Ethnie Mobilization and Minority Resources*

Home , Frisians

Forschungsnotiz Ethnie Mobilization and Minority Resources* Erik Allardt Universität Helsinki, Institut für vergleichende Soziologie Mariegatan 10 A 13, SF-00170 Helsinki 17, Finnland Comments to the ethnic mobilization and the Difficulties of Definition and Delineation renewal of ethnic agitation in Western Europe in the 1970’s are abundant. Yet, there is hardly This paper focuses on territorial, linguistic mi any agreement about the background and social norities in Western Europe. The ultimate aim sources of this ethnic revival. The centre-peri is to throw some light on the conditions under phery paradigm has been a natural point of de which ethnic minorities are likely to become parture in many analyses already because ethnic mobilized, activated or even militant as well as minorities as well as ethnic-regional separatist on the conditions under which minorities are movements almost by definition are located in likely to remain immobilized or even unaware the peripheries. Yet there are no satisfactory of their own predicament. generalizations concerning the interplay of eco nomic, political, cultural and geographic factors Needless to say there is an abundance of dif in the rise of ethnic activism. In particular there ferent types of minorities. It is by no means has been disagreement on the effects of the class easy to delineate the category of ethnic mino structure and on whether the economic and social rities in Western Europe. Despite the decision predicament of the ethnic minorities has improv to study territorial minorities only, i.e. minor ed or worsened. General descriptions of the poli ities which have long inhabitated the same tical actions by ethnic, territorial minorities (An geographical areas, there exists great variation derson 1978) and analyses with a heavy focus on in both factors such as size and historical im militant separatist movements (Williams 1980) portance and also in the collective ability of tend to render fairly different images. the minority to speak its own language. There are strongly militant minorities where only a This paper will not produce any comprehensive small portion of the population speaks the answers to problems concerning the relationship minority’s own language. The most remarkable between social and ethnic stratification. Nor will instances are perhaps those in which a previously it with certainty confirm any hypothesis con extinct language now is the focus of revival as cerning the effects on nationalist resurgence by is the case with some of the languages of the centre-periphery relations. The aim of the paper Celtic fringe of Great Britain, and to some ex is to present systematic statistical data concer tent with Occitania in southern France. ning the territorial, linguistic minorities in Wes tern Europe, and to submit the data to a prelim Almost all languages have dialectical varieties, inary multivariate analysis. It goes without say and it is sometimes difficult to distinguish bet ing that no thorough answers and results can ween speakers of a dialect and linguistic groups be found by such a preliminary and superficial claiming special minority status. It is usually analysis. Statistical data and analyses of the difficult to decide whether two varieties of the kind used here have to be combined with his same basic language should be considered dia torical studies and in-depth analyses of unique lects or distinct languages on the basis of purely cases. Yet it seems worthwhile to emphasize the linguistic criteria (Kloss 1978). The categoriza importance of the kind of statistics used here tion of a vernacular as a distinct and independent despite its great shortcomings. There is presently language is always based on historical circum almost an overproduction of interpretations stance and interpretation. Having a distinct lan based on selected cases. Systematical variation guage is not a prerequisite for recognition as a amongst ethnic minorities has only been ac linguistic minority if there is clear continuity in counted for to a minor extent. the history of the group. There is, however, a definite risk of the concept of ethnic or linguistic ♦ Paper presented at the colloquium on “Understand minorities becoming too broad, thus permitting ing Political Society” , Werner-Reimers-Stiftung, Bad the inclusion of very different groups under the Homburg, 1 8 -2 2 May 1981. concept of ethnic minorities. 428 Zeitschrift für Soziologie, Jg. 10, Heft 4, Oktober 1981, S. 427 - 437

The minority concept generally refers to the re The Minorities Compared lative size of the group, but it also holds many It can hardly be overstressed that it has proved problems for the study of ethnic minorities. It very difficult to gather data about the social is fairly common for a group which nationally composition of different ethnic groups. Questions is an ethnic minority to constitute a majority about language, on either the main language, within a certain region. The majority of the po mother tongue, most widely spoken language or pulation of that country thus constitutes a mi knowledge of languages are rather rare in Euro nority, within a given region. This, for instance, pean cencuses. In most cases, data refer to re is true of the Finnish speakers on the Aaland gions rather than to minorities proper. The likeli isles in the Baltic Sea. As a general rule, the hood of erroneous interpretation is naturally majority of the population of a country has not particularly great in cases where the ethnic group been considered as a minority. There are, how is only a small minority within its own region. ever, exceptions due to special historical circum Occasionally, the difficulty lies in the localization stances. The Flemish speakers of Belgium are at of a minority to or within a specific region. In present the largest linguistic group in Belgium. such it has sometimes been necessary to rely on Nevertheles, historically speaking, the group has approximations of the values of certain variables. so many features in common with other lin It is so to say in the nature of ethnicity that guistic minorities that it is treated here as a na operationalizations of membership in ethnic tional group comparable to a linguistic minority groups are very difficult. Different criteria for as it is in many other lists of European languages. membership are emphasized in different situa tions. There are sometimes political disagreements The problem of deciding which groups and na over which criteria for membership ought to be tionalities actually are linguistic minorities is by used, but the difficulty of defining membership no means the only difficulty encountered in a even persists in cases where there is political con statistical comparison of ethnic minorities. The information on minorities is sometimes based on sensus on the proper criteria. data about individuals, sometimes on data about The linguistic minorities analyzed are listed in the regions in which the minorities live. There table 1. They can all be localized to certain sub are additional problems regarding both individual national territories. This is why for example the and regional data. There are also various criteria Norwegian “Landsmal” is omitted from the ana by which individuals are regarded as members of lysis. They are also minorities within a certain minorities. The implications of regional data dif country. Thus, the Lapps in Finland, Norway, fer greatly depending on whether or not the mi and Sweden are counted as three separate mi nority being studied forms a majority in its own norities. Table 1 also lists the regions in which region. the minorities live. Unfortunately, these regions are not always defined and delineated in the Despite these many difficulties, we shall attempt same manner. This is mainly due to two circum to make a statistical comparison based on multi stances. Firstly, different sources define regions variate analysis. Even if the findings generally in different ways. Some data are taken from the point in the right direction, one cannot over regional statistics of the European Community, emphasize the fact that the decisions which have that is from the so-called EUROSTAT-publica- to be made on the definition of linguistic mi tions (Eurostat 1976) where regional divisions are norities, on which variables to include and on in some cases different from those used in na the categorization of those variables can lead to tional statistical publications. Secondly, in some grossly misleading results in single instances. It cases where it has not been possible to obtain is almost as if any decision must be wrong for data about the minority region proper, larger re one particular minority at least. Therefore, it is gions of which the minority region forms a part important to use such multivariate techniques have been taken into account. which enable each individual case to be identi fied. The use of factor scores based on factor The minorities listed are also given in handbooks analysis, for instance, is such a method. such as those by Grulich and Pulte (1975), E. Allardt: Ethnic Mobilization and Minority Resources 429

TABLE 1 Linguistic Minorities in Western Europe

Minority Region Nation-state

1. Slovenes Carinthia (Kärnten) Austria 2. Magyars Burgenland Austria 3. Croats Burgenland Austria 4. Flemings Flanders Belgium 5. Walloons Wallonia Belgium 6. Germans Liege Belgium 7. Gaels Highland region, Scotland Great Britain 8. Gaels Isle of Man Great Britain 9. Gaels Northern Ireland Great Britain 10. Welsh Wales Great Britain 11. Channel islanders Jersey, Guernsey, Alderny, Sark Great Britain 12. Germans North Schleswig Denmark 13. Faroe islanders The Faroe islands Denmark 14. Greenlanders Greenland Denmark 15. Swedish Finns The provinces of Uusimaa, Turku- Pori, Vaasa Finland 16. Swedish Finns Aaland isles Finland 17. Lapps Inari, Sodankyla, Utsjoki Finland 18. Occitans Centre-Est, Mediterranee, Sud-Quest France 19. Catalans of Rousillon Languedoc-Rousillon France 20. Basques Aquitaine France 21. Corsicans Corsica France 22. Alsatians Alsace France 23. Flemings of Westhoek Nord-Pas de Calais France 24. Bretons Bretagne France 25. Danes of South Schleswig Schleswig Holstein Germany (BRD) 26. North Frisians Schleswig Holstein Germany (BRD) 27. Gaels Ireland Ireland 28. Occitans Piedmonte Italy 29. Friulians Friuli-Venezia Giulia Italy 30. Ladins Trentino-Alto Adige Italy 31. Valdotains Valle d’Aosta Italy 32. South Tyroleans Trentino Alto Adige Italy 33. Slovenes Friuli-Venezia Giulia Italy 34. Sards Sardinia Italy 35. Greeks Calabria Italy 36. Albanians Calabria Italy 37. West Frisians Friesland The Netherlands 38. Lapps Troms, Finmark, Nordland Norway 39. Catalans Georna, Balearic islands, Barcelona, Spain Lerida, Tarragossa 40. Basques Alva, Guipüzcoa, Spain Navarra, Vizcaya 41. Galicians Lugo, Orense, Pontevedra, Coruna Spain 42. Lapps Norrbotten Sweden 43. Finns Tome valley, Norrbotten Sweden 44. Jurassians Ticino Switzerland 45. Jurassians Jura, canton of Bern Switzerland 46. Rhaetians Grau biinden Switzerland

Stephens (1976), Straka (1970) and Haarmann Great Britain, the Piedmontese from the Italian (1975). Groups which sometimes are mentioned Piemonte, and the Letze burgers of Luxembourg. among European ethnic minorities but which Ethnic minorities are constantly defined and re have not been included in table 1 are for in defined, and it is virtually impossible to reach stance the Cornish and the Lowland Scots from consensus as regards all of them. 430 Zeitschrift für Soziologie, Jg. 10, Heft 4, Oktober 1981, S. 427 - 437

The Variables Used in the Analysis a value of 2 if emigration and immigration were about equal strength, and a value of 3 if immi Not only it is difficult to reach consensus about gration was higher than emigration. which groups should be counted as ethnic mi (3) Unemployment norities but also the selection of the variables for The unemployment in per cent of the labor force. The comparabüity of the data from the EEC- comparison is a fairly hazardous business. The countries can be regarded as very good, whereas selection of variables was not guided by an ex there is uncertainty as regards the comparability plicit theory or a set of clearly formulated hypo of the data from other countries. theses. Yet, a general frame of reference con (4) The Proportion o f Persons Working in Primary cerning the relationship between ethnic and so Occupations cial stratification did guide the selection. The Persons working in the primary sector in per cent aim of the analysis was to explain ethnic mobi of all gainfully employed. lization and militancy on linguistic grounds. The (5) GNP per capita (in relation to the GNP of the dependent variables thus had to deal with poli whole country) The aim has been to create a variable measuring tical activity within the linguistic minorities: the relative economic well-being of the region. the formation of ethnically based political par Since it vyas impossible to obtain comparable data ties, voting for radical leftist parties, the incid from all regions, the GNP per capita - or in ence of ethnically motivated violence, etc. The some cases the mean income - was related to the GNP per capita for the whole country. The independent or explanatory variables were as minorities were coded in the following way: sumed to describe different kinds of resources. 1. GNP per capita is lower in the minority region In particular, it seemed important to account for than in the whole country. 2. GNP is about the resources related to the economy, the politics, same in the region and the whole country. and the position o f the minority language. 3. GNP is higher in the minority region than in the whole country. It is almost too easy to mention and conceive (6) The Development Level o f the Language factors which might have some kind of relation The variable is coded as follows: ship to ethnic militancy and activity. Many con 1. The minority does not have a developed writ ceivable variables had to be omitted because ten language, highly diverging dialects exist. 2. The written language is fully developed. of a lack of data. The final list of variables was based on both the objectives of the analysis (7) Linguistic Ability The variable denotes the extent to which members and the availability of data. The values of the of a minority know their own language. variables for each minority are presented in 1. Only a small portion is able to speak the lan table 2. It should be noted that most mea guage. sures were extremely crude and often used in a 2 The language is spoken and understood by dichotomous or trichotomous form. The data most of the minority, but not by all. are from 1975 or later, unless otherwise indi 3. The language is known both in oral and writ ten form by practically all members of the cated. minority. (8) The Uniqueness o f the Language The following variables were obtained:1 1. The language is spoken in one country only. 2. The language is spoken in several countries (1) Population density but is not the main language in any country. The number of inhabitants per square kilometer. 3. The language is the main language in another (2) Immigration country (countries). The variable denotes the mean values for 19 7 0 - (9) Autonom y 1975. It has the value o f 1 if emigration from The variable denotes whether the minority has the minority region was larger than immigration, some form of home rule. A minority is regarded as possessing autonomy even if there are several minorities within an autonomous region and the minority in question is not the largest 1 depicts 1 The data for the first five variables as regards the lack o f autonomy, whereas value 2 indicates that EEC-countries are obtained from the Eurostat-publi- the minority has at least some form of autonomy. cation. For other countries they have been compüed from handbooks and national statistical publications. (10) School Language A full account of all sources and operations as re 1. The minority language is not taught in schools. gards the statistical variables can be obtained from There might exist courses in which the lan the Research Group for Comparative Sociology, Uni guage is taught as a “foreign language” in rare versity of Helsinki. cases. E. Allardt: Ethnic Mobilization and Minority Resources 431

2. The language of the minority is taught as a (16) Voting for Radical Socialist Parties compulsory foreign language. The minority Social Democratic parties have not been re language is not the language of instruction. garded as parties of the radical left. Otherwise 3. The minority has, at least in some regions, this variable is identical with variable 15. schools in which its own language is the lan (17) The existence o f an Ethnic Party guage of instruction. An Ethnic party is any movement which tries (11) Status as an Official Language to safeguard and improve the rights of the mi The variable denotes the rights and opportunity nority and which has often had candidates in of minority to use its language in dealings with national elections. national and local government. 1. No ethnic party exists. 1. The minority language lacks official status. 2. There is an ethnic party. 2. The minority has a legal right to use its own (18) Political violence language within certain restricted realms. 1. No terrorist acts or similar violence have oc 3. The minority can use its own language in curred during the 1960’s and 1970’s. practically all official contexts. 2. Terrorist acts and similar violence actions (12) Existence o f Mass Media have occurred during the 1960’s and 1970’s. The variable describes whether the minority has its own newspaper(s), radio programmes in its own language, or television programmes in its The assigned list of variables and values is apt own language. to highlight many of the difficulties of creating 0. The minority has no mass media in its own data files for ethnic minorities. The above men language. tioned list could of course be improved on with 1. The minority has access to one of the three painstaking and laborious statistical groundwork. above mentioned forms of mass media. 2. The minority has access to two of the three The crude data about the minorities studied are above forms of mass media. presented in table 2. 3. The minority has access to all above kinds of mass media. It has often been difficult to decide whether a Correlation and Factor Analyses: Resources and minority should be considered having its own newspaper, or its own radio programmes. They Ethnic Mobilization come in many different formats. This is one reason why all three types of mass communica The correlations between the variables are given tion have been combined as one variable. Parti in table 3. The variables are presented in the cular emphasis has been laid on whether commu matrix in the same order they were listed above. nication occurs in the minority language. This means that the independent variables are (13) The Size o f the Minority presented first, and that the dependent variables Since it often is very difficult to exactly deter are presented at the bottom of the matrix. The mine the population size of a minority, the fol lowing crude division has been used: division into independent and dependent variables 1. - 9.999 is not watertight or clearcut, but it serves as a 2. 10.000 - 99.999 point of departure. 3. 100.000 -999.999 4. 1.000.000 and more No correlation in table 3 is very high but there are The size of the minority is in the first hand nevertheless some significant positive correlations estimated on the basis of how many know the indicating consistent patterns at least to some ex minority language to some extent at least. tent. It is notable that ethnic activity and mobili (14) The Growth o f the Minority Population zation, as measured by the existence of ethnic 1. There is a decrease in the number of persons parties and by outbursts of violence, are positive having the minority language as their mother ly correlated with the availability of resources of tongue (or main language). different kinds. Generally speaking, the active 2. The number of persons having the minority language as their mother tongue is constant and violent minorities are not those devoid of or on the increase. resources, rather, those are minorities which de finitely do have some means of expressing them (15) Voting for Socialist Parties selves. This result is not unexpected; it is very Support for Socialist parties in per cent of all eligible votes. Data are from the last national logical to assume that ethnic mobilization is elections for which information is available. facilitated by such factors as access to mass 432 Zeitschrift für Soziologie, Jg. 10, Heft 4, Oktober 1981, S. 427 - 437

TABLE 2 The Distribution of Resources and Patterns of Ethnie Mobilization among Territorial Linguistic Minorities in Western Europe

(1) (2) (3) (4) (5) (6) (7) (8) Population Immigration Unemploy Primary GNP/capita Level of Linguistic Uniqueness density ment Sector language ability of language

1. Slovenes (Austria) 133 1 4 12 1 2 3 3 2. Magyars (Austria) 67 1 1 18 1 2 3 3 3. Croats (Austria) 67 1 1 18 1 2 3 3 4. Flemings 410 3 3 5 2 2 3 3 5. Walloons 191 3 4 5 1 2 3 3 6. Germans (Belgium) 263 3 5 5 2 2 3 3 7. Gaels (Scottish Highlands 7 1 6 - 2 1 2 8. Gaels (Isle of Man) 100 3 - - - 1 1 2 9. Gaels (Northern Ireland) 109 1 7 8 1 2 1 2 10. Welsh 133 3 5 4 1 2 1 1 11. Channel islanders 665 3 1 1 3 12. Germans (Denmark) 76 3 9 18 - 2 3 3 13. Faroe islanders 28 1 - 24 - 2 3 1 14. Greenlanders 0 2 - 22 1 2 3 1 15. Swedish Finns 60 2 3 12 3 2 3 3 16. Aalanders 15 2 1 17 2 2 3 3 17. Lapps (Finland) 2 1 9 28 1 1 2 2 18. Occitans (France) 75 3 4 10 1 2 2 2 19. Catalans (France) 65 3 4 20 1 2 2 2 20. Basques(France) 61 3 4 14 1 2 1 2 21. Corsicans 25 3 6 5 1 1 2 1 22. Alsatians 184 3 2 6 2 2 3 3 23. Flemings (France) 312 1 3 7 1 2 3 3 24. Bretons 96 3 4 21 1 2 1 1 25. Danes (Germany) 165 3 4 6 1 2 2 3 26. North Frisians (Germany) 165 3 4 6 1 1 1 2 27. Gaels (Ireland) 45 1 10 25 - 2 1 2 28. Occitans (Italy) 179 3 2 11 3 2 2 2 29. Friulians 159 3 2 7 3 1 2 2 30. Ladins 64 2 2 21 3 1 2 2 31. Valdotams 35 3 2 15 3 2 3 3 32. South Tyroleans 64 2 2 30 3 2 3 3 33. Slovenes (Italy) 159 3 2 7 3 2 3 3 34. Sards 64 1 6 17 1 1 1 1 35. Greeks (Italy) 135 1 8 24 1 2 2 3 36. Albanians (Italy) 135 1 8 24 1 2 2 3 37. West Frisians (Netherland) 144 3 2 11 1 2 1 2 38. Lapps (Norway) 4 1 3 16 1 1 2 2 39. Catalans (Spain) 160 3 - 3 3 2 2 2 40. Basques (Spain) 133 3 - 9 3 2 1 2 41. Galicians 88 1 - 47 1 2 2 1 42. Lapps (Sweden) 3 2 - 14 1 1 2 2 43. Tome valley Finns 2 2 4 41 1 2 2 3 44. Ticinese 87 3 0 5 - 2 2 3 45. Jurassians 143 2 0 11 - 2 3 3 46. Rhaetians 23 3 0 14 - 2 2 2 media, level of language development, size of and access to mass media (.25). Political violence the group etc. The background variables having within and around linguistic minorities is most the highest correlation with the existence of an highly correlated with the size of the minority ethnic party are the size of the minority (.61), (.42), access to mass media (.35) and a high pro level of language development (.35), the use of portion of people not employed in the primary the minority language as school language (.25), sector (—.32). Existence of an ethnic party and E. Allardt: Ethnic Mobilization and Minority Resources 433

Continuation TABLE 2

(9) (10) (11) (12) (13) (14) (15) (16) (17) (18) Autonomy School lan Officialness Mass media Size of Growth of Socialist vote Radical Ethnic party Political guage of language minority minority socialist violence vote

1 3 2 2 2 1 53 2 1 2 1 3 1 0 1 1 52 0 1 1 1 3 2 1 2 1 52 0 2 1 2 3 3 3 4 2 24 2 2 2 2 3 3 3 4 2 43 6 2 2 1 3 3 2 3 2 - 2 1 1 1 1 1 2 1 19 0 1 1 2 1 1 0 1 2 0 0 1 1 2 3 1 2 - 1 0 0 2 2 1 2 1 3 3 1 50 0 2 2 2 1 1 0 2 1 00 0 1 1 1 3 1 1 2 1 37 4 2 1 2 3 3 2 2 2 26 0 1 1 2 3 3 2 2 2 0 0 1 1 1 3 3 3 3 1 40 15 2 1 2 3 3 3 2 2 0 0 2 1 1 1 1 2 1 2 37 27 1 1 1 2 1 0 4 2 48 23 2 1 1 1 1 2 3 1 60 32 2 1 1 1 1 2 2 1 37 10 2 1 1 1 1 0 3 1 24 14 2 2 1 3 1 2 4 2 22 8 2 1 1 1 1 1 3 1 53 26 2 1 1 1 1 3 4 1 38 16 2 1 1 3 1 1 2 1 44 0 2 1 1 2 1 0 2 1 43 0 1 1 2 3 2 3 2 2 7 0 2 1 2 1 1 - 3 1 54 38 1 1 2 1 1 1 3 1 42 25 2 1 2 2 2 2 2 1 26 13 1 1 2 3 2 1 2 1 36 1 1 2 3 3 3 3 2 26 13 2 2 2 3 2 2 2 1 40 29 2 2 2 1 1 0 4 1 50 36 2 1 1 1 1 0 2 1 48 33 1 1 1 1 1 0 2 2 48 33 1 1 2 3 2 2 3 1 41 4 2 1 1 3 1 - 2 1 22 16 2 1 2 2 1 3 4 2 71 20 2 2 1 1 1 2 3 1 43 15 2 2 1 2 1 - 4 1 24 6 2 1 1 2 1 2 1 2 65 12 1 1 1 1 1 3 2 1 65 12 1 1 2 3 - 3 3 2 48 4 - 1 2 3 3 - 3 2 31 1 2 2 2 3 3 3 2 1 14 0 1 1 outbursts of political violence have a correlation network of social relations. The size, or, using of .26. Both the dependent variables are most the term employed in social anthropology, the highly correlated with the size of the minority. scale needed for a more or less full social life in It seems reasonable to assume that a certain size a group varies from one historical situation to is necessary in order to enable a minority to another (Barth 1978). Nevertheless, it would live a full social life and to have a developed seem warranted to assume that ethnic mobili- 434 Zeitschrift für Soziologie, Jg. 10, Heft 4, Oktober 1981, S. 427 - 437

TABLE 3 Coefficients of Correlation between Variables Describing Linguistic Minorities in Western Europe

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

1. Population density X 2 Immigration 30 X 3. Unemployment - 07 -.29 X 4. Primary sector -.36 -.52 .22 X 5. GNP per capita .12 .35 -.36 -.19 X 6. Level of language development -.05 .01 - .11 .03 .04 X 7. Linguistic ability -.04 .05 -.41 -.05 .23 .36 X 8 Uniqueness of language .34 .03 -.23 -.16 .22 .30 .52 X 9. Autonomy .16 .17 -.29 -.12 .45 - .05 04 -.02 X 10. School language -.07 -.01 -.31 -.19 .15 .37 .54 .38 .26 X 11. Officialness of language .02 .02 ->.36 -.03 .29 .32 .54 .23 .48 .62 X 12. Existence of mass media -.16 .15 -.21 .03 .28 .41 15 .01 .24 .38 .49 X 13. Size of the minority .27 .34 -.17 -.25 .16 .22 .05 -.16 17 -.02 .08 .19 X 14. Growth of the minority .04 .07 -.01 -.06 .18 .08 36 .11 .38 .19 .41 .21 27 X 15. Socialist (Leftist vote) -.08 .06 -.05 -.10 -.03 .14 .13 .14 -.40 - 18 -.32 .02 .08 -.24 X 16. Radical socialist vote -.03 -.04 .18 .07 .21 -.14 -.01 - .05 -.13 -.54 -.35 -.20 .17 - 07 49 X 17. Ethnic party .05 .23 -.02 -.22 .12 .35 .13 -.07 .05 .25 .18 .25 .61 .19 .04 -.08 18. Political Violence .16 .19 -.04 -.32 .17 .19 .01 -.05 .15 .13 .17 .35 .42 .06 .04 -.07

TABLE 4 Factors and Factor Loadings Based on an sis. An orthogonal rotation procedure was used, Orthogonal Solution in a Factor Analysis of Variables and three factors were extracted (cf. table 4). Describing Resources of Linguistic Minorities in Western Europe It is easy to give reasonable verbal interpretations to the factors. The first factor can be labeled I II III Linguistic Resources. It has high loadings for the variables (7) Linguistic ability, (10) School lan 1. Population density .04 .59 -.1 1 guage, (8) Uniqueness of language, and (11) Of 2. Immigration -.0 9 .65 .28 3. Unemployment -.3 6 -.3 0 -.3 4 ficialness of language. 4. Primary sector -.0 7 -.6 7 -.0 6 5. GNP per capita .18 .35 .44 The second factor can be regarded as denoting 6. Level of language develop Economic Resources. The highest loadings are in ment .47 -.0 9 .18 7. Linguistic ability .76 .02 .09 (4) the proportion of people in the Primary sec 8. Uniqueness of language .68 .27 -.2 2 tor with a negative loading —.67, (2) Immigra 9. Autonomy .06 .20 .59 tion, (1) Population density, and (5) GNP/capita. 10. School language .70 -.0 3 .31 11. Officialness o f language .58 -.0 7 .56 The third factor can be described as Political Re 12. Existence o f mass media .25 -.1 3 .62 13. Size of the minority -.0 8 .34 .33 sources with high loadings in (9) Autonomy, (12) the Existence of mass media, and (11) Of zation usually presupposes a size large enough ficialness of language. to make possible functioning networks within the minority. A meaningful, or at least an intuitively interpret able result, is obtained when the size of the mi The patterns observable in the correlation matrix nority has a fairly high loading on both the se stand out more clearly when new composite cond and third factor. Sufficient group size variables are constructed with the help of multi enables development of the well-functioning net variate analysis. As a first step thirteen back works necessary to both economic and political ground variables were submitted to factor analy activity. E. Allardt: Ethnic Mobilization and Minority Resources 435

It should, however, be remembered that the re assumption that the relationship between re sults of factor analysis are of course dependent sources and ethnic mobilization is either positive on the variables used. There is no reason to be or non-existent. Many new national movements lieve that all important background factors have do indeed lean toward the left but this does not been accounted for. As was said before, the set imply that the Socialist and Leftist parties are of variables used in itself illustrates the difficul those which emphasize ethnic identity most ties and restrictions in quantitative comparisons strongly. of ethnic minorities. The strongest relationships are generally found Some composite, generalized variables can now between political resources and ethnic mobiliza be constructed on the basis of the three factors tion. The correlation between political resources with the use of factor scores. Factor scores are and political violence is .33. It would seem obtained by giving individual variables different reasonable to assume that this positive correla weights on the basis of factor loadings. The ra tion is due to autonomy often having been tionale for using factor scores and thus construc granted after a period of ethnically based poli ting new generalized variables lies in the assump tical unrest. This is, for example, true of the tion that factor scores describe the resources in Catalans in Spain, the South Tyroleans in Italy, a more reliable and comprehensive way than the and the Jurassians in Switzerland. It might there original, observed variables. fore be misleading to say that violence depends on political resources. Rather, it seems as if Table 5 presents the correlations between the some of the bombs thrown have had the desired new generalized variables, denoting different ty effect. pes of resources, and the dependent variables describing different forms of political mobiliza Table 6 ranks different minorities on the basis tion. Two summated variables are also intro of their resources. The minorities on the list are duced in table 5. Linguistic, economic and poli two large language groups of Belgium, the Fle tical resources are summated to a variable mea mings and the Walloons. They do not come first suring all resources by simply adding the factor on any individual type of resources, rather they scores of each minority for each of the three have values for each of the resource variables. resource variables. A general measure of ethnic mobilization is likewise obtained by simply ad Table 6 also clearly demonstrates some of the ding the two variables denoting the existence pitfalls of the analysis. When no information on of an ethnic party, and political violence. a particular variable was available, the minority in question was assigned the mean value of the TABLE 5 Correlations between Different Types of variable. In some cases this had led to misleading Resources and Forms of Ethnic Mobilization results. It is for example hardly in conformity with reality for the Channel islands to have been Linguistic Econom ic Political When summated resources resources resources All resources ranked First on economic resources. The ranking

Growth of the Minority 24 08 33 35 positions are astonishing in some cases even when Voting for Socialist parties 02 10 27 08 variable values have been known. One cannot Voting for radical Socialist parties 27 09 20 21 overemphasize the fact that in individual cases re tthnic party 11 17 32 33 sults can be — and are — misleading. Still, it Political violence 03 23 33 33 seems reasonable to assume that findings point Ethnic Mobilization (as summated on the basis of in the correct direction. the two above mentioned variables) .10 26 42 42

The correlations are fairly low. They are, how Conclusion: A Host of Unanswered Questions ever, mainly positive. This indicates clearly that ethnic mobilization is more common when re The ethnically most active and mobilized West sources are available than when they are lacking. European minorities are found among those that A special pattern prevails for the Socialist vote, a have good resources, but mobilized minorities variable, which tends to increase as resources de also exist among minorities with weak resources. crease. This pattern does not contradict the This is apparent when the variables denoting re- 436 Zeitschrift für Soziologie, Jg. 10, Heft 4, Oktober 1981, S. 427 - 437

TABLE 6 Ranking of the Minorities on the Gener TABLE 7 alized Resource Variables Resources (summated on the basis of linguistic, economic and political resources) Rank Minority Linguistic Economic Political Small Medium Great resources resources resources Bretons Basques (Spain) Flemings Corsicans Welsh Walloons 1 . Flemings 9 3 6 Gaels (Northern Slovenes (Italy) 2. Walloons 10 12 9 "§ X Ireland) Jurassians 3. Slovenes (Italy) 11 7 17 South Tyroleans 4. Jurassians2 6 18 11 Catalans (Spain) 5. Ticinese2 13 8 16 6. Germans (Belgium) 5 9 23 Catalans Friulians Germans (Belgium) 7. South Tyroleans 8 35 2 Gaels (Ireland) Danes (Germany) Swedish Finns 8. Swedish Finns 2 25 12 Basques Faroe islanders Valdotams 9. Valdotams 12 17 19 Ethnic | (France) 10. Aalanders 3 30 7 Mobili- ■§ Sards Slovenes (Austria) Aalanders 11. sation 3§ Galicians Occitans (France) Alsatians Rhaetians1 18 27 1 Croats (Austria) West Frisians 12. Catalans (Spain) 31 4 5 Germans (Den 13. Alsatians 14 5 29 mark) 14. Occitans (Piedmonte) 34 6 15 Flemings (France) 15. West Frisians 27 20 10 16. Friulians 41 2 20 Magyars (Austria) Ladins Ticinese 17. Ladins 28 26 8 North Frisians Channel islanders Rhaetians 18. Danes (Germany) 16 14 40 Lapps (Norway) Greenlanders Occitans (Italy) 19. Channel islanders3 37 1 41 Lapps (Sweden) Gaels (Isle of Man) Tome valley 20. Basques (Spain) 40 11 21

References:

Allardt, E., 1979: Implications of the ethnic revival of Haarmann, H., 1975: Soziologie und Politik der Spra modern industrialized society: A comparative study chen Europas. München: Deutscher Taschenbuch-Ver of the linguistic minorities in Western Europe. Hel lag. sinki: Commentationes Scientiarum Socialium 12 Kloss, H., 1978: Die Entwicklung neuer germanischer Anderson, M., 1978: The renaissance of territorial mi Kultursprachen seit 1800. Düsseldorf: Pädagogischer norities in Western Europe. West European Politics Verlag Schwann. 1: 128-143. Stephens, M., 1976: Linguistic Minorities in Western Barth, F., 1978: Introduction. In: F. Barth (ed.), Scale Europe. Llandysul, Dyfed, Wales: Gomer Press. and Social Organization. Oslo/Bergen/Troms^: Uni- Straka, M., (ed.), 1970: Handbuch der europäischen versitetsforlaget. Volksgruppen. Wien/Stuttgart: Wilhelm Braumüller, Eurostat 1976: Regional Statistics 1975: Population, Universitäts-Verlagsbuchhandlung. Employment, Livingstandards. Williams, C.H., Ethnie separatism in Western Europe. Grulich, R./Pulte, P., 1975: Nationale Minderheiten in Tijdschrift voor Economie en Social Geografie 71: Europa. Opladen: Heggen-Dokumentation 12. 142-158.