans le verbe lieben, aimer: conjugué cidessus, comme: Ich lieb, j'aime, ou ich thue lieben, je fais amour
8
Adelung appears to be unaware that one cannot use do in non-emphatic sentences in standard English (except of course in negative sentences and questions).
NILS LANGER
76
The examination of grammars from or into English is particularly interesting in this context because of the properties of tun as a fully grammaticalized auxiliary in English. The L2 grammar teacher will be aware of this, and depending on whether he considers this to be at odds with German or not, will teach the construction as either something that is also found in German or something that is ungrammatical in German. The evidence in Offelen (1686/7) is open to interpretation. On the one hand, Offelen paraphrases I did love as either Ich liebte or ich thate lieben, thus suggesting that he considers both the synthetic and the periphrastic indicative as possible constructions of the German grammar, in analogy to Anon 2 (1682) above. On the other hand, it may simply be the case that he translated the phrase word by word to illustrate the construction clearly, a practice frequently employed by grammarians. His ich thate lieben might therefore, even in his eyes, be ungrammatical, and the lack of any further explanation would be obvious to his reader, who, as a speaker of German, will know what is German and what is not. Clearly both interpretations are possible here. Offelen (1686/87:65)
[English grammar]
[...] darumb habe ich hier nicht setzen woellen; die unvollkommene zeit der anzeigende weise, imperfectum praesentis, hat allezeit, I did heist so viel als Ich thate I did love Ich liebte, oder ich thate lieben [...]
However, the following may suggest that Offelen did not consider auxiliary tun as a grammatical construction of German since in cases where he had to translate auxiliary do into German, he never used the auxiliary in German, as exemplified below: Offelen (1686/87:85)
[English grammar]
My Father doth not love me ? Mein Vatter liebt mich nicht. [...] Doth my Father love me ? Liebt mich mein Vatter ?
In sum, there is evidence in only two grammars of the corpus where tun is discussed as a grammatical construction. Whilst Anon 2 (1682) clearly accepts it as a possible feature in German, the evidence in Offelen (1686/87) is not as clear-cut and probably suggests that the tun is not part of his grammar of German. 4.3 Polynegation: usage The use of polynegation in the corpus is equally rare as the use of tun. However, given the relative scarcity of polynegation in other texts of the Early New High German period (Pensel 1976), this is less surprising than the rarity
EARLY FOREIGN LANGUAGE GRAMMARS
77
of tun. Polynegation is found in two cases, both from the first quarter of the 17th century. In de la Faye (1615), it is part of a phrasebook, whereas in Sumaran (1623) it is used to explain a particular rule of Italian spelling. Both cases suggest that the construction was not frowned upon and both cases were used by non-native speakers of German. To what extent the last point is indicative is open to discussion, for the construction, whilst rare language use, was not stigmatized as poor German before the mid-18th century. de la Faye (c. 1615:7)
[French phrasebook]
Geh hin trag sie wiederumb ins Waschhauß vnd gieb ihr einen guten außpützer / dann ich sie mit nichten nicht haben will.
Sumaran (1623:26)
[quadrilingual grammar & phrasebook]
k kein k haben sie gar nit / sondern anstatt deß k, brauchen sie das ch nemblich / che cosa, liß ke cosa
4.4 Polynegation: rule There are four comments which discuss the use of polynegation. Whereas English grammars are particularly interesting for an evaluation of tun because of the grammatical properties of English do, grammars of French and Greek are more interesting regarding the use of polynegation in German, because in both these languages two negative elements in a clause do not render positive reading, contrary to Latin, but also contrary to Modern German and Modern English. De la Faye 1621 (and Rayot 1643 in strikingly similar words) discusses the use of French negation by explaining that two elements are required, of which one, namely the pas ox point has no special meaning. Since he appears to emphasize this fact of French grammar, one might deduce that this is different from German in his view, i.e. polynegation is only part of French, not German. On the other hand, this result conflicts with the fact that the same author does use polynegation in his German (de la Faye 1615). de la Faye (1621:387f.)
[French grammar]
Ne, nicht / wird allezeit in Phrasibus, die da ein Ding verneinen gebraucht / vnnd gemeiniglich das pas oder point, so doch in diesen Reden keine sonderliche Bedeutung hat / auff das Verbum so nach ne gesetzet wird [...]
Rayot (1643:196)
[French grammar]
[...] das pas oder point, so doch in diesen reden keine sonderliche bedeutu[n]g hat / nach dem verbo gesetzet / als: [...]
In the much later Latin grammar by Stübel (1694), the use of two negatives in the target language is explained as yielding a positive reading. To what
NILS LANGER
78
extent this is meant to be a difference from the source language German is unclear. An assumed difference is possible, for Stübel chooses to translate non sum nescius without any negatives as ich weiß, but the opposite, namely that Latin and German are seen as behaving alike, can also be argued, since two negatives are hardly possible in his example (? ich unweiß nicht, ? ich nicht nicht weiß). Stübel (1694:33)
[Latin grammar]
Reg XVII. Duarum negationum. Zwey Verneinungs = Wörter in einem Commate bedeuten das Gegentheil; als: nemo non, iedermann non ignoro, oder / non sum nescius, ich weiß
This ambiguity for the 21st-century reader is further complicated by the evidence from Langjahr's Gründliche Anleitung where polynegation is clearly mentioned and seen as a perfectly normal property of German. Admittedly, Langjahr's example is rather poetic, and this type of usage can be found much later still, e.g. in Lessing and Schiller (Wustmann 31903:265f.). Furthermore, much in the spirit of the time, Langjahr supports his grammatical rule by referring to other languages, namely French and the biblical language Greek. Langjahr (1697:no page)
[German grammar]
Wenn zwey Verneinungs Wörter zusammen gesetzt werden / alsdem verneinen sie noch härter / wie auch bey den Griechen und Frantzosen. Als: Gerechter Leute Licht Verleschet nimmer nicht.
Again, despite the clear evidence from Langjahr that polynegation is an acceptable construction in German, it is striking that of the sixty grammars investigated, only very few actually comment on these features. This is somewhat puzzling in that it appears to indicate teachers were not aware of either of these constructions. However, as alluded to above, polynegation and auxiliary tun are exceptional in that both are strongly ungrammatical today and in that they have very different histories of stigmatization: tun was stigmatized early but remained a common property of language use until the very late 17th century; polynegation was already rare by about 1500 but not stigmatized until the mid-18th century. In any case, the apparent disregard of foreign language teachers for these constructions is somewhat unexpected, given our prediction above that L2 teachers would be particularly careful not only in using good German, but also in rejecting 'bad' constructions.
EARLY FOREIGN LANGUAGE GRAMMARS
79
5.
Conclusion In this article the case was made for an investigation of early foreign language grammars in connection with drawing up a history of standard German. It was argued that L2 grammars might be particularly suitable for examining how and to what extent recommendations made by prescriptive grammarians found their way into language use because of the metalinguistic reflection which language teachers are likely to subject themselves to prior to writing a foreign language grammar. In this way, L2 grammars are fundamentally different from other types of text which will produce language with much less reflection on the choice of constructions and words. The effectiveness of prescriptive grammarians might thus be measured on the basis of their success in influencing or convincing language teachers to use their, rather than somebody else's variety of German. On the evidence of two prominent morphosyntactic constructions, namely polynegation and auxiliary tun, sampled because of their ungrammaticality in Modern Standard German and drawn from a corpus of sixty L2 grammars, dictionaries and phrasebooks, this article showed that the evidence is heterogenous and partially ambiguous. Crucially, however, the rules and usages of L2 teachers do not fundamentally violate the recommendations of contemporary prescriptive grammarians, suggesting that language teachers were probably aware, as practical linguists, of the results of discussions by theoretical linguists.
References Primary literature Adelung, Johann Christoph. 2 1801. Versuch eines vollständigen gramm atisch-kritischen Wörterbuches der Hochdeutschen Mundart, mit beständiger Vergleichung der übrigen Mundarten, besonders der oberdeutschen. Vierter Teil, von Seb—Z. Leipzig: J.G.I. Breitkopf. (Reprint: Hildesheim: Olms, 1970). Anchinoandrius, Henricus Cornelius. 1616. Grammatica Italica Das ist Gründliche Unterrichtung wie die Italianische oder Welsche Sprach von den Deutschen in jhrem gnugsamten gelernet werden [...]. Hamburg: Heinrich Carstens. Anon 1. 1607. Dictionarium Latinvm, Gallicvm, et Germanicvm: Vna cvm formulis loquendi [...]. Mümpelgart: Jacob Foillet. Anon 2. 1682. La veritable et unique Grammaire Allemande, Exactement corrigée et augmentée de plus d'un tiers dans cette nouvelle Edition [...]. Strasbourg: Frederic Guiillaume Schnuck.
80
NILS LANGER
A.N.S. 1696. Grammatica Nova Anglicana, facillima methodo congesta, tertia vice revisa, auctior reddita, et syntaxi aucta. Marburg: Ephraim Böncke. Ebers, Johannes. 1794. Englische Sprachlehre für die Deutschen nach Sheridan's und Walter's Grundsätzen. Berlin: Gottfried Carl Nauck. Faye, Abraham de la. c.1615. Miroir des Actions [...] nützliche Gespräch [...]. Jena: n.p. Faye, Abraham de la. 1621. Institutiones Lingua Gallicae: Oder Gründliche Vnterweisung der Frantzösischen Sprachen [...]. Jena: Johannes Beithmann. Güntzel, Johannes. 1648. Haubtschlüssel Der Teutschen und Italiänischen Sprache. Daß ist Vollständiges Wortbuch aller Teutschen vnd Italiänischen Stamm [...]. Augsburg: Andreas Aperger. Klatowsky, Ondrej. 1603. Knjz'ka w Cjeském a Nemeckém Gazyku Slozena / [...] Ein Büchlein in Behmischer vnd Deutscher Sprach / wie ein Behem Deutsch / deßgleichen ein Deutscher Behemisch lesen / schreiben vnd reden/lernen soll. Prag: M. Danyel Adam von Weleslawin. J.J.M. 1763. Allgemeine Sprachkunst das ist Einleitung in alle Sprachen. Herausgegeben von J.J.M. Erlangen: Johann Caspar Müller. Langjahr, Johann Jacob. 1697. Kurtzgefaßte Doch Gründliche Anleitung Zu Leichter Erlernung der Teutschen Sprache / allen derselben Liebhabern zum besten aufgesetzet/[...]. Copenhagen: Johann Melchior Liebe. Offelen, Henricus. 1686-87. A double grammar for Germans to learn English and for Englishmen to learn the German tongue [...]. London: Nathaniel Thompson. Rayot, Pierre. 1643. Le Souhait des Alemans. Der Teutschen Wunsch [...]. Bremen: n.p. Seidel, Caspar. 1647. Kurtzer / Richtiger / Leichter vnd Anmuthiger Kinder=Donat Sampt angehenden Compendio Grammatica, [...]. Tübingen: Zacharias Hertel. Stübel, M. Andreas. 1694. Latinismus in Nue: [...] Das ist: Die gantze kurtz—gefasste Lateinische Sprach=Kunst / [...]. Leipzig: Friedrich Groscheffen. Sumaran, loan. Angelus von. 1623. Newes Sprachbuch/als nemblich Teutsch/ Frantzösisch / Italianisch vnd Spanisch gar leichlich lernen zu reden. Munich: Auctoris. Secondary Literature Besch, Werner. 1967. Sprachlandschaften Jahrhundert. Munich: Francke.
und Sprachausgleich
im 15.
EARLY FOREIGN LANGUAGE GRAMMARS
81
Davies, Winifred V. 1995. Linguistic Variation and Language Attitudes in Mannheim-Neckarau. Stuttgart: Franz Steiner. Davies, Winifred V. 1999. ,Geregeltes Miteinander oder ungeregeltes Durcheinander?' Versuch einer Beschreibung der sogenannten ,Umgangssprache' in Mannheim-Neckarau. Linguistische Berichte 178. 205-229. Gardt, Andreas. 1999. Geschichte der Sprachwissenschaft in Deutschland. Berlin, New York: de Gruyter. Hundt, Markus. 2000. ,Spracharbeit' im 17. Jahrhundert. Berlin & New York: de Gruyter. Jones, William Jervis. 1999. Images of Language. Six Essays on German Attitudes to European Languages from 1500 to 1800. Amsterdam & Philadelphia: John Benjamins. Konopka, Marek. 1996. Strittige Erscheinungen der deutschen Syntax. Tübingen: Niemeyer. Langer, Nils. 2000a. Zur Verteilung der tun-Periphrase im Frühneuhochdeutschen. Zeitschrift für Dialektologie und Linguistik 67:3. 287316. Langer, Nils. 2000b. On the Polyfunctionality of Auxiliary tun. German Studies towards the Millennium ed. by Christopher Hall & David Rock, 261-282. Berne: Peter Lang. Langer, Nils. 2001a. A Lesson in Linguistic Purism. German as a Foreign Language, III/2000. [URL: http://www.gfl-journal.de]. Langer, Nils. 2001b. Linguistic Purism in Action — Stigmatising the Auxiliary tun in Early New High German. Berlin & New York: de Gruyter. Müller, Johannes. 1882. Quellenschriften und Geschichte des deutsch sprachigen Unterrichtes bis zur Mitte des 16. Jahrhunderts. Gotha: Thienemann. (Reprint: Hildesheim: Olms, 1969.). Pensel, Franz. 1976. Die Satznegation. Zur Ausbildung der Norm der deutschen Literatursprache auf syntaktischer Ebene (1470-1730). Der Einfachsatz ed. by G. Kettmann, & J. Schildt, 285-326. Berlin: AkademieVerlag. Preston, Dennis R. & Nancy Niedzielski. 2000. Folk Linguistics. Berlin & New York: Walter de Gruyter. Polenz, Peter von. 1994. Deutsche Sprachgeschichte vom Spätmittelalter bis zur Gegenwart. Vol. IL Berlin & New York: de Gruyter. Reichmann, Oskar. 1988. Zur Vertikalisierung des Varietätenspektrums in der jüngeren Sprachgeschichte des Deutschen. Deutscher Wortschatz. Lexikologische Studien. L.E. Schmitt zu seinem 80. Geburtstag ed. by Horst
82
NILS LANGER
Haider Munske, Peter von Polenz, Oskar Reichmann & Reiner Hildebrandt, 151-180. Berlin, New York: de Gruyter. Takada, Hiroyuki. 1998. Grammatik und Sprachwirklichkeit von 1640-1700. Tübingen: Niemeyer. Wustmann, Gustav. 31903. Allerhand Sprachdummheiten. Leipzig: Grunow.
NORMS AND STANDARDS IN 16TH-CENTURY SWEDISH ORTHOGRAPHY ALEXANDER Y. ZHELTUKHlN Neurostyles Inc, Atlanta, USA Introduction Standardization is a process with an underlying social-political motivation attempting to promote uniformity and to suppress variability for reasons that are considered functional (Milroy 1992:130). It comes about for functional reasons and its purpose is to make a language serviceable for "communicating information-bearing messages over long distances and periods of time. It is imposed through its use in administrative functions by those who have political power" (Milroy 1992:129). Bartsch stressed that "standardization usually takes place together with literalization, and the development of a written language, or after different written varieties of a language have been developed, and printed techniques have been invented in order to achieve a widespread distribution of the literary works" (Bartsch 1987:251). The 16th century is the period in Swedish cultural history, when all the prerequisites of standardization are in place. Most importantly, book printing is introduced and a great quantity of religious and secular literature becomes widely available in print, laying the path to the standardization of orthography. Traditionally, 16th-century Swedish orthography has been viewed as chaotic and irregular, even in printing. In this paper it is argued that a considerable effort was made to achieve uniformity in spelling in 16th-century Swedish printed documents. Using empirical findings, combined with psycholinguistic and sociolinguistic theories, the claim is made that several community norms evolved over that period. One of them was elevated to the level of the orthographic standard, which was first and foremost followed by the royal print shop. Norms and standards Studies from the sociolinguistic perspective (Labov 1972; Thelander 1979; Milroy 1992) have convincingly demonstrated that variation could be
84
ALEXANDER Y. ZHELTUKHIN
structured. As variation in 16 -century Swedish orthography recalls that in modern spoken language, it seemed appropriate to apply sociolinguistic methods to the study of 16th-century orthographic patterns. The model discussed in this paper comprises two key components: 1) dynamic consensus norms; and 2) a standard, which is less susceptible to change than the norms. The dynamic norms are further divided into social and periodic norms. Consensus norms The term 'social norm' is borrowed from sociolinguistics, where it describes a set of rules followed within one community. The sociolinguistic concept of norm holds that norms are social in the sense that they depend on consensus among speakers within the community or communities concerned (Milroy 1992:6). From such a vantage point it follows that different social groups incorporate within themselves sets of recurrent and distinctive norms, community norms, through which they can be characterized. A periodic norm is a set of rules, which are followed by language users of different social affiliations over a period of time. Members of several communities could share a range of orthographic elements, which then as a whole constituted the periodic norm. Over time they could give up using this set of shared elements and adopt a set of alternative rules, thus giving rise to a new periodic norm. Despite the transitions from one periodic norm to the next, the social norm preserves its distinctive character and uniqueness in relation to the other contemporary community norms. The standard The standard is a relatively static prescriptive and/ or codified norm, in relation to which both types of consensus norms are subordinate, but independent (cf. Milroy 1992:82). The standard takes its origin in a norm, which evolved within a social group and is maintained by that group, with the social-political motivation to promote uniformity and to suppress variability for functional reasons (Milroy 1992:130). In this model, a social norm gives rise to the standard, which is then followed by several communities in certain contexts (e.g. in print). As the printed messages are long lasting and widely available for members of different social groups, the lifespan of the standard is longer than that of a norm. The standard is likely to outlive the norm that originated it, but eventually it is modified or fully replaced by a new standard for the same reasons as it came about originally.
16TH-CENTURY SWEDISH ORTHOGRAPHY
85
Orthographic representations and automatic writing Consistency in the spelling of individual community members is an important prerequisite for the rise of a norm. The earlier, relatively scarce, studies of 16th-century orthography claimed, however, that the 16th-century scribes displayed "a very lax attitude to spelling" (Scragg 1974:26). The Swedish orthography of the 16th century seems indeed chaotic due to considerable graphotactical variation and a large number of homophones at the level of phonemes. The verb hava ("to have") could be spelled in at least four different ways — , , , — because the modern Swedish medial was represented by four homophones — , , , — which are also referred to as orthographic variants below. According to the dual-route theory of spelling, there are two major subsystems in the mental lexicon that put together the orthography of words, following semantic activation: the phoneme-grapheme conversion system; and the orthographic output lexicon (e.g. Rapcsak & Rubens 1990; Roeltgen & Rapcsak 1993). Depending on how the orthography of the word is provided by the lexicon, there may be two groups of orthographic representations: one at the level of phoneme or phoneme clusters, residing in the phoneme-grapheme conversion system; and the other at the lexical level (as a whole word, e.g. ), residing in the orthographic output lexicon. The variants are mutually exclusive, as only one of them must be picked for the output at any given moment in time and the selection is automatic. The alternative connectionist theory of parallel distributed processes (PDP) provides a more detailed mathematical model of how orthographic representations are created and accessed, basing its core assumptions on the advances in neural networks modelling. Like its rival, the dual-route theory, it also holds that two pathways are necessary in, for example, reading, but not because different principles apply to entities of different types (regular words versus exceptions), but because different tasks must be performed. One pathway (phonological) performs the task of transforming orthographic representations into phonological representations directly. The other (semantic) pathway actually performs 2 tasks. The first is specific to reading, namely the transformation of orthographic representations into semantic representations. The second is a more general aspect of language, namely, the transformation of semantic representations into phonological representations (Plaut et al. 1996:100). In contrast to the dual-route theory, the connectionist model argues that there is a single, uniform procedure for computing a phonological representation from an orthographic representation, and vice versa, that is applicable to irregular words, non-words, and regular words. This single mechanism learns to process regular words, irregular words, non-words, and
86
ALEXANDER Y. ZHELTUKHIN
other types of letter strings through experience with the spelling-sound correspondences implicit in the set of words from which it learns (Seidenberg & McClelland 1989:525). The representations of different entities are encoded as different patterns of activity over a number of primitive neuron-like representational units and the processing is mediated by connections among the units (see Plaut et al. 1996 for details). Both theories imply that reading and, for that matter, writing and copying are automatic processes, which have evolved with experience, require little cognitive effort, and are difficult to turn off (cf. Logan 1997). The spelling ambiguities present in deep orthographies like that of Modern English or 16thcentury Swedish, are resolved by the competition of the mutually exclusive orthographic representations. The outcome of the competition is determined by the strength of the competing representations (or, using the connectionist terminology, by the strength of connections between the representational units). The strength (or the activation value) depends, in its turn, upon the frequency of exposure to the entities. The more frequently an orthographic variant is used, the greater chance there is that it will be provided by the lexicon for the output in the automatic writing or copying. This effect is further magnified by word frequency. Several experiments involving neural networks supported the claim of connectionists that the phonological pathway maintains an intrinsic sensitivity to word frequency (Plaut et al. 1996:100). Judging from the volume of preserved handwritten and printed material and the known routines of the Royal Chancery (Svalenius 1991), the writing practised by 16th-century professional scribes must have been automatic and the laws governing such practice must have determined orthographic output. Following exposure to available orthographic variants over the course of a career, a scribe would have established orthographic representations for each variant, each of them having a particular strength. For efficient and expedient writing in the automatic mode, particular orthographic representation must have retained the strongest value over time and consequently would have been retrieved more readily than alternative variants. A scribe was then more likely to spell the verb hava as, for instance, , if the variant or was the strongest representation and thus the most accessible for the output. This automatic preference for particular orthographic variants could have been shared within groups of several scribes (e.g. chanceries), forming social orthographic norms, which, in their turn, would prompt more orthographic consistency at the individual level. In an earlier study (Zheltukhin 1996), I provided evidence for orthographic consistency at the individual level by analyzing some 600 handwritten and printed documents from the first edition of the Swedish New Testament in
16TH-CENTURY SWEDISH ORTHOGRAPHY
87
1526 to the second edition of the Bible in 1618. The individual orthographic strategies were then described and identified using sets of orthographic variants (homophones at a single phoneme level and homophones representing phoneme clusters), which were selected based on their frequency of occurrence and internal variability (see Table 1). l)th-/dh-/d-
2) sk-/sch-/sc-
4) -V-/-VV-/ -Vh-/-hV7)-nC-/-nnC10)-t/-tt/-th/ -tth/-dt/-dtt
5) -ck-/-k-/-ch-
3) -ffu-/-ffw-/ -fu-/-fw6) -e-/-ä-;-o-/-å-
8) -Cn-/-Cnn11) endings of participles
9) -n/-nn 12) supine ending
Table 1: Sets of mutually exclusive orthographic variants In this paper the same material is used for identification and description of the dynamic social and periodic norms, as well as the 16th-century orthographic standard. 16th-century standard All religious documents in print, analyzed for this study, including the only 16th-century edition of the Bible (1541), several editions of the New Testament, and works of prominent scholars of the time, displayed a homogeneous orthographic pattern with respect to the selected variables. The recurrent set of variants of the selected orthographic variables characteristic of these documents is presented in Table 2. This set is referred to as the standard orthographic code. l)th4)-V7)-nC10)-t/-dt
2)sk5)-ck-/-k8)-Cn11) -t/-d/-dt
3)-ffu6)-e-;-o9)-n 12) -t/-d/-dt
Table 2: The standard orthographic code The standard orthographic code was first introduced in the New Testament printed in 1526 under the supervision of Olaus Petri (1493-1552) and Laurentius Andræ (14827-1562). Several elements of the standard code (e.g. <sk->, ) seem to have been borrowed from the Vadstena tradition.1 It 1
The influence of the 15th-century Vadstena tradition is considered in various papers. Carlquist describes it as a "written Bridgettine standard", referring to the homogeneity of
88
ALEXANDER Y. ZHELTUKHIN
is interesting that the standard orthographic code was also found in the available autograph letters by Olaus Petri and Laurentuis Andrae (Zheltukhin 1996:102). Along with the editors of the New Testament, the standard orthographic code was also displayed in the writings of other prominent clergymen, including Bishops Martinus Gestricius, Nils Knutsson, and Laurentius Petri, as well as clerics, who also served in the Royal Chancery as secretaries (e.g. Knut Håkansson (9 September 1575; KKMS 8), Petrus Michaelis Fecht, the author of the treatise on the history of the Swedish coat of arms (KS 22 May 1572), and Henrik Matsson, to name but a few). These examples indicate that the standard orthographic code was not an artificial construct, but a product of the evolution of the Vadstena tradition into the norm of the highly educated clergy. As the prominent members of the group were entrusted by royal power with the editing and the "distribution of the literary works"(cf. Bartsch 1987:250-251), this norm was elevated to the level of orthographic standard. Changes in 16th -century periodic norms The 16th century saw three major orthographic changes with respect to the selected variables. The →, <sch->→<sk->, →, and → transitions stand out as boundaries between long-lasting periodic norms. The → transition was the most drawn-out, spanning almost half a century from the late 1540s to the 1590s. While the primary reason for this transition is thought to be an earlier sound change, analogy could have been another driving force. The chronology of the → shift suggests a paradigmatic influence of homographic German forms, to which Swedish scribes were exposed to a considerable extent during 16th-century language contact. Figure 1 demonstrates that the transition affected the homographic forms (e.g. då(daa), derföre, der, den, dem, and des(s)) much
structure, style, and elaborate punctuation (Carlquist 1996:230-235). A similar attitude to the Vadstena tradition was expressed earlier in Wollin (1981) in respect of the translation strategies. Loman pointed out a number of morphological features in the Gustavus Vasa Bible and writings of clerics similar to those of the Vadstena language (Loman 1955; 1956). Similarities were even found in the production of the royal chancery, especially in the 1520s, when royal charters were normally compiled and penned by clergy (Loman 1956:111). Lindqvist discovered several Vadstena features in the orthography of Laurentius Andræ, who was secretary to the royal chancery during that time (Lindqvist 1929:10-17). The fact that some orthographic features in the New Testament, edited by Laurentius Andræ, and the Gustavus Vasa Bible were inherited from the Vadstena language, is also stressed by Svensson (Svensson 1981:30).
16TH-CENTURY SWEDISH ORTHOGRAPHY
89
earlier than words lacking the graphic association with German (e.g. de, det, devis, denne, dette).
Fig. 1: The / transition in Swedish words by orthographic similarity with German. The → transition was less protracted. It began around 1570, at the same time as the → transition gained ground across the board, and lasted for about quarter of the century until the mid 1590s. This change must have come about for purely practical reasons to make writing more expedient, by dropping superfluous elements. Finally, the <sch->→<sk-> shift, which in itself is not a clear transition from one variant to the other, but an almost sudden disappearance of one of the two competing variants, <sch->. Over the course of the 16th century, <sch-> coexisted with <sk-> as an equally popular variant, but was rather abruptly eliminated in the late 1580s. Unlike the two other major, relatively gradual, transitions, the <sch->→<sk-> shift suggests either a formal act of linguistic planning or some structural changes within the chanceries. Incidentally, the three changes seem to have peaked at approximately the same time — in the 1580s. To verify this, I conducted a statistical analysis of the distribution of the competing variants over the period 1570-1599, when these transitions overlapped. I compared 2 groups of 40 documents each, dating from 1570-1585 and 1586-1599 respectively. The documents in both groups originated in the Royal Chancery and represented all three pairs of competing variants. The results are summarized in Figures 2-4.
90
ALEXANDER Y. ZHELTUKHIN
Fig. 2: The t(h)/ d(h)-transition Over the period 1570-1585, the forms are still 5 times more common than the forms (F(l, 78) = 104.1, p < 0.01). After 1586, the royal scribes already display a non-significant trend in favour of (F(l, 78) = 2.2, p > 0.05), but the old variant is still heavily utilized (see Figure 2). Figure 3. The ff/f-transition
1570-1585
1586-1599
Fig. 3: The ff/f-transition.
16TH-CENTURY SWEDISH ORTHOGRAPHY
91
Figure 3 demonstrates a significant interaction between the average number of occurrences and the period. The short forms are virtually non-existent in the first period, 1570-1585 (F(l. 78) = 239, p < 0.01). During the second period, 1586-1599, they become almost 7 times more common than the old forms (F(l, 78) = 11.1, p < 0.01).
Fig. 4: The sk/sch-transition. The comparison of the <sch/sk> occurrences for the period 1570-1585 (see Figure 4) shows no significant difference in their distribution (F(l, 78) = 0.3, p > 0.05). In the second period, the balance shifts significantly in favour of <sk> (F(l, 78) = 132.7, p < 0.01). 16th-century social norms Besides the royal print shop, major contributors to the archive of dated and signed documents include independent chanceries, united under the name of the Royal Chancery. These chanceries, separated from each other in time or, in some instances, by the rivalry of their lords, are viewed as independent communities with their own set of rules and norms that reflect their identity. The analysis yielded smaller sets of the orthographic variants, which uniquely identified the chanceries considered as independent social entities. The data are summarized in Table 3.
92
ALEXANDER Y. ZHELTUKHIN
King Gustavus (Vasa) King Erik King Johan Duke Karl, King Karl IX King Sigismund
sk-/sch-,-ffu-/-ffw-,ffsch-,-dt,-nn(-) sk-,-ffu-/-fu-,-nn(-) sk-,-ffw-/-fw-,-ck-/-ch-,-n(-) sk-,-fu-,-å-,-ä-
Table 3: Social norms. Like the numerous individual strategies, the social norms were not impermeable to change. In the course of my analysis I found several changes within community norms that could be explained by historical facts, such as changes in royal power, internal changes in the chancery personnel, or other historical events. Most of the periods of strong variation coincided with changes in royal power, which normally led to structural changes within chanceries and considerable flow of personnel. One such period is the 1596-1598 civil war between King Sigismund (1566-1632) and Duke Karl (1550-1611). The preceding one, 1588-1592, falls in the last years of King Johan's reign (15681592), characterized by the influx of new secretaries to the chancery (Svalenius 1991:34-36). Almost all observed changes in royal power, also followed by certain changes of personnel in the Royal Chancery, are clearly marked by <sch/sk> variation: the <sk/sch> transition separates the reign of King Gustavus Vasa (1496-1560) from that of King Erik (1533-1577) around 1560, and its reversion falls around 1568, when King Johan (1537-1592) succeeded his brother Erik. Several orthographic changes during the reign of King Gustavus Vasa (1523-1560) may be associated with the flow of personnel in the chancery. One such development is the shift, around the turn of the 15301540s. Throughout the 1520s and for most of the 1530s was dominant. At that time the Royal Chancery was manned primarily by clerics (Loman 1956). Laurentius Andræ, one of the editors of the New Testament printed in 1526, in which the standard orthographic code was introduced, was in charge. As a result, the norm of the Royal Chancery coincided to a considerable extent with the norm of the clergy and, consequently, the standard, characterized by . The alternative came into continuous usage in the late 1530s, when the Germans Georg Norman and Conrad von Pyhy became secretaries. The very first <sch/sk> shift in 1542 might have been connected with the publication of the Gustavus Vasa Bible, in which <sk> was the dominant variant.
16TH-CENTURY SWEDISH ORTHOGRAPHY
93
Historical changes can also explain the varying orthographic pattern of Duke Karl's chancery in 1570-1572. It is known that after the fall of King Erik, King Johan ruled the country along with his brother Karl. They together presided over the meetings of the Parliament and State Council, and both signed the most important political acts. The resulting contacts between the chanceries caused the charters of the ducal scribes to be interspersed with features typical of the Royal Chancery. In December 1572, which marks the end of the period of strong variation, their co-operation collapsed. The break in relations led to the prolonged isolation of the chanceries (Söderqvist 1898:200-201) and made the differences in their orthographic strategies more pronounced. The early 1580s were characterized by a revival of the friendly relations between the brothers Johan and Karl (Söderqvist 1898:239), and this period is also clearly marked by the intrusion of orthographic features typical of the Royal Chancery, into the relatively homogeneous pattern of the ducal chancery. The shift from the original form <Johan> to the Polish <Jahan> in the address of royal charters can be directly linked to the efforts of the King and, even more so, of the Queen, Katarina Jagellonica (1526-1583) — daughter of Sigismund I of Poland (1467-1548) — to revive Catholicism in Sweden. This change in orthography was carried out over a very short period of time. The transition started in 1577, and already by the end of 1578 the new form <Jahan> dominated in the royal charters. The instantaneous nature of this change suggests an act of linguistic planning. It is interesting that the change happened immediately after the publishing of King Johan's Liturgia and the opening of the Catholic College in Stockholm (both in 1576). Group identity and provenance Despite the changes, each community norm preserved its uniqueness and remained distinguishable from other contemporary social norms. In the periods of severe confrontation between the groups, documents of each group could be easily identified even in print. In the early 1590s, after the death of King Johan, the royal chancellors followed the orthographic standard, slightly modified by the → transition. The norm of the ducal chancery was consistently adhered to in the printed charters signed by Duke Karl, while the Warsaw chancery of King Sigismund was faithful to its own norm. It is interesting that different norms could be kept separate even within one document. For example, in Acta Coronationis (19/11, 1594), Chancellor Sparre (1550-1599) composed the introduction, while Duke Karl was in charge of the second part. The two parts can easily be distinguished on an orthographic basis and associated with the
94
ALEXANDER Y. ZHELTUKHIN
opposing authorities: the first is characterized by the standard orthographic code, while the other has elements typical of the norm of the ducal chancery. The uniqueness of the community norms makes orthography a very useful provenance tool. The assumption is that the orthography of a copy comprises elements characteristic of the individual orthographic strategy of the scribe, who penned the copy, as well as orthographic variants from previous copies and, probably, the original. Identifying the author of a manuscript or a copyist involves the separation of the orthographic layers in a manner similar to archaeological excavation. Each orthographic element is related to known social or periodic norms and dated or attributed to a certain group or even a person. I describe this method in greater detail with examples in my earlier study (Zheltukhin 1996). One of the examples deals with an unidentified manuscript with a copy of the Swedish Chronicle by Olaus Petri. The orthographic strategy of the author, which, as mentioned above, was identical to the standard orthographic code, was known. First, the orthographic variants typical of the standard orthographic code were excluded. The set of remaining variants, introduced by the copyist(s), was then compared with known periodic and community norms. The comparison yielded a close match with the orthographic norm of King Sigismund's chancery stationed in Warsaw in the late 1590s. Several other texts contained in the codex were penned by the same hand and displayed the same orthographic strategy. The texts mentioned several dates, which confirmed the assumption that the manuscript came about in the late 1590s. The codex also contained evidence that it had been commissioned by Nils Rask, King Sigismund's tutor and advisor, providing additional support for the hypothesis that a scribe from King Sigismund's chancery could have been involved. Discussion This study demonstrates that some of the changes in periodic and social norms could be related to historical facts. Given the scarcity of historical material, however, it is difficult to establish if what is termed here the 16thcentury orthographic standard was the result of conscious attempts to achieve and enforce uniformity in spelling. There are several documents dated to the second half of the 16th century, which, among many other matters, are concerned with linguistic problems, language purity, and, in particular, resistance toward the influx of foreign words (Zheltukhin 1996:26). Some of them, judging from their chronology, might have prompted the major shifts in the periodic norms, discussed above.
16TH-CENTURY SWEDISH ORTHOGRAPHY
95
However, none of them explicitly mentions orthography, the 16 -century standard, or the attitudes of language users to it (Lindroth 1975:222).2 Our knowledge of 16th-century 'language-planning agencies' (cf. Bartsch 1987:250) is confined only to sporadic references to proof-reading like, for example, a paragraph from The New Church Rules of 1571 ("skal intedt wtgå på tryck medh mindre thet bliffuer aff någrom af Bisperna well öffuerseedt och gillatt" ["nothing should be printed without supervision and approval of the bishops"]). Both the source of the quoted regulation and the mention of bishops in it indicate that "intedt" is likely to refer primarily to religious literature, rather than the entire production of the print shop, including royal charters. The proof-reading of documents prior to printing is also indirectly supported by the existence of the Corrector's office, associated with the royal print shop. The archives have preserved very little information about the royal correctors, in which the assignments are formulated in a very general way. It is known, for example, that a corrector was once specially employed to supervise the printing of the catechism (Schück 1923:68). During the 16th century correctors were only created by royal appointment. All known correctors were clerics and some of them served at the same time as secretaries in the Royal Chancery, for example, Petrus Michaelis Fecht and after him Johan Bilius, both of whom were also professors at the Catholic college in Stockholm. The affiliation of the correctors with groups, characterized by the standard orthographic code, provides additional support for the claim that at least religious books with wide distribution were proof-read. While contributing to the uniformity of religious works in print, by applying the standard orthographic code to them, the correctors also acted as guardians of their own community norm. It is also obvious that certain adjustments to the standard must have taken place, as the orthography in the printed works overall is more homogenous than in the manuscripts. The standard orthographic code is, for instance, 2
Early Modern English material in this connection is far more revealing. The earliest known attempt to draw up rules and to devise new orthographic systems was The A.B.C. for Children, printed before 1558. It was followed by Thomas Smith's Dialogue concerning the Correct and Emended Writing of the English language (1568) and John Hart's An Orthographie (1569). Beside this work, John Hart produced two more, advocating spelling reform: The opening of the unreasonable writing of our inglish toung; and A Methode or comfortable beginning for all vnlearned, whereby they may bee taught to read English, in a very short time, with pleasure (see Barber 1976). A more considerable attempt at reform was made in 1580 by William Bullokar in Booke at Large, for the Amendment of Orthographie for English Speech, and later, in 1582, by Richard Mulcaster, the teacher of Spenser, in Elementarie, "the most extensive and the most important treatise on English spelling in the 16th century" (Baugh & Cable 1993:205).
96
ALEXANDER Y. ZHELTUKHIN
absent from the handwritten charters by King Erik's scribes, while it dominates the printed charters of the king for that same period. It is possible that the standard orthographic code in the printed secular documents was also the contribution of the royal chancellors and secretaries. Secretaries Rasmus Ludvigsson (1500?-1594) and Henrik Matsson, who adhered to the standard orthography in their writs, were active throughout the entire rule of Erik XIV. The orthographic strategy of Erik Sparre — the long-standing chancellor in Stockholm during the reigns of Kings Johan and Sigismund — deviated from the standard only because of the forms, reduced under the influence of the <jfu>→ transition. One document, in fact, provides evidence of A charter, signed by Chancellor Sparre's involvement in proofreading. Sigismund on 7 January 1598 and printed in Warsaw in accordance with the norm of the Sigismund's chancery, contains a number of spelling corrections made by the chancellor's hand. The same charter was then published several days later in Stockholm under the chancellor's supervision. It is interesting that the orthography of the Stockholm copy was even closer to the standard than Sparre's, because it contained the older forms, and no forms. If the Warsaw copy was indeed the original version of the charter, the Stockholm variant should be considered the result of several layers of proofreading, in which Chancellor Sparre took an active part. Conclusion Despite the differences in numerous variable orthographic norms, the 16thcentury orthographic standard, established by the country's spiritual leaders and sealed in the Gustavus Vasa Bible, was maintained in print and by several social groups of highly educated language users throughout the century. Toward the beginning of the 17th century, under the influence of the major shifts in the periodic norms, the standard became obsolete and the social group that established it was no longer influential. When the second edition of the Bible came out in 1618 under King Gustavus II Adolf (1594-1632), its orthography differed dramatically from the first edition. The orthographic code in it was based on the elements of the social norm formed in the chancery of Duke, and later King, Karl. With the publication of the Gustavus II Adolf Bible, this norm became the new orthographic standard.
References Primary sources The National Archives. (Riksarkivet, Stockholm). Kanslitjänstemänskoncept och mottagna skrivelser [KKMS], vols 7, 8, 9, 10.
16TH-CENTURY SWEDISH ORTHOGRAPHY
97
Kanslistilar (odelade kansliet) [KS], vol. 1. Ämbetsarkivet. Kungliga koncept, vols 5, 13, 14, 22. Riksregistraturet (svenskt) 1561-1594. Secondary sources Barber, C. 1976. Early Modern English. Edinburgh: Edinburgh University Press. Bartsch, R. 1987. Norms of Language. Theoretical and Practical Aspects. London: Longman. Baugh, A.C. & T. Cable 1993. A History of the English Language. 4 th revised edition. London: Routledge. Carlquist, J. 1996. De fornsvenska helgonlegenderna. Källor, stil och skriftmiljö. (= Samlingar utgivna av Svenska Fornskriftsällskapet, 262:81.) Uppsala: Svenska Fornskriftsällskapet. Labov, W. 1972. Language in the Inner City. Philadelphia: University of Pennsylvania Press. Lindqvist, N. 1929. Bibelsvenskans medeltida Ursprung. Uppsala: Appelbergs boktryckeri. Lindroth, S. 1975. Svensk Lärdomshistoria. Medeltiden, reformationstiden. Stockholm: Norstedt. Logan, G.D. 1997. Automaticity and Reading: Perspectives from the Instance Theory of Automatization. Reading & Writing Quarterly: Overcoming Learning Difficulties 13:2. 123-146. Loman, B. 1955. Försvenskning av latinska substantiv i reformations-tidens skriftspråk I. Nysvenska Studier 35. 84-137. Loman, B. 1956. Försvenskning av latinska substantiv i reformations-tidens skriftspråk II. Nysvenska Studier. 74-121. Milroy, J. 1992. Linguistic Variation and Change. Language and Society 19. Oxford: Blackwell. Plaut. D.C., J.L.A. McClelland, M.S. Seidenberg & K. Patterson 1996. Understanding Normal and Impaired Word Reading: Computational Principles in Quasi-Regular Domains. Psychological Review 103. 56-115. Rapcsak, S.Z. & A.B. Rubens 1990. Disruption of Semantic Influence on Writing following a Left Prefrontal Lesion. Brain and Language. 334-344. Roeltgen, D.P. & S.Z. Rapcsak 1993. Acquired Disorders of Writing and Spelling. Linguistic Disorders and Pathologies ed. by G. Blanken, J. Dittman, H. Grimm, J. C. Marshall & C.-W. Wallesch, 262-278. Berlin — New York: De Gruyter. Schuck, H. 1923. Den svenska förlagsbokhandelns historia. Festskrift författad på uppdraget av P. A. Norstedt & söner i anledning av dess 100 årsjubileum nittonhundratjugotre. Vols. 1-2. Stockholm: Norstedt.
98
ALEXANDER Y. ZHELTUKHIN
Scragg, D.G. 1974. A History of English Spelling. New York: Barnes & Noble. Seidenberg M.S. & J.L.A. McClelland 1989. Distributed Developmental Model of Word Recognition and Naming. Psychological Review 4. 523-568. Söderqvist, O. 1898. Johan III och Hertig Karl 1568-1575. Uppsala: Uppsala Universitet. Svalenius, I. 1991. Rikskansliet i Sverige 1560-1592. (= Skrifter utgivna av Svenska Riksarkivet, 7.) Stockholm: Riksarkivet. Svensson, L. 1981. Ett fall av språkvård under 1600-talet. Om historiskt a i andeiser i aldre nysvenskt sprak med särskild hänsyn till regieringen i tryckta kanslihandlingar under Gustav II Adolfs regering. Lund: Ekstrand. Thel ander, M. 1979. Språkliga variationsmodeller tillämpade på nutida Burträsktal. Uppsala: Uppsala Universitet. Wollin, L. 1981. Svensk latinöversättning 1: Processen. (= Samlingar utgivna av Svenska Fornskriftsällskapet, 251.) Uppsala: Svenska Fornskriftsällskapet. Zheltukhin, A. 1996. Orthographic Codes and Code-Switching. A Study in 16th Century Swedish Orthography. Stockholm: Almqvist & Wiksell International.
EMERGING MOTHER-TONGUE AWARENESS THE SPECIAL CASE OF DUTCH AND GERMAN IN THE MIDDLE AGES AND THE EARLY MODERN PERIOD
LUC DE GRAUWE University of Ghent, Belgium
A rare incunabulum, printed in 1482 — only 10 years before the 'discovery' of the New World — in the city of Gouda in the present-day province of South Holland, contains what is believed to be the first instance of the term Nederlands 'Netherlandic/-isch, Dutch'.1 It occurs as part of the formula also wel overlantsche als nederlantsche tale ende sprake (De Vooys 1950:120),2 "both 'Upperlandic' and 'Netherlandic' language and speech". This belief is partly right, but it is also partly wrong. It is right insofar as nederlantsch here undoubtedly refers to the form of language used in the Low Countries, on the eastern shores of the North Sea. It contrasts with the term overlantsch, for which 25 years earlier (in 1457) the synonym dat hoghe duutsche had already been used. The latter occurs in a breviary, also from Holland (Van Wijk 1910:239-240):3 "dese oefeninghe is ghetogen vanden hoghen duutsche int neder duutsche" ("this (spiritual) exercise has been translated from High 'Dutch' 4 into Low Dutch"). In contrast to today, the term neder duutsch here refers not only to the language of Northern Germany, but to all the speech forms spoken along the
1 I wish to thank esp. Torsten Leuschner, Andrew Cusack and Peter Flynn for carefully reading the manuscript and for their valuable suggestions. 2 For more details see De Grauwe 1992, whose views are cited with approval by von Polenz 2000: VII, 82-84, 100-101; cf. also Menke 2001:12-14. 3 In fact, the term is already found once 17 years earlier in the anonymous continuation of Jan van Boendaele's Brabantse Y eesten of Rijmkroniek van Brabant, book 7 (completed A.D. 1440), verse 2455 uut hoghen duutsche ghetransfereert (see the ed. by J.H. Bormans, vol. 3, Brussels 1869:109). 4 Still used by De Foe in 1721: "I spoke high Dutch": OED, vol. III:729c (sub B 1).
100
LUC DE GRAUWE
shores of the North Sea and the Baltic Sea, from Dunkirk to Reval (Tallinn).5 It is interesting to note that these contrasting terms occur first in this Holland breviary rather than in some text from Germany itself. There the first reference to a corresponding distinction is a famous passage in a sermon dating from ca. 1275 and attributed to Berthold von Regensburg (Pfeiffer 1965:250/38-251/2): Ir wizzet wol, daz die niderlender und die oberlender gar ungelîch sint an der sprâche und an den siten. Die von Oberlant, dort her von Zürich, die redent vil anders danne die von Niderlande, von Sahsen, die sint ungelîch an der sprâche. [You may know that the 'Niderlender' and the 'Oberlender' are quite different in speech and manners. People from the 'Oberlant', from Zürich, speak differently from those from the 'Niderlant', from Saxony; they are quite different in speech.]
Of course, the term 'Oberlender' refers not only to the inhabitants of Zurich, nor does 'Niderlender' refer to those from Saxony alone; these places are merely taken to represent the larger opposition between North and South to which the statement primarily refers. Similarly, a quarter of a millennium later, Martin Luther strove to write in such a way that "mich beide, ober- und niederlender verstehen mögen,"6 (that "both ober- and niderlender may understand me"). In my opinion, Luther might well have subsumed presentday speakers of Dutch under the latter. In Luther's day, they were the inhabitants of the Burgundian Kreis within the Holy Roman Empire, including such important duchies and counties as Brabant, Holland, Zealand and (since 1529) the county of Flanders, for most of its part a former French fiefdom. Medieval sources never clarify quite where the frontier between oberlant and niderlant lay. Whether it coincided with the famous Benrath isogloss (maken/ machen line), the northernmost limit of the territory affected by the High German or Second Germanic Consonant Shift, is rather questionable. However important from the point of view of the linguistic system, this isogloss appears to have been quite meaningless even in Old High German times (before 1000) as far as language awareness is concerned (cf. Klein 1990:40). In medieval times, Cologne was one of several places sometimes associated with the niderlant and sometimes (if more rarely) with the oberlant (cf. Schützeichel 1963).
5 On the other hand, the language of present-day Northern Germany was often called nedderlendisch, as well as nedersassesch, sassesch, sassesch diidesch and (incorporating an extra distinction) westfeles unde sassesch; cf. Lasch 1920:8-19 = Lasch 1979:218-231; De Grauwe 1992:193 note 9. 6 Cf. Feudel (1970); von Polenz (2000:166); de Grauwe (1992:195 note 24).
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
101
In fact, 'languages', in the modern sense of autonomous, standardized constructs used for communication and symbolism by well-delimited speech communities, were seldom recognized by medieval speakers. Instead, the contemporary view held that Western Europe was fundamentally divided into Romance and Germanic, as expressed succinctly by the Brabantine poet Jan van Boendale (d. ca. 1352): Want tkerstenheit es gedeelt in tween Die Walsche tongen die es een Dandre die Dietsche al geheel.7 [For Christendom is dealt in twain / the 'Welsh' tongues being one / the other, 'Dietish', all and hale.]
To describe the entirety of the 'Continental West Germanic' peoples, as they are known in historical linguistics, Boendale uses the form with -ie-, which as such was restricted to the counties of Zealand and Flanders and the adjacent part of the Duchy of Brabant (the language of the latter being under strong Flemish influence during the High Middle Ages).8 This -ie- form is a purely phonological variant of the forms with -ü- used elsewhere: diutesch in Southern and Middle Germany, düdesch in Northern Germany, duutsch in the remaining parts of the 'Low Countries' such as Holland, Utrecht, etc. The Boendale quotation illustrates moreover that the -ie- variant could refer not only to the language, population etc. of the area just delimited, but also to the entirety of the 'continental Germanic' territories as today's historical linguistics would call them; no distinction between dietsch "Dutch" and duutsch/ diutesch "German" was ever made. Further evidence (details in De Grauwe forthcoming a) of the anachronistic character of the 'Dutch/ German' distinction is found in Flemish charters from the end of the 13th century, in which the formula "van den dietschen huus van machgheline" denotes the House of the Teutonic Military Order in the city of Mechelen (Malines) — a specifically German order imported from the East into the Netherlands! The self-designation diutesch/düdesch/duutsch (which simply means "the people's language") covered the entire continuum of dialects and regiolects which extended from the North and Baltic Sea coasts to the Alps,9 and continues even today to run right across the borders of the modern states which have Dutch 7
Van den Derden Edewaert ("Of Edward III") II, vv. 1585-87, as cited by Weevers ((1939) 1948:370). 8 Cf. Van Loey (1965:61); Mooijaart (1992:56 note 14, 57-60, 133); Van den Berg & Berteloot (1993:38-39). 9 Recent literature is listed in Haubrichs (1994:22-25, 65-67) and Gardt (2000:40-46, 98-101).
102
LUC DE GRAUWE
and German as their respective official languages. Its traditional academic division into three large groups, traditionally called Middle Dutch, Middle Low German and Middle High German, in no way does justice to the medieval view, which combined a vague, generalized awareness of common linguistic roots (not of complete unity) with a belief in the autonomy of small-scale regiolects (von Polenz 2000:83). Accordingly, varieties elsewhere in the same continuum were simply called ander duutsch "different Germanic" (De Grauwe 1992:195 note 21; Menke 1995:216-217). Even in the later Middle Ages, the emerging Schreiballianzen or "writing alliances" (Besch 1985:1790a; von Polenz 2000:166) still fell far short of the modern standard languages, which took until the Baroque and Enlightenment periods to reach full structural maturity and geographical consolidation, and were not accepted as a model by the lower and rural classes until as late as the 19th or 20th century. All over Europe, standardization based on the selection and deselection (elimination) of variants was a long-winded process running parallel with the unification of nation-states (Van der Wal 1995:5-21). In this respect, the more centralized states such as France and England, with the well-established capital cities of Paris and London, had an obvious advantage over what we might term Continental Germania (what is now Germany, Austria, Switzerland and the Benelux countries), where political fragmentation did little to favour the development of unified languages. Nor indeed did rulers in the Low Countries ever aspire to break away from the Holy Empire during the Middle Ages. On the contrary: their ambition was to be among the mightiest princes within it (Avonds 1982:465). Not until 1384, when the Duke of Burgundy inherited the County of Flanders, did a process start in the northwest of the continent that was to lead gradually to a wider dynastic union with common state institutions. It was this area which the Habsburg rulers Maximilian I, husband of Mary Duchess of Burgundy, and their grandson Charles V, established in 1512 and again in 1548 as the Burgundische(r) Krei(t)s within the Holy Roman Empire (cf. above). Though nominally part of the latter, this 'Burgundian District' was proclaimed by the 'Pragmatic Sanction' of 1549 to be indivisible, subject to its own dynastic succession and exempt from Imperial legislation and jurisdiction. However, no unifying monarchy was founded in the area and the principle of personal union prevailed. Each Habsburg ruler considered himself Count of Flanders, Duke of Brabant etc., the individual provinces insisted on their privileges, and in matters concerning language and its use regiolects continued to flourish, despite early endeavours in the High and Later Middle Ages to develop a more unified language (Willemyns 1997:149, 179). The autonomous standard language today called 'Dutch' (Nederlands) only arose in the wake of
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
103
political autonomy, i.e. after the successful rebellion of the northern provinces, dominated by a Protestant ruling class, against Charles's son and heir, Philip II. (Even at that moment Prince William of Orange-Nassau, known as 'Father of the Fatherland' to the Dutch, opted for independence largely out of disappointment at his German fellow-princes' failure to come to his rescue against the Spanish: cf. Schmidt 1988). Following its proclamation as a Republic in 1581, the United Netherlands seceded officially from the Empire in the Treaty of Westphalia in 1648. This long-term evolution eventually resulted in the development of a language distinct from present-day Hochdeutsch or "High German", whose foundations were in turn laid in Maximilian's and Luther's day by the oberdeutsche (i.e. Austro-Bavarian/ Upper Franconian/ Upper Saxonian) 'writing alliance'. Note, by the way, that the same period also saw the defeat of two additional candidates for the status of autonomous standard language, viz. the Low German of the Hansa, now ousted by Hochdeutsch (Gabrielsson 1983; Menke 1995; Peters 1998:121-124) and a Swiss variety, which — at least in its written form — joined Hochdeutsch as early as the 16th century (Haas 1984:145 = 2000:125; Sonderegger 1993). In Northern Germany, Luther's Bible had been translated into the still prestigious Hansesprache even before it was published in its complete High German edition (1534), but this translation was soon superseded by the latter, thus creating one of the various causes of the decline of Low German after 1530 and its permanent replacement in official life by High German. In 1660 the grammar school teacher Jacob Schöpper from Dortmund (Westphalia) published a list of no less than 6000 words from southern German printed books so that his pupils could learn the "oberlendische Spraach" which he recommended as "ein außbundt und mutter unser Deutschen sprach" ("a model and mother of our German language" (Gabrielsson 1983:143)). This statement implies that there is a single German language consisting of various varieties, of which the oberlendische is the better one. Similarly Switzerland, which detached itself from the Reich as early as 1499, quickly gave up its own Schwyzer Schreibdialekt ("written dialect") marked by monophthongs (Zwingli mîn niiwes Ms vs. Luther mein neues haus). From 1530 onwards, printers used the diphthongs instead; in fact, they had always considered their tütsche und eydtgenössische, helvetische landsprach (i.e. regiolect, in the words of Renward Cysat, town clerk of Luzern) to be a variety of High German (Haas 1984:131 = 2000:114). This is implied, for instance, by the remark added by Adam Petri, printer in Basle, to the glossary in his pirate edition (March 1523) of Luther's New Testament (cited in Haas 1984:141= 2000:122):
104
LUC DE GRAUWE
Lieber Christlicher Leser / So ich gemerckt hab / das nitt yederman verston mag ettlich wörtter im yetzt gründtlichen verteutschten neuwen testament / doch die selbigen wörtter nit on schaden hetten mögen verwandlet werden / hab ich lassen dieselbigen auff unser hoch teutsch außlegen vnd ordenlich in ein klein register / wie du hie sihest / fleißlich verordret. [Dear Christian reader, since I have noticed that not everyone can understand some words in the New Testament, which has been rendered newly and thoroughly into German, while on the other hand the same words could not have been changed without causing damage, therefore I have had these words explained in our High German and properly and diligently indexed in a short list, as you can see.]
It is significant that just after the Thirty Years' War, when Switzerland broke away from the Reich under international law (1648), the official dominance of the regional Swiss printers' standard ceased (Gut-Haas 1998:19). Thus, from the middle of the 17th century onwards only two languages remained, viz. today's German and Dutch, each of which went on to develop a standard of its own and to drift apart from the other. Low German and Swiss German, on the other hand, remained as regiolects (Regionalsprachen) (cf. for the former De Grauwe 2001) or even (groups of) dialects and so were out of the running for the status of future standard language. Especially between 'Dutch' and 'Low German', as these terms are understood now, there was no clear frontier during the Middle Ages (Goossens 1994:82-83). This is well illustrated by the expression tusschen dier Elve entier Zomme ("between the Elbe and Somme rivers"), used by Reynard the Fox in the famous Flemish adaptation of the epic of the same title (13th century):10 the Elbe formed the eastern border of the ancient territory of the Saxons, the Somme had once been the western border of Flanders when this county was at its largest (10th century), and the entire area between the two rivers could be designated as dat nederlant. In 1557, for the first time, a book printed in the city of Emden (Ostfriesland, Germany) claims to have been translated into the westersche nederlantse sprake from the oostersche sprake (De Grauwe 1992:202); the latter term refers to the language known today as Low(er) Saxon or Low German and is probably an ellipsis for oostersche nederlandsche sprake. Together the two terms refer to nothing less than the entire dialect continuum between Dunkirk and Reval. Viewed from coastal franconized Holland, the term Oostersch included the language of those Saxon dialect areas which had come to be part of the Netherlands when Charles V conquered them in the first half of the 16th century. Linguistically these provinces, viz. Groningen, Drent(h)e, Overijs(s)el 10
Ms. A, v. 2444 in De Keyser's 1984 edition.
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
105
and (East-)Geldria and Zutphen, with such important Hanseatic towns as Deventer and Zwolle, were of course closer to Low German properly speaking; they were the cradle of the important religious movement of the 'Devotio Moderna' of Geert Groote and Thomas à Kempis, and their literary language occupied the middle ground between a form prevalent in western Holland and a north German form (Peters 2000:1417b-1418a; cf. also Berns 2001). It is interesting to note that the terms in question here were normally used without the definite article during the Middle Ages: in Dietsche spreken, in dietsche(r)/duutscher/ duytscher tale(n)/ spraken (all said with reference to the Netherlands), in diutscher diute, (lit.) "in Germanic interpretation" (with reference to Germany). Modern expressions like Nederlands spreken/ Deutsch (deutsch) sprechen/ to speak English, auf deutsch, in deutscher Sprache continue this usage, which is also found in Luther, cf. Biblia, das ist die gantze Heilige Schrijft Deutsch. Like latine loqui, these expressions actually mean "in a (Latin, Germanic etc.) way" (Coseriu 1988:254-255; De Grauwe 1992:204205; De Grauwe forthcoming a, esp. note 7). They do not yet refer to a particular, well-established, more or less standardized language, as in 'the English/ German/ Dutch language.' In many cases, the uncertainty about the geographic extent of one's own language was expressed by the phrase in onser talen "in our language": already Notker von Sankt Gallen (10 th century) uses the expression in unsera uuîs, "in our way (of speaking)", variants of which can be found throughout the whole Middle Ages, the Early Modern period and even nowadays, e.g. in (Slavonic) Macedonia and in the Swedish Tornedal, where a Finnish minority calls its local variety meänkieli "our language" (De Grauwe forthcoming b, notes 50-52). As we saw earlier, developments which were political in the first instance, encouraged, if not caused, the parallel standardization and separation of two distinct Kultursprachen across Continental Germania from the Early Modern period onwards. The internal process of unification had its starting point in the respective prestige varieties. In Germany, Austria and Switzerland, this was the so-called Meißnisch, i.e. the eastern Kanzleisprache or "chancellery language" mentioned earlier as the upper German writing alliance, which Martin Luther's writings did much to promote. In the Netherlands, it was the speech of the regenten, the leading members of the merchant class in the main towns of Holland such as Amsterdam, the Hague, Leiden and Haarlem, whose ranks included numerous upper-class religious refugees from the southern provinces of Flanders and Brabant. As these provinces remained under Spanish-Habsburg rule and were therefore Roman Catholic, Protestant refugees exported their language to the north and integrated many of its forms into the new national language then coming into being (Van Leuvensteijn
106
LUC DE GRAUWE
1997). A landmark in the autonomous standardization of modern Dutch was the famous Statenvertaling, the Bible translation ordered by the (northern) provincial authorities, first published in 1637 (cf. now Berns 2001). In both Germany and the Netherlands, these 'high' varieties came to serve as points of reference for the speakers of the many coexisting duutsch/ dietsch dialects and regiolects, leading to the emergence and survival of only two standard languages, each 'covering' several 'dependent' sub-varieties. It is very important to realize, however, that the initiative was not taken primarily by the latter (such that they might have started to develop towards a 'standard' form), but rather by (users of the) the leading varieties who were eager to mark out their own area of common usage (De Grauwe 1992:203). Thus, Dutch now also covered those Low Saxonian dialects still spoken in the north-eastern part of the 'Kingdom of the Netherlands' along the German border (cf. above), and High German came to cover the adjacent plattdüütsch dialects of the northern Hanseatic area that had formerly been covered by the Low German (semi-)standard. These considerations show that, contrary to the opinion still set out in most textbooks, the beginnings of standard Dutch and German should be described as a 'bifurcation' (rather than a 'separation' of Dutch from German), and that the joint study of the two languages prior to that bifurcation should best be treated as a single sub-discipline for which I propose the name 'theodistics' {Theodistik). The term 'separation' is much more apt for the evolution of Lëtzebuergesch from the Middle Franconian dialects of the Grand Duchy (and within the limits of that country only) into a distinct standard language after the Second World War. After considerable efforts of language planning, Lëtzebuergesch is now treated on a par with French and German in the administration and the schools, has its own standard and orthography based on the central dialects spoken around the capital, and in 1984 was actually promoted by law to langue nationale des Luxembourgeois (Ammon 1995:398404; Gilles 1998: esp. 22 note 6; Newton 2000; Newton in this vol.). Dutch, by contrast, never had to 'separate' from German in this sense, and the frequent comparison with Switzerland, which also happened to split from the Empire in 1648, is misguided because in Switzerland, High German continues to fulfil functions (covering the closely related Alemannic dialects summarily called Schwyzertüütsch) which it never had in either the Netherlands or Northern Belgium. The misconception was fostered by the fact that High German has always claimed the larger territory and has moreover (perhaps as a result) been known by the oldest, most general and, in terms of word formation, simplest of names: Deutsch. The fact that there is evidence of the composite term Nederlands Duyts in the Low Countries until at least 1688, proves that an
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
107
awareness of the joint linguistic heritage continued to survive there. In fact, in the northwest too the most current term until about 1550 was Duitsch, then Nederlandsch Duitsch or simply Nederduitsch; it took until the early 19th century for the elliptical term Nederlands to prevail (De Grauwe 1992:194195, with references in notes 6, 7, 16), whereas Germany stuck to the simplex Deutsch. Yet things might as well have gone the opposite way: Duits could have survived in the Netherlands (cf. the English name of the language, Dutch), and what we call 'German' might conceivably have been known and taught under the name of Oberländisch today. It may be interesting to note that both 'Netherdutch' and 'Low Dutch', which are found in bilingual textbooks in the period 1580-1740, were mainly used by native speakers of English living and working in the Dutch Republic, as Loonen (1993) points out. In the modern Netherlands there is one remnant of duytsch (Duits) in the original broad sense of 'Germanic', viz. in the national hymn, the 'Wilhelmus', which dates from the 16th century and opens with the verse "Wilhelmus van Nassouwe, ben ick van duytschen bloed" (cf. Grijp 1998). Here the prince simply states that his roots are Germanic rather than Romance — in spite of his being Prince of Orange in France as well. Incidentally, the Prince's attitude to the Holy Empire was less negative initially than in later years; like his first sovereign, Charles V, he did not originally intend to separate the Burgundian Kreis from it. As noted earlier, his opting for independence for the United Provinces from the German-Habsburg supremacy was mainly driven by disappointment with his fellow German princes (Schmidt 1988; Leerssen 1999:29). Besides the fact that the terms duytsch, nederduytsch and nederlandsch duytsch were long maintained in the Netherlands, many explicit statements by 16th-century linguists also testify to a continuing awareness of special linguistic ties with the (south)east. One such statement was made by the mathematician Simon Stevin (1548-1620), born in Bruges, whose book Uytspraeck vande Weerdicheit der Duytsche Tael ("Eulogy on the dignity of the Dutch language") was published in Leiden in 1586. In this and other writings Stevin praises his own Duytsche Tael above Romance languages such as French, Italian and Spanish, yet despite taking a purist stance and rejecting any loan words from the latter, he makes no corresponding references to the German language of his day. This has rightly been regarded as evidence that in Stevin's eyes High German and Dutch, if not actually identical, were at least varieties of one and the same language (Damsteegt 1982:32; Van der Wal 1995:24, 29). This also emerges clearly from a treatise probably written by Hendrik Laurensz Spiegel (Peeters 1989) and published two years earlier under
108
LUC DE GRAUWE
the title Twe-spraack van de Nederduitsche Letterkunst ("Dialogue of the Dutch grammar") on behalf of the Amsterdam Rederijkerskamer ("Chamber of Rhetoric") In Liefd' bloeyende ("Flourishing in Charity"). This famous work contains the following dialogue: R. ghy zeyde flux dat de Duytse taal by haar zelven bestaat/ ick heb my wel laten segghen dat onze spraack uyt het Hóóghduyts zou ghesproten zyn. G. Ick spreeck [...] int ghemeen vande duytse taal/ die zelve voor een taal houdende/ dóch dat de zommighe wat te hóógh/ andere wat te laegh spreken/ [...] /wel iet wat inde uytspraack verschelende/ maar zó niet óf elck verstaat ander zeer wel/ tis kenlyck dat de Griexe taal/ die zó waard gheacht is/ óóck haar verscheyden Dialectos had. (Caron 1962:62-63). [R. You've just said that the Dutch language exists by its own right; I've heard somebody say that our language is supposed to be an offspring of High German. G. I'm talking [...] about the Germanic language in general, considering it as one; however, some speak rather high, others rather low [...], differing from each other somewhat in pronunciation, though not all that much, so that each understands the other very well. It is obvious that the Greek language, which is so highly esteemed, had different dialects of its own, too.]
On the German side, the purist Philipp von Zesen, who lived in the Netherlands for 27 years, praised Spiegel for his "eifer zur reinligkeit der Deutschen Muttersprache", his "zeal for the purity of the Germanic mother tongue" (Peeters 1989:48). Von Zesen and Spiegel are obviously referring to the same language, if different varieties of it. Yet with von Zesen we have already passed the crucial date of 1648, after which the standardization of Dutch, now felt more and more to be a separate language, was to gain momentum, as did that of German on the other side of the frontier. The correct chronology thus leads from political autonomy or independence to standardization, and only then to the perception of a separate 'native' language in its own unified territory (which in turn may be recognized and taught as a 'foreign' language in other, similarly unified territories). Given that from a Eurocentric perspective, any nation-state needs a national language as one of its major symbols, the subsequent political consolidation of Germany, Austria and Switzerland on the one hand and of the Netherlands and Belgium on the other helped widen the linguistic gap and establish two present-day sister languages, closely related yet recognized as both clearly distinct and completely autonomous. As for the present, one might wonder whether the spread of the modern German and Dutch languages over more than one state (Germany, Austria, Switzerland on the one hand, the Netherlands and Belgium on the other) might not favour the adoption of different names for the respective languages in each
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
109
country. Neither in Switzerland nor in Austria is there any widespread support for re-naming the varieties of German spoken there, let alone turning them into separate languages. At best, these varieties could be described as different 'standard or national varieties' of a single German language (Ammon 1995, 1996 and 2000; Hogan-Brun 2000). As Sonderegger (1985:1931b) points out, Swiss speakers of German have only one mother tongue, das Deutsche, though this consists of two varieties, viz. a local Schwyzertüütsch dialect and Schweizerhochdeutsch. In Austria there is an österreichisches Deutsch (which is by no means uniform!), but sociolinguistically speaking this does not clash with the existing awareness of autonomy or "Eigenständigkeit" (Wiesinger 1997:50 with note 12; cf. Wolf 1994; von Polenz 1999:435-437). Supporters of a totally 'independent' Austrian language {Österreichisch) are a minority (Hessmann 1995; Wiesinger 2000 with references). In the southern part of the area where Dutch is an official language, viz. Northern Belgium or Flanders," some — perhaps slightly more than in Austria — would similarly favour a Flemish (< Flanders) language. Unlike the Austrians and the Swiss, the Flemish elite underwent large-scale Frenchification, beginning with the separation of the Spanish and later Austrian Netherlands from the Northern Provinces (the present-day Kingdom of the Netherlands) from 1585 and 1648 onwards. Common people, by contrast, stuck to their local Germanic dialects until, in a lengthy process following the independence of Belgium in 1830, the Flemish Movement regained the lost ground (Willemyns 1992; Geerts 1997).12 This process culminated in the federalization of the Belgian state according to the territorial principle (cuius regio, illius et lingua) in 1993, with the capital Brussels (though largely Frenchified) becoming officially bilingual. The long-term isolation and domestic comeback of Dutch in Belgium brought about simultaneous processes of convergence and divergence. The official standard variety Belgisch Nederlands (Belgian Dutch) does not differ fundamentally from Nederlands Nederlands (as the variety used in the Netherlands is sometimes called by linguists), but is seldom actually spoken except in the media and in highly formal situations. In informal contexts a rather wide range of regiolects or local dialects is used instead; in writing, special lexemes are sometimes used; the syntax is almost the same as in the Netherlands, spelling and morphology 11
Flanders is in fact a pars pro toto designation for the whole northern part of the Kingdom, named after the ancient county between the North Sea coast and the rivers Scheldt and Dender, but now including also the greater part of the old Duchy of Brabant (with cities like Brussels, Antwerp, Louvain) and the present-day Belgian province of Limburg. 12 Similarly, Swedish was the predominant language of the Finnish elite from the Middle Ages until as late as the 20th century.
110
LUC DE GRAUWE
totally so — a pattern very similar to Austria and Switzerland vis-à-vis Germany (cf. De Grauwe 2001:306-307 with references). An important difference, however, is that from the High and Late Middle Ages onwards people from the entire 'Low Countries' — whose boundaries were still imprecise — were referred to as Flemings all over Europe (De Smet 1973:318, 325-326). The reason was that the county of Flanders, situated in the southwestern corner of the Low Countries, was one of the most advanced, prosperous and powerful territories of the whole of Western Europe, and cities such as Ghent, Bruges and Ypres competed in wealth and power with their Northern Italian rivals (Venice, Milan, Genoa) or even Paris and Cologne. Not surprisingly, the old term Flemings remained for centuries the usual designation for people from the northern part of what is now Belgium (in 16thcentury Italy the term fiamminghi included even the French-speaking people of the southern part, nowadays known as the Walloons; De Smet 1973:325), and for their language, too, the name Flemish survived without problems. However, today's informal 'Flemish' is, as already mentioned, far from uniform and scarcely standardized. Its lack of 'systemic distance' from (Belgian) Dutch, its exclusion from higher registers and formal contexts, its lack — in spite of the glorious past of the county of Flanders — of a Sprachmythos comparable to that of Low German as Hansesprache in Northern Germany — all these factors make it impossible to characterize it as an autonomous language (De Grauwe 2001:306-308). Our final conclusion must therefore be that there are only two Germanic standard languages (not counting Frisian) in present-day continental Europe, i.e. German and Dutch — each with several national varieties.
References Ammon, Ulrich. 1995. Die deutsche Sprache in Deutschland, Österreich und der Schweiz. Das Problem der nationalen Varietäten. Berlin & New York: Walter de Gruyter. Ammon, Ulrich. 1996. Typologie der nationalen Varianten des Deutschen zum Zweck systematischer und erklärungsbezogener Beschreibung nationaler Varietäten. Zeitschrift für Dialektologie und Linguistik 63. 157-175. Ammon, Ulrich. 2000. Sprache — Nation und die Plurinationalität des Deutschen. Gardt 2000. 509-524. Avonds, Piet. 1982. Brabant en Limburg 1100-1403. Algemene Geschiedenis der Nederlanden, deel 2, 452-482. Haarlem: Fibula-Van Dishoeck.
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
111
van den Berg, Evert & Berteloot Amand. 1993. Waar kwam Jacob van Maerlant vandaan? Verslagen en Mededelingen van de Koninklijke Academie voor Nederlandse taal- en letterkunde, 30-77. Berns, Jan B. 2001. Die niederländische Bibelübersetzung von 1637 und der Mythos der regionalen Variation. Peters, Pütz & Weber 2001. 31-40. Besch, Werner. 1985. Die Entstehung und Ausformung der neuhochdeutschen Schriftsprache/ Standardsprache. Sprachgeschichte. Ein Handbuch zur Geschichte der deutschen Sprache und ihrer Erforschung ed. by Werner Besch, Oskar Reichmann & Stefan Sonderegger. Zweiter Halbband, 17811810. Berlin & New York: Walter de Gruyter. Caron, W.J.H. 1962. Hendrik Laurensz. Spiegel, Twe-spraack, Kort Begrip, Ruygh-bewerp, Rederijck-kunst (= Trivium, 3). Groningen: J.B. Wolters. Damsteegt, Boudewijn Cornelis. 1982. Simon Stevin: taalspiegeling en taaldaad. Tussen intuïtie en weten, zes grote denkers op het raakvlak tussen exacte en geesteswetenschappen ed. by Max Wildiers et al., 27-46. Muiderberg: Coutinho. Feudel, Günter. 1970. Luthers Ausspruch über seine Sprache (WA Tischreden 1,524). Ideal oder Wirklichkeit? Beiträge zur Geschichte der deutschen Sprache und Literatur 92. 61-75. Gabrielsson, Artur. 1983. Die Verdrängung der mittelniederdeutschen durch die neuhochdeutsche Schriftsprache. Handbuch zur niederdeutschen Sprach- und Literaturwissenschaft ed. by Gerhard Cordes & Dieter Möhn, 119-153. Berlin: Erich Schmidt. Gardt, Andreas, ed. 2000. Nation und Sprache. Die Diskussion ihres Verhältnisses in Geschichte und Gegenwart. Berlin & New York: Walter de Gruyter. Geerts, Guido. 1997. Nederlands in België. Van den Toorn, Pijnenburg, Van Leuvensteijn & Van der Horst 1997. 563-595. Gilles, Peter. 1998. Die Emanzipation des Lëtzebuergeschen aus dem Gefüge der deutschen Mundarten. Zeitschrift für deutsche Philologie 117. Sonderheft. 20-35. Goossens, Jan. 1994. Normierung in spätmittelalterlichen Schreibsprachen. Niederdeutsches Wort 34. 77-99. de Grauwe, Luc. 1992. Das historische Vcrhältnis Deutsch-Niederländisch 'revisited'. Zur Nicht-Existenz von Einheitsarealen im Sprachbewußtsein des Mittelalters und der beginnenden Neuzeit. Amsterdamer Beiträge zur älteren Germanistik 35. 191-205. de Grauwe, Luc. 2001. Eigenständigkeit einst und jetzt. Zu einigen Parallelen in Geschichte und Gegenwart des 'Flämischen' und des Niederdeutschen. Peters, Pütz & Weber 2001. 299-313.
112
LUC DE GRAUWE
de Grauwe, Luc. Forthcoming a. 'Oem, walschedi? spreect jeghen mi in dietsche' (Reynaert I, 1458-60). Zu den Eigenbezeichnungen der Volkssprachen in den Niederlanden im hohen Mittelalter. The Dawn of the Written Vernacular in Western Europe ed. by Michèle Goyens & Werner Verbeke. (= Mediaevalia Lovaniensia. Series I. Studia, 22). Leuven: Leuven University Press. de Grauwe, Luc. Forthcoming b. Theodistik. Zur Begründung eines Faches und ein Plädoyer für eine kontinentalwestgermanische Sicht auf die neuzeitliche Bifurkation Deutsch/ Niederländisch. Die deutsche Schrift sprache und die Regionen. Entstehungsgeschichtliche Fragen in neuer Sicht ed. by Raphael Berthele, Helen Christen, Sibylle Germann & Ingrid Hove. (= Studia Linguistica Germanica). Berlin & New York: Walter de Gruyter. Grijp, Louis, ed. 1998. Nationale hymnen: het Wilhelmus en zijn buren. Nijmegen & Amsterdam: Sun. Gut, Judith & Walter Haas. 1998. Zur regionalen Sprachgeschichte der Eidgenossenschaft. Zwei Fallbeispiele. Zeitschrift für deutsche Philologie 117. Sonderheft. 4-19. Haas, Walter. 1984. Die deutschsprachige Schweiz. Die viersprachige Schweiz ed. by Jachen C. Arquint et al., 71-160. Zürich: Buchclub Ex Libris. (1982: Cologne: Benzinger.). Haas, Walter. 2000. Die deutschsprachige Schweiz. Die viersprachige Schweiz. Zweite, neu bearbeitete Auflage ed. by Hans Bickel & Robert Schläpfer, 57-138. Aarau-Frankfurt am Main & Salzburg: Verlag Sauerländer. Haubrichs, Wolfgang, ed. 1994. Deutsch — Wort und Begriff (= Lili, Zeit schrift für Literaturwissenschaft und Linguistik 24, Heft 94). Hessmann, Pierre. 1995. Österreichisches Deutsch oder Standardsprache Österreichisch? Germanistische Mitteilungen 41. 19-23. Hogan-Brun, Gabrielle, ed. 2000. National Varieties of German Outside Germany. Oxford, Bern, Berlin, Brussels, Frankfurt am Main, New York, Wien: Peter Lang. de Keyser, Paul. 1984. Van den vos Reynaerde, uitgegeven, ingeleid en verklaard. 10e druk. Antwerpen: De Nederlandsche Boekhandel. Klein, Thomas. 1990. Zur Nordgrenze des Althochdeutschen und zu germ. ß,Y, im Altmittelfränkischen. Beiträge zur Geschichte der deutschen Sprache und Literatur (Tübingen) 112. 26-54. Lasch, Agathe. 1920. Sassesche Sprake. Über die älteren Bezeichnungen für die niederdeutsche Sprache. Zeitschrift für Deutschkunde 34. 8-19. Lasch, Agathe. 1979. Ausgewählte Schriften zur niederdeutschen Philologie ed. by Robert Peters & Timothy Sodmann. Neumünster: Karl Wachholtz Verlag.
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
113
Leerssen, Jo. 1999. Nationaal denken in Europa. Een cultuurhistorische schets. Amsterdam: Amsterdam University Press. van Leuvensteijn, J.A. 1997. Vroegnieuwnederlands (circa 1550-1650). Externe taalgeschiedenis. Van den Toorn, Pijnenburg, Van Leuvensteijn & Van der Horst 1997. 227-272. van Loey, A. 1965. Middelnederlandse Spraakkunst. II. Klankleer. Vierde herziene druk. Groningen: J.B. Wolters & Antwerpen: De Sikkel. Loonen, P.L.M. 1993. A Revaluation of the Term 'Low Dutch'. Leuvense Bijdragen 82. 49-60. Menke, Hubertus. 1995. Zwischen sprachlichem Selbstbewußtsein und Inferioritätsgefühl. Die Hansesprache in der Eigen- und Fremdeinschätzung. Slawen und Deutsche im südlichen Ostseeraum vom 11. bis zum 16. Jahrhundert ed. by Michael Müller-Wille et al., 211-222. Neumünster: Karl Wachholtz. Menke, Hubertus. 2001. 'Een' Spraak is man bloots een Dialekt, de sik to Wehr setten kann.' Nachlese zur Diskussion um die Europäische Sprachenschutzcharta. Niederdeutsch. Sprache und Literatur der Region ed. by Ursula Föllner (= Literatur—Sprache—Region, 5). Frankfurt am Main, Berlin, Bern, Brussels, New York, Oxford, Wien: Peter Lang. Europäischer Verlag der Wissenschaften. Mooijaart, Marijke. 1992. Atlas van Vroegmiddelnederlandse taalvarianten. Diss. Leiden. Newton, Gerald. 2000. The Use of German in the Grand Duchy of Luxembourg. Hogan-Brun 2000. 139-160. OED = The Oxford English Dictionary, 1st edition. Peeters, L. 1989. Auteurschap en tekst van 'Spiegels' Twe-Spraack (1584). L. Peeters. Taalopbouw als Renaissance-ideaal ed. by G.R.W. Dibbets, J. Noordegraaf & M.J. van der Wal, 40-50. Amsterdam: Buijten & Schipperheijn. Peters, Robert. 1998. Zur Sprachgeschichte des Niederdeutschen. Zeitschrift für deutsche Philologie 117. Sonderheft. 108-127. Peters, Robert. 2000. Soziokulturelle Voraussetzungen und Sprachraum des Mittelniederdeutschen. Sprachgeschichte. Ein Handbuch zur Geschichte der deutschen Sprache und ihrer Erforschung. 2. Aufl. ed. by Werner Besch, Anne Betten, Oskar Reichmann & Stefan Sonderegger, 2. Teilband, 1409-1422. Berlin & New York: Walter de Gruyter. Peters, Robert, Horst P. Pütz & Ulrich Weber, eds. 2001. Vulpis Adolatio. Festschrift für Hubertus Menke zum 60. Geburtstag. Heidelberg: C. Winter. Pfeiffer, Franz, ed. 1965. Berthold von Regensburg. Vollständige Ausgabe seiner Predigten. 1. Bd., Neudruck. Berlin: Walter de Gruyter.
114
LUC DE GRAUWE
Pijnenburg, Willy J.J. 1997. Vroegmiddelnederlands (circa 1200-1350). Van den Toorn, Pijnenburg, Van Leuvensteijn & Van der Horst 1997. 69-145. Polenz, Peter von. 1999. Deutsche Sprachgeschichte vom Spätmittelalter bis zur Gegenwart. Band III: 19. und 20. Jahrhundert. Berlin & New York: Walter de Gruyter. Polenz, Peter von. 2000. Deutsche Sprachgeschichte vom Spätmittelalter bis zur Gegenwart. Band I: Einführung, Grundbegriffe, 14. bis 16. Jahrhundert. 2. überarbeitete und ergänzte Auflage. Berlin & New York: Walter de Gruyter. Schmidt, Georg. 1988. Des Prinzen Vaterland? Wilhelm von Oranien (1533— 1584) zwischen Reich, deutscher Nation und den Niederlanden. Deutschland in der Neuzeit. Festschrift für Karl Otmar Freiherr von Aretin, 1. Halbband ed. by R. Merville et al., 223-239. Stuttgart: Franz Steiner. Schützeichel, Rudolf. 1963. Köln und das Niderland. Zur geographischhistorischen Stellung Kölns im Mittelalter. Groningen: J.B. Wolters. de Smet, Gilbert. 1973. Die Bezeichnungen der niederländischen Sprache im Laufe ihrer Geschichte. Rheinische Vierteljährsblätter 37. 315-327. Sonderegger, Stefan. 1985. Die Entwicklung des Verhältnisses von Standardsprache und Mundarten in der deutschen Schweiz. Sprachgeschichte. Ein Handbuch zur Geschichte der deutschen Sprache und ihrer Erforschung ed. by Werner Besch, Oskar Reichmann & Stefan Sonderegger, Zweiter Halbband, 1873-1939. Berlin & New York: Walter de Gruyter. Sonderegger, Stefan. 1993. Frühneuhochdeutsch in der Schweiz. Versuch einer Standortbestimmung. Vielfalt des Deutschen ed. by Klaus J. Mattheier, Klaus-Peter Wegera, Walter Hoffmann, Jürgen Macha & Hans-Joachim Solms, 11-36. Frankfurt am Main: Peter Lang. van den Toorn, M.C., W.J.J. Pijnenburg, J.A. van Leuvensteijn & J.M. van der Horst, eds. 1997. Geschiedenis van de Nederlandse taal. Amsterdam: Amsterdam University Press. de Vooys, C.G.N. 1950. Nederlantsche tale. De Nieuwe Taalgids 43. 120. van der Wal, Marijke. 1995. De moedertaal centraal. Standaardisatie-aspecten in de Nederlanden omstreeks 1650. Den Haag: Sdu Uitgevers. Weevers, Theodoor. (1939) 1948. The Use of Dietsch and Duutsch in Middle Dutch Historical Works between 1280 and 1450. London Mediœval Studies 1:3.354-381. Wiesinger, Peter. 1997. Zu Ulrich Ammons Buch 1995. Zeitschrift für germanistische Linguistik 120. 45-52. Wiesinger, Peter. 2000. Nation und Sprache in Österreich. Gardt 2000. 525562.
EMERGING MOTHER-TONGUE AWARENESS IN DUTCH AND GERMAN
115
van Wijk, Nicolaas. 1910. Hochdeutsch, Niederdeutsch. Zeitschrift für deutsche Wortforschung 12. 239-240. Willemyns, Roland. 1992. Taalontwikkeling in de Zuidelijke Nederlanden na de politieke scheiding. Verslagen en Mededelingen van de Koninklijke Academie voor Nederlandse Taal- en Letterkunde. 99-115. Willemyns, Roland. 1997. Laatmiddelnederlands (circa 1350-1550). Van den Toorn, Pijnenburg, Van Leuvensteijn & Van der Horst 1997. 147-219. Wolf, Norbert Richard. 1994. Österreichisches zum Österreichischen Wörterbuch. Zeitschrift für Dialektologie und Linguistik 61. 66-76.
TWO HUNDRED YEARS OF LANGUAGE PLANNING IN BELGIUM
JETJE DE GROOF F. W. O. - Vlaanderen Vrije Universiteit Brussel, Belgium
1.
Introduction A clear understanding of the structure and nature of language planning activities in Belgium over the past two hundred years is vital to the understanding of the present linguistic situation in that country. The successive regimes that have ruled the territory now known as Belgium since 1795 had a number of different ideas concerning corpus and status planning. Alongside these official language policies, a number of 'non-official' associations, action committees and even individuals have contributed to the language planning debate. Although Belgian language policy and planning have raised much academic interest from historians and politicians, these issues have thus far seldom been treated from a purely (socio)linguistic perspective. In this paper, the peculiar process of language planning and policy in Belgium and its effects upon standardization will be described against a background of different language planning models. The Belgian situation will serve as a case study for the evaluation of these models. 2.
Theoretical background
2.1 Language planning and language policy: delineations A first important question for the researcher of language planning and policy is just what is understood by the term 'language planning'. I see language planning as every conscious effort to intervene in the evolution of a language. Still, this delineation of 'conscious intervention' is not always easy to make. It is sometimes difficult to tell whether anyone has attempted to influence the course of a language's development, or whether societal factors (e.g. the shift of the economic centre of gravity) brought about language
118
JETJE DE GROOF
change without deliberate intervention. And even if someone did intervene, it is difficult to tell whether it was a conscious interference. Did the intervening person, organization, administration or government act as part of a plan, or was the intercession an ad hoc solution to a problem? The meaning of the term language planning has changed throughout the years. This is not surprising considering it was first coined by Haugen at the end of the 1950s and that the world has changed considerably since. One of Haugen's early definitions of language planning was the following: As I define it, the term language planning includes the normative work of language academies and committees, all forms of what is commonly known as language cultivation [...], and all proposals for language reform or standardization. (Haugen 1969:701)
The definition was limited to normative work. Soon after, the meaning of the term was broadened: Kloss (1969) made a distinction between status planning, which would come to mean the functional allocation of languages or language varieties, and corpus planning, which included actions concerning the language itself (its structure). In addition to this, there have been other shifts in the meaning of the term. Macro-sociopolitical changes have affected the way language planning is viewed. The collapse of the Soviet Union, the end of apartheid in South Africa, and the expansion of the European Union have confronted us with new linguistic realities. Moreover, epistemological paradigms have changed: today's postmodern research models are different from former structuralist models (see also Ricento 2000). First of all, there has been a broadening in the sense of how 'planned' the planning has to be in order to be called language planning. Research used to focus on initiatives starting from an explicit plan, which in its ideal form entailed different steps: in a first phase all the data from a certain context are collected, out of which a structured plan is drawn up in a second phase. In a third phase the plan is implemented, to be evaluated in a fourth phase (Rubin 1971). But in most cases, language planning does not start from a detailed scheme. In many instances of language policy and planning, people start with the implementation phase, without any previous research into the sociolinguistic situation of the country. Therefore, a consensus has emerged in the literature that language planning entails all kinds of conscious intervention, whether more or less traditionally planned. Secondly, the term language planning is no longer limited to activities of authorities and academies, as in Haugen's early definition. Language planning
LANGUAGE PLANNING IN BELGIUM
119
can be and indeed is the work not just of authorities, but also of groups, organizations and even individuals. The Belgian case clearly supports this point of view. A definition reflecting these changes is one that does not exclude any possible acts of language planning or policy. The following definition (Cooper 1989:45) incorporates this: "Language planning refers to deliberate efforts to influence the behaviour of others with respect to the acquisition, structure, or functional allocation of their language codes". These last three are the fields of application of language planning: acquisition planning (influence on acquisition), corpus planning (influence on structure) and status planning (influence on functional allocation). Thirdly, although language planning can be initiated by all kinds of actors, it tends to have the greatest effect when issued by those in power. Since the 1980s and 1990s, language planning research has increasingly come to criticize the uncritical way in which earlier scholars carried out their research. Social stratification, the social and political consequences of language contact, and the importance of ideology have drawn increasing attention in language planning research. For Tollefson (1991:16), "language planning-policy means the institutionalization of language as a basis for distinctions among social groups (classes)". 2.2 Models of language planning For the description of a specific language planning situation, like the Belgian one, models of language planning can be an important guide. I will now describe different language planning paradigms that are, in my opinion, of use for the description of the Belgian situation. They are dealt with chronologically and also reflect the changing contents of the concept 'language planning'. 1. Haugen's (1983) scheme, describing the different phases of a language planning process, is without any doubt the best known (see Table 1). His model allows a clearer view of the description of the language planning process. In a first phase a language is chosen, and from this language the variety which is to be codified. In some cases, this step is self-evident, as there are no languages competing for the status of 'norm'. In other cases, this choice can be difficult and hazardous. This is the case when competing groups in society each want their language to be selected. After this selection procedure has taken place, the stage of standardization follows. This process consists of three aspects, which do not necessarily have a set sequence: a graphization phase, during which oral languages are turned into literate ones or already existing
JETJE DE GROOF
120
scripts are reformed; a grammatication phase, involving "the extraction and formulation of rules that describe how a language is structured" (Baldauf & Kaplan 1997:41), and a lexication phase, referring to the selection and development of an appropriate lexicon. After the standardization stage, a phase of implementation follows, with a focus on adoption and spread of the selected and codified norm, mainly relying on the educational system and language legislation. In a last phase, the language is further elaborated: terminology is modernized, and the stylistic properties of the language are adapted to new needs (see also Baldauf & Kaplan 1997).
Society (status planning)
Language (corpus planning)
Form (policy planning) 1. Selection (decision procedures) a. Identification of problem b. Allocation of norms 2. Codification (standardization procedures) a. Graphization b. Grammatication c. Lexication
Function (language cultivation) 3. Implementation (educational spread) a. Correction procedures b. Evaluation 4. Elaboration a. Terminological modernization b. Stylistic development
Table 1: The phases of language planning according to Haugen (1983) This language planning process model offers a logical succession of steps, starting from a dialectal, multilingual situation, without any language regulation, and ending in a situation in which a standardized language is the norm for a language community, and in which this norm can be used for all situations of public life. Obviously, this ideal scenario is not often followed in real-life situations. Language planning measures are not taken in a vacuum. They are elaborated in a society and are thus deeply influenced by cultural, political and religious conflicts. The non-inclusion of the latter elements in most language planning models has been, as mentioned above, an important point of criticism with regard to language planning theory (see also Calvet 1996; Tollefson 1991). Haugen's model, although a target of regular criticism, is an important point of reference for any description of language planning, as it embodies the ideal language planning process, from which real situations can differ in various degrees. The whole can be seen as an ongoing process, as language is changing constantly, and standardization never really reaches an end point. 2. Another guiding paradigm is that of Cooper (1989:89). His well-known question, "What actors attempt to influence what behaviours of which people
LANGUAGE PLANNING IN BELGIUM
121
for what ends under what conditions by what means through what decisionmaking process with what effect?", gives an account of what issues are vital to the understanding of a language planning process. The importance of the societal context in which language planning occurs is incorporated into the model. Moreover, Cooper asks the question: "for what ends?" This search for the motivations underlying language planning and policy is fundamental (see also Ager 2001). Language planning is very often used to attain extralinguistic goals. A description of language planning thus cannot exclude the search for motivation. 3.
'Belgium ' before 1830
3.1. 'Belgium' under Spanish and Austrian rule: laissez faire In the Middle Dutch period, the Dutch language area was dialectally diverse. As Flanders and later Brabant became prominent trade centres, the dialects spoken in these areas were considered prestigious. In the 16th century, the Netherlands revolted against the Catholic Spanish rulers. While the North of the Netherlands managed to gain independence and become a Protestant state, with Dutch as its national language, the Southern part remained under Spanish rule. Dutch became the language of the blooming culture of the North, where Amsterdam grew to be the prominent trade centre. Although some writers complained about the fact that the language used in the North was a bookish language, Dutch was used in all public spheres and underwent a spontaneous development and standardization on the basis of the Hollandic dialect. The Dutch used in the South, now cut off from the language area's prestige region, remained in its dialectal situation (Van de Craen & Willemyns 1988). Rather than a linguistically related standard language, French was used in situations where a standard language was required (High functions), whereas the different dialects were used only in intimate and 'Low' domains. The South was ruled first by Spain, then by Austria, both clearly favouring French over Dutch. Two elements are important in this situation. Firstly, up to the end of the Austrian period, no language was imposed on people. Although the bon ton language of that time was French and although higher administrations were expected to use French in their correspondence with the central administration, the lower levels of administration were allowed to do whatever they wanted and continued using Dutch — meaning the local dialect, as the Dutch used in the Southern Netherlands was far from being standardized. A second important element is that the higher classes were more than eager to use French, because it was the fashionable language of that time (Deneckere 1954).
122
JETJE DE GROOF
3.2 Incorporation by the French and the introduction of a language policy The policy of laissez faire came to an end when the French incorporated the Southern Netherlands in 1795. Their one nation, one language policy entailed a marked rupture with the past. Already in 1790, Abbé Grégoire had distributed a questionnaire in order to obtain an overview of the linguistic situation of the Republic. In 1794, the report on this questionnaire revealed that six out of twenty million French citizens had no knowledge whatsoever of the language of the Republic. This linguistic ignorance was equated with the counter-revolution: "Federalism and superstition speak Lower Breton, emigration and hatred speak German, the counter-revolution speaks Italian, and fanaticism speaks Basque" (Barère de Vieuzac as cited in de Certeau et al. 1979:295).' The republican ideology was established in language legislation. In July 1794, a language law was passed stating that French was to be used in courts and administration at all levels of public life. In September, the enactment of the law was postponed, as its practical implementation was impossible. Between September 1794 and June 1803, no real language legislation was operative, and the initiative was left to local decision makers. These local functionaries, appointed by the central government, were either French citizens or local supporters of the revolution. In both cases, French was favoured over Dutch. From 1803 onwards, the tide had turned to the extent that the language legislation, which it had been impossible to enact in July 1794, could now be passed. The language law of 1803 stipulated that from 1804 onwards, all official documents were to be written in French. The Napoleonic period also meant the beginning of a directed language policy against press, literature and theatre in Dutch. 3.3 The United Kingdom of the Netherlands (1814-1830) After the defeat of Napoleon, it was decided at the Vienna congress that the Netherlands were to be reunited under William I, the Dutch monarch. The latter tried to turn the tide and, by promulgating several language laws, reimplemented the Dutch language in official domains: education, administration and court. Opponents and advocates of this new language policy emphasized different points. Adversaries indicated the status and prestige of French abroad, while pointing out the fragmentation and poor development of the Flemish dialects, having no connection whatsoever with the cultivated 1
"Le fédéralisme et la superstition parlent bas-breton; l'émigration et la haine parlent allemand; la contre-révolution parle italien, et le fanatisme parle basque."
LANGUAGE PLANNING IN BELGIUM
123
language spoken and written in the North. It was stated that (Defrenne 1929:7) "a great number of inhabitants do not speak that language and [...] almost everyone also speaks French" [emphasis in original].2 Moreover it was asserted "that our Flemings from the Southern provinces understand with difficulty, or do not understand at all, the Dutch of the Northern provinces, and this is mutual" (Defrenne 1929: 7).3 The advocates, on the contrary, started from the common background of both languages (the Northern and Southern varieties of Dutch). The Dutch language was assigned valuable structural features like simplicity and abundance of morphological forms. In particular, the ease of word formation was the centre of attention. New words are coined in Dutch by combining existing words and adjectives, which makes the newly coined word's meaning very transparent. This is not the case with French, which: resorts to the Latin and Greek Language, from which it borrows many words: but these words have the disadvantage that they have to be translated for the French to understand them.4 (Mulle 1815 - 1819:8-9)
Still, William I could scarcely find support for his language policy in Belgium (De Jonghe 1967). This was due to two factors. On the one hand, the Liberals, who were in favour of his economic policy, were thoroughly frenchified in language use and attitude and objected to the use of Dutch in the public domain. They stated: that we do not see in the south of the kingdom how the obligatory use of Dutch or Flemish can have a beneficial influence on our external relations; we confess moreover, that we view this usage as destructive for our indigenous well-being [...].5 (Defrenne 1829:13)
On the other hand, the clergy, who might have been an ally in his language policy, because especially the lower clergy had always continued to use Dutch (or rather a Flemish dialect), were opposed to Holland's Protestantism. The 2
"une foule d'habitans [sic] ne parlent pas ce langage, et [...] presque tous parlent aussie [sic] le français." 3 "que nos Flamands méridionaux entendent difficilement, et n'entendent même souvent pas du tout, le hollandais des provinces du Nord; ce qui est réciproque." 4 "tot de latynsche en grieksche Taelen haeren toevlucht neemt, waer uyt zy menigvuldige gemaekte woorden ontleent: maar deze gemaekte woorden hebben dit ongemak en geven die bezwarenis, dat zy voor de franschmannen zelve, namige en woord-grondige bepalingen noodig hebben [...]." 5 "que nous n'apercevons pas dans le midi du royaume, quelle influence salutaire l'usage obligé du hollandais ou du flamand peut exercer sur nos relations à l'exterieur; nous confessons de plus, que nous envisageons cet usage comme subversif de notre bien-être indigène [...]."
124
JETJE DE GROOF
Dutch language was equated with Protestantism and it was said that "Dutch, written even by learned and irreproachable people, always carries in it the elements of heresy" (Bergmann in a letter to Van Maanen, Minister of Justice, as cited by De Jonghe 1967:308).6 William I failed to take any language planning measures which might have stimulated a sense of language unity. This made it all the easier to state that Dutch and Flemish were separate languages, an argument used eagerly by the opponents of William I: the pro-French Liberals and the anti-Protestant clergy. They joined forces and organized petitions against the language legislation. William I had to abandon his language legislation before the country based on language unity collapsed completely. 4. Belgium from 1830 onwards 4.1 'Freedom of language ' The new Belgian government directed its language policy in yet another direction. The Constitution elaborated 'freedom of language' as the guiding principle of its language policy. Still, by stating that French was to be the language used in courts and in official publications, as Flemish was dialectally too diverse, it became clear that the undertone of 'freedom of language' was in fact a policy of frenchification. All sectors of public life became very intensively frenchified, as French was de facto the language of administration, the court, the army and education. The elite, favouring the French language, did not object to this, nor did the lower classes, who had other more pressing matters on their mind, like scraping a living. Still, a new, emerging, non-frenchified middle class felt limited in its upward social mobility. It is from this segment of society that the first protests against the frenchification of public life emerged, giving rise to the Flemish Movement in the 1830s. 4.2 A first phase of status planning: from personality to territoriality The Flemish Movement did not act on a political level straight away. The first years were characterized by literary activity. There was an increase in works published in Flemish. Some of the major works of literature from the Middle Ages, the time when Dutch was still a prestigious language, were republished by a special commission founded by the government.
6
"le hollandais, écrit même par des personnes savantes et irréprochables, porte toujours avec lui les germes de l'hérésie."
LANGUAGE PLANNING IN BELGIUM
125
The petitions of 1840, in which the first political demands of the Movement were formulated, were the opening shot of the political Flemish Movement. It was at this point that the government ceased to view the Flemish Movement as harmless. The latter started striving for language legislation to be put in place, to ensure that everybody had the basic right to be able to speak in their own language in administration, education, court and the army. The struggle for the establishment and the enactment of these language laws mirrored the factions in public life. A very important societal division was that of Catholics against non-Catholics, making any union of 'champions of the Flemish Movement' impossible, the religious divide being too great to overcome. Throughout the 19th century, the language legislation maintained the 'personality principle' for Flanders, meaning that people had the choice between speaking French or Dutch in all spheres of public life (see also De Rammelaere 1974). The language laws of 1873 (court), 1878 (administration) and 1883 (education) enforced language rights for speakers of Dutch in Flanders. In 1898, the Equality Law was passed, which made Dutch an official language of Belgium, alongside French. But the demands of the Flemish Movement evolved from seeking limited language rights for Flemings in Flanders, to complete equality of French and Dutch in Flanders. Toward the end of the 19th century, demands became more aggressive and were directed towards language rights for Flemings in the Walloon (French-speaking) part of Belgium and establishing Dutch as the sole official language of the 'Flemish' provinces. The demand for a Dutchlanguage university (to be located in Ghent) became a pressing matter too. Although the First World War meant a setback for the Flemish Movement (some Flemish Movement champions had collaborated with the Germans), the implementation of the 'one man, one vote'-system in 1919 made it possible to achieve these aims.7 This meant not only that the first all-Dutch university opened its doors in Ghent in 1930, it also meant that the territoriality principle (McRae 1976) was introduced by language legislation in the thirties (see also Van Ginderachter, Witte & Van Velthoven 1998). These language laws made Dutch the sole language of education (1932), administration (1932), the court (1935), and the army (1938) in the Flemish provinces. 'In Flanders Flemish', the old slogan of the Flemish Movement, had become reality. Language legislation had provided a secure structure in which the standardization of Dutch in Flanders could progress (see also Haarmann 1988). 7
Women were not granted universal suffrage until 1948.
126
JETJE DE GROOF
4.3 Aspects of 19 th-century standardization of Dutch in Flanders 4.3.1 Orthography: integration towards the North. The government did not at first participate significantly in the standardization of Dutch in Flanders. Only at the level of graphization was there any involvement in corpus planning by the government (see also Couvreur & Willemyns 1998). The question of orthography has been a lively one from the very beginning of the existence of Belgium. Societal issues were drawn into the discussion and at certain points they really made up the entire debate. In order to regulate the orthographic chaos that ruled in Belgium, the government appointed a committee in 1836, which decided to adopt the Dutch spelling in most cases. This decision was made official in a Royal Decree. At this point, a fierce battle between proponents of a separate Flemish system (called 'particularists') and adherents of the Dutch one ('integrationists'), arose. Both groups preferred different orthography systems. Vandenbussche (this volume) gives an overview of these spelling models. 4.3.2 The Dutch Congresses: contact with the North. The champions of the Hollandic norm realized that the prestige of the language in the North could be used against opponents of Hollandic in Belgium. Stressing that the language used in Flanders and Holland really was the same language, they countered the very persistent argument that Flemish could never be made official because it was an amalgam of dialects. It was in this context that the 'Dutch Congresses' came into existence (Vanacker 1982). These conferences on Dutch language and literature were to be organized every two years from 1849 onwards. Dutch as well as Belgian participants were invited. Even though the titles of the Belgian speeches often seemed to refer to corpus planning issues, they were primarily about the status of Dutch in Flanders. Apparently, this is a constant for such congresses. Fishman (1993:337) points out the following concerning the agenda of language congresses: "[...] at most congresses the corpus-planning emphases merely serve as rather transparent masks for the status-planning passions that are just a little below the surface". 4.3.3 Dictionary of the Dutch Language. The idea of compiling a Dictionary of the Dutch Language originated from the first congress in 1849. Work on this enormous task started a few years later under the guidance of Matthias de Vries, and was only completed in 1998 (Van Sterkenburg 1992). Both Dutch and Flemish lexicographic work would form the basis of the Dictionary.
LANGUAGE PLANNING IN BELGIUM
127
The Dictionary was a crucial step in the standardization of Dutch, as it was to be the standard work determining which words were part of the language and which not. A crucial matter in this collaboration between both countries was the question as to what would be done with typically Flemish, thus 'peripheral' vocabulary. Van Sterkenburg (1992:26) has summarized De Vries's viewpoints on peripheral vocabulary: The Southern Dutch idiom must be consulted with care and used as an enrichment to the dictionary. Besides, one will have to beware of adopting the kind of expressions, that, having arisen from French influence, are at variance with the rules of the Dutch language.8
Still, quite a few Flemings were soon disappointed about the amount of Southern vocabulary actually finding its way into the dictionary. In 1891, the newly founded Flemish Academy, which I discuss below, even initiated a commission to investigate this matter. The Academy invited its members to offer more lexical material to the commission of the Dictionary. Wils (1955— 1956) also points in this direction: although the Flemings did complain, they did not always suit the action to the word and were quite slow to excerpt Southern vocabulary. 4.3.4 Further discussions on the norm. Although the first regulation of spelling was already a clear victory for the integrationists, driving the particularists into opposition, the latter did not disappear from the scene at all. The search for a norm for vocabulary and pronunciation would be the cause of fierce battles between a second generation of particularists and integrationists. A central point in this discussion was the belief that the exclusion of the Flemish everyday language from the standard language would lead to a stiff, bookish language, no longer reflecting the 'spirit' of the people. This attitude need not surprise us. Most of these people were strongly influenced by figures like Jacob Grimm who were deeply influenced by Romanticism. As Thomas (1991) underlines, one of the consequences of the Romantic Movement was the interest accorded to dialects and other vernaculars. Dialecticisms were highly valued in those days. Still, at this point a very important distinction must be made within this second generation of particularists. On the one hand, there were those who strove for Southern influence in the instruments of standardization, like the 8
"Het Zuidnederlandse taaieigen moet met zorg geraadpleegd worden en ter verrijking van het woordenboek aangewend worden. Daarbij zal men zich echter wachten voor het opnemen van zodanige uitdrukkingen, die, uit Franse invloed ontsproten, met de Nederlandse taalwetten in strijd zijn."
128
JETJE DE GROOF
Dictionary of the Dutch Language. But on the other hand, there were also those for whom the hidden agenda of their polemics was more important than their actual linguistic activity. Central in this group was the Guild of SinteLuitgaarde, a group of religious fanatics who targeted language and since a change in linguistic habits was one of the most 'recognizable' characteristics of societal change, this might explain why language was going to play such a dominant part in their strategy. Intuitive feelings rather than scientific analysis kept them going and nostalgia for the 'good old days' seems to be what underlies their complaints on the 'persecution of the church' as well as the 'threatening' linguistic evolution. (Willemyns 1997:287)
4.3.5 The Flemish Academy: establishment o f an authority. The demand for a Flemish Academy, operating separately from the national, Francophone Academy, had been prominent from the very beginning of the 'Flemish Movement'. Still, it would take until 1886 to see the establishment of this Academy. Its task would be varied (Rombauts 1998). Due to the specific situation of Dutch in Belgium, its activities would include linguistic science, Dutch literature, actions directed at improving language rights through language legislation, and the stimulation of the emancipation of the Flemish people. The Academy aspired to a fully accepted standard language in Flanders and thus stressed the language union of North and South. Still, it did also stress respect for Flemish material in this standard language and thus tried to find the middle ground between the various positions at that time. 4.3.6 Results of 19th-century standardization. The extension of the functions of Dutch in Flanders by language legislation was the main goal of the Flemish Movement in this period. But not many people realized that the Dutch spoken and written by the inhabitants of Flanders was not structurally adequate for this task. Calvet (1996:45) emphasizes that in this kind of situation "their [the languages'] deficits will have to be reduced, they [the languages] will have to be equipped to play this role" [emphasis in original].9 But this did not happen. The debate on the norm, concerning whether the Northern or the Southern language was more apt to become the norm, seemed unending and remained very theoretical. The result of the fierce discussions was that the necessary next step, the focus on the adoption and spread of a codified norm, was not taken. A handful of teachers, linguists and other 'language lovers' were involved in the practice of purification, pinpointing the frequent 'mistakes' resulting 9
"il faudra bien réduire leurs déficits, les équiper pour qu'elles puissent jouer ce rôle."
LANGUAGE PLANNING IN BELGIUM
129
from the language contact situation with the French language on the one hand, and dialects on the other. Although they did not have the means to implement their norm for the masses, their discussion probably increased the awareness of the existence of a norm for certain layers of society. This was achieved through hard work — an amazing amount of literature was published on this subject. The problem of the norm was on the one hand discussed in newspapers and journals for the general public, and on the other hand in specialized magazines for people interested in language matters. Moreover, pamphlets were published. Entering the 20th century, the first reference materials were published (Suffeleers 1979). 4.4 The consolidation of the territoriality principle and federalization The language laws of the 1930s (see 4.3), which had established the territoriality principle in Belgium (with the exception of Brussels, where the personality principle was maintained), were constantly violated due to the fact that there was no clear, official demarcation between the Dutch language area and the French language area. Still, the Flemish Movement experienced a setback after World War II: again, a considerable number of Flemings collaborated with the Germans, which put the Flemish Movement and its demands in a bad light. Nevertheless, language had become such a conscious separating factor between Walloons and Flemings that further measures had to be taken. In the sixties, the language border was officially marked, and the territoriality principle was further enshrined in a series of language laws. At the end of the sixties there was a last important upheaval, concerning the dutchification of the University of Louvain, which was situated in what was now a Flemish area, but which was still partly French-speaking. The University was dutchified, which assured the linguistically homogeneous nature of the language areas. The presence of a clear demarcation between two linguistically homogeneous territories led the way to later federalization. In 1970, the first phase of federalization was set in motion, and in 1994, the entire process reached its final stage (Senelle & Van de Velde 1998). Belgium is now a federal state, consisting of three communities (Dutch, French and German) and three regions (Flanders, Wallonia and Brussels—Capital). The communities are based on language and are responsible for cultural and educational matters, meaning that they deal with the language policy in their own area. The regions are in charge of environmental planning, health care, economic development and other matters. Although federalization has made relations more peaceful, there are still elements that cause trouble. Some of the districts in Brussels and near to the language border, which have a mixed language regime, continue to be at the centre of attention.
130
JETJE DE GROOF
4.5 Spread of the standard language The period from the thirties onwards has brought a breakthrough in corpus planning issues. At this stage people were made aware en masse of what the standard was like, which meant that at last efforts were being made to implement what had been the object of so much quarrelling. This evolution was stimulated to a considerable degree by the increasing economic weight of Flanders compared to the Walloon area. Organizations like the "Society for Civilized Dutch Pronunciation" (Vereeniging voor Beschaafde Nederlandsche Uitspraak) (founded in 1913) actively propagated the norm in its magazines. The publication of suitable handbooks, teaching the 'right' Dutch, was another important stimulus. Moreover, the government engaged in corpus planning. This eventually led to a collaboration between the Flemish and the Dutch government through the "Dutch Language Union" (Nederlandse Taalunie), established in 1980. Official language cultivation and advice functions of both the Dutch and the Flemish governments were transferred to one representative body. The press, finally, has also been used to make people aware of what cultivated language 'should' be like. Contributions on the radio, television and in newspapers were of prime importance here (Willemyns & Haeseryn 1998). 4.6 The standard today As this paper has primarily provided a historical account of the evolution of the standard language through language planning, I will briefly look into its recent evolutions. Although codification is now primarily in the hands of the Dutch Language Union (dictionaries, grammars), there is still a diverging movement between the Northern and Southern realizations of standard Dutch, most prominently in pronunciation and the lexicon (see also Van de Velde 1996). Not only is the North evolving away from the norm, but Flemish substandard varieties are also starting to put pressure on the standard language in the South. This is an interesting case of interaction between language culture and language nature. Although language planning can lead a language to evolve in a certain direction, language nature, language usage in vivo, can turn the tide yet again. 5.
Conclusions 1. Language planning has been an important aspect of Dutch language standardization in Belgium: individuals, organizations, academies, political parties and the government have constantly worked to influence the language situation. These interventions or attempts to intervene have occurred in the most diverse ways, under the most disparate conditions, initiated against a
LANGUAGE PLANNING IN BELGIUM
131
background of 'hidden agendas'. Language planning as it occurred was not always 'planned' in the traditional sense. In most cases, planning was more an ad hoc decision to try to intervene in the evolution of Dutch in Flanders. Still, theory and definitions of language planning and policy have evolved in such a way that this is now also considered an act of planning. The same counts for the fact that such planning was carried out by actors ranging from individuals to authorities. The term 'language planning' is no longer reserved for actions intiated by governments and authorities. 2. The ruling frenchified class opted for a de jure freedom of language, which entailed de facto a policy of laissez faire and thus frenchification. The Flemish Movement of the late 1800s provided Belgium with language legislation giving the speakers of the Dutch language in Belgium increased language rights. The peculiar aspect of the whole situation was the failure to understand that corpus planning measures were necessary to assure the success of status planning. Although it was decided quite early that the norm was to be the Northern Dutch one, the various issues of standardization caused too many ideological debates in socially divided Flanders for the standard to be implemented and used. As Vikør (1993) points out, this is not an unusual situation. The principles underlying corpus planning are many and varied, but those derived from societal ideologies tend to be most important. Hostile attitudes towards the Dutch norm and the traditionalism of many 'language lovers' caused debates which retarded the language planning process. It is clear that the idealized form of standardization through language planning, as we saw in Haugen's scheme, does not hold true for Belgium. The phase of selection of a norm was followed immediately by a phase in which the emphasis was on language legislation. The norm was neither codified nor elaborated, which meant that the language to be implemented through language laws was not up to its new functions. Moreover, the people who were to use the Dutch language in its new functions did not have the necessary competence for this task. 3. Very typical of the Belgian situation was thus the initial non-emphasis on the linguistic structure of the language. One of the protagonists of the early Flemish Movement, J.B. David, illustrates this situation very nicely, saying that the champions of the Dutch language had to build the language with one hand (standardization of Dutch in Flanders), while holding a sword in the other (expanding its functions in Belgium). Once the sword became unnecessary, more time could be spent on building (Bon 1841:22). For the latter to happen, a relative pacification of the language conflict between speakers of French and Dutch was necessary, and this took the form of
132
JETJE DE GROOF
the territoriality principle. Increased government involvement, more frequent contact with the language spoken in the North through the media, and the shift of the economic centre of gravity to Flanders, brought about the eventual spread of the standardized form of Dutch in Belgium. Corpus planning became at that moment a true activity and was no longer used as a strategy for achieving status planning aims.
References Ager, D. 2001. Motivation in Language Planning. Clevedon: Multilingual Matters. Baldauf, R.B. & R.B. Kaplan. 1997. Language Planning: from Practice to Theory. Clevedon: Multilingual Matters. Bon, F. 1841. Deuxième suite aux lettres pour servir de matériaux à l'histoire des deux introductions du système linguistique néerlandais en Belgique. Bruxelles: M.-E. Rampelbergh. Calvet, L.-J. 1996. Les politiques linguistiques. Paris: Presses Universitaires de France. Cooper, R.L. 1989. Language Planning and Social Change. Cambridge: Cambridge University Press. Couvreur, W. & Willemyns, R. 1998. Spellingoorlog. Nieuwe Encyclopedie van de Vlaamse Beweging. Database. Tielt: Lannoo. De Certeau, M., D. Julia & J. Revel. 1979. Une politique de la langue. La Révolution française et les patois. Paris: Éditions Gallimard. Defrenne, J. 1829. Quelques idées sur l'usage obligé de la langue dite nationale, au Royaume des Pays-Bas, par un belge, ami de la justice et de la vérité. Brussels: Dumont. De Jonghe, A. 1967. De taalpolitiek van Willem I. Bruges: Darthet. Deneckere, M. 1954. Histoire de la langue française dans les Flandres (17701823). Ghent: Rijksuniversiteit. De Rammelaere, L. 1974. Het ontstaan van de drie eerste taalwetten (1873— 1883). Ghent: unpublished dissertation. Fishman, J.A. 1993. The "First Congress" Phenomenon: Arriving at some General Conclusions. The Earliest Stage of Language Planning: the "First Congress" Phenomenon, ed. by J.A. Fishman, 333-348. Berlin & New York: Mouton de Gruyter. Haarmann, H. 1988. Sprachen- und Sprachpolitik. Sociolinguistics. An International Handbook of the Science of Language and Society. Vol. 2 ed. by U. Ammon, N. Dittmar & J.K. Mattheier, 1660-1679. Berlin: Walter de Gruyter.
LANGUAGE PLANNING IN BELGIUM
133
Haugen, E. 1969. Language Planning, Theory and Practice. Actes du Xe Congrès International des Linguistes: Bucarest, 28 Août - 2 Septembre 1967 ed. by A. Graur, vol. 1, 701-711. Bucarest: Éditions de l'Académie de la République Socialiste de Roumanie. Haugen, E. 1983. The Implementation of Corpus Planning. Theory and Practice. Progress in Language Planning. International Perspectives ed. by J. Cobarrubias & J.A. Fishman, 269-291. Berlin: Mouton de Gruyter. Kloss, H. 1969. Research Possibilities on Group Bilingualism. A Report. Quebec: International center for research on bilingualism. McRae, K. 1976. The Principle of Personality and the Principle of Territoriality. International Journal of the Sociology of Language 4. 35-54. Mulle, J.B.F. 1815-1819. Letterkundig tydverdryf over het bewys, dat de Vlaemsche tael beter is als de Fransche, en eenige belangryke aenmerkingen tot voorspoed van den landbouw en verdere staetshuyshoudelyke bedenkingen. Ghent: Goesin-Verhaege. Ricento, T. 2000. Historical and Theoretical Perspectives in Language Policy and Planning. Ideology, Politics and Language Policies (Impact. Studies in Language and Society.) ed. by T. Ricento, 9-24. Amsterdam & Philadelphia: John Benjamins. Rombauts, W. 1998. Koninklijke Academie voor Nederlandse Taal- en Letterkunde (KANTL). Nieuwe Encyclopedie van de Vlaamse Beweging. Database. Tielt: Lannoo. Rubin, J. 1971. Evaluation and Language Planning. Can language be planned? ed. by J. Rubin & B.H. Jernudd, 217-253, Honululu: University Press of Hawai'i. Senelle, R. & Van de Velde, E. 1998. Staatshervorming. Nieuwe Encyclopedie van de Vlaamse Beweging. Tielt: Lannoo. Suffeleers, T J . 1979. Taaiverzorging in Vlaanderen. Een opiniegeschiedenis. Bruges: Orion. Thomas, G. 1991. Linguistic Purism. London: Longman. Tollefson, J.W. 1991. Planning Language, Planning Inequality. Language Policy in the Community. New York: Longman. Vanacker, H. 1982. De "Nederlandsche Taal- en Letterkundige Congressen" en de vernederlandsing van het onderwijs. Ghent: Cultureel Documentatiecentrum 't Pand. Van de Craen, P. & Willemyns, R. 1988. The Standardization of Dutch in Flanders. International Journal of the Sociology of Language 73. 45-64. Van de Velde, H. 1996. Variatie en verandering in het gesproken StandaardNederlands (1935-1993). Catholic University of Nijmegen: dissertation.
134
JETJE DE GROOF
Van Ginderachter, M., E. Witte & H. Van Velthoven. 1998. Taalpolitiek- en wetgeving. Nieuwe Encyclopedie van de Vlaamse Beweging. Database. Tielt: Lannoo. Van Sterkenburg, P.G.J. 1992. Het woordenboek der Nederlandsche taal. Portret van een taalmonument 's Gravenhage: SDU-uitgeverij. Vikør, L.S. 1993. Principles of Corpus Planning — as Applied to the Spelling Reforms of Indonesia and Malaysia. Language Conflict and Language Planning, ed. by E.-H. Jahr, 279-298. Berlin & New York: Mouton de Gruyter. Willemyns, R. 1997. Religious Fundamentalism and Language Planning in 19th Century Flanders. Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis. 2:2. 281-302. Willemyns, R. & R. Haeseryn. 1998. Taal. Nieuwe Encyclopedie van de Vlaamse Beweging. Database. Tielt: Lannoo. Wils, L. 1955-1956. Vlaanderen en het tot stand komen van het "Woordenboek der Nederlandsche Taal". Nova et Vetera 23. 373-382.
POLITICAL INFLECTIONS GRAMMAR AND THE ICELANDIC SURNAME DEBATE
KENDRA WILLSON University of California, Berkeley, USA 1. Introduction Iceland is famous for its preservation of the patronymic system. Only some 15% of contemporary Icelanders bear surnames inherited in a fixed form (Tomasson 1975:281). A person's first name remains his or her primary name, while the indication of whose son or daughter he or she is is viewed by Icelanders less as a name than as a secondary descriptive label. The fact that the Icelandic telephone catalogue is organized by given name is a source of wonder to foreigners and a locus of national pride for many Icelanders. These Icelanders may not, however, remember that this system has been maintained by centralized planning as well as by the will of the people. The same stringent name laws which, notoriously, long obliged foreigners seeking Icelandic citizenship to assume Icelandic names, also restricted the adoption and transmission of surnames. A hundred years ago, the future of the patronymic system was far from assured. A heated discussion went on around the start of the twentieth century as to whether Iceland should retain the patronymic system or adopt a system of fixed surnames, as the other Nordic countries were doing.1 A central point in the debate was whether it was possible to create surnames which would be consonant with the Icelandic linguistic system, and how such surnames should be formed. The insular Nordic languages, Icelandic and Faroese, differ from the Mainland Scandinavian languages in the conservatism and consequent richness of the insular inflectional system. The exaggerated view that the Icelandic language has not changed since medieval times occupies a central position in Icelandic patriotism, and linguistic purism has enjoyed remarkable success in Iceland. (See in general Kjartan G. 1 Surnames were mandated in Denmark and the Faroes by a law of 1828, in Sweden in 1901, in Finland in 1920 and in Norway in 1923.
136
RENDRA WILLSON
Ottósson (1990), and a shorter treatment in English by Halldor Halldórsson (1979).) At the same time there was a general understanding that surnames should have identical forms in all cases (save perhaps for a genitive -s) and be underspecified for gender — following the model of surnames in Denmark and other nations in Western Europe and North America where languages with less elaborate nominal inflection were spoken. The occasional voices advocating fully declinable surnames on the Roman or Slavic model (e.g. Johannes L.L. Jóhannsson 1909, Holger Wiehe 1917) seem not to have gained much support. Literature from the Icelandic surname debate presents numerous examples of arguments from grammar, with both sides adopting patriotic rhetoric about the integrity of the Icelandic language. Those who opposed the adoption of surnames based their opposition on the supposed impossibility of creating surnames which would be "Icelandic" in their morphological composition. They stressed the purported risks to the Icelandic language posed by alien structures. Those who favoured surnames of a particular type justified their position by appealing to analogies with patterns existing in the language, while remaining conspicuously silent on the more obvious relationship to Danish or other foreign models. The linguistic arguments served to deflect focus from the central, sensitive issue of Iceland's cultural autonomy vs. identification with the outside world, particularly the colonial power Denmark, from which Iceland was at that time emerging.2 I will present a sample of contributions to the surname debate in Iceland at the turn of the 20th century. The writers involved in the debate vary in their degree of linguistic expertise, but all use linguistic arguments to support their positions on an issue which was at bottom less a matter of grammar than of national identity. 2. Background The history of surname use in Iceland is summarized by Guörun Kvaran and Siguröur Jonsson frá Arnarvatni (1991:70-81). Medieval texts seem to indicate that some by-names were transferred to the original bearers' descendants for a few generations. However, surnames entered Modern Icelandic usage via Icelandic men of letters who found it desirable to have a by-name or surname during their studies abroad. The oldest surname in continuous usage (adopted in the 17th century) is Vidalín, derived from a Latinized form of the place name Vididalir (Guörun Kvaran & Siguröur Iceland received home rule in 1904 and full independence in 1944.
POLITICAL INFLECTIONS
137
Jonsson frá Arnarvatni 1991:70). Over the following two centuries, the assumption of surnames by members of the upper and upwardly mobile classes became more and more common. Some surnames were formed from Icelandic names or patronymics, often in a Latinized or Danicized form (hence Thor lacius < Porláksson, Thoroddsen < Póroddsson); others were based on placenames (Blöndal < Blöndudalur, Briem < Brjânslœkur), likewise often in a Danicized form (Svavar Sigmundsson ms.). Some Modern Icelandic surnames may also reflect names imported to Iceland by families who moved there (Hafstein, Leví). Later, some Icelanders would adopt surnames based on bynames of Celtic origin taken from medieval literature (Kvaran, Kamban). The first official attempt to stem the tide of surnames was a proposal presented to Alpingi, the Icelandic parliament, in 1881 by Jon Jonsson and Jon Olafsson (Frumvarp 1881). This law would have required Icelanders to obtain royal permission before adopting a surname, as well as exacting a fee of 500 crowns for the privilege of having such a name and an annual nafnbótarskattur ('title tax') of 10 crowns per syllable of the last name. Eptirleiöis ma ekki skíra neinn mann ættarnafni, nema konunglegt leyfi sje til pess. Ekkert ættarnafn m5 enda á "son". Fyrir ættarnafnsleyfi skal borga 500 krónur, sem renna í landssjóö. Hver sá, sem skrifar sig ættarnafni, skal par aö auki greiöa árlegan nafnbótarskatt, 10 krónur fyrir hvert atkvæoi, sem f nafninu er. [Henceforth it is forbidden to baptize any person with a surname without royal permission. No surname may end in "son." Permission to take a surname shall cost 500 crowns, which shall be added to the national treasury. Anyone who writes his name with a surname shall, in addition, pay an annual title tax of 10 crowns for each syllable in the name.] {Frumvarp 1881:1)
The syllable count is perhaps the most distinctive aspect of this proposal. Not only is a family name a taxable luxus, but Thorlacius is a greater extravagance than Thors. Some later contributions to the surname debate (Jóhannes L.L. Jóhannsson 1909, Islenzk mannanöfn 1915) express opinions as to whether shorter or longer surnames are preferable (discussion below). Short surnames tend to correlate with omission of inflectional endings and hence with indeclinability, as well as with a closer resemblance to common bourgeois names in Mainland Scandinavia. The 1881 proposal was not ratified. Surnames continued to multiply; in 1910 there were 297 different surnames borne by persons born in Iceland (Guorun Kvaran & Sigurour Jonsson frá Arnarvatni 1991:70), compared to 108 in 1855. The future of Icelandic naming practices became an issue of heated debate.
138
KENDRA WILLSON
The writer Guomundur Kamban (1908) and the linguist Jóhannes L.L. Jóhannsson (1909) both argue that Iceland will inevitably adopt a system of fixed surnames. The progressionist Guömundur Kamban welcomes this development, while the more conservative Johannes L.L. Jóhannsson views it as an unfortunate necessity (ascribing the same view to Guömundur Kamban as well). The two represent opposite stands on the issue of whether surnames should be inflected. Both writers make extensive reference to grammar and exploit rhetoric about the integrity of the language. While Johannes L.L. Jóhannsson's presentation is, predictably, the more philologically sound, Guömundur Kamban's linguistic arguments are at times the more original. 3. Guömundur Kamban The playwright and novelist Guömundur Jonsson Kamban (1888-1945) lived abroad for much of his life, in the U.S. and Germany but primarily in Copenhagen. He wrote in Danish as well as Icelandic, and aspired to become known as a writer in a world language, English or German (Kristinn E. Andrésson 1949:221-231). Hence Guömundur Kamban clearly identified with an international community and aimed for international renown. In the course of this career, he himself chose to adopt a surname (a by-name probably of Celtic origin taken from Old Icelandic sources). The philosophy apparent in Guömundur Kamban's (1908) essay advocating surnames is progressionist, forward-seeking and optimistic. Guömundur Kamban (1908) views the adoption of surnames in Iceland as a necessary development if Iceland is to be recognized as a nation of culture and civilization: Ef mentaðir menn með öðrum pjóðum vita um eitthvert land, að par séu menn alment nefndir föðurnafni sínu, aö ættarnöfn séu par engin til, pá dettur peim f hug um leið, aö sú pjóö sé ekki stödd a hau menningarstigi. [If educated men in other nations know of some nation where people are generally called by patronymics and there are no surnames, then they think that this nation is not in an advanced state of culture.] (Guömundur Kamban 1908:165)
The emphasis in Guömundur Kamban's linguistic rhetoric is on the semantic integrity of the language, as opposed to its morphological integrity. He takes as a starting assumption, almost a natural law, that "Hjón vilja eiga sér samheiti" ["Married couples wish to have a name in common"] (Guömundur Kamban 1908:169). In the patronymic system, women do not change their names upon marriage, but some felt that a woman who had borne an inherited surname before marriage was obliged to abandon this name after
POLITICAL INFLECTIONS
139
she was married. A Danish woman whose Icelandic husband did not have a surname was thus supposedly forced to use her husband's patronymic as a surname and call herself, say, fru Karen Höskuldsson ('Mrs Karen Höskuldsson'), a collocation Guömundur Kamban finds aesthetically unpleasing. Worse, Guömundur Kamban envisions that in the future Icelanders will adopt the European practice of referring to a woman by her husband's first name as well as his last name, hence frú Einar Jonsson; this would lead others to imagine that poor Einar himself was being ridiculed by being called Mrs. The reason such (hypothetical) collocations are an affront to Icelandic sensibilities is their violation of the "logic" of meaning: speakers of Icelandic know that a woman cannot be anyone's son. Islenzkan er sjálfsagt eitt hiö rökréttasta mal, sem nú lifir a vörum nokkurrar pjoðar. Af hverju? Af pví að hun hefur reynst trûust sjálfri sér, af pví að hún hefir varðveitt eðli sitt bezt allra tungna. En hvad litid sem brotið er móti eðli tungunnar, er pað altaf nógu mikið til pess, aö pað sljófgar næmleikann viö fegurö hennar. Og paö er aö mispyrma tign og eöli hverrar tungu, aö lata orö merkja alt annaö eöa gagnstætt pví sem pau merkja áður ad allra vitund. [Icelandic is of course one of the most logical languages which now live on the lips of any nation. Why? Because it has proven most true to itself; because it has preserved its nature best of all tongues. And however little the nature of the language is violated, it is always enough to blunt sensibility to its beauty. And it is an abuse of the dignity and nature of any language to make words mean something entirely different from or opposite to that which everyone has previously known them to mean.] (Guömundur Kamban 1908:170)
The only way to avoid this offence to the Icelandic ear, according to Guömundur Kamban, is to introduce surnames. As methods for deriving such, he first mentions genitive forms of men's names, already in colloquial usage as abbreviated forms of patronymics. A Jón Sveinsson would henceforth call himself Jon Sveins, his wife Gudrun Sveins, his daughter Pórdis Sveins, etc. This "heflr [sic] einn kost og enga fleiri. paö er islenzkt" ["has one advantage and only one. It is Icelandic"] (1908:171); but it will only remain Icelandic insofar as people consistently use the first name as well as the last. Guömundur Kamban predicts, however, that people (especially foreigners) will not bother with both names, but will start to call wife and daughter frú Sveins ('Mrs Sveins') and jungfru Sveins ('Miss Sveins') respectively. This is also inaesthetic; "ekkert vit er f pví aö nefna menn eignarfallinu einu saman" ["it's silly to call people by the genitive alone"] (Guömundur Kamban 1908:172).
140
KENDRA WILLSON
However, there is a solution: Icelanders can learn not to hear these forms as genitive: Vér munum geta vanist af að lfta a pessi nöfn svo sem væri pau eignarföll, af að vér höfum vanist af pvf urn önnur orð f málinu. Vér hugsum oss endingu peirra afleiðsluending í stað fallendingar. Svo er t.d. aö taka um oröin lax og hams (flt. hamsar), aö fyrra oröiö er ekkert annaö en eignarfall af lag og hiö síðara af hamur. [We will be able to train ourselves not to view these names as if they were genitive forms, as we have done with other words in the language. We think of their ending as a derivational ending rather than as a case ending. It should be mentioned, e.g., regarding the words lax ('salmon') and hams ('fruit peel, snakeskin'), that the former word is nothing other than the genitive of lag ('layer') and the latter of hamur ('skin'). (Guömundur Kamban 1908:172)
These are not recognized etymologies for these particular lexemes. Guömundur ultimately rejects the genitive suggestion as unsatisfactory, both because such examples as those cited above are rare, and because the resulting surname system would be limited and monotonous. However, the idea remains that it is easier to train oneself to reinterpret an inflectional morpheme as a derivational one than to redefine the significance of the lexical morpheme son. The same notion that morphological categories are more easily redefined the more distanced they are from "meaning" is also evident in Guömundur Kamban's next suggestion. Vér eigum aö velja pjál og falleg nöfn a hverju sem er, hlutum, hugtökum, mönnum. Og vér eigum aö sleppa beygingum peirra, og halda eftir rótinni einni. Meöal margs annars fyrir pà sök, aö kynferði orösins sést venjulegast a beygingu pess. [We should choose supple and beautiful names for everything — objects, concepts, people. And we should omit their declensions and stick to the root alone. Among many other reasons, because the gender of the word is most often seen from its declension.] (Guömundur Kamban 1908:173)
There is a semantic conflict if the gender of a name and that of its bearer are in discord. In Guömundur Kamban's view, this is more essential to the nature of the language than are case endings. Guömundur Kamban justifies his proposal with a progressionist appeal to a universal tendency toward morphological simplification in language change, expressing no little admiration for English: "I öllum malum er einhver tilhneiging til aö gera oröin einfaldari meö pessum hætti. Tilhneiging sem vex, jafnóöum og máliö proskast. Allir vita urn enskuna, eitt hiö langproskaöasta tungumál heimsins" ["In all languages there is some tendency to make the
POLITICAL INFLECTIONS
141
words simpler in this way. A tendency which grows as the language develops. Everyone knows about English, one of the most developed languages in the world"] (Guð mundur Kamban 1908:173). This is directly at odds with the preservationist, puristic tradition which dominates Icelandic language politics. Guömundur Kamban nonetheless exploits the trappings of philology to justify indeclinable surnames through analogies to existing patterns in the language. This is a type of argument which recurs in the legal report (Islenzk mannanöfn 1915) discussed below. Guömundur Kamban argues that the way is already paved for indeclinable names in Icelandic, citing classes of nouns which for various reasons (e.g. assimilation of the nominative -r to a preceding consonant, shortening of final -rr to -r, analogy, late borrowing) lack overt endings in most cases. A sample: Nefnifall. Oil karlkynsorð, sem enda a -ar: annar, hamar, Gunnar, o.s.frv. Nokkur karlkynsorð ja-stofna með stuttri rótarsamstöfu: byr, bær, gnyr, her, o.s.frv. Mörg ustofna orð: ár, as, björn, knörr, son, vin, örn, o.s.frv. Mörg a-stofna orð: akur, fugl, hals, hrafn, fs, karl, kurr, skafl, o.s.frv. Lysingarorð t.d.: hvass, laus, sagnorð t.d. les, vex. [Nominative. All masculine words which end in -ar. annar ('other'), hamar ('hammer'), Gunnar (personal name) etc. Some masculine ja-stem words with short roots: byr ('favourable wind'), bær ('farm'), gnyr ('din'), her ('army') etc. Many ustem words: ár ('year'), ás ('ace'), björn ('bear'), knörr ('ship'), son ('son'), vin ('friend'), örn ('eagle') etc. Many a-stems: akur ('field'), fugl ('bird'), hals ('throat'), hrafn ('raven'), is ('ice'), karl ('man'), kurr ('grumbling') skafl ('snowdrift') etc. Adjectives: e.g. hvass ('sharp'), laus ('loose'), verbs: e.g. les ('reads'), vex ('grows').] (Guömundur Kamban 1908:173)
It is not clear why these last present-tense verb forms are included in the paragraph on nominatives. The stem classes are in general irrelevant to the assimilations which lead to the absence of an overt nominative ending, but are presumably cited so as to lend a scientific air to the presentation. Guömundur Kamban goes on to cite the example of the possessive pronouns okkarr ('our'), ykkarr ('your (dual; now plural)') and ydvarr ('your (plural)') which have become indeclinable (or, more technically, the adjectival possessive pronouns replaced by the genitive forms of the personal pronouns) on the way from Old to Modern Icelandic. Finally, he mentions the lack of a genitive ending on the foreign title frú ('Mrs') "til fru Guörúnar, í stað: til frûar G" ["to frú Guörúnar, instead of: to frûar G"] as evidence "aö malið veröur pví óbrotnara aö öllum beygingum, pví meir sem paö proskast og lengra líður fram" ["that the language will become less fragmented by all inflections, the more it develops with the passage of time"] (Guömundur Kamban 1908:174).
142
RENDRA WILLSON
The dominant position in Iceland, however, was to view loss of inflections, not as evolution, but as decay. Guömundur Kamban's praise of less inflected languages offends, among others, Johannes L.L. Jóhannsson (1909:62), whose response (1909) to Guömundur Kamban (1908) appeared in the following issue of Skírnir. 4. Johannes L.L. Jóhannsson In contrast to Guömundur Kamban, Jóhannes Láms Lynge Jóhannsson (1859-1929) spent his entire life in his home country. He trained to be a pastor, and served as such in rural Iceland before moving to Reykjavik with a grant from the national treasury to compile an Icelandic dictionary.3 J≤hannes L.L. Jóhannsson contributed to the Swedish journal Arkiv för nordisk filologi a response (1926), in Icelandic, to Björn K. pórólfsson's (1926) review of Johannes L.L. Jóhannsson's (1924) book, but the bulk of Jóhannes's works were published in Iceland. These include a monograph on phonological changes between Old and Modern Icelandic (1924), which remains an insightful and useful handbook. Johannes L.L. Johannsson was thus fully aware that the Icelandic language had indeed changed since medieval times. At the same time, he endorsed the Romantic view that the language's morphological conservatism was central to its beauty and dignity. Johannes L.L. Johannsson's (1909) reply to Guömundur Kamban (1908) contains numerous examples of patriotic rhetoric about the ornateness and antiquity of the Icelandic language. He takes exception to Guömundur Kamban's suggestion that Icelanders look to Norway for an example of a rich and varied system of surnames, because Norwegian is a daughter to Icelandic, not a sister: hence a less "mature" language, rather than, as in Guömundur Kamban's view, a more "developed" one. Að vitna f norskuna og önnur beygingarsnauð mal dugir ekki f pessum efnum. pað er systerni a milli nynorðrænu mâlanna, t.d. norsku og sænsku, en alls eigi milli norsku og íslenzku, pví pær eru mæðgur; hin fyrri er dóttirin en hin síðari móðirin og myndríkt fornaldarmál, er eigi verður borið saman við hitt, með petta. [It will not do to cite Norwegian and other inflectionally impoverished languages in these matters. There is sisterhood among modern Scandinavian languages, for example, Norwegian and Swedish, but not at all between Norwegian and Icelandic, for they are mother and daughter: the former is the daughter and the latter the mother, an ornate, archaic tongue, which cannot be compared to the other in this.] (Jóhannes L.L. Jóhannsson 1909:59-60)
3
http://www.aknet.is/oskarutd/kirkja/prestar/sud/jllj.htm
POLITICAL INFLECTIONS
143
Johannes L.L. Jóhannsson (1909) would prefer to see Iceland retain the traditional patronymic system. Given, however, that it seems that Icelanders will not be dissuaded from the fad of taking surnames, he proposes a few types of formation which he feels will be more harmonious with the structure of the language than those cited by Guðmundur Kamban (1908). Jóhannes L.L. Jóhannsson (1909) cites the short length of Guömundur Kamban's examples as a flaw in their design. He seems, perhaps naively, to assume that the morphological mutilation seen in Guömundur Kamban's proposed name types is a consequence of the attempt to keep them short, rather than the opposite. pessi 40 synisdæmi hans hafa pann gallann, að pau öll eru ein- eða tvikvœð, en foröast aö nefna prí- og ferkvœðu orðin. En slíkt gefur beint efni til ófslenzkulegra styttinga og afbakana a íslenzkum orðum og færir pví niður f sama feniö sem gömlu ættarnofnin hér a landi. Maöurinn yröi t.a.m. nefndur Finnur Fróðstad fyrir F. Fróðastadur, Björn Bíldal fyrir B. Bildudalur og annaö pvílíkt. Mega ní allir sjá, aö hiö fyrra er ljót útlenzka, en hiö sföara hrein íslenzka. [These 40 examples of his have the flaw that they are all mono- or disyllabic, while he shies away from tri- and tetrasyllablc words. This leads directly to un-Icelandic truncation and deformation of Icelandic words, and thus leads to the same mire as the old surnames in this country. A man would be called, for instance, Finnur Fródstad instead of F. Fróðastaður, Björn Bildal for B. Bildudalur, and such like. It is plain for all to see that the former is ugly outlandish, the latter pure Icelandic] (Jóhannes L.L. Jóhannsson 1909:58)
Johannes L.L. Jóhannsson proposes that Icelanders should look elsewhere than to Denmark, Germany or the English-speaking world for models of surname formation and use. Fully-inflected nouns and adjectives of all genders have been used as by-names in Iceland throughout its history, and Jóhannes L.L. Jóhannsson (1909:63) feels that such terms as Ljónshjarta ('Lion's heart') would make suitable Icelandic surnames. He also (1909:64) commends the Russian practice of using patronymics in addition to surnames. Jóhannes L.L. Jóhannsson suggests that Icelanders look to the ancient Romans for a model of how to form and use surnames in an inflected language. Rómverjar eru sü pjóö, er fyrst kom ættarnöfnum í fast skipulag, enda eru ættarnöfn peirra bæði fögur og pjóðleg. pau eru alveg samvaxin tungumáli peirra og hafa fulla beygingu sem hver önnur latnesk orð. par er pví agætt dæmi, sem peim ber aö taka til fyrirmyndar og samanburðar, er koma vilja ættarnöfnum inn í íslenzkuna, en vera eigi aö vitna í pað, sem alls eigi a viö. [The Romans are the nation that first systematized surnames, and their surnames both are beautiful and have national character. They are fully integrated into their
144
KENDRA WILLSON
language and have full declension, like any other Latin words. This is thus a fine example, which those who wish to bring surnames into Icelandic should take as a model and comparandum, rather than citing things which are irrelevant.] (Jóhannes L.L. Jóhannsson 1909:62)
This suggestion implies that Icelanders and Romans constitute a natural class, ancient peoples who speak languages of dignity and civilization. Bjarni Jonsson frá Vogi (1923:100-101) takes exception to Johannes L.L. Jóhannsson's (1909:62) suggestion that an onomastic system as sexist as the Roman one, in which women bore no names of their own, would be a suitable role model for the Icelanders. In general, however, proposals like Jóhannes L.L. Jóhannsson's seem to have been largely ignored. Those who were involved in creating surnames wanted them to be indeclinable. They did not consider name formation in relation to Roman or Slavic models, and only marginally in connection with Old Norse by-names, but rather in terms of exactly the nations Jóhannes L.L. Jóhannsson (1909) views as unsuitable comparanda because their languages are "inflectionally impoverished": Mainland Scandinavia, Germany, England and North America. 5. Islenzk mannanöfn (1915) A law regulating personal names, chiefly concerned with surnames, was passed by the Alpingi in 1913 (Islenzk mannanöfn 1915). This required permission from the government for an individual to adopt a surname (1915:2), and levied a charge of two crowns per surname in addition to the ten-crown fee for a name change (1915:4) (there is no per-syllable charge, nor an annual tax to the name-bearer). A surname adopted before January 1 1901 "telst viðurkend eign pess, er paö hefir notað" ["is regarded as the recognized possession of the one who has used it"] (1915:3), whereas those taken up since 1901 would only be acknowledged if royal permission were obtained (1915:4). The law also specifies that the government should have compiled and published a "skrá yfir orð og heiti, sem fallin pykja til að hafa að ættarnöfnum" ["register of words and terms which are regarded as suitable for use as surnames"] (1915:3). The committee appointed to construct this register consisted of the author and editor Einar Hjörleifsson (later Kvaran); the philologist, essayist, editor and (later) national librarian Guômundur Finnbogason; and Pâlmi Pálsson (1857-1920), who was head teacher at the Reykjavik school and produced some editions of medieval Icelandic texts (Pall Eggert Ólason 1948-1976, vol. 4 (1951):150). Einar Hjörleifsson Kvaran (1859-1938) belonged to a progressionist, internationalist paradigm similar to that represented by Guômundur Kamban. Einar H. Kvaran likewise spent a substantial amount of time abroad, residing
POLITICAL INFLECTIONS
145
for five years in Copenhagen and ten in Winnipeg before returning to Iceland in 1895. Einar was influenced by reading Darwin and Huxley (porsteinn Jonsson 1957:342). Like Guömundur Kamban, Einar H. Kvaran assumed a surname based on a Celtic by-name attested in Old Norse literature (cf. Gils Guömundsson 1997:218). Guömundur Finnbogason (1873-1944) was a prominent figure in many areas of Icelandic life and letters (cf. Guömundur Gíslason Hagalín 1966, Stefán Einarsson 1943). While Guömundur was involved in the linguistic purism movement (cf. Baldur Jónsson 1976), Stefan Einarsson (1948:170) feels that "In Guömundur Finnbogason's work the emphasis was less on preserving than on improving the language". Guömundur was active in the creation of neologisms. His essay "Hreint mal" ["Pure Language"] (1928) advocates the use of transparent native terms over opaque foreign ones. At the same time, I find in Guömundur Finnbogason's work a librarian's devotion to inclusiveness and descriptive thoroughness, which extends even to aspects of language use of which Guömundur does not necessarily approve. Another essay, "Bölv og ragn" ["Cursing and Swearing"] (1927), concerns profanity in Icelandic. While Guömundur Finnbogason states up front that he feels that Icelanders curse more than they should, he nonetheless views these blasphemies as data worth recording and cataloguing. It would thus be consistent for Guömundur to accept the challenge to design a system of surnames in accordance with demand, independent of whether he personally felt such names to be necessary or desirable. The 1915 report Islenzk mannanöfn [Icelandic Personal Names] contains many suggestions and arguments which seem bizarre and "un-Icelandic", and which lent themselves to biting satire (e.g. Arni Pálsson 1947; Bjarni Jonsson frá Vogi 1923; cf. discussion in Baldur Jonsson 1976:30-40). This strangeness does not stem from ignorance of Icelandic philology, but almost the opposite. Language-internal parallels are found in the effort to represent implausible generalizations as having precedents within the linguistic system, even while different historical stages of the language and components of the lexicon are deliberately conflated. The aim is to produce indeclinable, disyllabic surnames which are underspecified for gender, while justifying these as being consonant with the general patterns of the language. The report states outright that "aö vorum dómi væri pað lang-æskilegast, aö ættarnöfn væru alveg óbeygjanleg" ["in our judgement it would be by far the most desirable for surnames to be completely indeclinable"] (1915:13). It does not provide a motivation for this view, but simply justifies the idea of indeclinable nouns with a language-internal parallel: "í mâlinu er til fjöldi óbeygjanlegra nafnoröa, sem engan óskunda hafa gert og hver maöur kann aö
146
KENDRA WILLSON
nota" ["the language contains a large number of indeclinable nouns, which have done no damage and which everyone knows how to use"] (1915:13). The report is conspicuously silent on the role of models from Mainland Scandinavian or elsewhere. Some of the proposals for types of indeclinable surnames represent clever linguistic argumentation. The committee suggests that surnames be formed from non-compound place names or masculine personal names by means of the indeclinable suffix -an, hence Snorran from the personal name Snorri, Bakkan from the place name Bakki. This ending is ostensibly taken from such adverbial formations as héðan ('hence, from here') and nordan ('from the north'), in which the suffix seems to denote movement from or origin at a place. Both Ami Pálsson (1947) and Holger Wiehe (1917) find the extension of this -an to surnames implausible; names are not adverbs. A secondary parallel for the use of -an is found in medieval by-names such as bjólan, kamban and keikan. These names are probably Celtic, and a type which apparently disappeared in Iceland shortly after the settlement (cf. Finnur Jónsson 1908), until their revival nearly a thousand years later by such persons as Guðmundur Kamban. Masculine nominals in -ann ( < EE. *-onos) such as pjóðann ['prince, ruler'] are anetymologically conflated into the mix. The committee further endorsed some suffixes which do not even assume an Icelandic form, but a hybrid of Icelandic and Proto-Germanic. Surnames based on names of the weak feminine declension are to be formed with -on, the Proto-Germanic form corresponding to Icelandic -a; hence Tungon from Tunga. The base for surnames derived from place-names in -fjördur is -fer, the Germanic root found in Icelandic fjördur. Place names in -stadur or -stadir are to be rendered by surnames in -star, where -sta is the root in stadur and -r "algeng afleiðsluending f málinu" ["a common derivational ending in the language"] (1915:12). The committee also suggests that the dative and genitive forms of place names be adopted as new nominatives. They cite the Icelandic convention of addressing an envelope with the addressee's name in the nominative followed by the address in the dative. Utanáskriftin er t.d. Eirfkur Ólafsson Brúnum eöa Magnus Sigurðsson Skardal. Að taka bæjarnafnio f pâgufalli og hafa aö ættarnafni, er pà ekki annaö en aö auökenna sig f ræðu og riti eins og maöur er auðkendur a bréfum. [the letter is addressed e.g. Eirfkur Ólafsson Brûnum [i.e. at Brunir] or Magnús Sigurösson Skardal [in Skardalur]. To take the farm name in the dative and have it as a surname is simply to identify oneself in speech and writing as one is identified on letters.] (1915:11)
POLITICAL INFLECTIONS
147
Such names based on dative forms would naturally be indeclinable. Indeclinable proper names transparently based on oblique case forms are known from elsewhere, e.g. Medieval Latin Hiberione ('Ireland'). Hence this is a possible linguistic development, but it is not one which had occurred in Icelandic. Many place names in Iceland are compounds; often the first element appears in the genitive case. The committee proposes that such place names can be turned into surnames by replacing the genitive compounding with stem compounding and truncating the name to two syllables, one from each of the members of the compound, "með pví að sleppa miðhluta peirra eða endingu fyrri hlutans" ["by omitting the middle section, or the ending of the first part"] (1915:12), hence Jakdal for Jökuldalur. The result is thus not dissimilar to older names like Blöndal < Blöndudalur which might appear to be derived by "Danicization". The report emphasizes, however, that "höfum vér farið par að dæmum mâlsins sjálfs" ["there we have followed the language's own example"] (1915:12), viz. certain demonyms in -ingur and -dœlir, such as Súgfirdingur for a resident of Súgandafjörður. This type of templatic truncation of compound names is, however, restricted to these particular suffixes, and appears no longer productive even with these (cf. Willson 1995). Hence the language-internal model mentioned in the report is far more marginal and less salient than the corresponding Mainland Scandinavian pattern of surnames, which they do not cite. This latter is described by Tegnér (1930:51-52): Ett sammansatt namn får ej innehålla mer än två sammansättningsleder [...] Beståndsdelarna i namnet få ej vara förbundna genom ett genetiviskt -s- [...] De både i sammansättningen ingående orden maste, med inskränkningar [...] vara eller göras enstaviga. [A compound name may not contain more than two compositional elements [...] The components of the name may not be joined by a genitive -s- [...] Both of the words which form the compound must, with some restrictions [...] be or be made monosyllabic] (Tegnér 1930:51-52)
To be sure, the Mainland Scandinavian surname compounds did evolve in languages closely related to Icelandic. It would require further investigation to determine to what extent the development is dependent on the loss of inflections in the Mainland Scandinavian languages, and to what extent morphological categories such as those cited in Islenzk mannanöfn (1915) as possible analogical models played a role in the development of these name types in Mainland Scandinavia.
148
KENDRA WILLSON
The committee cites a syllable limit as part of the justification for truncation: Vér höfum eigi myndað lengri ættarnöfn en tveggja atkvæða. pau nöfn pykja oss fegurri og pjálli meöferöar, enda nógur úrkostur ættarnafna, pó að eigi séu höfð lengri. [We have not formed surnames longer than two syllables. We feel these names to be prettier and easier to use, and there are plenty of surnames to choose from even if they are not made longer.] (1915:12)
This may relate to the association of name length with ostentation seen in the first proposed name law (Frumvarp 1881), but it also correlates with indeclinability and stem truncation. The committee suggests that surnames ending in -son be formed by stem rather than genitive compounding, hence in many cases to contain a single s rather than ss, to aid in distinguishing them from patronymics. Both stem and genitive compounding are common in Icelandic, so that this is a comparatively structure-preserving piece of linguistic planning. However, it does not seem to be a contrast which would have been likely to emerge in the language spontaneously. The immediate model here, in addition to Danish surnames in sen, is probably names in -son which are common in the English-speaking world. Many Icelanders in Canada and the U.S. mutated their patronymics into surnames of this type. The report directly confesses its international orientation in the discussion of the letter p (thorn, representing the voiceless dental fricative /0/). The committee avoided creating names which contained this letter, as it was not used in other contemporary Western European languages. Here, no justification from the structure of Icelandic is possible. Vér höfum engin ættarnöfn latiö hafa p i sér, pvi aö fiestum mun óljüft aö veröa aö breyta stöfum f nafni sfnu, ef peir hafa einhver viöskipti viö önnur lönd, eöa pá fa pað afbakaö. Ütlendir menn pekkja langfæstir p, og stafa pví p fyrir p f íslenzkum nöfnum, t.d. Porkelsson fyrir porkelsson. pess vegna hefir p í ættarnöfnum hingaö til veriö breytt i th, sem er alveg óíslenzkt. Og pótt pað sé aö visu satt, aö ættarnofn flestra Islendinga mundu aö jafnaöi aö eins hljóma hér a landi, pá hafa líklega fæstir pann metnaö, aö ættar peirra veröi urn aldir ada aldrei aö neinu getiö fyrir utan landsteinana. [We have not formed any names which contain p, because most people will find it unpleasant to have to change letters in their name, or else have it distorted, if they have any dealings with other nations. Most foreign people do not know the letter p, and thus write p for p in Icelandic names, e.g. Porkelsson for porkelsson. For this
POLITICAL INFLECTIONS
149
reason, p in surnames has hitherto been changed to th, which is completely unIcelandic. And though it is admittedly true that the surnames of most Icelanders will for the most part only resound in this country, nonetheless probably few have the ambition that their line will in the course of history never be mentioned outside the border.] (Islenzk mannanöfn 1915:13)
Islenzk mannanöfn (1915) aroused heated debate; some of the more exotic formation types were subject to ridicule. Arni Pâlsson (1947) objects that "petta er volapük, en ekki islenzka" ["this is Volapük, not Icelandic"] (1947:285). Bjarni Jonsson fra Vogi (1923) envisions an alternative history in which Iceland's original settlers and the heroes of medieval literature had in accordance with the committee's borne "Kleppskinnunöfn" recommendations. Hefði pórvarör pórarinsson verið kallaður Th. Önfer og porgerör Egilsdóttir verið nefnd Th. Brundebjalvesen, eða Ingólfr Arnarson kallaöur I. Sunnfer, eöa Olafr feilan nefnzt O. Breiöfer eöa O. Hvammon eöa pá O. Thorstensen, eöa heföi bræöur Unnar djúpügöu veriö menn eignarfallsins og heitiö Flatnefs, pá heföi landar vorir eigi purft aö fyrirveröa sig fyrir nöfn sín. [If pórvarör pórarinsson had been called Th. Önfer and pogerör Egilsdóttir been named Th. Brundebjalvesen, or Ingólfr Arnarson called I. Sunnfer, or Ólafr the shy been named O. Breiöfer or O. Hvammon or O. Thorstensen, or if the brothers of Unnr the deep-minded had been men of the genitive and been called Flatnefs, then our countrymen would not have had to be ashamed of their names.] (Bjarni Jónsson frá Vogi 1923:98)
Nor did these suggestions prove popular. While a number of people adopted -an names, only one person took up a dative form as a surname (Studium) during the years that the 1913 law remained in effect. Only two names in -on were adopted (Blandon and Maron), one in -fer (Gilfer < Gilsfjördur) and none in -star (Svavar Sigmundsson ms.). Those name-types which multiplied were largely the patterns that had been in existence before the report, such as disyllabic names in -dal. Àrni Pâlsson (1947) takes the fact that a committee of experts has not been able to do better by the task of designing Icelandic surnames than the bizarre proposals in Islenzk mannanöfn (1915) to indicate that the language is inherently unsuited for the creation of surnames. Eg vona, að augu allra munu nú opnast fyrir pví, að pað muni ekki auögert aö smíöa ættarnöfn viö hæfi íslenzkunnar, pegar einum lærðum malfræðingi og tveimur velmenntuöum og venjulega smekkvísum rithöfundum hefur ekki tekist betur en petta.
150
KENDRA WILLSON
[I hope that the eyes of all will now be opened to the fact that it will be no easy task to form surnames suitable for the Icelandic language, when one learned linguist and two well-educated and generally tasteful writers have not succeeded better than this.] (Àrni Pálsson 1947:288)
The difficulty arose because the committee was not in fact devoted to the task of forming surnames consistent with the Icelandic language. Rather, it aimed to justify as consonant with the structure of the language surnames which were felt by many to be desirable specifically for their relation to other linguistic and onomastic systems. 6. Conclusion The morphological gyrations undertaken by the surname advocates in order to prove the Icelandic character of indeclinable surnames did not succeed in dislodging the general understanding that patronymics were the native tradition, surnames a foreign innovation. The subsequent name law, which was ratified in 1925 and remained in effect until 1991, flatly forbade the adoption of new surnames: "Ættarnafn má enginn taka sjer hjer eftir" ["Henceforth no one may assume a surname"] (Guðrun Kvaran and Sigurður Jonsson frá Arnarvatni 1991:78). The surname opponents had won, for the time being. Though the name law was not infrequently violated, the patronymic system seems in general to have survived the rapid population increase, urbanization, internationalization and technological revolutions of the 20th century. Nor has the existence of a surname-bearing minority, hence a dual system, proved as unsustainable as Jon Jonsson and Jon Olafsson had predicted (Frumvarp 1881:2). However, the more liberal personal name policy of 1996 makes provision for an Icelander to bear a millinafn ('middle name') in addition to one or two first names.4 Millinöfn are not used in place of patronymics but in addition to them, nor are they automatically inherited. However, like fixed surnames, millinöfn can be borne by both men and women. They are to be derived from Icelandic stems, but may not contain nominative endings. The millinöfn which have been registered since this law took effect are generally similar to the fixed surnames already in existence in Iceland (Svavar Sigmundsson ms.). The indeclinable, gender-neutral family name may have been readmitted by the back door.
4
http://www.althingi.is/lagasofn/nuna/1996045.html
POLITICAL INFLECTIONS
151
References Ârni Pálsson. 1947. Um ættarnöfn. (Erindi flutt fyrir Al]pyðufræðslu Stúdentafélagsins 1916). Á víð og dreifi Ritgerdir by Ârni Pâlsson, 269290. Reykjavik: Helgafell. Baldur Jónsson. 1976. Mâlyrkja Gudmundar Finnbogasonar. Reykjavik: Bókaútgafa Menningarsjóðs. Bjarni Jónsson fra Vogi. 1923. Nöfnin. Skirnir 97. 96-106. Björn K. pórólfsson 1926. Jóhannes L.L. Jóhannsson. Nokkrar sögulegar athuganir urn helztu hljóðbreytingar o. fl. í fslenzku, einkum í miðaldamalinu (1300-1600). Reykjavik 1924. Arkiv för nordisk filologi 42, N.F. 38. 77-81. Finnur Jónsson. 1908. Tilnavne i den islandske oldlitteratur. Særtryk af Aarbøger for nordisk Oldkyndighed og Historie, 1907. Kjøbenhavn: H. H. Thieles Bogtrykkeri. Frumvarp til laga, um nöfn manna. 1881. Reykjavfk: Prentsmiðjan Isafoldar. Gils Guðmundsson. 1997. / nœrveru sálar. Einar Hjörleifsson Kvaran. Maðurinn og skâldid. Reykjavik: Setberg. Guðmundur Finnbogason. 1927. Bölv og ragn. Skirnir 101. 48-61. Guðmundur Finnbogason. 1928. Hreint m l . Skirnir 102. 145-155. Gumundur Gfslason Hagalin. 1966. Guömundur Finnbogason. Merkir Íslendingar, Nyr flokkur, 5, 221-243. Reykjavik: Bókfellsütgafan. Guömundur Kamban. 1908. Ættarnofn. Skirnir 82. 164-177. Guörun Kvaran & Siguröur Jonsson frá Arnarvatni. 1991. Nöfn Íslendinga. Reykjavik: Heimskringla. Halldor Halldórsson. 1979. Icelandic Purism and its History. Word 30. 76-86. Holger Wiehe. 1917. Enn urn ættarnofn a Islandi. Skirnir 91. 286-295. Íslenzk mannanöfn. Lög, nefndarálit og nafnaskrâr 1915. Gefiö út aö tilhlutun Stjórnarraös Islands samkvæmt lögum nr. 41, 10. nóv. 1913. Reykjavik: Prentsmiöjan Gutenberg. Jóhannes L.L. Jóhannsson. 1909. Um ættarnöfn. Skirnir 83. 53-64. Johannes L.L. Jóhannsson. 1924. Nokkrar sögulegar athuganir um helztu hljóöbreytingar o.fl. i islenzku, einkum i miâaldarmâlinu (1300-1600). Reykjavik: Prentsmiöjan Sigfüsar Eymundssonar. Jóhannes L.L. Jóhannsson. 1926. Svar. Arkiv för nordisk filologi 42, N.F. 38. 276-280. Kjartan G. Ottósson. 1990. Íslensk mâlhreinsun. Sögulegt yfirlit. (= Rit Íslenskrar mâlnefndar, 6.) Reykjavfk: Islensk málnefnd. Kristinn E. Andrésson. 1949. Íslenzkar nûtímabókmenntir 1918-1948. Reykjavfk: Mal og menning. Páll Eggert Olason, ed. 1948-1976. Íslenzkar œviskrâr frâ landnâmstimum til ârsloka 1940. Reykjavfk: Hi Islenzka bókmenntafélag.
152
KENDRA WILLSON
Stefan Einarsson. 1943. Gumundur Finnbogason sjötugur. Tímariî PjódrœknisfélagsIslendinga 25. 44-61. Stefan Einarsson. 1948. A History of Icelandic Literature. New York: The Johns Hopkins Press for The American-Scandinavian Foundation. Svavar Sigmundsson. N.d. Urn ættarnöfn a Islandi. Ms., pjóminjasafn Islands. Tegnér, Esaias. 1930. Svenska familjenamn. Ur språkens värld. Fem uppsatser by Esaias Tegnér, 147-158. Stockholm: Albert Bonniers förlag. Tomasson, Richard F. 1975. The Continuity of Icelandic Names and Naming Patterns. Names 23. 281-289. Willson, Kendra. 1995. Kartafla, eða íbüaheiti a Islandi. Paper presented at the RASK conference, Reykjavik, 28 January 1995. porsteinn Jonsson. 1957. Einar H. Kvaran. Merkir islendingar: ævisögur og minningargreinar, ed. by porkell Jóhannesson, 6, 336-364. Reykjavik: Bokfellsútgâfan.
STANDARDIZATION, LANGUAGE CHANGE, RESISTANCE AND THE QUESTION OF LINGUISTIC THREAT 18TH-CENTURY ENGLISH AND PRESENT-DAY GERMAN
PETER HOHENHAUS University of Freiburg, Germany 1. Introduction This paper aims at being comparative in at least two ways. First, on a historical level, and second interlingually: developments in 18th-century English will be compared with present-day developments in German. Perhaps I should say England and Germany — in so far as some of the discussion goes beyond linguistic questions proper and includes cultural and political aspects pertaining to the two countries. Furthermore, the paper will be comparative in an intralinguistic sense, as it were, looking at different linguistic levels of English and, more so, German, especially spelling, lexis, semantics and grammar. As regards the questions of purism and resistance, several parts of this paper are of a sociolinguistic nature. However, we will also look at issues that go beyond that, namely issues of language change. In particular when it comes to claims of linguistic threat, or even language death, these have to be checked against findings of linguistics about the actual nature and mechanisms of language change. Given the scope of these issues, I must be selective as regards the choice of linguistic details and depth of analysis. A broad overview will be provided but I will restrict myself to picking out only samples from the huge array of data and arguments, especially as far as English is concerned. I will place more emphasis on issues concerning current German. 2. Standardization and purism It is useful first to distinguish two possible readings of the term 'standardization', a) as a process noun, and b) as an action noun. Standardization in the first reading, i.e. the issue of how Standard English has
154
PETER HOHENHAUS
developed (if the very concept is accepted at all),1 has recently attracted much attention from various angles (cf. Wright 2000; Bex & Watts 1999; and also several other papers in the present volume). Here, however, only the second reading of 'standardization' will play a role: aspects of how a standard is, or can (or should) be, brought about or maintained by deliberate means; i.e. by language planning, by prescriptivism in language criticism, and in particular: purism. Note that I will often use the word 'purism', not in its most literal sense, i.e. as efforts to 'purify' a language from foreign elements, but in a rather wide sense to include other aspects of language criticism and prescriptivism as well. Inherent in purism, in that wide sense, may be a certain 'ideology' (cf. Milroy 2000) - although this may have the connotation of 'conscious set of beliefs', whereas I would hold that purist feelings about and attitudes towards language (at least in 'folk linguistics') are often rather hazy. However, there is a recurrent set of mostly tacit assumptions discernible in prescriptive/ purist writing, and it is to these that we will turn next. 2.1 Conservation, purism, prescriptivism — an age-old tradition It must be pointed out at the outset that, although standardization in English reached particular force in the 18th century, many of the underlying assumptions go back to the very beginnings of Western writing about language.2 Thus, the oldest systematic grammar in the European tradition that has come down to us in complete form (short as it is, though, comprising a mere 400 lines), namely Dionysius Thrax's Techne Grammatike was written around 100 BC as a purist response to perceived language decline, seen as corrupting the language so much that the old classics of Greek literature might become unintelligible if the development was not halted (cf. Weite 1985:45ff, 128). Similar attitudes towards language in general persist to this day and are usually present implicitly in the modern 'complaint tradition' (cf. Milroy & Milroy 1999:24ff), especially in lay people's criticisms about language.
1 Indeed there is some debate about whether there is such a thing as 'the Standard' or whether this is rather an ideological myth in itself. It is beyond the scope of this article to pursue this issue; instead I refer the reader to e.g. Bex & Watts (1999) and, in particular, the extensive critical review of that book by Pennycook (2000). 2 Outside Western Europe even older examples of such writing can be found, most importantly Panini's grammar of Sanskrit dating back to the 6th4th century BC, but these had no influence on the earlier development of the Western tradition of grammar writing.
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
155
From the perspective of grammar-writing in particular, Weite (1985:114155) collates the tacit assumptions that have dominated attitudes towards English through history; those that are relevant here as well include: (a) linguistic value judgements are possible and desirable: 'good' vs. 'bad' usage (b) speakers need guidance (preferably by some authority): normative rules have to inform people how they should speak/ write (c) the model of tradition (including in the most extreme version the notion of linguistic perfection): 'older' = 'better' (d) the primacy of written over spoken language: language = writing (e) confusion of language and classical logic: 'logical' vs. 'illogical' language (f) the word as the basic unit: language = words The first three aspects often combine to constitute what is known in sociolinguistics as the 'Golden Age' assumption (cf. for example Milroy & Milroy 1999:40; Watts 2000:35). We may want to add another separate item to the above list: (g) the model of the language of literature: literary language is 'exemplary', it is the 'highest' form of language. This is also known as the 'best speakers/ 'greatest writers' tradition of prescriptivism. In fact it is a combination of assumption (d) with (a) and (b), sometimes with the addition of (c), i.e. the written language of literature (in particular older classics) is seen not only as good, but as the ideal form of language which speakers should aspire to in general. 2.2 18th-century calls for standardization of English and the role of prominent figures All of these basic assumptions played important roles during the stage in the development of English that is widely held to be the strongest phase of standardization and codification in the language's history. The 18th century was an age of standardization in general, especially for the sciences. Standards had to be laid down, so some authority to do so was in demand, and authority, in turn, is linked to the status of prominent figures of the time. In the next two sections I will concentrate more on the latter aspect, rather than on linguistic details of standardization. 2.2.1 Complaints about language change and calls for standardization and authority, especially by prominent writers. Prominent figures of the time regularly quoted include writers such as Jonathan Swift, Daniel Defoe, John Dryden, Alexander Pope etc., who can be regarded as having had some
156
PETER HOHENHAUS
authority themselves,3 but their primary role in our context was that they voiced complaints about the state of language — often coupled with calls for authoritative measures, in particular some sort of official academy. The best known example of this, indeed "the great classic of complaints literature" (Milroy & Milroy 1999:27), is Jonathan Swift's Proposal for Correcting, Improving and Ascertaining the English Tongue, published in 1712 (in Bolton 1966:107-123). After complaining about "decay", "decline", "corruption" (p. 111) and "Barbarity" (p. 115) etc., he ultimately makes it clear that: "[...] what I have most at Heart is, that some Method should be thought on for ascertaining and fixing our language for ever" (p. 117). While such sentiments were thus largely directed against language change, i.e. were more purist than normative, an area where there was a much more real need for normative standardization was spelling. Before the 18th century, spelling in English was unregulated and seemingly chaotic, as decried, for example, in the following statement: we write by guess, more than any stated rule, and form every man his diction, either according to his humour and caprice, or in pursuance of a blind and servile imitation. (T. Stackhouse in Reflections on the Nature and Property of Language in General, on the Advantages, Defects, and Manner of Improving the English Tongue in Particular (1731:187), cited in Weite 1985:66)
Indeed, the fixing of English spelling during the 18th century is regarded by many as one of the key achievements during that time. As already indicated, the general call for standardization was often linked to calls for an authoritative institution to look after such measures: a language academy, like the Académie Française in France (founded in 1635) or the Accademia della Crusca in Italy (founded in 1582). Such calls began even before the start of the 18th century. Cf. for instance Daniel Defoe's An Essay upon Projects of 1697 (in Bolton 1966:91-101), in which he expresses high hopes for the impact of such an academy. The fact that in England no such academy ever came into existence to wield authoritative power has sometimes been attributed to the sense of liberty of the English; however, it may simply be that the need for such an institution had diminished thanks to the achievements of individuals' efforts (cf. Weite 1985:72; Wardhaugh 1999:81) — to whom we will turn next.
3
Even though some of these writers were, in turn, criticized themselves for 'improper' usage (cf. e.g. Fitzmaurice 2000:199).
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
157
2.2.2 Famous practitioners of standardization efforts. Indeed, a number of linguistic scholars were highly influential in 18th-century standardization efforts, i.e. by more directly setting standards, actually 'laying down the laws' of the language. Amongst these scholars, some achieved a status of considerable eminence themselves. Samuel Johnson must be named first and foremost here. His Dictionary of the English Language, published in 1755, is generally regarded as the most significant milestone in the emergence of an authoritative standard in English with regard to vocabulary and especially orthography. An important predecessor, however, was Nathan Bailey, who set the stage, as it were, for efforts to assess the vocabulary (and etymology) of English comprehensively (cf. Hughes 2000:254). In America, Noah Webster enjoyed huge success with his American Spelling Book published in 1783 (i.e. before his American Dictionary of the English Language of 1828, for which he is most famous). As far as prescriptivism in grammar is concerned, the most widely adduced example is that of Robert Lowth's A short Introduction to English Grammar, published in 1762. Many of the classic prescriptive notions about English can indeed be found in this particular grammar (e.g. that of double negation being illogical, or equivalent to an affirmation). However influential Lowth's grammar itself may have been, an important multiplying factor for the spread of its prescriptive stance was another grammar, namely Lindley Murray's English Grammar, adapted to the different classes of learners published in 1795, which was largely based on Lowth's and went to a staggering 200 editions over the following five decades or so (cf. Weite 1985:69f). The works of these scholars were not only successful and thus influential, they were sometimes quite explicitly regarded as authorities. Occasionally this went to almost bizarre extremes. With regard to Johnson, for instance, Lord Chesterfield, demanding "good order and authority", stated in 1754: We must [...] chuse a dictator. Upon that principle, I give my vote for Mr. Johnson to fill that great and arduous post. And I hereby declare that I make a total surrender of all my rights and privileges to the said Mr. Johnson during the period of his dictatorship. (cited in Weite 1985:67)
Appreciation of the 'standardizes' did not generally go quite that far, but it is clear that by the end of the 18th century the norms set by the work of Johnson, Murray and others were quite firmly established. This was facilitated by the fact that a mass of less well-known publications written to instruct people on grammar, spelling and style proliferated in the wake of the work of the 'big names', and also through the education system (cf. e.g. Fitzmaurice 2000:196). Arguably, the firm entrenchment of so many prescriptive myths, coupled with
158
PETER HOHENHAUS
the traditional questionable assumptions in 'folk linguistic' attitudes towards language, can indeed be traced to a large degree to these developments in the 18th century (cf. Pinker 1994:373ff). 3. Germany today — purism in old and new guises Despite a much more enlightened picture painted by modern linguistics, commonly held beliefs and assumptions about language amongst lay people (and occasionally not so 'lay' people) remain strikingly similar. And they can still play a similar role in public debates of linguistic matters. For English, this has been widely documented (e.g. Wardhaugh 1999 or Milroy & Milroy 1999 offer general overviews). At this point, however, I will make the big leap not only in time, but also across the Channel and turn to German. More specifically, two issues that have recently been publicly debated will be focussed on: the spelling reform and the anglicisms debate. 3.1 The spelling reform The German spelling reform, which was officially introduced in 1998 (allowing a transitional period until 2005 during which both old and new spelling are valid) after a long series of discussions and revisions, met severe resistance in the German public, albeit rather belatedly, after it was officially passed in 1996. The reform itself can be seen partly as an effort of increased standardization, as one of its aims was to reduce the number of exceptions and generally 'streamline' the rules of spelling with respect to more systematic phoneme-grapheme relationships, on the one hand, and a more consistent application of the morphemic principle of spelling on the other (although, admittedly, it also allows for more variation at the same time, especially regarding the spelling of loan words). The rejection of the reform by large parts of the public can thus be seen as resistance to increased standardization efforts by means of official, normative 'language planning' from above. This resistance did not stop at mere complaints but also included petitions and lawsuits. Aspects of the cultural politics of the reform and the resistance against it are well documented (see e.g. Johnson 2000). Many arguments against the reform displayed elements of purism: most generally, equation of the spelling system with 'the language', as well as fears about an imminent decline or limitation of expressive means, impairing 'logic' and so forth. This is discussed at some length in, for instance, Langer (2000). What I would like to highlight in our context is that prominent figures, in particular writers, were again involved in, or got drawn into, the discussion about the reform. In particular this happened in what the German news magazine Der Spiegel (issue 42, 14 October 1996, 262-275) called the
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
159
"writers' revolt" ("Der Aufstand der Dichter"). This somewhat exaggerated expression referred to the signing of a petition against the reform by a number of well-known writers, including Günter Grass, Martin Walser, Hans-Magnus Enzensberger and others. These were clearly considered to be the 'true' experts — as they were seen as the professional producers of the 'highest' forms of language. Again we can observe an instance of the traditional 'best writers' assumption. When asked about their rejection of the reform, however, most of these 'experts' displayed a profound lack of knowledge of what the reform was actually about or what it consisted of. Even some comments meant to qualify the criticism are often totally beside the point. Consider for instance the following remarks by Martin Walser (in an interview in Der Spiegel (loc. cit., p. 270): Bei uns im Süden hat man schon immer 'Gämse' gesagt und, unter hochdeutschem Druck 'Gemse' geschrieben. [...] Dem 'Hass' tut die Verschärfung gut. [We in the south have always said Gämse (= "chamois") but under pressure from High German had to spell it Gemse. The fact that Hass (= "hatred") is made harsher (voiceless) is a good thing.]4
This is bafflingly nonsensical: the spelling change from Haß to Hass has not had the slightest impact on pronunciation, nor could it have: the final devoicing rule of German generally rules out a non-"scharfes" <s>, i.e. a voiced pronunciation as /z/ is never possible word-finally, no matter what the spelling. Neither could the opposition <e> vs. reflect a difference in pronunciation in German short vowels in the word mentioned: in both spellings it would be [ε] anyway.5 At the same time, the members of the committees that drew up the reform were called, rather insultingly, a "bunch of self-appointed experts",6 implying they were not really experts at all (or at least not as much as the writers were 4
The translation is my own, as is the case for all translations in brackets in the remainder of this article. I am indebted to Sally McMullen, however, for some assistance in those translations, as well as for proof-reading the whole draft of this text. Any faults remain my own, of course. 5 Only in the corresponding long vowels is there a dialectal difference. The pronunciation /e:/ contrasting with /e:/, however, is assumed to have come from northern Low German dialects, but the loss of that opposition (i.e. the collapse of /£:/ to /e:/) has generally spread; it is retained only in Leseaussprache or when a salient distinction, such as sehe vs. sähe, is being emphasized, often even over-emphasized to /æ:/ (see Hakkarainen 1995:48ff). 6 "Eine Clique von selbsternannten Experten" (Hans Magnus Enzensberger in Der Spiegel, loc. cit., p. 266).
160
PETER HOHENHAUS
considered to be), despite being Germanist linguists, who not only had the relevant qualifications to justify their expertise but were also appointed for the job quite officially. Admittedly, the spelling reform does contain problematic aspects (cf. e.g. Augst et al. 1997), but those were hardly ever focussed on in the public debate. Instead, a wild mix of opinions and ultimately a pseudo-political resistance to the 'decreed' change as such was voiced. As this slowly calmed down, however, another linguistic topic took centre stage: 3.2 Fears concerning English influences on German While resistance to the spelling reform was mostly (said to be) directed against too much 'state interference' in the language, complaints about another development have rather taken the opposite stance: that too little is being done by authorities to maintain a standard, namely with regard to the strong influence of English on German. 3.2.1 The current anglicisms debate. Over the last few years, public debate about a perceived 'anglicization' of German has been increasing. A recent culmination of this debate was a call for legislation to protect the German language {Sprachschutzgesetz), which was instigated in particular by the Berlin senator for the interior {Innensenator) Eckehard Werthebach in the Berliner Zeitung, 29 January 2001. Other politicians getting involved in the argument include such prominent figures as Wolfgang Thierse, the President of the Federal Parliament {Bundestag), and even the Federal President Johannes Rau, who called the fashionable use of anglicisms "silly" and "stupid". This reflects a general popularization of complaining about the 'anglicization' of German. However, while the general complaints about the use of anglicisms and the state of the German language seem to be widely agreed upon, the suggestion of governmental legislation met instant opposition.7 In fact, it seems that once the call for legislation was made, this rather triggered a certain turn in the public argument, which could be summarized informally as: 'OK, anglicisms may be bad, but a language law? No, thank you!' As with the spelling reform, concrete authoritarian measures of language planning appear to be viewed by a large part of the public as going too far. On the other hand, the belief that there might be a need for something to be done, if not by means of laws, then, say, in education, still seems to be widespread. And the continued coverage of the topic in the media is clear 7 See e.g. the following newspapers: Süddeutsche Zeitung, 1 February 2001; Die Welt, 11 Feb.; Nordwest Zeitung, 14 Feb.; der Standard, 14 Feb.; Badische Zeitung, 17 Feb.; Lippische Landeszeitung, 28 Feb. (Links at http://vds-ev.de/presse/pressespiegel01-a.php).
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
161
evidence that there is increased public concern in Germany about the 'epidemic' spread of anglicisms. Another significant signal of public concern was the founding in November 1997 of the VDS, the Verein Deutsche Sprache ("Association for the German Language"); initially it had the more openly purist name Verein zur Wahrung der Deutschen Sprache ("Association for the Preservation of German"), but after some negative press it was re-named with the more neutral sounding shorter version. The association has grown to some 10,000 members, organized in several regional branches, and over the last few years has increasingly been raising its voice calling for resistance to anglicisms and annually naming and shaming a Sprachpanscher des Jahres ("language adulterator of the year"), a sort of anti-award for what the VDS sees as the worst and most unnecessary use of English in German. The VDS has certainly gained some recognition among the general public. Furthermore, its founder and chairman, Walter Krämer, even took part in a panel discussion at the 36th annual meeting of the IDS (Institute for the German Language — the 'official' Germanist professional linguists' research organization) in March 2000. The Berlin branch of the VDS formed a sub-association to mark the European Year of Languages 2001. This Berliner Sprachbündnis ("language alliance of Berlin") is particularly cynical in calling their target "BSE" — standing for "bad simple English" (sometimes also: "badly spoken English"), which they see developing in German. It must be noted that none of the members are in fact experts, i.e. Germanist linguists. Again, what seems to be seen as relevant is their status as prominent figures of public life. Amongst those who signed the Berlin 'alliance' are, apart from politicians, journalists and academics from non-linguistic fields, the well-known actor Michael Degen and the songwriter Reinhard Mey. The Berlin 'alliance's' declaration8 mixes the usual complaints about possible decline with the call for authority (with explicit reference to France's and Poland's language policies — without mentioning the very limited success those countries' measures have had), and in addition contains the bizarre notion of an 'intact' language shared by all citizens as necessary for a 'living democracy' [sic!].9 It argues that such language would serve as a Bindeglied ("joining link") between general language and specialist registers. This begs the question whether there had 8
Formerly (early 2001) at: www.vds-ev.de/aktuelles/berliner-sprachbuendnis/7thesen.htm. This is indirectly also an insult to countries like Switzerland where there are several official languages coexisting. Going by the VDS's argument, that would mean by implication that Switzerland cannot be a 'living democracy'. It would be interesting to see how this sort of argument would go down in that country, which, after all, claims to be one of the oldest and one of the most exemplary democracies in the world. 9
162
PETER HOHENHAUS
ever been such a 'linking function' and whether this would have anything to do with language being 'intact' (which is probably to be read as 'pure'). Moreover, it is rather Latin, not English, that characterizes technical registers. If German were in fact constitutionally enshrined as the common language for all (as the VDS demands), would that mean that science was prohibited from using anything else? This just shows how muddled the arguments can get. In yet another recent campaign, the VDS attempted to interfere more directly in politics, namely in the run-up to the March 2001 elections in the federal state of Baden-Württemberg. It approached all parties with a set of Wahlprüfsteine zur Entwicklung der deutschen Sprache ("election touchstones regarding the development of the German language"), which included such manipulative questions as "what measures could you think of to halt this cultural decline?" (the 'decline' being the increasing influx of 'unnecessary' anglicisms). Perhaps not all that surprisingly, the strongest support came from the right-wing Republikaner,10 while the Greens refused to reply at all, and the larger main parties, the Christian Democrats (CDU) and the Social Democrats (SPD) replied rather half-heartedly, partly admitting that the concerns raised were legitimate but also rejecting legislative measures. Only the Liberals (FDP) had their response made by a trained Germanist who, for instance, rightly pointed out that the high frequency use of anglicisms in advertising is not representative of usage in general, and that those anglicisms that are indeed 'unnecessary' or 'unintelligible' were, for precisely those reasons, unlikely to become fully established.11 Of course it must be noted that worries about too strong an influence of English on German are not at all new. Hilgendorf (1996) refers to works dating back a century (e.g. Dunger 1909). The complaints may currently be resurgent in a particularly widespread fashion; however, the most strongly purist view I have so far encountered came, interestingly, not out of Germany but England, over two decades ago. Heald (1978), in his assessment of the state of German, unleashes the full blast of purist polemic, as in the following: [T]he German language [is] reflecting the general corruption of language in all walks of life, has in fact rejected the best of the old while assimilating the very worst of the new. It has [...] been poisoned by the all-pervasive spread of a spurious 'pop culture' [...] (Heald 1978:139)
10
It has to be noted, however, that the VDS has repeatedly stressed that it does not associate itself with right-wing politics, no matter how conservative its aims may be in linguistic matters. 11 Quotes from the replies by the parties were gathered (early 2001) from: www.vdsev.de/aktuelles/zum_ereignis.htm.
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
163
[T]he linguistic inflation in which too many empty words chase too few genuine ideas or intelligible concepts is an evil [...] this unattractive [...] pot-pourri of meaningless or misused [loans], its 'Anglicised' elliptical syntax, bears little, if any, resemblance to good German as it should be written or spoken. (Heald 1978:140)
In such statements we find hallmarks of prescriptivism/ purism par excellence'. the equation of certain linguistic surface phenomena with mental deficiencies, the equation of 'old' with 'good', and consequently 'new' with 'bad' (here even 'evil'), the linking of the alleged corruption in language with a general cultural decline (all these combined implying the 'Golden Age' leitmotif in purism, cf. section 2.1). When Heald (1987:140) complains about expressions that "convey nothing intelligible or nothing worth making intelligible," in particular "that repellent and, one hopes, ephemeral phenomenon 'punk rock'," it becomes obvious that his case is ultimately not about language at all, but about content and culture.12 Admittedly, the present-day anglicism debate only rarely approaches this sort of extreme purism. Nor can it be denied that there is a considerable influence of English on German, especially in as much as English is indeed exceedingly fashionable in some areas, such as advertising.13 However, whether or not this is tantamount to threatening the future of German in general is a question worth a closer look. 3.2.2 Dieter E. Zimmer's (1997) hypotheses concerning the future of German. One such claim, in my view the most influential such claim about German being endangered, is made by Zimmer (1997). He develops, at great length, a grim outlook for German — and does so on the basis of arguments that would at first appear to be more grounded in genuine linguistic reasoning rather than constituting 'blind' purism.14 However, Zimmer's (1997) arguments can be shown to be rather pseudo-scientific and often inconclusive in themselves. Moreover, behind the linguistic arguments, a good deal of (hidden) purism can 12 Heald's last remark about punk is, with the benefit of hindsight, especially ridiculous with respect to German, since the Punk and New Wave movement also sparked off a renewed use of German in pop music in the 1980s Neue Deutsche Welle. 13 The diminishing significance of German as a language of science must be noted here as well; see e.g. Ammon (1998). 14 It should perhaps be pointed out that Zimmer, although originally trained as a linguist, is not a professional in linguistics proper. As a journalist, he is a regular contributor to the German weekly Die Zeit, reporting on advances in various sciences; he is also the author of several books of a 'popular science' nature, as well as a literary translator. His 1997 book on the German language, however, and in particular the chapter referred to here, proved quite influential, not only among the general public but also within Germanist linguistics (cf. Eisenberg 1999:18).
164
PETER HOHENHAUS
be detected (which we will pick up in more detail in the next section). Zimmer's (1997:8) central hypothesis is the following: Die Sprachentwicklung [...] hat eine Richtung eingeschlagen, die den Fortbestand [...] des Deutschen [...] in Frage stellt. [The development of the language has taken a course which casts doubts upon whether German will continue to exist.]
This is a very strong claim. The phenomena under attack on which Zimmer rests his case include: (a) loan words such as Editorial, Joint venture, Weekend, Bike, Hotline, Trash Look: (b) hybrid word formation, e.g. Antiklau-Code ("anti-theft code"), Politthriller, Open-air Gefühl ("open-air feeling"), Mini-Abo Service ("mini subscription service"); (c) loan translations and changes in lexical semantics, e.g. ein Thema adressieren ("to address an issue"), kontrolliert fahren ("controlled driving"), in Plätzen wie Sarajevo ("in places like Sarajevo");15 (d) spelling of the genitive as in English, e.g. Ossi's Grill', or of compounds as separate words, e.g. Technische Unterstützungs Abteilung ("technical support department") (e) uncertainty in, or lack of, grammaticalization/ morphologicalization in particular in adjective inflection, e.g. ein easyes Leben, ein easyeres Leben ("an easy life, an easier life"), so that often inflection is dropped altogether, as in ein easy Leben', or in the area of verb inflection the uncertainty about the form of the past participle, e.g. backupen ("to back up") — backuped?, backupt?, gebackupt?, upgebackt? This is just a brief summary — for a comprehensive and detailed critical discussion of these points and other linguistic arguments of Zimmer's I refer the reader to Hohenhaus (2001); as regards the latter point see also Eisenberg (1999), and from a broader perspective on issues of loan word integration Eisenberg (2001). Here just a brief note: of the 'problems' listed above, only the last one relates to the grammatical system — however, these problems already exist within German (cf. ein super Leben vs. *ein superes Leben', or bergsteigen ("to mountaineer"): *berggestiegen, *gebergstiegeri), so they 15 In fact, these examples can rather be regarded as mistranslations (Zimmer widely uses examples from translated texts); however, similar phenomena of meaning transfer from related words in English have become more generally established in German, e.g. realisieren in the wider sense of English "to realize" ("perceive as real, notice"), whereas in the past it was restricted to the narrower sense of "convert (a plan etc.) into fact" only.
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
165
cannot be attributed to English influence. A further aspect worth mentioning in our historical comparative context is the fact that, say, the dropping of case, gender and number markings on adjectives may signal a change in grammar, but does not threaten the language's existence. After all, this is more or less exactly what happened in English long ago (case was largely lost somewhere between the 10th and 13th centuries during the development of Old English into Middle and Early Modern English). This has changed, but obviously not 'killed' English. Moreover, even if the dropping of adjective endings were to spread in German, this would probably not be enough to trigger the loss of the case, gender and number marking as a whole, as these would still be marked elsewhere.16 The main points of my criticism of Zimmer (1997) can be summarized as follows: (a) A large proportion of the phenomena Zimmer complains about only affect orthography, but not the German language. (Cf. section 2.1 — the same misunderstanding was found in complaints about the spelling reform.) (b) The extreme 'scare' examples often adduced by Zimmer are hardly representative of general usage. (c) A large proportion of Zimmer's examples of high-density anglicization are limited to certain specialized registers. (d) In particular, ('sloppy') translation must be distinguished from 'free' language use. (e) Such phenomena will not normally be available to children during the most critical phases of language acquisition, thus making Zimmer's worst-case scenario of damage to the structure of German through pidginization and subsequent creolization implausible. (This point will be taken up again in section 3.4.) Despite the flaws in Zimmer's arguments, a number of them have apparently become quite popular; in fact some of his terminology recurs in both the press and statements by the VDS. For instance, the latter's Berlin 'language alliance' uses precisely Zimmer's words, when it claims that German had lost its Assimilationskraft (roughly "capacity for integrating loans"). Likewise, Zimmer's odd term Tiefencode ("deep code"), strangely reminiscent of "deep structure" and/ or Bernstein's use of "(elaborated/ restricted) code", is gaining wider usage (cf. Eisenberg 2001:206; Hohenhaus 2001:4), even though it is far from clear exactly what it is meant to refer to. It 16
There are, however, sometimes prescriptive complaints about the genitive being lost in German; e.g. instead of das Auto meines Vaters ("my father's car") some registers/ dialects prefer mein Vater sein Auto (literally "my father his car"), without any overt case marking. But even that is still clearly German.
166
PETER HOHENHAUS
is often used together with a more common metaphorical style of describing change/ decline as e.g. an 'erosion' of the language system, that it is 'crumbling' or 'corroding'. In fact, Zimmer (1997), and perhaps even earlier remarks to the same effect that he had voiced in the weekly Die Zeit (also picked up in part by Ammon 1995), may well have been the original source of the now widely held belief that the very core of German, namely its Tiefencode is under threat. We will pursue this crucial point further in section 3.4. First, however, let us round off the discussion of purism by relating the traditional set of assumptions outlined in section 2.1 to the German anglicization debate of today. 3.3 Elements of prescriptivism/purism underlying the arguments Much of the public debate about anglicisms shows quite evident elements of purism, as indicated in section 3.1 above, and to some degree they can be detected in Zimmer (1997) too, albeit not as directly. However, even under the surface of supposedly objective reasoning, various characteristics of Zimmer's (1997) arguments are at least highly reminiscent of classic features of prescriptivism/ purism and familiar tacit assumptions: 3.3.1 Good vs. bad. The notion of 'good' vs. 'bad' language, and that 'good' usage should be supported, is probably the one most generally agreed upon. Even those who are critical of any legal or other official interference in language still concede that it is important to uphold 'quality of expression', especially in what these days is called 'literacy' in education, as well as to encourage clarity and avoidance of too uncommon words wherever possible (cf. the 'Plain English' campaign in Britain). In that sense there is some justification in the distinction 'good' vs. 'bad'. Often, however, it is employed in sweeping value judgements meant not to encourage any precise strategies, but rather to condemn the perpetrators of 'bad' anglicization. Cf., again, the VDS: "das ärgerlichste Kauderwelsch stammt von Leuten, die weder gutes Deutsch noch richtig englisch sprechen" ("the most annoying gibberish is produced by people who can speak neither good German nor real/ correct English").17 Zimmer (1997) also uses the notion of 'good' German, mostly in relation to literature (see section 3.3.5); occasionally it appears rather in the variant of häßlich vs. schön ("ugly" vs. "beautiful"). Zimmer (1997:82) concedes this is ultimately subjective, yet he strangely links it to the supposed 'deregulation' of German, claiming that in such a 'deregulated' state language was "urn die 17
This statement was part of the Berlin alliance's 'seven propositions' (cf. section 3.2.1).
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
167
Fähigkeit gebracht, schön zu wirken" ("deprived of the ability to convey an effect of beauty"). 3.3.2 Older = better. This next prototypical assumption in prescriptivism/ purism is implicitly present in all complaints about 'decline' — whatever metaphorical form these take (see section 3.3.9). Whether directly expressed or only implicit, the 'Golden Age' vein of purism is ubiquitous in current complaints by the VDS or the press about the development of German. Zimmer (1997), even though generally much more aware of the inevitability of language change, still occasionally alludes to the 'good old days' in a familiar way. If not linked to literature (see section 3.3.5), it takes the form of the notion of a geschichtliche ("historical") language, which might be lost. In this context, Zimmer (1997:85) ultimately even predicts that one day Germans will 'envy' the French or the Poles etc. for their efforts to preserve the good old state of their languages (through academies/ legislation). That seems rather unlikely. We do not normally envy speakers of languages at earlier stages of their development (and that's what 'preservation' actually means, preservation of a stage that has already passed) — speakers take their own, current language as the point of reference, despite perhaps some hazy admiration for the past. The notion of 'envy', however, appears questionable. 3.3.3 Language = words. Current language criticism in Germany (as in the past) generally focusses on words; most examples of 'anglicisms' given are words, and there is a call for equivalent words in German instead. Explicit complaints about loan translations or shifts in collocation etc. are much rarer. And where reference to language as a grammatical system can (rarely) be found, this is usually shrouded in hazy, general metaphors (see section 3.3.9). Zimmer (1997) refers to linguistic levels other than that of words somewhat more clearly. However, he too discusses words at far greater length; and the number of examples which are simple or complex words by far outweighs the number of examples of phonological, semantic or grammatical elements of anglicization. The notion of language as primarily words becomes particularly apparent when Zimmer (1997:21), discussing sentences in which virtually every content word is indeed anglicized, refers to the non-lexical elements as mere Füllmaterial ("filling material"). From the point of view of the language system, however, one could argue exactly the other way round: the grammar (inflection, word-order, function words), the syntactic grid that makes the sentence a German sentence, is still entirely German, only the 'slots' for content words are 'filled' with anglicized material. (That is: without the theoretical primacy of 'words', the picture appears much less dramatic.)
168
PETER HOHENHAUS
3.3.4 Language = writing. In most current purist comments, aspects pertaining to spoken language are largely absent. Occasionally a vague remark is thrown in to the effect that 'many people don't know how to pronounce many anglicisms', but that is usually it. As is traditionally the case, much more attention is given to questions of meaning and style — in written language. In Zimmer (1997) the predominance of written data is also striking. Mostly, these are not only taken from written language but are even restricted to particular text-types alone (such as advertising or 'localized' software manuals). However, he briefly raises the issue of pronunciation of English elements in German, namely the question of 'phonological code-mixing' in English-German hybrid word formations, where 'uncertainties' regarding the pronunciation arise. On closer inspection, however, his examples again turn out to be primarily written expressions that would only have to be pronounced if read out, which will normally not be necessary. It would have been more interesting to ask questions more systematically related to phonology. For instance: whether today's phonological system of German must be viewed as having been extended by originally English elements; e.g. whether the English diphthong in words such as rave or baby is integrated into present-day German or still replaced by /e:/; or what the status of English /r/ is in German, and so forth. (See Eisenberg 2001:186ff, and the literature cited there, for some discussion of questions of phonological integration.) Zimmer indicates the existence of phonological problems, but does not really address the linguistic issues — instead he often prefers to mock German pronunciations of English loans, e.g. of airbag, which most Germans pronounce in the way that Zimmer (1997:36) 'transcribes' as Ehrbeck. In such cases he appears to be opposed to phonological integration, in others his view remains rather unclear. In any case, he devotes much more space to written forms and, more directly, to questions of orthography. 3.3.5 'High' literature as a point of reference. When written language is the focus, literature is generally the point of reference for 'good' usage. In the German context, Goethe is a name commonly encountered. Even the linguist Harald Weinrich (2000:13), in criticizing the choice of the term Bachelor, for the three-year university degree being suggested in Germany, does so on the grounds that Baccalaureus could have been chosen instead because it is well attested in Goethe's Faust! The purist leitmotif of older literature as 'exemplary' can frequently be found in Zimmer (1997) as well. For instance, with regard to hybrid word formations (e.g. Mini-Abo Service "mini subscription service"), Zimmer (1997:24) claims that because they have "keine Geschichte" ("no history") they
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
169
were "noch auf lange Zeit für jede Literatur ungeeignet" ("unsuitable for any kind of literature").18 In this cultural criticism of anglicized German, Zimmer (1997:83) even goes so far as to indicate that for speakers of 'future German' there may actually be no demand for literature at all any more. The latter is simply cultural pessimism in its most extreme form. Here the 'Golden Age' myth comes back in a slightly different guise: not only is older language taken to be better, but also the people themselves, in as much as, allegedly, they used to be better able to appreciate literature. It is doubtful, however, whether there has ever really been a sizeable number of average speakers able to fully understand and appreciate the high literary forms of language adduced as exemplary. Zimmer (1997:84) himself predicts that a speaker of the German of the future may be just about able to understand a sentence in the works of e.g. Heine or Enzensberger, but: er wird nicht mehr in der Lage sein, zu erkennen, daß er gut war und was an ihm gut war, wird es weniger können als wir angesichts eines Satzes von Walther von der Vogelweide. [he will no longer be able to see that it was good, and what was good about it, less so than we could with respect to a sentence by Walther von der Vogelweide.]
One might ask to what extent the same could be said about English with regard to the language of, say, Thomas Hardy, Shakespeare, Chaucer, or Beowulf. Still, English is thriving. Why the state of affairs with regard to literature should, according to Zimmer, be so damaging in the one language but not in the other remains totally unclear. Ultimately, it has always simply been natural for works of literature to age linguistically as well (remember that this was the motivation for Dionysius Thrax's grammar of Greek — see section 2.1). It is therefore to be expected that the ability of the average speaker of contemporary German to understand and appreciate Goethe will be more limited than is often assumed in those references to this 'landmark' icon of 'good' German. What such views fail to acknowledge is that a) high literature has at no point been fully representative of general linguistic abilities, neither in its own time, nor centuries later, and b) that this has never been a particular 'deficiency' on the general speaker's part.
18
Of course, it remains open in what sense the two parts of this argument are connected in the first place, i.e. there is no indication given why a word should 'have history' before it could be used in literature. This is not only a naïve notion as such, it would probably be rejected by writers of literature themselves, as it implies a restriction of linguistic innovation in literature.
170
PETER HOHENHAUS
3.3.6 Logic/ plausibility. Traditionally, purist comments on language often involve (pseudo-)arguments based on logic, clarity or practicality, when certain expressions are being judged as 'superior'/ 'inferior' (cf. Weite 1985:139-144). Zimmer (1997) in fact rarely uses arguments explicitly based on logic; however, such reasoning does seem to play a part when Zimmer (1997:42, 27) judges certain expressions as "weder häßlicher noch unlogischer" ("neither uglier nor more illogical") or "einleuchtend und praktisch" ("plausible and practical"). Similarly, Zimmer's (1997:79) criticism of the word order in Ski total is based on "Sprachlogik" — as this "language logic" (whatever that is meant to be) would be unable to determine whether the relation between the two words was attributive ('what kind of skiing?') or adverbial ('to ski in what manner?') (which wouldn't make a great difference anyway, cf. Hohenhaus 2001:20f). 3.3.7 Authority/guidance. The traditional call for authority, in particular in the form of legislation laid down by 'language academies', has recently been by far the most noticeable purist/ prescriptivist element in the language debate in Germany — cf. the founding of the VDS and, in particular, the recent calls for a Sprachschutzgesetz (see section 3.2.1). Zimmer (1997) obviously did not foresee such a development, in fact he decries its absence while assuming that people actually want guidance from an authority (see next section). Zimmer (1997:8) notes that even though there are "allerlei Akademien und Vereinigungen" ("all sorts of academies and organizations"), none of these could speak with "unverdächtiger Autorität" ("authority above suspicion"). Indeed, rather than explicitly demanding the setting up of an official authority, he laments that this does not appear possible in Germany, because such a suggestion would instantly attract the accusation of nationalism, or politically incorrect Deutschtümelei.19 However, as so many purists past and present have done (cf. sections 2.2.1 and 3.2.1), Zimmer (1997:44ff) praises the French example of the Académie française. The German Duden, on the other hand, which at least in matters of spelling does have some generally recognized authority, is regularly criticized for being too submissive (cf. Hohenhaus 2001:12). This is linked to the next point: 3.3.8 Modern linguistics leaving the language and its speakers in the lurch. A rather more recent trend in the purist 'complaints tradition' is that "sometimes 19 The latter is another of Zimmer's terms that seems to have caught on with purists. Numerous purist voices today use it when distancing themselves from being misunderstood as nationalist.
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
171
[...] the scapegoat is now the discipline of linguistics itself' (Milroy & Milroy 1999:28). As Wardhaugh (1999:24, 182) points out, "[l]nguists themselves must carry some of the responsibility for being treated like this", as most linguists prefer to stay out of the public debate and rather hide in their 'ivory tower' of linguistic theory.20 This is understandable, however, in the light of some of the gross misunderstandings of modern linguistics's non-normative, descriptive approach. In extreme cases this is misinterpreted as linguistics being an 'enemy' (cf. Honey 1997 for Standard English); normally it is somewhat less aggressively seen 'merely' as 'permissiveness'. With regard to anglicisms in German, but also aimed at any rejection of prescriptivism, Heald (1978:139) calls this descriptive stance "the 'anything goes' lobby" and its arguments "pseudo-rational" and "defeatist". Others argue that such 'admitting defeat' does not necessarily follow from the tenet of descriptivism. This is also partly the point which the scientific advisory board {wissenschaftlicher Beirat) for the VDS tries to make, in addition to accusing linguistics of having 'shirked its responsibility' for the German language.21 Zimmer (1997:7f, 83f) also notes somewhat resignedly that "die Sprachwissenschaften haben längst allem 'Normativen' abgeschworen" ("linguistics has long renounced all 'normative' function") and thus 'merely' observes phenomena of language contact "mit grenzenloser Duldsamkeit" ("with boundless tolerance"). The idea of people being left in the lurch becomes apparent, then, in Zimmer's (1997:83f) odd assumption that it is many people's Bedürfnis ("wish, desire") to master a certain 'code' which is 'correct', even though nobody could claim to be an 'ideal speaker'. This last point is linked with the 20
There are exceptions, perhaps increasingly so at the moment. A noteworthy one is Gauger (1999), who even stepped right into the lion's den, as it were. In his paper "Die Hilflosigkeit der Sprachwissenschaft", given at a conference held by the Deutsche Akademie für Sprache und Dichtung precisely on the issue of the threat to German, Gauger made an effort to explain to an at least potentially purist audience why linguistics can hardly be interested in petty little quarrels such as settling an argument about two alternative spellings. 21 "Die Sprachwissenschaft [hat sich] aus ihrer Verantwortung für unsere Sprache gestohlen" (taken early 2001 from the board's declaration at the VDS's website, at: www.vdsev.de/aktive/wiss_beiratthesen.htm; updated revised versions can be found at: http://vdsev.de/denglisch/index.php). This Beirat (including a handful of Germanist linguist members) is, it must be admitted, the VDS's link to linguistics, its only link, even though little trace of objective linguistic reasoning is to be found anywhere else in the VDS's domain. Even the board's 'scientific' declaration simply claims that "es kann schlechterdings nicht bestritten werden, daß unsere Sprache erheblich beschädigt worden ist" ("it simply cannot be denied that our language has been seriously damaged") without supplying any evidence whatsoever. Their defence against being 'condemned' as 'undifferentiated' and 'unscientific' thus remains hardly convincing.
172
PETER HOHENHAUS
common misunderstanding of the linguistics term 'ideal speaker' as a value judgement, which in its original sense in generative grammar it is not.22 3.3.9 Negative metaphors. In general, 'folk linguistic' metacommunication is characterized by typical metaphors — see e.g. Weite & Rosemann (1989:83136) for a detailed account. In complaints about language, such metaphors naturally tend to be negative ones, e.g. of 'attack' on the 'body' of language, or of its substance 'crumbling', 'deteriorating' etc. (cf. Hohenhaus 2001:3, esp. note 5). This is by no means limited to 'folk' linguistics. An example from within linguistics of arguments directed against anglicisms which is based extensively on negative metaphor is Schaefer (1999:127), where Germans' 'dietary behaviour' is criticized as an unhealthy 'appetite' for English elements which are 'devoured' despite the danger that they may not be 'digestible'. Zimmer (1997) also repeatedly falls into the purist routine of using emotional metaphorical expressions such as verstümmelt ("mutilated") or Sprachschutt ("linguistic rubble/ debris") etc. to describe certain phenomena (e.g. Zimmer 1997:23f). Perhaps it is not entirely appropriate to judge Zimmer (and others) too harshly on this. It is not easy in matters of language change/ variation to keep one's emotions in check. Nobody, including professional linguists, is immune to such emotions.23 This is neatly demonstrated in Pinker (1994:402): first he "confesses" that the use of disinterested to mean "apathetic" makes him "go in a rage" and argues, purist style, why it should only mean "unbiased", but then debunks his own argument by pointing to the fact that "uninterested" was in fact the "original, 18th-century meaning" and that it "makes grammatical sense". In short: one is naturally 'allowed' to have emotions of dislike about certain phenomena, but that must not filter through into a line of argument if it is to be scientifically sound.
22
In that context it is merely a scientific abstraction. The misinterpretation of this important theoretical notion is in fact not uncommon even within linguistics, in particular, it has to be said, in sociolinguistics (cf. Botha 1989:65-67 and the references given there; cf. also Milroy 2000:12f; Milroy & Milroy 1999:5If; Pennycook 2000:120f). 23 I could add an example from my own experience. For many years during the 1970/80s, I openly resisted the spreading usage in German of causative weil-clauses with main clause verb position as 'incorrect', but gradually adopted this 'wrong' word-order myself (i.e. only in informal conversational German where it is simply normal). Also, I certainly do not particularly 'like' many of the anglicization phenomena under discussion here either, but those emotional feelings must be kept separate from linguistic arguments.
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
173
3.4 Thinking the unthinkable seriously: is German on 'death row'? Unlike most of the usual, 'blindly' opinionated purists, Zimmer (1997) purports to reason scientifically24 — he explains the reasons for his pessimism about the future of German in (semi)linguistic terms of language acquisition and language change. It is thus worth taking a brief look at this part of his arguments separately. In Zimmer's (1997:70) view, the speed and force of current language change has caused German to lose its capacity for assimilating loans, resulting in 'interferences' and areas of uncertainty. Could some degree of 'interference' and 'lack of assimilation' really threaten German, then? Zimmer argues that this is the case. In his view, the actual danger for German in the future lies not so much in the individual phenomena as such, but rather in the following hypothetical scenario: that 'pidginization' will lead to 'creolization' by children exposed to this Neu-anglodeutsch ("New Anglo-German") during the critical phase of language acquisition, which will thus make the 'damage' irreversible. Zimmer (1997) repeatedly alludes to this scenario, but the idea remains rather superficial and dubious in the light of actual linguistic research on language change, creolization and language death (see e.g. McMahon 1994: 284-313). To start with, Zimmer seems to understand 'pidgin(ization)', not as the technical term it is in linguistics, but as a polemic term (as in some informal, everyday language usage). Moreover, his argument of 'creolization' is seriously flawed: the necessary conditions in the 'primary linguistic data' (PLD) in language acquisition for this sort of 'catastrophic change' to take place are not really given in German today (nor in the foreseeable future).25 As noted briefly in section 3.2.2, the kind of registers that Zimmer draws his examples of anglicization from are highly unlikely to be the kind of language children are exposed to during the critical phases. Nobody talks to small children like a marketing director or a fashion designer. Nor would children even have a chance of accessing the bulk of the linguistic examples cited by Zimmer, as these are written language, mostly of highly specialized text types at that.26 24 Thus he does not easily fall into either of the two categories in the 'complaints tradition' that Milroy & Milroy (1999) discuss. Rather, reading Zimmer (1997) amounts to a purist encounter of the third kind, as it were. 25 For a detailed theoretical discussion of 'catastrophes' in language change see Lightfoot (1999), and McMahon (2000:118-128) for some general criticism. 26 Zimmer (1997:72) claims that the Entwicklungsfenster ("window of development") for determining the Tiefencode of a child's language extends to the age of about ten to fourteen, at which age children will be immersed in the anglicized 'world' of pop culture, computers and sport, and that their Sprachgefühl ("linguistic intuition") will thus be damaged. This is not in line with most linguists' accounts of language acquisition, which place much more importance
174
PETER HOHENHAUS
However, at least implicitly, there is another interpretation of 'death': speakers deliberately deciding not to speak German and opting for English instead. This would first presuppose (near) bilingualism, which would then give speakers a 'choice' between two languages so that they could opt for only the one (more prestigious) language. There are speculations that German may indeed be pushed out in the sense that English may assume a status amounting to that of an official language.27 In such a scenario, German would still survive at least as a language of the private sphere and for everyday affairs. That is, speakers would (need to) be bilingual and capable of code-switching, but German would not be lost. Such a situation could be likened to that in parts of Wales, where Welsh continues to be used as a native language despite the strength of English, or parts of the world such as India, where English is an official language without threatening the indigenous languages. As far as the future is foreseeable on the basis of current developments, however, I do not see any such scenario threatening German at all. Even though many Germans may speak English well, and some borrow from it wholesale in special registers such as marketing, advertising, etc., "that isn't to say that that is what they want to relax with in the evening", as Bill Bryson (1990:181) put it in a nutshell. In fact, books, films, television etc. in Germany are still by far predominantly in German, i.e. even for American best-selling novels to be economically successful in Germany, they must be translated into German. Blockbuster movies are still dubbed (unlike in Scandinavian countries, where subtitles are the norm), and the same goes for American soap operas on television, even if these days the titles often remain in English (a distinction that does not sufficiently come across in Hilgendorf 1996:5). Furthermore, German viewers of non-German television channels (e.g. CNN) constitute only a very small minority. Moreover, it is still quite artificial indeed for Germans to engage in 'Smalltalk' using English instead of German amongst themselves. In short: no matter how 'fashionable' some elements of English may be, the situation in Germany (or Austria) is far from being a bilingual one which could be setting the scene for imminent catastrophic language shift.
on the earlier stages, especially with regard to the development of grammar (while the lexicon remains much more open). It is somewhat surprising that Zimmer rests his case about the future of German on such a distorted conception of language acquisition as he himself, in Zimmer (1986), had given a much more adequate, detailed account (aimed at a general readership) of the results of linguistic research on acquisition. 27 E.g. by the director of the IDS, Gerhard Stickel, in a radio interview on 18 January 2001 (transcript at: http://vds-ev.de/presse/zeitungsartikel/archiv/deutsche-welle_l8_01_01.txt).
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
175
4. Conclusion For English, there is clearly no 'danger' today.28 Internal changes which may sensibly be predicted are comparatively minor (see Kortmann 2001 for some realistic speculations). Its status is stronger than ever — it is the dominant global language. This does pose a threat to many languages (up to 90%, as pessimistic estimates go — cf. Crystal 1998; 2000). Thus some sort of resistance and language planning may well be needed in many cases if such languages are to be saved.29 However, German, with its approximately 100 million native speakers and a status that can still be described as "thriving" (Hughes 2000:399), is certainly not 'on death row'. The purists' 'panic' these days is far out of proportion — and it bears a striking, and telling, resemblance to the traditional elements of purism/ prescriptivism (including, as pointed out in section 2.2, that of England in the 18th century). The view of contemporary (professional) Germanist linguistics regarding the state of German and its future is generally much more realistic and relaxed, both within Germany30 and from outside — see e.g. Russ (1994:248ff).31 In the same vein, this article can be seen as a contribution to a more sober discussion of the allegedly 'dangerous anglicization' of German. In conclusion, then: 1. The changes in German under the influence of English are nowhere near as dramatic as claimed by the critics of the current 'flood' of anglicisms. Despite some lexical changes, German is still German; the grammatical system, in particular, is intact and under no threat. 2. No matter how fashionable the use of (elements of) English may currently be in some walks of life, such as IT or marketing, English is far from having assumed the status of everyday language in normal conversation for Germans; no 'catastrophic' language shift is in sight. As long as that continues to be the case (and I am still confidently optimistic about that) German is under no serious threat of being pushed out by English.
28 This despite the continued 'complaints tradition' about its 'decline' too, as exemplified by Honey (1997), cf. also Bailey (2000:133). 29 This is becoming a pressing issue worldwide for a large number of minority languages (see e.g. Nettle & Romaine 2000). 30 See for instance: www.ids-mannheim.de/pub/sprachreport/sr00-2a.html. 31 In fact, Görlach (1999:756) criticizes a recent book on the state of German 'in the real world', Stevenson (1997), for not even mentioning the current concerns about the impact of English on German.
176
PETER HOHENHAUS
References Ammon, Ulrich. 1995. Eine Gefahr für die deutsche Sprache? Zeitschrift für den Kulturaustausch 45:4. 569-575. Ammon, Ulrich. 1998. Ist Deutsch noch internationale Wissenschaftssprache? Berlin: de Gruyter. Augst, Gerhard et al., eds. 1997. Zur Neuregelung der deutschen Orthographie. Begründung und Kritik. Tübingen: Niemeyer. Bailey, Richard W. 2000. Review of A Cultural History of the English Language by Gerry Knowles (London: Arnold, 1997). Language in Society 29:1. 131-133. Bex, Tony & Richard J. Watts, eds. 1999. Standard English: the Widening Debate. London: Routledge. Bolton, W.F., ed. 1966. The English Language. Essays by English and American Men of Letters 1490-1839. Cambridge: Cambridge University Press. Botha, Rudolf P. 1989. Challenging Chomsky. Oxford: Blackwell. Bryson, Bill. 1990. Mother Tongue. London etc.: Penguin. Crystal, David. 1998. English as a Global Language. Cambridge: Cambridge University Press. Crystal, David. 2000. Language Death. Cambridge: Cambridge University Press. Davidson, T. 1874. The Grammar of Dionysius Thrax, translated by T. Davidson. Journal of Speculative Philosophy 8. 326-339. Dunger, Hermann. 1909. Engländerei in der deutschen Sprache. 2 n edition. Berlin: Allgemeiner Deutscher Sprachverein. Eisenberg, Peter. 1999. Stirbt das Deutsche an den Internationalismen? Der Deutschunterricht 3. 17-24. Eisenberg, Peter. 2001. Die grammatische Integration von Fremdwörtern — was fängt das Deutsche mit seinen Latinismen und Anglizismen an? Neues und Fremdes im deutschen Wortschatz- Aktueller lexikalischer Wandel. (Jahrbuch 2000 des IDS) ed. by Gerhard Stickel, 183-209. Berlin & New York: de Gruyter. Fitzmaurice, Susan. 2000. The Spectator and the Politics of Social Networks. Wright 2000. 195-218. Gauger, Hans-Martin. 1999. Die Hilflosigkeit der Sprachwissenschaft. Meier 1999. 85-101. Görlach, Manfred. 1999. The German Language in the 1990s. Linguistics 37:4. 753-763. Hakkarainen, Heikki J. 1995. Phonetik des Deutschen. Munich: Fink. Heald, David. 1978. 'Deutschlisch' or 'Engleutsch' — has German a Future? Modern Languages 59. 139-142.
STANDARDIZATION, CHANGE, RESISTANCE, LINGUISTIC THREAT
177
Hilgendorf, Suzanne. 1996. The Impact of English in Germany. English Today 47. 3-14. Hohenhaus, Peter. 2001. 'Neuanglodeutsch' — zur vermeintlichen Bedrohung des Deutschen durch das Englische. German as a Foreign Language 1/2001. 57-87. At: http://www.gfl-journal.de/l-2001/hohenhaus.html. Honey, John. 1997. Language is Power: the Story of Standard English and its Enemies. London: Faber & Faber. Hughes, Geoffrey. 2000. A History of English Words. Oxford: Blackwell. Johnson, Sally. 2000. The Cultural Politics of the 1998 Reform of German Orthography. German Life and Letters 53:1. 106-125. Kortmann, Bernd. 2001. In the Year 2525 ... Reflections on the Future Shape of English. Anglistik 12:1. 97-114 Langer, Nils. 2000. The Rechtschreibreform — A Lesson in Linguistic Purism. German as a Foreign Language 3/2000. 15-35. At: http://www.gfljournal.de/3-2000/langer.html. Lightfoot, David. 1999. The Development of Language — Acquisition, Change, and Evolution. Maiden, Mass. & Oxford: Blackwell. McMahon, April. 1994. Understanding Language Change. Cambridge: Cambridge University Press. McMahon, April. 2000. Change, Chance, and Optimality. Oxford: Oxford University Press. Meier, Christian, ed. 1999. Sprache in Not? Zur Lage des heutigen Deutsch. Göttingen: Wallstein. Milroy, James & Lesley Milroy. 1999. Authority in Language. 3 rd edition. London & New York: Routledge. Milroy, James. 2000. The Ideology of the Standard Language. Wright 2000. 11-28. Nettle, Daniel & Suzanne Romaine. 2000. Vanishing Voices. The Extinction of the World's Languages. Oxford: Oxford University Press. Pennycook, Alastair. 2000. Disinventing Standard English. Review of Bex & Watts 1999. English Language and Linguistics 4:1. 115-124. Pinker, Steven. 1994. The Language Instinct. London etc.: Penguin. Russ, Charles V.J. 1994. The German Language Today. London & New York: Routledge. Schaefer, Ursula. 1999. Ausgepowerte Lerner and Other Guests: Some Remarks on Recent Influences of English on German. Form, Function and Variation in English: Studies in Honour of Klaus Hansen, ed. by Uwe Carls & Peter Lucko. Frankfurt/M.: Peter Lang. 127-139. Stevenson, Patrick, ed. 1997. The German Language and the Real World. Revised edition. Oxford: Clarendon.
178
PETER HOHENHAUS
Wardhaugh, Ronald. 1999. Proper English — Myths and Misunderstandings about Language. Malden, Mass. & Oxford: Blackwell. Watts, Richard J. 2000. Mythical Strands in the Ideology of Prescriptivism. Wright 2000. 29-48. Weinrich, Harald. 2000. Deutsch in Linguafrancaland. Passé und mega-out? Zur Zukunft der deutschen Sprache im Zeitalter von Globalisierung und Multimedia. Cologne: Deutsche Welle Kommunikation, November 2000. 6-16. Weite, Werner. 1985. Die englische Gebrauchsgrammatik. 1. Geschichte und Grundannahmen. Tübingen: Narr. Wright, Laura, ed. 2000. The Development of Standard English, 1300-1800 — Theories, Descriptions, Conflicts. Cambridge: Cambridge University Press. Zimmer, Dieter E. 1986. So kommt der Mensch zur Sprache. Munich: Heyne. Zimmer, Dieter E. 1997. Neuanglodeutsch — Über die Pidginisierung der Sprache. Deutsch und anders by Dieter E. Zimmer. Reinbek: Rowohlt. 7104.
THE STANDARDIZATION OF LUXEMBOURGISH GERALD NEWTON University of Sheffield, England Introduction The present-day national language of the Grand Duchy of Luxembourg belongs genetically to the West Moselle Franconian group of German dialects, having the same origins as the varieties of German spoken between Siegen in Germany itself, and Arlon in Belgium. Historically, the focus of these dialects converged on Trier, the ancient capital of the Romans north of the Alps. However, in early modern times the importance of Luxembourg began to outstrip that of Trier, through the elevation of the Luxembourg dynasty (13091437) to become ruling House of the Holy Roman Empire. Prior to and during this period, Luxembourg territory expanded greatly, moving away from its German origins, and into the wider sphere of European politics. However, from the 16th century on an opposite process began to occur, and the territory gradually diminished in size, a large part of it ceded to France in 1659, another part to Prussia in 1815, and a third part to Belgium in 1839. This latter partition split Luxembourg into two units: a francophone western section, now the Belgian Province de Luxembourg, and a germanophone eastern section, the present-day Grand Duchy.1 The language situation The western territory is better designated 'francophone' rather than 'French-speaking', in that not only French, but Belgo-Romance ('Walloon') is in use there. Similarly the east is not 'German-speaking' in the true sense of Germany or Austria, but only insofar as Luxembourgish is historically linked to German. German in the standard sense exists in the Grand Duchy only in a system of complex diglossia alongside standard French, both of which are written rather than spoken languages, while Luxembourgish itself exists as the chief spoken medium of the Grand Duchy (Hoffmann 1979; Berg 1993; 1
For further detail on the general history of the Grand Duchy, see Newcomer 1984.
180
GERALD NEWTON
Newton 1996). How far this spoken medium - Lëtzebuergesch - has moved towards standardization may be examined at the external level of social motivation, and at the internal level of actual linguistic change. Internal change At the internal level, various scholars have considered phonological, morphological and lexical normalization to have been in progress in Luxembourgish since the beginning of the 20th century. The first such scholar was René Engelmann (1880-1915), a native Luxembourger, who in 1910 published a study of the vocalism of Vianden (north-eastern Luxembourg). Engelmann chose to refer to this levelling-off variety as the Luxembourg koine. This term, used of the various important regional forms that had arisen in Ancient Greek, may point either to a dialect that absorbed provincial and foreign elements as it spread out from its base (Macedonian koine), or rejected them (Mycenaean koine) in order to mould a language of administration and culture (Palmer 1954). It was in this latter sense of rejection that Engelmann understood the term koine for Luxembourg, perceiving that a common form of Luxembourgish was being smoothed out for use in the businesses of Luxembourg City by the people whose work brought them there from the line of the Alzette river.2 Another scholar, the Belgian-Luxembourger, Hary Godefroid (18771942), was more specific, and dated the origin of the koine to the decade around 1890, when the Luxembourg steel industry and its subsidiaries were benefitting greatly from German investment during the period following the Franco-Prussian War (Godefroid 1938). Further evidence for a koine was found by Richard Huss, who, from surveys conducted by himself and others, determined that dialect recession had taken place in the north between 1889 and 1924, while language in the central part of Luxembourg over the same period had remained stable (Huss 1927:60). It seemed therefore that the more pronounced features of dialectal Luxembourgish were receding in favour of the form most prevalent in the central region of Luxembourg, where the greatest number of speakers of the language was concentrated. This was also the main area of commuting by rail (Ettelbruck to Luxembourg City and the industrial south, along the line of the Alzette), on which passenger-traffic underwent a twelvefold increase between 1875 and 1913 (Statec 1990:294-295). It was also the area in which most of the establishments of higher secondary education were located. This 2
The Alzette is the main river of Central Luxembourg. It rises at Russange (France) and flows through Esch/Alzette, Luxembourg City, Mersch, before discharging into the Sûre at Ettelbruck.
THE STANDARDIZATION OF LUXEMBOURGISH
181
interaction resulted in what has been called the Duerchschnëtts-sprooch, "the average type of language" (Schmitt 1984:174). Here, extreme variants, such as northern velarization, or southern palatalization, did not appear.3 Yet neither did items too closely identified with Luxembourg City itself.4 Many northern preterites were also replaced in the koine by perfects (Russ 1978).5 Engelmann himself (1910:9) reported that elements from the Alzette river valley such as final [n] for [n], were becoming increasingly frequent even in Vianden (Walsdorf), where grein ("green"), the form used by the older generation, was being rejected by the younger generation in favour of grey. In 1948, Palgen noted that much of the old dialect of Esch-sur-Alzette was being replaced by the forms of Central Luxembourg (1948:3), while in 1981 Hoffmann attributed part of this to a desire to move away from the open vowels of this dialect, which were considered to be vulgar (1981:17).6 Emulation of the koine can also occur through fear of embarrassment or lack of job security (Flener 1978:31; Hermes 1993:13). There are however degrees of emulation, and Luxembourgers can easily recognize whether a speaker is from Luxembourg City or from elsewhere (Rinnen 1991:32). Internal change has most recently been discussed by the German researcher, Peter Gilles, who replaces the concept of koine with one of 'Central Luxembourgish', and, in a largely acoustic analysis, concludes that the reduction in regional variation often takes place in favour of the variant of greatest geographical validity (Gilles 1999:256). External change As for the external changes, these have been variously motivated. One, indisputably, was the sense of national unity that came to the country on 19 April 1839, when it gained a measure of independence from the ties that since 1815 had bound it to the Dutch monarchy in The Hague. The new selfawareness resulted in 1843 in the first study of the vocalism of a dialect of Luxembourgish (Hardt), in 1847 in the first dictionary of Luxembourgish (Gangler), in 1855 in the first attempt to produce a standardized spelling for Luxembourgish (Dicks), as well as in the same year the first sociolinguistic study of the language (Klein). These works served to consolidate the language
3
E.g. (velarization): Lekt for Leit ("people") (LSA 130); (palatalization) Fënschter for Fenster ("window") (LWB I:323). 4 E.g. ons ("us") for äis (LSA 84) and dongen ("do") for din (LSA 158). 5 Cf. LSA 19: northern ich mouch ("I made") for koine ech hu gemaach ("I have made") (= I made). 6 E.g. där ("to you"), gääss ("eaten") for koine dir, giess.
182
GERALD NEWTON
and to make it the first line of defence against any encroachment on its territory and culture by France and Germany. This feeling of solidarity with the language did not however proceed at that time beyond a certain level, and the attempt in the Chamber of Deputies, the Luxembourg Parliament, to introduce Luxembourgish in 1896 as a parliamentary debating language alongside French and German, was ruled out of order by the Conservative majority, who did not wish to open the Chamber in this way to the socialist masses (Hoffmann 1987). Education Act of 1912 An alternative path for the promotion of the language was however provided by the Education Act of 1912, which introduced Luxembourgish for the first time as a subject to be taught in schools. Under this heading, Nicolas Welter (1871-1951), the Minister of Education, understood a non-examination subject, imparting Luxembourgish culture to schoolchildren by means of familiarizing them with well-known texts and songs in Luxembourgish, and also by introducing them to a spelling system for Luxembourgish, though only passively. Unlike the complex spelling system of 1855, this had to be simple enough for children to recognize, and since Luxembourg children learnt to read and write in school by use of the first foreign language, German, it was considered that German orthography would provide the best starting-point for Luxembourgish. This system, because it had originally been devised by Engelmann and modified by Welter, became known as the 'Welter-Engelmann orthography'. 'Welter-Engelmann' was incorporated into the new Luxembourgish textbook for schools (Welter 1914), and this remained a standard work until 1947. Through passive assimilation, therefore, schoolchildren were for the first time given the opportunity to see their native vernacular in its written form, a process which might therefore be viewed as a first stage towards the standardization of Luxembourgish. Radio The next impetus towards standardization was provided by radio broadcasting, which began as a medium of popular entertainment at the beginning of the 1920s. Luxembourg had its own local transmitter from 1928, which for a few hours each evening gave a selection of music and information in French, repeated in German. Since, however, the programmes were not of a high intellectual quality, Luxembourgers generally listened to France or Germany. German was easier for Luxembourgers to understand, and introduced them to living spoken German in a way that had not previously
THE STANDARDIZATION OF LUXEMBOURGISH
183
been possible, since on a day-to-day basis there was little contact in the Grand Duchy with German native speakers. The result of listening to German radio was that many words and phrases 'washed over' from German into Luxembourgish, provoking a reaction from purists, who then published lists of forms to be avoided; thereby, if somewhat negatively, providing a further measure of standardization (Newton 1996:185-186). Luxembourg Linguistic Society During this period another organization also came into being: the Luxemburgische Sprachgesellschaft ("Luxembourg Linguistic Society") (1924), the largest undertaking of which was to compile a comprehensive new dictionary of Luxembourgish, with definitions in standard German (Tockert 1925). This was to replace Gangler's dictionary of 1847, and the later, somewhat flawed, dictionary of 1906 {Wörterbuch 1906). The Society collected data directly from the field, and hoped to have this task completed by 1940. 1940 The immediate issue for the standardization of Luxembourgish in the late 1930s revolved however once more around the spelling. The WelterEngelmann spelling of 1914 was, in view of the rise of National-Socialism in Germany, considered by the Luxembourg Ministry of Education to be far too German in appearance (Newton 1996:195-196). Consequently, in November 1939, Nicolas Margue (1888-1976), the Minister of Education at that time, commissioned a new system from the Luxembourg phonetician, Jean Feltes (1885-1959). This, for use in schools and in official communications, was to be based on strict phonemic principles. However, the decision came too late to be effective, for on 10 May 1940 Germany began its invasion of France and the Low Countries. Luxembourg was unable to resist, and fell on the first day, as it had once before, in August 1914, at the commencement of the First World War. The difference from 1914 was however that during the First World War the Germans had little other intention for Luxembourg than denying its railways to the French. In 1940 this was not the case, and once the initial troops had left, the country found itself in the grips of the Gestapo and the German Civil Administration, whose goal it was to germanize Luxembourg as quickly as possible, and to wipe the country from the map, attaching it henceforth to Gau Moselland, a Nazi administrative unit based on Koblenz. The result was civil protests, national strikes, resistance movements, clandestine publications in Luxembourgish (a language which the Nazis
184
GERALD NEWTON
prohibited in printed form in 1942), broadcasts in Luxembourgish from the BBC in London by the Grand Duchess and the Government in Exile, and, as a Nazi backlash, summary executions, deportations to concentration camps, and relocation of the general populace in Czechoslovakia. At the end of the War, all these events served to bring the Luxembourgish language forward as the link that had held the nation together during the time of oppression. It was clearly the national language, and deserved to be recognized as such. Yet it was not a standardized language, only a semi-standardized one, one that the German linguist Heinz Kloss (1904-1987) would in 1952 first call eine Halbsprache, "a semi-language", and then, in 1978, eine Ausbausprache, a language "undergoing elaboration", and possibly emerging eventually as an Ab Standsprache, a language so different from German as to be recognized as separate from it (such as Dutch and Afrikaans). These processes would involve orthography, purism, dictionaries, Luxembourgish in schools, grammars, and even the setting up of a Luxembourg Linguistic Academy, to rule on matters of usage. In reality very little of this has so far come about. Spelling As regards spelling, the phonemic system commissioned in 1939 was introduced in 1946, and declared by Grand-Ducal decree to be "official" (7 September). The 'Margue-Feltes' orthography, as it is called, was incorporated into a new school textbook, Lezebuurjer Gedichter a Proosashteker fiir ons Shoulen (Luxembourgish Poems and Prose Passages for our Schools) (1946). This received its last reprinting in 1951, after which it became obvious that the system was so far removed from the spelling of German, the basic language of instruction in Luxembourg schools, that it was held to be unteachable. Accordingly, a move was then made towards the spelling devised in 1950 for the new dictionary by the Luxembourg scholars Joseph Tockert (1875— 1950) and Robert Bruch (1920-1959), which was then, little by little, used for all publications in Luxembourgish. However, the Tockert-Bruch system was not yet official, and in consequence books were often reprinted in their original spelling, even where this might date from the 19th century, thus producing a situation of great uncertainty amongst those who wished to write in the language.7
7
In 1985, the Letzeburger Lidderbuch, a collection of 101 songs collected and published by Guillaume Stomps in 1898, was re-issued with spelling unchanged, for use in schools.
THE STANDARDIZATION OF LUXEMBOURGISH
185
Purism Another major consideration to some in 1945 was that all germanisms be expunged from the language as swiftly as possible (Newton 1996:202). This current has persisted up to the present day, where "good Luxembourgish" is contrasted with "bad Luxembourgish" (for one such list, see Roth 1998). Much of this purism has, however, been ignored by the nation generally, although the wider implications still provoke acerbic exchanges in the correspondence columns of the Luxembourg national press (cf. Luxemburger Wort, 9 and 30 December 2000). The Luxembourgish Dictionary In the matter of the dictionary, work continued steadily from the publication of the first fascicle of Volume I in 1950, to the publication of the last in 1977. The dictionary did not however seek to standardize the language, and indeed was at pains not to repeat the mistake of the dictionary of 1906, which had been to take the dialect of the city of Luxembourg and its environs only. Thus the 1950 dictionary included many regional variants, together with maps of locations visited in fieldwork and a chart of phonological variation. In addition to this, the dictionary was compiled on the same strict principles that had been used in the 19th century by scholars of medieval German (listing, for example, words beginning in V with those beginning in F). The result was an expensive set of five volumes that was extremely frustrating for the novice to handle, and which was made even more confusing when, after the publication of Volume 3 (1965), slight spelling changes were introduced. This was scarcely the basis on which to set up a linguistic academy in Luxembourg, and consequently none has been set up, despite periodic demands that there should be one. Nor can the dictionary of 1950 now be considered anything more than a museum piece. Republished in 1995 as the Lëtzebuerger Dixionär, it immediately ran into controversy, for when the original materials were collected in the interwar years, there was no embarrassment about including phrases derogatory to the Jews, or to feminist groups, or to foreigners generally, and these were not subsequently suppressed in the 1995 reprint. The complaints from offended groups were such that the Luxembourg Ministry of Education eventually ordered the pulping of all the copies as yet unsold, and commissioned work on a new dictionary, to be compiled using electronic methods (in progress 2002: Institut Grand-Ducal). Another positive step has been the investment of money from various funds in the compilation of a computer spelling-checker for Luxembourgish,
186
GERALD NEWTON
for use in schools and local government administration (in progress 2002: Centre Universitaire).8 Grammars of Luxembourgish A guideline to the grammar of Luxembourgish for use in the understanding of poetry had first appeared in 1845 (Meyer). This was followed in 1928 not by a grammar, but by an impassioned plea that one should be published (König 1928), which then eventually did appear in 1955, when Robert Bruch produced his Luxemburger Grammatik in volkstümlichem Abriss (A People's Grammar of Luxembourgish). This was in reality far from easy for the people to use, and was in no way prescriptive towards the koine. Bruch's grammar has subsequently appeared in various editions, but is still largely unmodified from the original, and does not serve to aid foreigners in the acquisition of the Luxembourgish language. Other attempts at this latter task have been more successful, and combine the grammar with basic language teaching (Christophory 1974), or with a bilingual French-Luxembourgish dictionary (Rinnen 1988), while a structural analysis of Luxembourgish grammar was published by Pierre Schmitt in 1984. Luxembourgish as a school subject Expansion of the teaching of Luxembourgish in schools was the great hope of 1945, when regulations were introduced that the subject should be made compulsory for all 12 and 13 year-olds. Yet the issue foundered on problems of spelling and textual choice. In the event, what generally happened was that, as before, a few songs were taught to 12 year-olds, and the rest of the time was used for teaching French or German (GjpR 1989). Eventually, however, national considerations and puristic movements combined in 1971 to produce the Actioun Lëtzebuergesch, the "Campaign for Luxembourgish", which in 1975 was able to persuade the Chamber of Deputies to discard the spelling of 1946 as the official system, and instead to establish the Tockert-Bruch spelling in this role. This was then made obligatory for use in schools and government administration. Exemplification followed this in the form of two new school textbooks, Lëtzebuergesch an der Schoul (1976) and Lëtzebuergesch Texter fir postprimär Schoulen (1978). Further pressure on the Luxembourg Parliament produced a constitutional revision in February 1984, under which Luxembourgish was declared the "national language" of the Grand Duchy. 8
Projet CORTINA (= Correction ORThographique INformatique Appliquée à la langue luxembourgeoise).
THE STANDARDIZATION OF LUXEMBOURGISH
187
Reaction to the official orthography and a move to standard forms Yet the only thing that had standardized even by 1984 was the spelling, and what is more, many Luxembourgers did not, and do not, believe that there was or is any standard official spelling; and worse, that if there is, then this marks the beginning of a linguistic tyranny over the speakers that can be seen well enough operating with German, French and English, where pedantic putdowns of those unable to spell or use correct grammar have become a way of establishing social control. Hence on the one hand hackles rise at the very thought of koine forms becoming the only forms acceptable to "good Luxembourgish", while on the other hand there is pressure to incorporate the vast number of foreigners currently working in the Grand Duchy, particularly Portuguese, into Luxembourg society by providing them with a "bridge" to the national language (Moien 1999). Yet this can scarcely be done without a standard form that may be learnt without reference to regional variants. Conclusion In the meantime, the Luxembourgish language in the Grand Duchy remains strong, in whatever variant it might be. It shows no signs of falling victim to either French or German, and since 1984 has been increasingly used in its written form. Collectively and notwithstanding the dialects, Luxembourgish is one of the pillars of national identity, but one that strongly defies homogenization. Despite this, however, outsiders must be provided with a standard form in order to be able to learn the language. For this the teachers generally adhere to the forms of Central Luxembourg, which, whether or not this represents an emerging koine, may be taken to be universally understood throughout the whole of the Grand Duchy.
References Berg, G. 1993. «Mir wëlle bleiwe, wat mir sin». Soziolinguistische und sprachtypologische Betrachtungen zur luxemburgischen Mehrsprachigkeit. (= Reihe germanistische Linguistik, 140.) Tübingen: Max Niemeyer. Bruch, Robert. 1955. Précis populaire de Grammaire Luxembourgeoise/ Luxemburger Grammatik in volkstümlichem Abriss. Luxembourg: Pierre Linden. (2 nd edition: Revised by L. Senninger, 1968.). Christophory, Jul. 1974. Mir schwätze Lètzebuergesch. Nous parlons Luxembourgeois. Abécédaire luxembourgeois. Guide bilingue de Luxembourgish grammaire et de lecture. We speak Luxembourgish.
188
GERALD NEWTON
Primer. Bilingual Guide to Grammar and Reading. Luxembourg: Saint Paul. Dicks [= Edmond de la Fontaine]. 1855. Versuch über die Orthographie der Luxemburger-deutschen Mundart. Luxemburg: V. Bück. Eis Sprooch. Zeitschrift für Luxemburgischer Sprache, neue Folge. 1971-1993. Ed. by Henri Rinnen. Féiz (Luxembourg): Kremer-Muller. Engelmann, René. 1910. Der Vokalismus der Viandener Mundart. Gross herzogliches Gymnasium zu Diekirch. Programm herausgegeben am Schlusse des Schuljahres 1909-1910. 1-44. Diekirch: J. Schroell. Flener, Will. 1978. Eng kleng Prouf vum Mamer Dialekt. Eis Sprooch 9. 31. Galerie. Revue Culturelle. 1982ff. Foetz (Luxembourg): Kremer-Muller. Gangler, Jean François. 1847. Lexicon der Luxemburger Umgangssprache (wie sie in und um Luxemburg gesprochen wird) mit hochdeutscher und französischer Uebersetzung und Erklärung. Luxemburg: J. Lamort. (Unveränderter Neudruck: Wiesbaden: Sändig, 1973.). Gilles, Peter. 1999. Dialektaus gleich im Luxemburgischen: Zur phonetischphonologischen Fokussierung einer Nationalsprache. (= Phonai: Texte und Untersuchungen zum gesprochenen Deutsch, 44.) Tübingen: Max Niemeyer. Godefroid, Hary. 1938. Lezeburger sprôch a literatur. Jong-Hémecht 12:1. 210. GjpR [= Rinnen, Guy-J.P.]. 1989. D'Lëtzebuergeschstonn op 7ième. Eis Sprooch 21. 15-16. Hardt, Matthias. 1843. Vocalismus der Sauer-mundart. Programm, herausgegeben beim Schlusze des Schuljahres 1842-43, KöniglichGroßherzogliches Progymnasium zu Echternach. 1-29. Trier: Fr. Lintz. Hermes, Monique. 1993. Gréiwemaacher erënnert sech [...] Jos. Hurt (1892-1962). Eis Sprooch 37. 7-14. Hoffmann, Fernand. 1979. Sprachen in Luxemburg. (= Deutsche Sprache in Europa und Übersee, 6.) Wiesbaden: Franz Steiner. Hoffmann, Fernand. 1981. Zwischenland. (= Germanistische Texte und Studien, 11.) Hildesheim: Georg Olms. Hoffmann, Fernand 1987. Spoo und die Folgen. Galerie 5:1. 40-61. Huss, Richard. 1927. Die luxemburgische Mundart vor vierzig Jahren und heute. Studien zum Luxemburgischen Sprachatlas. (= Beiträge zur Luxemburgischen Sprach- und Volkskunde. Vorarbeiten zu einem wissenschaftlichen Wörterbuch der Luxemburgischen Mundart, herausgegeben von der Luxemburger Sprachgesellschaft, 2), 57-68. Luxemburg: Linden & Hansen. Jong-Hémecht. Zeitschrift für heimatliches Theater, Schrift- und Volkstum. 1926-40. Esch-Alzette (Luxembourg): Victor Neuens.
THE STANDARDIZATION OF LUXEMBOURGISH
189
Klein, Peter. 1855. Die Sprache der Luxemburger. Besonderer abdruck aus dem zehnten jahresheft des Vereins für vaterl. geschichts- und alterthumskunde. Luxemburg: V. Bück. Kloss, Heinz. 1952. Die Entwicklung neuer germanischer Kultursprachen von 1800 bis 1950. München: Pohl. (2nd enlarged edition: (= Sprache der Gegenwart, 37.) Düsseldorf: Schwann,1978.). König, Lucien. 1928. Auf dem Wege zu einer Grammatik der luxemburgischen Mundart. I. Teil. Ein Beitrag zur Geschichte der Lux. Dialektliteratur. Letzeburg: Nationalverlag Ernster / Worré-Mertens. Lëtzebuergesch an der Schoul. 1976. Lëtzebuerg: Joseph Beffort. Lëtzebuergesch Texter fir postprimär Schoulen. 1978. Lëtzebuerg: Ministère de l'Éducation Nationale. (Revised 1990.). Letzeburger Lidderbuch. Gesammelt an erausgin fum Guill. Stomps. 1898ff. Luxembourg: J.-P. Krippler-Muller. Lezebuurjer Gedichter a Proosashteker für ons Shoulen. 1946. Lezebuurch: Munshausen. LSA = Luxemburgischer Sprachatlas. Luxemburger Wort. Newspaper. 1848ff. Luxembourg. Luxemburger Wörterbuch. 1950/1954-1977/1978. 5 vols. Luxemburg: Linden. (Republished as Lëtzebuerger Dixionär. 2 vols: Luxemburg: Quintus, 1995.). Luxemburgischer Sprachatlas. 1963. Marburg: N.G. Elwert. LWB = Luxemburger Wörterbuch. Meyer, Anton. 1845. Luxemburgische Gedichte und Fabeln nebst einer grammatischen Einleitung und einer Wörtererklärung der dem Dialekt mehr oder weniger eigenartigen Ausdrücke, von Gloden. Brüssel: Delavigne & Callewaert. Moìen. Actes Colloque. Eis Sprooch: eng Bréck, keng Barrière. 1999. Niederanven (Luxembourg): Ondine CONSEIL. Newcomer, James. 1984. The Grand Duchy of Luxembourg. The Evolution of Nationhood. Lanham: London: University Press of America. (2nd edition: Luxembourg: Emile Borschettel995.). Newton, Gerald, ed. 1996. Luxembourg and Lëtzebuergesch. Language and Communication at the Crossroads of Europe. Oxford: Clarendon Press. Palgen, Hélène. 1948. Studien zur Lautgeographie Luxemburgs. (= Beiträge zur Luxemburgischen Sprach- und Volkskunde, 6.) Luxemburg: P. Linden. Palmer, Leonard R. 1954. The Latin Language. (= The Great Languages.) London: Faber and Faber. Rinnen, Henri. 1988. Dictionnaire Français-Luxembourgeois. Luxembourg: St. Paul.
190
GERALD NEWTON
Rinnen, Henri 1991. Linguistik und Sproochwëssenschaft. Eis Sprooch 31. 3237. Roth, Lex, ed. 1998. Gréng a Rout Lëschten. (= Eis Sprooch, Extra-Serie, 20.) Lëtzebuerg: Sankt-Paulus-Dréckerei. Russ, Charles V. J. 1978. The morphological features of the verb in Luxemburgish. Zeitschrift für Dialektologie und Linguistik 45. 30-41. Schmitt, Pierre. 1984. Untersuchungen zur luxemburgischen Syntax. (= Deutsche Dialektgeographie, 84.) Marburg: N.G. Elwert. Statec Luxembourg. 1990. Statistiques Historiques 1839-1989. Luxembourg: Service Central de la Statistique et d'Études économiques. Tockert, Joseph. 1925. Über luxemburgische Lexikographie. Ein Wort zum Geleit. Jahrbuch/ Annuaire. 30-51. Luxemburg: Luxemburgische Sprachgesellschaft. Welter, Nicolas. 1914. Das Luxemburgische und sein Schrifttum. Luxemburg: Gustave Soupert. Wörterbuch der Luxemburgischen Mundart. 1906. Luxemburg: M. Huss.
LANGUAGE PLANNING IN NORWAY A BOLD EXPERIMENT WITH UNEXPECTED RESULTS
ARTHUR O. SANDVED University of Oslo, Norway
Introduction It is common knowledge that there are two officially recognized standards of written Norwegian.1 The present-day designations for these are Bokmål and Nynorsk. What is not commonly known is that there is a third standard, which — although not officially recognized — is, or comes very close to, what the majority of Norwegians use in writing. In this paper an attempt will be made to try to explain and comment on this rather bizarre situation. In order to do that, it is necessary to take a very brief glance at some aspects of the linguistic history of modern Norway.
From 1536 until 1814 Norway was in union with Denmark, and the official written language of the country was Danish. But the difference between Danish and Norwegian was small enough for literate Norwegians to be able to read Danish without too much difficulty. A language movement in the 19th century resulted in the creation of a new form of written Norwegian based on a selection of (relatively conservative) Norwegian dialects. In the early days of its history this new language variety was generally referred to as Landsmål. Towards the end of the 19th century the language question became a burning political issue. In 1885 the Norwegian Parliament {Storting) requested the Government to take the necessary steps towards placing Landsmål on an equal footing with Dano-Norwegian, which eventually came to be referred to as Riksmål, possibly a calque on German Reichssprache. During the 1890s the 1 I would like to put on record my gratitude to John Ole Askedal and Tor Guttu for reading a first draft of this paper and for suggesting several alterations and additions.
192
ARTHUR O. SANDVED
Storting passed several acts concerning the use of Landsmål in schools, in teacher-training colleges and at the university, and in 1899 a chair in Landsmål was established at the University of Oslo. During the last two decades of the 19th century it became increasingly clear that Landsmål, which was being strongly promoted by active politicians, could become a challenge, or even a threat, to Riksmål. The time was ripe for the formation of an organization to meet the challenge. This organization was formed in 1899. Ten years later it had grown into a nationwide organization and was from 1909 known as Riksmålsforhundet. The Riksmålsforhundet, then, was formed to strengthen the position of Riksmål in the face of the Landsmål threat, and during the early years of its existence the front-line was drawn between Landsmål and Riksmål. Later developments have brought about a change here, but more about that below. In the early stages there was a strong element of nationalism in the Norwegian language conflict. The main argument from the Landsmål camp was that their language variety was "Norwegian", whereas that of their opponents was "Danish". One of the first tasks confronting the Riksmål supporters, therefore, was to introduce a spelling reform which distanced Riksmål orthography from the Danish. During the second half of the 19th century Riksmål had become increasingly norwegianized, and after a spelling reform in 1907 there was no doubt that Riksmål was a Norwegian language variety, not Danish. In the 1920s the two language varieties were given new official designations by the Storting, Landsmål being replaced by the term Nynorsk, which means "New Norwegian", and Riksmål by Bokmål, which literally means "book language". This was a purely terminological change. There was no language reform in the 1920s. The terminological change from Riksmål to Bokmål has created a certain amount of confusion in as much as the former term, although officially discarded, continued to be used unofficially — and in course of time came to be used with a different semantic content. I shall have more to say about that below. For the time being I shall refer to the nonNynorsk language variety as Riksmål/Bokmål. This situation — with two rival language varieties existing side by side — was fraught with problems and potential conflicts. And conflicts did arise. Since the two varieties were in fact no more than that — two varieties of one and the same language (although in many ways rather different from each other) — the idea arose of gradually bringing them closer together and eventually merging them. This idea began to make serious headway relatively soon after 1900 and received support from political parties on the Left. It should be constantly borne in mind that in Norway language codification is,
LANGUAGE PLANNING IN NORWAY
193
and has been for more than one hundred years, a political matter. The final decision rests with the Storting. During most of the 20th century it was the policy of varying Norwegian governments to try to bring the two varieties together, and the ultimate goal was their fusion into a "Common Norwegian", Samnorsk. Behind this desire to bring together the two standards Riksmå/Bokmål and Nynorsk one can discern at least two motives. There was, as already mentioned, a nationalistic impulse to steer RiksmåU Bokmål, whose origin, it will be remembered, was Dano-Norwegian, in the direction of the more distinctively Norwegian variety Nynorsk. There was also a socio-political motive: to base the new Common Norwegian on the spoken "language of the people", as it was commonly phrased. To the extent that RiksmåU Bokmål can be said to have had a spoken basis, this was the language of urban upper and middle class speakers. The means by which Norwegian authorities tried to bring about the fusion of the two varieties of written Norwegian was a series of language reforms. There was one in 1917 and a second, a very radical one, in 1938. The strategy was this (here looked at it from a RiksmåU Bokmål point of view): a great many new spellings and inflectional forms which were alien to the old RiksmåU Bokmål standard were introduced into RiksmåU Bokmål — to begin with as alternative forms existing side by side with the traditional forms. These alternative forms were either simply taken from Nynorsk or were forms reflecting widespread popular usage, "the spoken language of the people". (Cf. the nationalistic and the socio-political motives mentioned above.) Later these new alternatives were to be made compulsory and the traditional RiksmåU Bokmål forms were to be banned. An example will make this clearer. The definite article in the feminine gender in Norwegian takes the form of a suffix attached to the base of the noun. In traditional RiksmåU Bokmål this suffix is -en (which is the common gender ending), while Nynorsk has the specific feminine ending -a: (1)
Riksmål/Bokmål: rosen boken
Nynorsk: rosa (the rose) boka (the book)
In the 1917 reform rosa and boka were introduced alongside the RiksmåU Bokmål forms as alternatives, and in the 1938 reform the old RiksmåU Bokmål forms rosen and boken were dropped. It was the same thing with a great many isolated words, which occurred in two different forms in the two varieties. For instance:
194
(2)
ARTHUR O. SANDVED
Riksmål/ Bokmål: efter sten abbor
Nynorsk: etter (after) stein (stone) åbor (perch, the fish)
The forms in the second column here are Nynorsk forms introduced as alternatives into Riksmål/ Bokmål in 1917, and in 1938 the traditional Riksmål/ Bokmål forms were again dropped. The 1938 language reform met with very strong opposition in several quarters. A very important event followed in the wake of the 1938 language reform, viz. the decision by the school-board of the City of Oslo in 1939 to adopt in school-books the so-called 'radical' forms, i.e. forms which were common to both varieties of written Norwegian. This meant that Nynorsk forms admitted to Bokmål in 1917 and 1938 were to be preferred to the traditional Riksmål/ Bokmål forms. During the Second World War the language conflict largely subsided, but relatively soon after the war it flared up again. From now on it will be convenient to make a slight alteration in our terminology. By 1938 the official Bokmål had come a very long way from the traditional Riksmål, which had had its official name replaced by Bokmål in 1923.2 In the remainder of this paper the term Bokmål will be used for the official, state-supervised variety of nonNynorsk, and the term Riksmål will be used for that special conservative brand of non-Nynorsk which came to be promoted by the Riksmålsforbundet (and by the Norwegian Academy for Language and Literature; cf. below) in the 1950s. This altered terminology is a reflection of a change of policy on the part of the Riksmålsforbundet. In its early days, it will be remembered, the Riksmålsforbundet had turned against Landsmål/ Nynorsk. By the 1950s the chief enemy was no longer Nynorsk — which could no longer be regarded as a threat to the existence of Riksmål — but the new Bokmål, which was the outcome of various Governments' amalgamation policy. This is made abundantly clear in a booklet by the poet Arnulf Øverland (1889-1968), chairman of the Riksmålsforbundet from 1947 till 1956, in which he sums up the situation with biting sarcasm: We had two adequate languages, each with its adequate literature. But two languages were too many, so the Storting decided that one more should be made. Now we have three. The third language has no books written in it, so it is called 'book language'. No one can speak it and no one can write it, so the children are 2
Textbooks usually give the date 1929 for this change, but recent research (Langslet 1999:137) has revealed that the change from Riksmål to Bokmål was in fact made in 1923.
LANGUAGE PLANNING IN NORWAY
195
made to learn it. At the same time Riksmål, the language spoken, read, and written by the overwhelming majority, is abolished by law. (Øverland 1948:9-10; here quoted from Haugen 1966:173)
And in an article in the Riksmålsforbundet' s magazine Ordet (The Word) in 1950 Øverland maintained that the policy of enmity towards Nynorsk, which had been pursued by the Riksmål camp, had been erroneous and harmful. An important event in the Government's post-war attempts to direct the linguistic development was the establishment of the Norwegian Language Commission (Norsk språknemnd) in 1952. The Commission was given the mandate to "promote the approach to each other of the two written languages on the basis of the Norwegian folk language" (Haugen 1966:182). The creation of the Language Commission with this mandate greatly exacerbated the language conflict in Norway. Perhaps the strongest opposition against the Commission was voiced by what became known as "The Parents' Action against Samnorsk" which raised a most vigorous campaign in the autumn of 1951. The opposition against the Language Commission was combined with a demand for children's textbooks with "moderate" rather than "radical" Bokmål forms. (Cf. what was said above about the Oslo school-board's decision in 1939 to use "radical" Bokmål forms in textbooks.) On 10 December 1951 all ten Oslo newspapers carried on the front page a three-column, eight-inch advertisement headed: "Have You Read Your Children's Textbooks?" All eligible voters were asked to register their protest by signing the form provided in the advertisement, which was signed by seven well-known members of the Riksmålsforbundet. Parents were warned that the Storting was about to establish a Language Commission which would mix Nynorsk and Bokmål into a new language called Samnorsk. Three days later, when the debate began in the Storting, some 92,500 signatures were placed on the Speaker's table in protest on behalf of the Parents' Action. When the collection of signatures closed in 1953 the number had risen to more than 400,000. In the same year the protest took another, and more drastic, form, in that parents were asked to go through their children's textbooks and "correct" the radical Bokmål forms to traditional Riksmål ones. To begin with, prominent Labour politicians (and others) labelled the organizers of the Parent's Action a clique of West-End snobs, but by 1953 it had become clear that more than 400,000 protesters from all over the country could not be dismissed as a mere clique of snobbish West-Enders, and that among them were a great many Labour supporters. The Labour Party, which was in power at this time, was getting cold feet and began to have serious doubts about the wisdom of its language policy.
196
ARTHUR O. SANDVED
And a few years later the party made a volte-face. A truce was called. A committee (variously known as the "Vogt committee" — after the name of its chairman, Hans Vogt (1903-1986), principal of the University of Oslo, — and the "language peace committee") was set up with instructions "to consider the whole language situation in our country and propose measures which [...] may serve to protect and develop our Norwegian linguistic heritage" (Haugen 1966:273). One result of the committee's work was that the Language Commission was done away with and replaced in 1972 by The Norwegian Language Council (Norsk språkråd). Before we take a closer look at the Language Council, let me just mention very briefly a later, very important stage in the development that has been outlined above. This stage was reached in 1981. In that year there was a new spelling reform, in which many, but not all of the banned Riksmål forms were reintroduced into Bokmål. If we look at the 1981 reform purely in terms of number, the number of words and forms affected, it cannot be said to have been a very extensive reform. But looked at from a different point of view, it is no less than a turning-point in the history of Norwegian language standardization, because it runs counter to the tendency which had dominated Norwegian language planning ever since the beginning of the 20th century. It marks the demise of the amalgamation policy. The 1981 reform is generally referred to as the 1981 "liberalization". What, then, is the Norwegian Language Council and what does it do? It was established, as already mentioned, in 1972. It is the Norwegian government's advisory body in matters pertaining to language and language planning in Norway. It is composed of 38 members organized in two different sections of 19 members each, one for Nynorsk and one for Bokmål. Its members are appointed for a period of four years by a number of institutions and organizations: the universities, the schools, the press, the Norwegian Broadcasting Company, various societies and unions representing a variety of language users, authors, actors, translators. Among these is The Norwegian Academy for Language and Literature, which elects two members of the Bokmål section. More will be said about the Academy below. What is the purpose of the Language Council, and what does it do? As has already been hinted, the Council was meant to serve a purpose rather different from that of the 1952 Language Commission. There had been a very marked change of climate in the period between 1952 and 1972. A brief glance at the terms of reference for the Language Commission and the Language Council reveals this change very clearly. Whereas the Commission had been directed to "promote the approach to each other of the two written languages on the basis of the Norwegian folk language", the Council shall:
LANGUAGE PLANNING IN NORWAY
197
(a) protect the cultural heritage represented by the Norwegian written and spoken language, promote initiatives to increase the knowledge of the Norwegian language, its history, and distinctive quality, promote tolerance and mutual respect among all users of Norwegian in its different varieties, and protect the right of each citizen with regard to the use of the Norwegian language (b) follow the development of written and spoken Norwegian and on this basis promote co-operation in the cultivation and codification of our two language varieties and support trends which in the long run will bring the varieties closer together (c) give advice to the authorities in matters pertaining to the Norwegian language, in particular as regards the use of Norwegian in schools, in the Norwegian Broadcasting Corporation and in government bodies, submit statements on the principles of codification of the written language and place names, and propose legal measures in matters regarding the Norwegian language (d) give advice and guidance to the general public (e) promote and participate in Nordic co-operation for language cultivation, and (f) make the results of its work publicly known. (Norwegian Language Council Act of 18 June 1971; official translation.)
It will be seen that there is still a trace of the old idea of bringing the two varieties closer together. But the aim is much more modest now than it was in 1952. The fusion is now something that is to be achieved "in the long run" and only if trends in that direction are already discernible. And even this more modest aim has later been abandoned. In a report to the Storting in 1998 the Government states explicitly that it is no longer official policy in Norway to try to amalgamate Nynorsk and Bokmål An alternative goal is set up in the report: Instead of emphasizing trends in the development of [Bokmål and Nynorsk] which in the long run will bring the two varieties closer together, language cultivation should be concerned with the question of how Bokmål and Nynorsk separately [kvar for seg] can develop in the best possible way. {Report to the Storting No. 13 (1997-98) from the Ministry of Culture: 17; my translation.)
Very recently (December 2001) the Ministry of Culture has gone one step further and announced its intention of submitting to the Storting in the near future a proposal that the clause about supporting "trends which in the long run will bring the [two language] varieties closer together" should be removed from the Language Council Act. For a number of years now the clause has in practice been null and void. Still, its removal would, in the eyes of the Ministry, be "an important symbolic act" because it will be the "final confirmation" that the policy of amalgamation has now been abandoned {Report to the Storting, No. 9 (2001-2002):20).
198
ARTHUR O. SANDVED
In practical terms the most important work of the Language Council since its establishment in 1972 has been concerned with the codification or standardization of the two official language varieties, Bokmål and Nynorsk. As has already been made clear there is also a third standard of written Norwegian, and the rest of the paper will be devoted to this standard. In particular it will be concerned with the questions: When and why did it come into being?; and What is its relation to the official standards? The answer to the first of these questions is that it hasn't really "come into being". It has been there all along. This may sound a bit odd, so let me try to explain what is in fact a somewhat complex matter. Before 1923 the term Riksmål was, as pointed out above, a term used for the Dano-Norwegian variety of Norwegian. The replacement of this term in 1923 by the term Bokmål did not imply the demise of Riksmål, neither the thing itself nor the term used for it. The point is that the variety of Norwegian to which the term Bokmål was applied had, over the years, been drastically changed through the official codification process, especially through the 1938 language reform. Since the Second World War the term Bokmål has come to be commonly used for the officially codified, state-supervised variety of written Norwegian, a variety which in many ways is rather different from the variety referred to as Riksmål until 1923. But this new official Bokmål standard was simply not accepted by a large part of the Norwegian population. In the main they continued to write Riksmål. In other words the Riksmål standard continued to exist, and continued to be referred to as Riksmål. But now as an unofficial standard. Unlike the official standard Bokmål, Riksmål was codified not by the Storting, but to begin with by an academy-like society called the Riksmålsvernet. This society was established in 1919 and was triggered by the controversial spelling reform of 1917, which was passed in the Storting by the Speaker's casting vote. Since vern means "protection", perhaps we could translate its name as "The Society for the Protection of Riksmål". The formation of the Riksmålsvernet is of some importance because it was the first step towards establishing a body whose task it would be to codify an alternative standard, i.e. a standard different from the official, state-supervised one. Its first task was to begin work on the large dictionary Norsk Riksmålsordbok {Dictionary of Norwegian Riksmål). The Riksmålsvernet considered its task performed when its main project, the Norsk Riksmålsordbok, was completed; and from the mid 1950s the codification of Riksmål has been in the hands of The Norwegian Academy for Language and Literature, with which the Riksmålsvemet eventually merged. Today the Academy has 45 members, most of whom are authors, actors or scholars.
LANGUAGE PLANNING IN NORWAY
199
What is the raison d'être of the third standard, the Riksmål standard? Why has it survived and why does it continue to exist in spite of the fact that it is still not officially recognized? In order to explain that, it is necessary to take a closer look at the principles behind official language standardization in Norway. This standardization has taken a rather different form from what we find in countries where the language situation is different. More usually an academy, or some other codifying body, will regard it as its task to preserve and cultivate an already existing written tradition. In Norway the codification of the two official standards has had almost the opposite effect — or would have had the opposite effect if it had been successful. The goal of amalgamation could only be reached through a break with tradition, or at least a strong modification of the two traditions that were to be fused. The guiding principle was: What in the written tradition can be changed? not as in more normal circumstances: How do we preserve our tradition? There was opposition to the amalgamation policy on both the Nynorsk and the Riksmål side. And Nynorsk has its own academy-like institution, The Norwegian Society for Language Cultivation (Norsk Måldyrkingslag). But the opposition was particularly strong on the Riksmål side. A break with a literary tradition that included such writers as Henrik Ibsen (1828-1906), Bjørnstjerne Bjørnson (1832-1910) and Knut Hamsun (1859-1952) appeared to many people to be outrageous, even almost sacrilegious. It is probably true to say that it was the establishment of the Norwegian Language Commission in 1952 that triggered the formation of The Norwegian Academy for Language and Literature, which was established in 1953. The Academy formulated its main aims in more or less explicit opposition to the Commission. The Academy aims to "protect the Norwegian [Riksmål] and its literature [...] to work against any artificial or compulsory approach of the two language forms [...] and to work for a free linguistic development" (quoted from Haugen 1966:217). The situation today, then, is the following: we have two official written standards — Bokmål and Nynorsk — both codified by the Norwegian government (and ultimately by the Storting) on the recommendation of the Language Council, and a third unofficial standard, Riksmål codified by The Norwegian Academy for Language and Literature. Finally, what is the relation between Riksmål and the other two standards? That is a rather complex question, which I think must be approached in a roundabout way. As a result of the process towards fusion, the official standards, both Nynorsk and Bokmål, came to be extremely "loose" in the sense that a great many words and inflectional forms had two or more alternative spellings: one traditional form and another which was intended to bring the
ARTHUR O. SANDVED
200
standard in question closer to the other. As has been pointed out above, that was the means by which fusion was to be achieved. This means that within e.g. the Bokmål standard it is possible, by choosing the new alternatives, to write a "radical" Bokmål i-e. one that is closer to Nynorsk, and, by choosing the traditional forms, to write a "moderate" variety of Bokmål. Again an example might not be out of place. (The example has been provided by Mr. Tor Guttu; the English translation is mine.) A passage like this: (3) She advanced slowly towards the house and stopped at the foot of the stairs. At long last she was able to sit down and have a rest, enjoy a bit of peace and quiet, be on her own for a little while. would read like this in moderate Bokmål (after 1981): (4) Hun gikk langsomt frem mot huset og stanset nedenfor trappen. Endelig kunne hun sette seg og hvile, nyte roen og stillheten, være alene med seg selv en Uten stund. (This sentence is identical with its Riksmål equivalent.) And in radical Bokmål. (5) *Ho *gjekk langsomt fr am mot huset og stansa nedafor trappa. Endelig kunne *ho *setta seg og kvile, nyte roa og stillheta, *vcera aleine med seg sjøl ei stund. (This sentence is almost identical with its Nynorsk equivalent.) Starred words are forms that the Language Council in February 2000 suggested should be removed from the Bokmål standard, but the Government has not yet acted on this suggestion. Something similar is true also of the Nynorsk standard. Thus in both Nynorsk and Bokmål you get a radical and a moderate or conservative subvariant. The Riksmål standard on the other hand has always been much more homogeneous. What all this amounts to then is a cline of Norwegian varieties and sub varieties: 1. 2. 3. 4.
Conservative Nynorsk Radical Nynorsk (which approaches radical Bokmål) Radical Bokmål (which is very close to radical Nynorsk) Moderate Bokmål
LANGUAGE PLANNING IN NORWAY
201
5. Riksmål The 1981 liberalization mentioned above brought 4 and 5 much closer. And since 1981 minor changes have been made in both 4 and 5, which have reduced the distance between them even further. The result is that the differences between 4 and 5 are now so small that many people think there is little or no reason why different designations should be used for them. The differences that do exist can be summed up by saying that No. 5 is generally somewhat more conservative than No. 4. And the main reason for that is that the 1981 Bokmål liberalization stopped short of admitting all the traditional Riksmål forms which were then (and to a large extent still are) in fairly common use. On this point the Language Council did not follow the recommendation of the Vogt committee, which had criticized the policy of excluding from the official norm traditional Riksmål forms which are alive in speech and literature. A very few examples of such forms are given below. There are few if any systematic differences between 4 and 5. Perhaps the nearest thing to a systematic difference is found in the conjugation of weak verbs, where Riksmål in a number of cases has -et in the preterite while moderate Bokmål has some other suffix, often -te or -de, e.g.: (6)
Riksmål: samtykke — samtykket svelle — svellet øve — øvet
Bokmål: samtykke — samtykte ("agree") svelle — svelte ("swell") øve — øvde ("practise")
One rather conspicuous difference is found in the case of a handful of numerals: (7)
syv tyve tredve enogtyve toogtyve etc.
sju tjue tretti tjueen tjueto
("seven") ("twenty") ("thirty") ("twenty-one") ("twenty-two")
A number of individual words are spelt differently in the two sub-varieties, among which are a small group of words with initial (silent) h- (dropped in Bokmål), e.g.:
ARTHUR O. SANDVED
202
(8)
almue grunde hylde hvalp hveps hverken [...] eller
allmue ("(the) common people") grunne ("speculate") hylle ("hail", "pay homage to") valp ("puppy") veps ("wasp") verken [...] eller ("either [...] or")
Another rather marked difference between Riksmål and moderate Bokmål is that the former is considerably more reluctant than the latter to norwegianize the spelling of loan-words, e.g.: (9)
fondue protegé schnitzel spaghetti souvenir telefax toddy
fondy protesjé snitsel spagetti suvenir telefaks toddi
Now what about the relative strength of these five varieties? To what extent are they used? That is a difficult question. For one thing the figures 1-5 represent a cline; they are not compartments. Secondly we have no statistics to help us. We do have some idea of the percentage of the population using Nynorsk (1 and 2 together). In private use Nynorsk is written by about 10-12% of the adult population. No official estimate of the extent to which No. 3 is used is available, but it is probably true to say that this variant is very rare, perhaps one could say extremely rare. In its most extreme form it is not even used by its own supporters in the organization Language Unification (Språklig Samling). In other words it is a fair assumption that 4 and 5 are used by the vast majority of Norwegians. Probably more than 90% of Norwegian newspapers, books, and magazines use this variety.
Conclusion We have to conclude, then, that the official language policy of varying Norwegian governments through most of the 20 th century, which aimed at a forced and speedy amalgamation of two language standards into a new form of written Norwegian, Common Norwegian (Samnorsk), has resulted in a situation where there are three standards, two of which are so "loose" that they
LANGUAGE PLANNING IN NORWAY
203
can be said to exist in different varieties. Hardly a result that anybody foresaw when the process towards amalgamation was set in motion, and certainly not one that many then would have wanted.
References The following list contains not only works directly quoted but also one or two other publications which I have found particularly useful in preparing this paper: Guttu, Tor. 1993. Noen aktuelle spørsmal i normeringen av riksmalet. Tanke og omtanke. Festskrift til Asbjørn Aarnes på syttiårsdagen ed. by Hans Kolstad, 327-353. Oslo: Aschehoug. Guttu, Tor. 1997. Riksmålsforbundet — en 'underlig forening'?. Ordet 1. 3032. Haugen, Einar. 1966. Language Conflict and Language Planning: The Case of Modern Norwegian. Cambridge, Massachusetts: Harvard University Press. Langslet, Lars Roar. 1999. I kamp for norsk kultur: Riksmalsbevegelsens historie gjennom 100 år. Oslo: Riksmålsforbundet. The Norwegian Language Council Act of 18 June 1971. Report to the Storting No. 13 (1997-98) from the Ministry of Culture. Report to the Storting No. 9 (2001-2002) from the Ministry of Culture. Selberg, Ole Michael. 1993. Enhetsvisjon og språkpluralisme i Norge. Standardspråk og dialekt. Seminarer i Oslo 1991 og 1992, 179-195. Bergens Riksmålsforening og Det Norske Akademi for Sprog og Litteratur. Øverland, Arnulf. 1948. Hvor ofte skal vi skifte sprog? Oslo: Aschehoug.
'DEMOCRATIC' AND 'ELITIST' TRENDS AND A FRISIAN STANDARD ANTHONIA FEITSMA Grou, The Netherlands
1.
Introduction Frisian and English historically belong to the North Sea or coastal Germanic (or Ingveonic) branch of the West-Germanic language group, whereas Dutch and German belong to the Continental branch of WestGermanic. Frisian still shows a considerable similarity to English, especially in its articulation, for example by its frequent palatalization and the deletion of nasals before [s], [f], and [6/ð]. To give some examples of palatalization: West Frisian tsjerke (English church), brêge (English bridge), rein (English rain); the deletion of the nasal before [s], [f], and [0/ð] can be exemplified by West Frisian us (English us), goes (English goose), oar (English other), fiif (English five). Different Frisian languages are spoken in different regions along the North Sea coast in the Netherlands and Germany. In the Dutch province of Friesland in the Netherlands, West (or Westerlauwers) Frisian is spoken by about 350,000 people; in Germany, East Frisian is still spoken by about 1000 people in Seelterlound (Saterland) and in North Friesland along the west coast of Schleswig, south of the Danish border, about 10,000 people still speak one of the dialects of North Frisian. The Frisian languages are mainly spoken in the countryside and are nowadays more and more influenced by the national languages of the Netherlands and Germany. This paper outlines some perspectives on the standardization of West or Westerlauwers Frisian (from now on called Frisian) from the Middle Ages up to the present day. The Frisian language has developed along two lines: a literary, archaic — or 'elitist' — norm, and the people's vernacular or 'democratic' norm, the term democratic here used in its ideological rather than in its statistical sense. At the same time, the Frisian standard has oscillated between a more Frisian and a more Dutch norm. The development of the language standard and the language ideology are linked to these contrasts and to the interaction between
206
ANTHONIA FEITSMA
them. In this interaction Dutch, as the more powerful language, plays the dominant role. The social position of a language in a community is of crucial importance to its standardization. From the Middle Ages onwards Friesland had a mainly rural character; it was relatively autonomous but not monolingual. It lost its independence in 1498 and written Frisian disappeared during the course of the 16th century. The standardization of the Frisian language was thus left to private enterprise. Linguistic standards have essentially quite different origins. As a rule standardization moves in the direction of a prestigious urban standard. However, it is said that modern standard Italian is "to a considerable extent virtually the single-handed creation of one man", Dante (Padley 1988:245; although Joseph (1987:64-65) considers the Italian standard rather as a partly standardized form of Florentine). The Frisian poet Gysbert Japicx (16031666), for want of an exemplary Frisian literature, created a Frisian standard language by himself. In the 19th century period of Romanticism following the 'awakening' of European nations, this Frisian standard was consolidated on the basis of the spoken vernacular of the countryside. This was a basis similar to that of the Nynorsk (Haugen 1966:27-61), Icelandic (Joseph 1987:58-60, 114) and Serbo-Croat languages (Gvozdanovic 1987:73-82). It was the linguist and popular Frisian writer J.H. Halbertsma (1789-1869) who acted as a pioneer in this respect, although Halbertsma himself proclaimed the superiority of the natural vernacular, without imposed literary and linguistic rules. Old Frisian (11th century -1500/1550) In the Middle Ages Frisian was used as a written language along with Latin and Dutch. The oldest Frisian texts date from the 11th century although the surviving manuscripts date only from about 1300. Old Frisian literature principally consists of legal texts. Their quantity is limited. The oldest Frisian texts are written in a rather homogeneous and archaic language influenced by Latin. There must have been a prior development towards some sort of linguistic and literary standard. Later on, the language became much more heterogeneous, with a greater influence brought to bear upon it from Middle Low German and Middle Dutch. The Frisian deeds date from the 14th and 15th centuries. The first of these, from 1329, was written in Frisian because of its intelligibility, as stated in the deed. In addition, these deeds in Frisian provide evidence of a certain sentiment of independence. Apart from the formulas and alliterative expressions in testaments etc., their style is not high and more resembles colloquial and everyday language (Looper 1984). 2
STANDARDIZATION OF FRISIAN
3.
Gysbert Japicx
207
(1603-1666)
3.1 Background In general, during the Reformation it was only the vernacular which had a contemporary written tradition supported by social, political and economic power that was able to survive as a language for school, church and administration (Feitsma 1989b:2). Frisian, however, no longer had a current written tradition, so that the development of Frisian in a more formal and standardized direction was frustrated. The language was mainly spoken in the countryside (Vries 1993). Frisian had lost its power and its written tradition: it represented the old and old-fashioned cultural tradition. In the somewhat larger Frisian towns the main spoken language for the upper echelons in society and, later on, even for others, had become some variety of Urban Frisian which lay somewhere between Frisian and Dutch. However, the existence of Friesland as a separate political entity, the former mediaeval written tradition and the functioning of actual spoken Frisian, mainly in the countryside, seem to have preserved the Frisian sense of identity and language ideology. Such circumstances left room for the poet Gysbert Japicx to create modern literary works of high quality and a new orthography in line with the generally accepted rules for the cultivation and regulation of the language at the time of the Dutch Renaissance. The aspirations of Gysbert Japicx as a language builder became evident in several different respects (Gysbert Japicx 1668). 3.2 Orthography and lexicon Gysbert Japicx criticized the Old Frisian orthography and based his new orthography on the sound values attributed to the different characters in Dutch. Within this context he tried to create a more systematic whole: he systematized the use of Old Frisian and English combinations (like uw and oa; ea and oa), whereas the circumflex (in cases like â, ô, êa and od) seems to point in the direction of French origins. The Gysbert Japicx orthographic standard represents a rather consistent phonological orthography that closely adheres to the pronunciation of Frisian. One cannot but admire the consistency of his orthography and the efforts the author made to improve it. As for the lexicon, Gysbert Japicx imitated the language of contemporary Dutch poets to a high degree by way of loan translations and new compounds. Most conspicuous are the sometimes very long and complicated compounds, especially in his more elevated and learned texts.
208
ANTHONIA FEITSMA
On the other hand, in later versions of some texts the poet altered some words in a more Frisian direction. Some of these words become more specific, more marked and as such more Frisian, but it is difficult to decide whether all these Frisian words were actually used in contemporary daily speech or whether some were new creations of the poet himself. The tendency towards the ordinary spoken language also found expression in the use of popular proverbs and proverbial expressions. Collecting proverbs was, for that matter, a prevailing occupation for the writers of the day. In order to characterize country people, curses or primitive expressions are sometimes used, so that these texts not only demonstrate differences in dialect but even indicate differing ranks of society. Both the imitation of the higher style in Dutch and French and the use of dialectal words correspond to the literary trends of the time. For some French writers of the Pléiade in the 16th century, for example, the use of dialectal forms represented a component of their literary programme. 3.3 Aim and impact of the works of Gysbert Japicx In prefaces addressed to his Frisian readers the poet himself explained that he wanted to contest the opinion that the Frisian language could not be written and that to write poetry in Frisian was impossible. The linguistic and literary pretensions of the work of Gysbert Japicx created a new symbol for Frisian identity and, indeed, a Frisian standard. His works had a great impact upon the then modest production of Frisian literature, particularly of the more serious sort. This influence continued even up until the beginning of the 19th century. Gysbert Japicx started out from the (or a) language spoken in his time. His followers, also his later followers, regarded his written language and orthography as a norm which they had to obey, though in the meantime the spoken language had changed. However restricted the impact of Gysbert's linguistic norm may have been in the more popular domain, his language carried authority with the more ideologically orientated Frisian public. In more popular and/ or comic pieces of Frisian literature a more simple orthography was often used. When, however, in 1777, an author such as Eelke Meinerts (1732-1810) wished to write for the ordinary people in a more direct way, some readers criticized him because of his simple orthography, which was not in accordance with the orthography of Gysbert Japicx and thus with Frisian language ideology. The poem concerned had to be reprinted in the Gysbert orthography.
STANDARDIZATION OF FRISIAN
209
3.4 Johannes Hilarides (1649-1725) Johannes Hilarides, who mainly wrote in Dutch, created a Frisian orthography (as part of a universal orthography), which, like that of Gysbert Japicx, corresponded to a large extent with the pronunciation. He shared Gysbert's interest in dialects, but nevertheless his orthography turned out to be very different. In addition, whereas Gysbert Japicx created a literary, more elitist standard, for Hilarides language was, in line with his enlightened ideas, mainly a means of communication with the ordinary people and thus his standard was more democratic. All the same, the ideas of Hilarides had very little impact on the further development of the Frisian language and orthography. 4. J.H. Halbertsma (1789-1869) The first decades of the 19th century gave rise to new Romantic literary genres and more modern linguistic forms than had been current in the 17th century. There were two new trends in orthography: a simpler one (as in the first Frisian texts of Halbertsma in the 1820s), and another more related to Old Frisian. However, the prestige of Gysbert's work was so great that in practice the influence of his orthography prevailed. This trend was corroborated by Joast Hiddes Halbertsma, who characterized his own orthography (of 1834) as a compromise between the orthography of Gysbert Japicx, which essentially corresponded with the Dutch one of his day, and current Country Frisian, as it was spoken in his own time, so that the readers of modern Frisian would be able to understand the language of the old poet too, and he [Halbertsma] thought that he had succeeded in that goal tolerably well. (1861:872).
In discussing the origin of the present Frisian linguistic standard, the influence of the different Frisian dialects is often considered to be the central question. I think rather that the impact of Gysbert Japicx should be the initial consideration. Only after first having determined the contributions of Gysbert Japicx and Halbertsma does it make sense to investigate the part played by what is referred to as 'dialect selection' in the present linguistic standard. Not only was Halbertsma a linguist, but also, together with his brother Eeltsje, he wrote a lot of popular literary texts in Frisian, dating from 1822 onwards. The work of these two brothers developed into a great classic collection of popular Frisian literature (Rimen en Teltsjes 1871). Halbertsma thought it necessary for the spoken Frisian vernacular to be supported by the written form of the language since the impact of Dutch was becoming ever greater on account of the growing influence of written Dutch. Halbertsma's later spelling became more 'grammatical' and archaic, but his starting point
210
ANTHONIA FEITSMA
remained everyday speech, enriched by dialectal and somewhat older forms which were in danger of disappearing. He used a great many colourful Frisian expressions, but he was not a purist. In general Halbertsma corroborated the colloquial contemporary norm of 'Country Frisian' and the popular writers of the 19th century followed his lead. The considerable number of Frisian booklets (with texts, plays and songs) for the people contributed a great deal to the literacy of Frisians in their own language and thus to a Frisian standard. Since the origin of that standard is mainly to be found in the language of the ordinary people, the middle class in the countryside, and in the more or less popular literature of the 19th century, the Frisian standard is strongest in these domains and weaker or less developed in the domains of higher literature and formal language. Apart from in the work of Gysbert Japicx, these latter domains hardly, for that matter, existed in Frisian. When Frisian was used in these domains the impact of Dutch was often clearly visible. There are some specimens of Bible translation in Frisian dating from the 19th century. Halbertsma's translation of the Gospel of St Matthew (1858) is written in a language quite close to the Frisian spoken vernacular, roughly according to the principle of what several Bible translators would now propagate as 'dynamic equivalence'. Such a translation should not aim to be a literal reproduction of the original text, but rather to transfer the meaning of the words in the source language to the target language. Halbertsma thought that Frisian was better suited to translation of the Gospel of St Matthew than was Dutch. Indeed, in his opinion the authentic vernacular used by ordinary people was superior to a formal aristocratic language, a language of 'convention'. Other Bible translators preferred a more elitist standard, particularly for purposes of Bible translation. It was not considered respectful to use the spoken vernacular when religion was discussed, and this did not apply only to orthodox people. Some Frisian authors, such as S.K. Feitsma (1850-1918), even characterized and caricatured orthodox people by imitating their language: full of Dutch and Dutch-coloured biblical terms when they spoke about religious matters. Here the elitist and the democratic standard clashed in ordinary speech. S.K. Feitsma, for that matter, criticized the use of more Dutch-coloured expressions even in more colloquial texts (S.K. Feitsma 1991: I:21, 23-24; II:[58]). Halbertsma, however, proclaimed his Romantic ideology concerning the authentic and democratic Frisian vernacular, with its various dialectal forms and its everyday speech. In his opinion, language, as the common creation of all people, be it high or low, was by nature democratic par excellence (Halbertsma 1851:6). In Frisian there existed, argued Halbertsma, no gap between spoken and written language (Halbertsma 1865:8). It is partly due to
STANDARDIZATION OF FRISIAN
211
his influence that the Frisian standard was and is less stylistically elevated than is its Dutch counterpart, and dialectal expressions in Dutch often find parallels in the Frisian standard language. A still more radical supporter of a democratic norm was Foeke Buitenrust Hettema (1862-1922). He used the term 'democratic' here in a more statistical way. He opposed every linguistic norm for Frisian, claiming that the everyday speech in Friesland constituted the general Frisian language. Development in the language should be considered positive, even if it was a development towards Dutch and away from Frisian. This was an exceptional point of view; all other authors, be they adherents of the elitist or of the democratic movement for the establishment of norms, always tried to preserve as many Frisian features as possible. 5. Harmen Sytstra (1817-1862) Some people, however, held quite different opinions about language. In their opinion there was (and ought to be) a substantial difference between the spoken and written language. They even held the view that Gysbert Japicx had failed to respect the true grammatical rules of Frisian. The poet had only written in the spontaneous Frisian of the countryside and not in the complete and truly regulated language, with consistent use of gender and endings, as in written Dutch. Attempts to arouse townspeople's interest in Frisian and to revive in their social circles "a language that had been far too much forgotten" (the goal of the Friesch Jierboeckjen (1829-1835) published by the Friesch Genootschap) also demanded a cultivated written language. Only such an elitist standard could fulfil the requirements of this public. In the opinion of the Danish linguist Rasmus Rask (1787-1832) the Frisian language had died out (Rask 1825). Frisian had not existed since 1300; only remnants, traces of it were left. Some Frisians blamed the Frisian orthography for this 'misunderstanding' on the part of Rask. The solution was simple: if modern Frisian had been written in the Old Frisian way, Rask, and others of his persuasion, could have been easily convinced that they were wrong. In his grammar Harmen Sytstra tried to create a general written Frisian language that corresponded to the 'regulated, pure, old' language. He elaborated a grammatical system and an orthography that covered the whole spectrum of Frisian, both historically and geographically (including East and North Frisian). Particularly conspicuous are the restored Old Frisian endings and a great number of diacritical signs. All civilized nations, Sytstra argued, had such a general written language or — as he called it — a 'notional' dialect. Frisian had failed to develop such a complete language and this was what Sytstra's grammar was meant to put right.
212
ANTHONIA FEITSMA
Harmen Sytstra's rather archaic Iduna orthography was considered too difficult to read and thus not appropriate for the promotion of Frisian literacy. Authors of popular literature, and later authors of learners' grammars, used the contemporary spoken language as the basis for spelling. On the other hand, these grammarians often fought against what they referred to as 'patois'. Thus, in principle, their linguistic standard conformed more or less to the elitist tradition of Sytstra. Later, in the second half of the 19th century, more explicit and practical language activities developed. In 1879 a committee of the Selskip foar Fryske Taal en Skriftekennisse agreed on a 'new' orthography, which was to be used for more than half a century. In 1872/1876 there appeared the Lexicon Frisicum (A-Feer), the first part of Halbertsma's never completed Frisian-Latin dictionary which was subsequently more or less resumed by Waling Dykstra's Frisian-Dutch dictionary (1900-1911). This dictionary was funded by the provincial government of Friesland. These activities marked the beginning of a more concrete and pragmatic Frisian language policy. 6. After 1915 The Young Frisian Movement (founded in 1915) originated from a more intellectual background than that of the older writers. Their ideals were more international and cultural, in contrast to those of 19th-century popular literature. The Young Frisian leader and English specialist Douwe Kalma (1896-1953) claimed that Frisian literature should be a general and complete literature. "Friesland and the World" was the Young Frisian slogan. Hence Frisian should, according to them, be a complete language of culture, and they wanted a linguistic standard based on a written language with purisms, neologisms and archaisms, removing it some distance from everyday speech. Moreover, their goals involved the official use of Frisian in administrative affairs and in governmental bodies. Kalma's adversary, the journalist and dialectologist Jan Jelles Hof (18721958), advocated a Frisian literature which expressed specific Frisian features: Frisian literature should be complementary in respect to Dutch literature and the Frisian language should mainly function as a vernacular, as the people's language, with typically Frisian features and idioms. He was an expert on the Frisian language, but did not like the idea of an elitist norm. This is where we see the old contrast between the elitist and the democratic standard emerging in a new form. The AFUK (a private educational institution founded in 1925/1928 for the organization of Frisian courses and examinations) was often considered the authority that ought to establish the Frisian standard because it brought
STANDARDIZATION OF FRISIAN
213
together people who were at the heart of Frisian linguistic life and knowledge. Notwithstanding their quite different opinions concerning the language standard, within AFUK both parties more or less respected one another and were able to co-operate. Regional novels of the 1920s and 1930s were being written in a language that came close to the generally accepted Frisian standard — both elitist and democratic. This was the contemporary language of the countryside, omitting some non-Frisian elements and certain vulgarisms. Furthermore, authentic Frisian linguistic features in terms of vocabulary and syntax occur perhaps more frequently in these books than in ordinary speech. This language, in turn, had a definite impact upon the standardization of Frisian. It was only now that action got underway to initiate the teaching of Frisian in schools. 7. Some aspects of the present situation Having had very little or no administrative support to begin with, Frisian has nevertheless attained a certain position of strength within the Frisian community. The Reformation and Renaissance had failed to bring to bear any positive influence on the position of the Frisian language, its only strong support coming from the work of the Frisian 17th-century poet Gysbert Japicx, followed by 19th-century Romanticists such as Halbertsma. Meanwhile, Frisian has managed to attain a (rather flexible) linguistic norm of its own. Nowadays Frisian receives administrative support that, though modest, is nevertheless greater than ever before and certain goals in Frisian school teaching have been attained during the course of the 20th century. There exists a literature of a certain level and of a certain breadth. There is considerable scholarly activity, in the last decades even in the sociolinguistic domain. Perhaps this sociolinguistic activity represents a principal innovation regarding the approach to Frisian linguistic problems. As to the character of the present-day Frisian linguistic standard: it is rather flexible and leaves room for several different opinions and practices. Furthermore, the Frisian norm and the Frisian written style contain more elements derived from the spoken language and from different dialects than does the norm of the Dutch written language. The present-day Frisian linguistic standard bears traces of its democratic and Romantic origins (Joseph 1987:120) in, for example, its flexibility, its dialectal variety, its close connection to the spoken language and in the frequency of connotative features. These are all non-standardized elements which the Frisian standard has in common with authentic vernaculars. This situation illustrates the interposition of Frisian between 'folk speech' and 'standard language' (Sjölin 1976; Feitsma 1980 and 1987). That these elements are more or less
214
ANTHONIA FEITSMA
contradictory to the pursuit of linguistic standardization expresses the paradox characteristic of the ambivalent Frisian attitude in this regard. The authentic vernacular remains an essential component of Frisian language ideology. Moreover, part of the process of standardization for the Frisian language is, of course, the tendency to affirm its identity and to set it apart discretely from Dutch. However, the situation for both spoken and written Frisian is complicated by the strong influence of Dutch. It should be kept in mind that school teaching in Friesland mainly takes place in Dutch, whereas the teaching of the Frisian language occupies a very modest place in the curriculum. Hence ideas about the linguistic standard in Friesland are strongly influenced by Dutch. The Dutch norm is quite unconsciously internalized as the linguistic norm, even for Frisian. This is another reason why various characteristics of Frisian are in danger of being suppressed by a hidden Dutch norm whilst many Frisians will not even be aware of the process. Moreover, the language of incomers from non-Frisian networks of communication affects the spoken language in a Dutch direction. In this connection it is worth noting that Frisian utterances with relatively many Dutch features are often considered 'modern', whereas, seen from this viewpoint, utterances carrying more Frisian features tend to sound old-fashioned or archaic. Some linguists (and some non-linguists too, for that matter) regard a more Dutch-coloured Frisian as the normal colloquial variety, in contrast to a more correct linguistic usage which might seem less appropriate in contact with certain people. The question is, however, whether the use of such linguistic variety depends on social factors, such as different interlocutors, or on greater or lesser mastery of the language. In a language like Frisian, which is minimally regulated by a standardized style, the latter possibility might be more probable. Nowadays it is, indeed, less a matter of course than it was in earlier days to speak and write according to the current rules. This applies to language use in general and to Frisian in particular. In addition, the dominant Dutch language has always had a relatively important influence on Frisian, but it was not until the 1960s that the impact of Dutch on spoken and written Frisian began to increase substantially. The immigration to Friesland of many non-Frisians and the growing impact of the media and of the Dutch written and spoken language, rendered the Frisian countryside and its communication networks less homogeneously Frisian (cf. Joseph 1987:75). This caused a serious deterioration in the position of the Frisian language. In order to survive, the language must be consciously supported by speakers, teachers and administrators. Yet even linguists often hold the rather Romantic view that language ought to develop unconsciously, in a 'natural' way (cf. Joseph
STANDARDIZATION OF FRISIAN
215
1987:9-11). Thus the progress towards a true Frisian standardization still meets with serious difficulties. However, I am unable to deal here with all the various implications and implementations of Frisian standardization. The Frisian Language Movement had its centre for centuries in literary activities. Such a base turned out to be too narrow for the survival of Frisian. Too little attention was paid to the material infrastructure of the Frisian community and to the social infrastructure of the Frisian language. Frisian authors often overestimate the influence of literary and linguistic work and fail to consider these within the framework of the social situation as a whole and hence of a struggle for power in which Frisian remains far weaker than its Dutch assailant. Ultimately power is the deciding factor, even in matters of language and linguistic norms. A sociological approach which might contribute to an analysis of the situation and of language attitudes represents an essential condition for the defence of the position of the Frisian language. 8. Conclusion I will follow Joseph's description (1987:50-56) of the process of standardization and apply it to the Frisian situation. In the last resort, language standardization is dependent upon prestige. Latin has long been the model of a standard language, a high language (H). Speakers of other languages tried to imitate the model of Latin (or other high languages) in order to create their own standard: they developed the standard L. At first the L elite must learn H, with the goal of showing themselves the equals of their H-community models. Later, the avant-garde of acculturation within the group of L-speaking Hlearners desires to perform H-functions in L. In this situation it becomes possible for them to create a new standard language L. Yet this development has a paradoxical character which simultaneously represents a central difficulty in attaining the standardization of their language: the speakers of the new L standard resent H at the same time as envying its prestige. They oppose H and they want to cultivate L, while they are obliged to borrow various elements from H. This paradoxical situation roughly opens up three possibilities. Firstly, an L-speaker can try to ameliorate his position by learning and speaking the Hlanguage (and 'forgetting' the L-language). Secondly, he may try to develop the L-language within the context of the same functions and in the same elevated style as the H-language, an option which implies a great impact of the H-language on the L-language. And, thirdly, he may develop and cultivate the L-language as a real authentic vernacular on the same level as the H-language. Of course, there exist many intercurrent possibilities between the three mentioned here. Nevertheless, the tripartite division roughly characterizes the
216
ANTHONIA FEITSMA
possible attitudes to Frisian. The second attitude is perhaps best exemplified by Harmen Sytstra and Douwe Kalma, the third by J.H. Halbertsma and J.J. Hof. As a rule, the prestige of the standard is connected with urban centres of power and upper-class speech, but other factors may play an important role too. Joseph (1987:58) mentions the example of the Icelandic language, in which the standard is based on lower-class rural speakers because their language was thought to be closer to the 'pure' Icelandic of an earlier era than that of the Danish-influenced upper-class speech of the time at which the standard was formed. In the Introduction are mentioned the examples of Nynorsk and Serbo-Croat. The Frisian case shows a similar development. The standardization of Frisian is characterized by a strong vernacular element. Therefore we observe in the Frisian standard, for example, uncertainty between the high and low standardization model and the significant contribution of the vernacular and of use of the language in the countryside. Another important characteristic of the Frisian situation is the presence of a considerable language ideology in spite of the rather weak infrastructure. Indeed, the want of a State clearly symbolizes an obvious weakness in the material infrastructure.
References Breuker, Ph. H. 1989. It Wurk fan Gysbert Japix. Ljouwert: Fryske Akademy (3 vols.). Dijkstra, Waling. 1898-1911. Friesch Woordenboek {Lexicon Frisicum) (I-IV). Leeuwarden: Meijer & Schaafsma. Feitsma, Anthonia. 1980. The Frisian Native Speaker between Frisian and Dutch. Linguistic Studies offered to Berthe Siertsema ed. by D.J. van Alkemade [et al.], 335-339. Amsterdam: Huis aan de drie grachten. Feitsma, Anthonia. 1987. Geschiedenis en karakter van de Friese norm. Variatie en Norm in de Standaardtaal ed. by J. de Rooij, 57-70. Amsterdam: Publikaties van het P J . Meertens-Instituut. Feitsma, Anthonia. 1989a. The History of the Frisian Linguistic Norm. Language Reform. History and Future, vol. IV ed. by István Fodor and Claude Hagège, 247-272. Hamburg: Helmut Buske Verlag. Feitsma, Anthonia. 1989b. The Reformation and the Vernacular. Fourth International Conference on Minority Languages II: Western and Eastern European Papers ed. by Durk Gorter et al., 1-10. Clevedon — Philadelphia: Multilingual Matters.
STANDARDIZATION OF FRISIAN
217
Feitsma, S.K. 1991. Ut 'e underste lagen (oarspronkelike realistise forhalen) mei in ynlieding fan Tr. Riemersma. I-H. Bolsert: Koperative Utjowerij. Gvozdanovic, J. 1987. Variatie en Norm in het Servokroatisch. Variatie en norm in de standaardtaal ed. by J. de Rooij, 71-84. Amsterdam: P J . Meertens-Instituut. Gysbert Japicx. 1668. Friesche Rymlerye. Bolsert: Samuel fen Haringhouk. (Second printing: Leuward: Kerst Tjallings in Gerrit Heegslag, 1681.). Halbertsma, J.H. 1834. Friesche Spelling. Friesch Jierboeckjen foar it Jier 1834 5. 1-54. Halbertsma, J.H. 1851. Aanteekeningen op het vierde deel van den Spiegel Historiael van Jacop van Mae riant (Inleiding). Deventer: J. de Lange. Halbertsma, J.H. 1858. Het Evangelie van Mattheus, vertaald in het LandFriesch door J.H. Halbertsma. Londen: Impensis Ludovici Luciani Bonaparte. Halbertsma, J.H. 1861. [Review of] B. Bendsen, Die Nordfriesische Sprache nach der Moringer Mundart (Leiden: E.J. Brill 1860). De Gids 1. 870-885. Halbertsma, J.H. 1865. De vertaling des Euangeliums van Mattheus in het Land-Friesch. De Vrije Fries 10. 1-78. Halbertsma Hiddonis filius, Justus. [1872]. Lexicon Frisicum A-Feer. Ex typographeo Jani de Lange Daventriensis. (Editions 1874/1876 Hagae Comitis Martinus Nijhoff.). Halbertsma, Broárren [ = Joast, Eeltsje & Tsjalling]. 1871. Rimen ind Teltsjes. Dimter: J. de Lange. Haugen, Einar. 1966. Language Conflict and Language Planning. The Case of Modern Norwegian. Harvard: Harvard University Press. Joseph, John Earl. 1987. Eloquence and Power. The Rise of Language Standards and Standard Languages. London: Frances Pinter. Loon Jz., J. van (et al.). 1879. De Fryske Boekstavering, in Hantlieding for hwa yn 't Frysk skriuwe wolle. Utjown troch it Selskip for Fryske Tael en Skriftenkennisse. Hearenfean: N.A. Hingst. Looper, Bert. 1984. Ta in better begryp. Latyn en folkstaal yn de oarkonden ut de Fryske lannen tusken Fly en Wezer 1200-1400. It Beaken 46. 1-14. Meinderts, Eelke. 1777. Volle Gelok en Wolkomst Winsk, oon [...] Willim de Vijfte, [...]. Op den 28 Augustus 1777. Dockum: H. Groenje. (First and second printing ed. by Chr. Stapelkamp: Groningen: Frysk Ynstitut oan de R. U. To Grins, 1953.). Padley, G.A. 1988. Grammatical Theory in Western Europe 1500-1700. Trends in Vernacular Grammar II. Cambridge: Cambridge University Press. Rask, Rasmus. 1825. Frisisk Sproglære udarbejdet efter samme Plan som den islandske og angelsaksiske. København: Beekens Forlag.
218
ANTHONIA FEITSMA
Sjölin, B. 1976. "Min Frysk". Een onderzoek naar het ontstaan van transfer en "code-switching" in het gesproken Fries. Amsterdam: Bijdragen en Mededelingen dialectencommissie KNAW L. Sytstra, Harmen. 1854-1862. Friesche Spraakkunst. Leeuwarden: D. Meindersma, Wz. and (1862) H. Kuipers. Vries, Oebele. 1993. "Naarploeg en koestal vluchtte uw taal". De verdringing van het Fries als schrijftaal door het Nederlands (tot 1580). Ljouwert/ Leeuwarden: Fryske Akademy.
YIDDISH NO STATE, NO STATUS — NO STANDARD?
ANE KLEINE University of Trier, Germany
The title of this paper contains a question-mark: "Yiddish: No State, No Status — No Standard?" However I should admit at the outset that I am quite sure that there is a standard, as I will seek to demonstrate here — even though Yiddish has no state, nor any governmental forces to strengthen any standard variety. Little research has been done so far on the standardization of modern Yiddish, especially with regard to pronunciation (Kleine 1998:201ff.) — the focus of this paper. Here I will try to point out in what ways various kinds of language contact may have contributed to establishing a supraregional standard Yiddish language. Central to the discussion is the differentiation of two types of language contact, interlanguage contact and intralanguage contact, which must be considered in their respective (socio-)historical contexts. By interlanguage contact I mean the coexistence of different languages in one territory, whereas by intralanguage contact I mean the coming together of Yiddish speakers with different dialectal backgrounds. Before elucidating the factors responsible for the rise and establishment of a Standard Yiddish (in pronunciation), I will give a brief introduction to the Yiddish language for those who are not familiar with it (and for those who are conversant with the discussion about the origins of Yiddish, it will make clear which scholarly tradition I follow). Yiddish has always coexisted with several other languages, and its speakers have not only lived alongside other vernaculars but right in the middle of them (doubtless needing to get at least a smattering of those languages). However, at no time did Yiddish discard the use of the Hebrew alphabet. Yiddish offers a peculiar insight into standardization phenomena, because of negative criteria: its evolving a standard language in spite of the lack of circumstances which are common to most other Germanic languages — and to
220
ANE KLEINE
almost all European languages. It has evolved its modern standard language in fairly recent times and without a political agency taking responsibility for it, even without an independent political quasi-national unity being envisaged or advocated (for its territory was always larger than any one of the European countries where it was spoken, and no one country was able to claim a majority of Yiddish speakers). The Yiddish Utopia was a transnational 'cultural autonomy'. Because of this extraordinary situation, Yiddish has played a role in helping (socio-)linguists either acquainted with or numbering themselves among younger participants in the standardizing process to pinpoint factors of a more private nature which may fuel an ongoing standardization. Because of this exceptional situation, Yiddish still remains a valuable test for many assumptions about standardization processes. The focus on the influence of various linguistic contacts allows a long-term view of the impact of the geographical dissemination over a great many countries which makes the Yiddish speaking community a very numerous overall population, while still being everywhere a minority group; that is, it allows a unified view of this aspect over the whole of Yiddish history. From its beginnings in the late 9th century, based on middle- and upperGerman dialects, the language of Ashkenazic Jewry has been a fusion language, incorporating Hebrew-Aramaic as well as Romance elements into its German component. With its continuing spread, Yiddish was increasingly shaped by numerous co-territorial languages. First it was influenced by local varieties of the surrounding German. With the great migratory movements to Eastern Europe during the 'Black Death' and the Crusades, German then gave way to the Slavic languages of the new home countries as a potential source of language innovation. Vocabulary, but also grammar, syntactic features and the system of phonemes were affected. As a result, Eastern Yiddish began to develop differently from the Western variety. Over the centuries, Yiddish has been the language of communication and literature in Germany, the Netherlands, Switzerland, Austria and Northern Italy — that is roughly the territory of Western Yiddish — and Poland, Hungary, Romania, Czechoslovakia, the Baltic countries, the Ukraine and Belarus, where Eastern Yiddish was spoken. While Western Yiddish almost vanished in response to the Enlightenment (Haskalah) furthered by the German philosopher Moses Mendelssohn,1 Eastern Yiddish grew and prospered on Slavic speaking territory. It is perhaps appropriate to note here that before World War II Yiddish was the third largest Germanic language, with about 12 million native 1
Moses Mendelssohn (1728-1786) and his followers advocated language assimilation of the Jews to Standard German, as a prerequisite to emancipation.
YIDDISH: NO STATE, NO STATUS — NO STANDARD?
221
speakers. It is natural that such a wide spread over a territory with different vernaculars gave rise to distinct dialects. Nevertheless, unifying tendencies — which are the first steps towards a standard — can be observed at any time in the development of Yiddish. Where do they come from? As noted above, I will focus on the impact of language contact in standardizing and language planning, because I am convinced that on the one hand long-term language contact was a potential source for the emergence of distinct dialects. On the other hand, even when the momentum of dialectal differentiation is at its highest, unifying tendencies can always be observed (see Kleine 2002 for a fuller discussion). First we must recall that Jews were frequently forced to migrate and were limited to special occupations, such as that of itinerant traders. Consequently Yiddish speakers of different dialects often met and had to find a way to communicate, which meant some sort of dialect levelling. Furthermore the necessity of written communication over long distances in a diasporic situation was a major factor leading to the development of a cross-dialectal, written Yiddish. Already in early times, after the Eastern Yiddish variety had begun to separate form Western Yiddish, an early form of a cross-dialectal iiterary' Yiddish can be observed. Max Weinreich was the first to describe this phenomenon and called this first supraregional language Shraybshprakh Alef ("Writing-Language A"). He lists various characteristics of this early form of 'literary' Yiddish (1973:II:397; 1980:733 and passim). Here printers played a dominant role, for the book market had to attend to customers in the Western Yiddish speaking territories too. Thus dialect markers were eliminated, and typical Eastern Yiddish words were replaced by such words as could be understood everywhere (often more archaic variants of Western Yiddish origin), as can be seen in Western reprints of books written in the East. Slavic words were obvious candidates for elimination or substitution by neutral synonyms. According to Dov-Ber Kerler's recent monograph on the origins of modern literary Yiddish (Kerler 1999),2 many authors were careful to avoid too local expressions in the first place. Simon Neuberg (1999:89-115) showed that the author of Zenerene, the most popular Yiddish paraphrase of the Bible, was very much aware of diatopic differences and tried to avoid Eastern Yiddish terms which might not have been common among Western Yiddish speaking Jews.3 But it was not just printers, forced to use a supraregional language 2
Kerler (1999) compares different literary texts and different Eastern editions in chronological order. For the disappearance of Slavic in reprintings, see also Shmeruk (1988:129). 3 For example, the author, Jacob b. Isaak Aschkenasi, avoided the more common expressions for "praying", oren, which is restricted to Western Yiddish, and the Eastern Yiddish davenen.
222
ANE KLEINE
intelligible for readers of the whole Yiddish speaking territory, who tried to suppress occurrences of local terms. Some awareness of pronunciation differences between distant communities can also be recognized. Dovid Katz (1993:52ff.) lists hypercorrect spellings in printings, but also in private letters, which clearly show a compensatory attempt and the idea of 'correct' and 'incorrect' pronunciation. Only the decline of Western Yiddish made way for the rise of an independent Eastern Yiddish variety. I will not go into further detail about historical developments, because my main object is the modern standard language. But I would like to emphasize that already here we find both interlanguage contact and intralanguage contact involved. Eastern Yiddish on Slavic speaking territory develops differently from Western Yiddish on German speaking territory thanks to interlanguage contact — the consequences are differentiation and segregation. And we find intralanguage contact, making the people very much aware of emerging differences in their local varieties and therefore prompting them to (try to) find a way out of that situation — that is, dialect levelling. This is something that can be observed repeatedly in the course of Yiddish history. I will now focus on more recent phenomena in unifying and standardizing Eastern Yiddish towards a 'Modern Standard Yiddish', viz. during the last 150 years. On the one hand these years are characterized by an increase of contact situations between speakers of different Eastern European languages — mostly Slavic, but also others such as Hungarian or Estonian — and Yiddish speakers. This also meant an increase in the number of multilingual Jews (as members of the Yiddish speaking community) in Eastern Europe regularly speaking Russian, Rumanian, Lithuanian etc. on the streets or at work. As would be expected, an acceleration in forming local varieties — willingly adding localisms — occurred. Before the destruction of the Jewish communities in Eastern Europe by the German Nazis, interlanguage contact had reached a peak, with interference from numerous Slavic languages. In Jewish centres with a dense Jewish population, like Warsaw (31%), Bialystok (76%) or Berdichev (80%) and many others, speakers could indulge in speaking exclusively their local variants. Enough people spoke exactly this specific dialect and were able to adhere to it. It was, therefore, not essential for these speech communities to level regional differences. On the other hand, the famous Yiddish writers Mendele Mokher Seforim, Shalom Aleichem, and I.L. Peretz created a widely acknowledged modern written Yiddish where there was a give and take in localisms. Just like the He used instead the periphrastic expression tfil ton, common to Western and Eastern Yiddish readers.
YIDDISH: NO STATE, NO STATUS — NO STANDARD?
223
writers and printers 200 years earlier, they had in mind scattered readers from different dialectal backgrounds — and wanted to supply a large market, encompassing great parts of Eastern Europe, with a supraregionally comprehensible Yiddish. In contradistinction to the earlier Western-based literary language, this 'linguistic' compromise is called Shraybshprakh Beys ("Writing-Language B"; M. Weinreich 1973; 1980 as above). This early modern Standard Yiddish, which emerged first in literature, gave its name to the corresponding pronunciation, known as the 'literary pronunciation'. The Eastern Yiddish dialects can be distinguished by their differing pronunciation of the vowels. In the South-Eastern Yiddish dialects some standard [ay] phonemes merge into [a], some [o] merge with [u] etc., while the traditional spelling is coherent with standard pronunciation. Only in the Northern Yiddish variety does a conservative pronunciation obtain, corresponding to the traditional spellings. On the other hand, the Northern dialect often merges the standard diphthongs [ey] and [oy], which both become [ey]. The spelling system clearly discriminates these sounds, as does the corresponding standard pronunciation. With this we find ourselves right in the middle of intralanguage contact. And we may add some social ingredients to the standardizing efforts too. The Jewish intellectuals of Lithuania, the Litvakes, were the most vocal advocates of the 'literary pronunciation', since its more or less conservative representation of the vowel system closely resembled the Lithuanian Yiddish dialect (Mark 1951:7f.). Despite the demographic dominance of Southern Yiddish speakers, the higher prestige of Northern Jewish communities in Eastern Europe — as far as secular learning is concerned — made this northern oriented pronunciation more likely to win out over all others. Furthermore, its use in such prestigious functions as the language of school instruction and public discussion contributed to the acceptance of this supraregional variety (Katz 1993:48), since training of teachers was carried out in Lithuania, where the initiative of a secular school system had been born.4 This remained the case even though another functional and supraregional variety entered into competition: the so-called bine-oysshprakh ("theatre pronunciation") (U. Weinreich 1951:28f.), a southern oriented way of speaking which had been accepted on the stage and in Yiddish film production. Like the 'language of the Litvakes\ which was associated with 'progress', 'modern intelligence' and secular learning, it functioned as a model, but its dissemination was restricted to the stage and to artistic performers. 4
The impact of the Yiddish secular school in Eastern Europe on Yiddish as a high language is discussed in several articles in Fishman (1981).
224
ANE KLEINE
Another factor which contributed to a northern oriented Standard Yiddish was the founding of YIVO, the Institute for Jewish Research, in 1925, located in Vilna, Lithuania. This institute in Vilna then became the home of many standardizing efforts; in particular, modern Yiddish spelling was defined there (takones fun yidishn oysleyg, 1937 [= "Rules for Yiddish spelling"]).5 As a result of growing linguistic interest in Yiddish, modern scholars at the beginning of the 20th century started to recognize Yiddish as a language in its own right (Weinreich, Max [1923] 1993:Chapter IV and passim). Language families were investigated using diachronic and synchronic methods (Kleine 1999:60f.). As a reaction to the undeniably close relationship between Yiddish and German, which came into focus at that time, a need for clear distinction and differentiation arose. Dialectal differences within Yiddish were pushed into the background while focussing on supraregional common features. In this situation, phenomena such as 'language loyalty' — as Uriel Weinreich (1970:99) calls it — spread, encouraging speakers to concentrate on the standardization of the language. These facts are not new, but they have not previously been seen in relation to language contact resulting in a Standard Yiddish pronunciation. Such beginnings of a modern standard pronunciation in the Eastern European Yiddish speaking territory show a peculiarity (for Europe): that this language has achieved a standard without any national or governmental anchorage. The socio-cultural setting of language contact and its linguistic effects were radically different in the new exile communities after World War II. Migration overseas both at the turn of the century and in response to the Nazi crimes in Europe all but extinguished the language in its historical territory, and the new Yiddish speaking centres around the world were marked by dialect multiplicity. In these new communities, speakers of different dialectal backgrounds suddenly lived next to one another. Earlier I noted centres with dense Yiddish speaking populations, which could — as I put it — "indulge in speaking exclusively local variants". But in the overseas communities no cohesive forces strengthened any local variety; no cohesive hinterland supported any one dialect any more. After the war the impact of intralanguage contact thus reached a new peak. New exile varieties, bearing a mixture of the original Yiddish dialects, were rapidly evolving and gaining in importance. In this situation the Yiddish speaking community rallied to draw attention to its medium of communication, and, in due course, its speakers aimed at a standardized variety. Moreover, interlanguage contact 5
However, Russian Yiddish adhered to a completely different spelling system, called sovyetisher oysleyg ("Soviet spelling") and undermined efforts to establish a writing system of universal validity.
YIDDISH: NO STATE, NO STATUS — NO STANDARD?
225
was at its height: the need to become familiar with English, Spanish, or other languages of the new home-countries provided potential sources for language innovation. This process very much resembles the impact of the Slavic languages at an earlier period, but the renewal process was much more rapid than it had been a few centuries earlier. The effects can easily be perceived in the speech behaviour of many Yiddish speakers today with dialect mixing (due to intralanguage contact) and the gaining of new components (as a result of interlanguage contact). Somewhat mockingly, the new variants are known as: 'Castidish' in Argentina or 'Potato Yiddish' in the USA, and so on. To illustrate this, I will give two examples. In Buenos Aires I was frequently told by Argentine Yiddish speakers that it is almost impossible to understand the US-American Yiddish speakers, who find it quite normal to say: Debby, di neksdorike fun daunsters un ir mister makhn biznes ("Deborah, the next door neighbour from downstairs, and her husband are doing good business"). My Argentinian Yiddish speakers meanwhile found it fairly normal to describe the way to Callao Road as follows: Du forst mit der kotshe oyf kashe korientes tsvey kvadres, doblirst bay der eskine, vestu tsukumen oyf kashe kashao ("With your coche [= car] you follow Calle Corientes [= Corrientes Road] for two quadras [= blocks], then you doblar [= turn around] at the esquine [= corner], and you will find yourself on Calle Callao [= Callao Road]"). Furthermore, language shift is a phenomenon often observed in (partly) Yiddish speaking families. (This language shift had already begun in the former Eastern European Jewish communities, where the trend to secularisation was accompanied by a language shift, although this did not lead to the total loss of the Yiddish mother tongue in a family or a whole generation. However, it is reported that vernacular Russian or Polish was used even in several groups consisting only of Yiddish speakers.) The new generation overseas now learns Yiddish as a second or even as a foreign language, leading to a breakdown into subgroups using various languages as their predominant idiom. In response to impending language shift, priority is now being given to preserving the threatened language, which is obviously vanishing from everyday life. The acquisition of Yiddish in academic programs is furthering the standardized version of that language — the language of teaching. After the YIVO's migration to New York in 1941 it was again very involved in promoting the development and spread of this new standard, especially Standard Yiddish pronunciation. The best known textbook, College Yiddish (U. Weinreich 1949), was printed in 1949 under the aegis of YIVO; it has already been reprinted in more than twenty editions and seems to be the basis of language instruction worldwide. It provided the first (modern) teaching material for Yiddish as a foreign language and therefore devoted a full chapter
226
ANE KLEINE
to the proper pronunciation of a form of Yiddish which can undoubtedly be traced back to earlier discussions on Yiddish language planning. To this day a great many descriptions of modern Yiddish pronunciation in various other textbooks are based on College Yiddish. Nowadays we still do find Yiddish as a language of communication, and within some Hassidic circles it is to this day the 'everyday language'. And it is true that this speech community is far from speaking what is called 'Standard Yiddish'. But this is no reason to reject my thesis that language contacts further standardization, as the modern standard is increasingly promulgated through second/ foreign language learning. On the contrary, it bears witness to it. Owing to a lack of intralanguage contact, the Hassidim of Mea Shearim (the orthodox neighbourhood in Jerusalem) or of Crown Heights or Borough Park in Brooklyn have little interest in dialect levelling and hardly any interest in secular learning, scientific interchange or academic programs. Therefore this group cannot be affected by corpus planning or by the dynamic standardizing effects following language contact. To sum up: Yiddish, the language of Ashkenazic Jewry, has no state nor any governmental forces to strengthen a standard, but nowadays the term 'Standard Yiddish' is widely accepted (Kleine 1998:202f.). However, I should like to conclude the discussion by pointing out that despite the acceptance of the notion of Standard Yiddish, there are in the 21 st century already several standard varieties of Modern Yiddish pronunciation. The largest exile communities — that is: the USA, Argentina, France, Canada, South Africa, Australia and Israel (plus Russia, which can — to some degree — be considered an exile community in this respect) — appear to have attained their own standards. Generally they follow the (phonological) 'rules' for the Standard Yiddish pronunciation, but the system of phonemes from the surrounding languages is somehow manipulating the speakers' attitude towards them. Let me illustrate this with a few examples. In Standard Yiddish both the uvular [R] and the alveolar [r] are accepted. A speaker may either articulate this trill in the alveolar region when speaking Standard Yiddish or may use the alternative uvular allophone. In the new home countries usage varies according to the surrounding coterritorial languages. In Argentina, among the generation of the former immigrants there are still a few speakers who prefer the uvular [R]. But the younger ones, who learned Yiddish parallel to Spanish, and who have attended Yiddish lessons at school, in their broad majority prefer the alveolar trill, the duration of which follows the same distribution pattern as
YIDDISH: NO STATE, NO STATUS — NO STANDARD?
227
in Argentinian Spanish.6 On the other hand US-American Yiddish speakers generally seem to prefer the uvular [R]. The fact that the American English Ixl is rather the vocalized approximant [J] might otherwise easily lead to a substitution of the Yiddish trill (or at least tap/ flap with one contact of the tongue at the alveolar ridge) by its English counterpart. It seems as if a preference for the uvular [R], non-existent in the English phoneme system, arose in order to avoid a merge of different phoneme systems. The fricative [h] may serve as another good example to compare Argentinian Yiddish with another 'standard variety', Russian Standard Yiddish. While in proximity with modern Argentinian Spanish the articulation of the glottal fricative is remarkably weak, in Russian Yiddish there seems to be a tendency to sharpen this sound by narrowing the vocal tract and articulating a sound between [x] and [h]. Even though these characteristics may indicate the speakers' provenance, all the phenomena discussed now may occur in 'Standard Yiddish' and cannot be viewed as dialectal patterns of any single Eastern Yiddish dialect. For methodological purposes it will be necessary to distinguish modern varieties of Standard Yiddish as 'Argentinian Standard Yiddish', 'USAmerican Standard Yiddish', and so on, where each seems to have developed its own tradition. In this linguistic regard the various Standard Yiddish pronunciations may be likened to the different standard varieties of German in Germany, Switzerland, Austria and Liechtenstein. My own study of one Standard Yiddish pronunciation, Argentinian Standard Yiddish, will provide the first instrumentally-aided phonetic analysis of Yiddish, and is near completion. It is to be hoped that it can be contrasted with other varieties in the near future.
References Fishman, Joshua A., ed. 1981. Never Say Die! A Thousand Years of Jewish Life and Letters. (= Contributions to the Sociology of Language, 30.) The Hague: Mouton. Katz, Dovid. 1993. tikney takones. fragn fun yidisher stilistik. [= Amended Amendments. Issues in Yiddish Stylistics] Oxford: oksforder yidish.
6
That is: short duration in articulation, performed as a tap or flap at the end of a word or in the middle of a syllable or following a plosive, but long (trill articulation) at the beginning of a word or a syllable if the previous syllable ends with a consonant. (Argentinian) Spanish also distinguishes two r-phonemes: short vs. long (or double) Ixl (pero /perro). The long [r] is not realized in Yiddish.
228
ANE KLEINE
Kerler, Dov-Ber. 1999. The Origins of Modern Literary Yiddish. Oxford: Clarendon Press. Kleine, Ane. 1998. Toward a 'Standard Yiddish Pronunciation'. An Instrumentally Aided Phonetic Analysis. Historical Linguistics 1997. Selected papers from the 13tl International Conference on Historical Linguistics (ICHL), Düsseldorf 10-17 August 1997 ed. by Monika S. Schmid, J.R. Austin & D. Stein, 201-211. Amsterdam & Philadelphia: John Benjamins. Kleine, Ane. 1999. Florilegium zur jiddischen Phonetik. Eine Zeitreise. Jiddische Philologie. Festschrift für Erika Timm ed. by Walter Roll & S. Neuberg, 51-63. Tübingen: Niemeyer. Kleine, Ane. 2002. Sprachplanung in der Diaspora. Soziolinguistische Aspekte bei der Normierung der standardjiddischen Aussprache. Sprachwissenschaft auf dem Weg in das dritte Jahrtausend. Akten des 34. Linguistischen Kolloquiums in Germersheim 1999 / Linguistics on the Way into the Third Millennium. Proceedings of the 34th Linguistics Colloquium, Germersheim 1999 ed. by Reinhard Rapp, Part II, 183-192. Frankfurt: Peter Lang. Mark, Yudl. 1951. vegn a klalishn aroysreyd. [= On the Standard Pronunciation of Yiddish.] yidishe shprakh 11. 1-25. Neuberg, Simon. 1999. Pragmatische Aspekte der jiddischen Sprachgeschichte am Beispiel der >Zenerene
