A sociolinguistic investigation of multilingualism in the Canton of Ticino Switzerland [Reprint 2017 ed.] 9783111678696, 9783111292823

Citation preview

JANUA LINGUARUM STUDIA M EMORIAE N I C O L A I VAN W I J K D E D I C A T A edenda curat C. H. V A N

SCHOONEVELD

Indiana

University

Series Practica,

241

A SOCIOLINGUISTIC INVESTIGATION OF MULTILINGUALISM IN THE CANTON OF TICINO SWITZERLAND

by

J Ü R G E N Β. H E Y E Pontificia

Universidade Católica Rio de Janeiro

1975

MOUTON THE H A G U E · PARIS

© Copyright 1975 in The Netherlands Mouton & Co. N.V., Publishers, The Hague No part of this book may be translated or reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publishers.

LIBRARY OF CONGRESS CATALOG CARD NUMBER: 74-78499

Printed in The Netherlands by Mouton & Co., The Hague

FOREWORD

Part I of this study investigates the multilingualism in the Canton of Ticino, the Italian speaking part of southern Switzerland. The investigation was conducted during the summer months of 1968 by means of questionnaires and personal interviews. The questionnaires were designed to determine (a) socioeconomic status, (b) non-native language performance, and (c) language attitudes of the respondents. The questionnaires were prepared in both German and Italian and mailed to a random sample of the population of the three major urban areas, Lugano, Bellinzona and Locarno. The results were then analyzed and correlated to provide an overall picture of the multilingual situation. I am particularly grateful to my mentor, Dr. Robert J. DiPietro, who gave freely of his time in helping me during the various stages of this study. His criticism was always constructive and kind. My gratitude extends also to my readers, Dean Robert Lado and Dr. Helena Hrabikova. I would like to express my sincere thanks to all those people in the Canton of Ticino who, by taking the time to complete the questionnaires, made this study possible. Among them, I would like to mention especially the following: Dr. Antonio Giaccardi and Dr. Emilio Mordasini, of Bellinzona, Mr. Carlo Bianchoni, Dr. Mario Forni, Dr. Edo Galli and Dr. Antonio Marazzi, of Locarno, Dr. Edo Rossi and Dr. Adriano Soldini, of Lugano — all of whom are connected in various functions with the school system of the Canton of Ticino — the staff members of the Department of Public Education and the Cantonal Office of Statistics, in Bellinzona, and the Cantonal Library in Lugano; my informants, Mrs. Anna Ferregutti, Miss Maria Milesi, Mr. Renato Sciolli, Mr. Bernardino Baroni, of Pura and Agno, and Dr. Gianfranco Rossi, of Brusino Arsizio, who submitted patiently to many hours of questions. My most heartfelt gratitude is directed to Mrs. Cesar A. Hidalgo and Annegret Heye who were equally instrumental for the completion of this study.

TABLE OF CONTENTS

Foreword List of Tables Introduction 1 Objectives 1.1 General Objectives 1.2 Specific Objectives 1.2.1 Socioeconomic Status 1.2.2 Language Performance 1.2.3 Language Attitudes

5 9 11 13 13 13 13 14 14

2

Significance of the Study 2.1 Literature 2.2 Definitions and Delimitations 2.2.1 Socioeconomic Status (SES) 2.2.2 Non-native Language Performance 2.2.3 Language Attitudes 2.3 Specific Definitions

16 17 17 18 18 18 19

3

Methods and Procedures 3.1 The Questionnaire 3.2 Percentage of Returns 3.3 Normality of Sampling 3.4 Reliability of the Questionnaires 3.4.1 Reliability of Part I (SES) 3.4.2 Reliability of Part II (Non-native Language Performance) . . 3.4.3 Reliability of Part III (Language Attitudes) 3.5 Validity of the Questionnaire

20 20 22 22 23 23 24 24 25

4

Results of Part I of the Questionnaire 4.1 Calculation of the Coefficient of Correlation 4.2 Demonstration of the 'Product-Moment' Method 4.2.1 SES Indices 4.2.2 Correlation : Income and Education

26 26 27 27 29

8

TABLE OF CONTENTS

4.3 4.4 4.5 4.6

4.2.3 Correlation: Income and Profession 4.2.4 Correlation : Profession and Education Additional Results of Part I Languages Studied in School Native Languages of the Sample Population SES Averages of the Sample Population

29 29 30 34 36 36

5

Results of Part II of the Questionnaire 5.1 Self-Evaluation Scales for Non-native Language Performance . . . 5.1.1 Oral Comprehension 5.1.2 Speaking 5.1.3 Reading 5.1.4 Writing 5.2 Statistical Evaluation of Part II of the Questionnaire 5.2.1 Correlations of Two Sets of Skills and Reliability of Part II .

38 38 39 40 40 41 43 43

6

Results of Part III of the Questionnaire 6.1 Ratings of Language Attitudes 6.2 Language Use in Interpersonal Communication 6.3 Reading as an Indication of Language Attitudes 6.4 Subjective Attributes of Languages 6.5 The Relation between Standard Language and Dialect 6.5.1 Complementary Questions 6.5.2 Learning Non-native Language Dialects 6.5.3 Relative 'Difficulty' Associated with Dialects 6.6 Standard Languages in Restricted Situations 6.7 Statistical Evaluation of Part III of the Questionnaire

45 45 46 48 49 51 52 54 55 56 57

7

Results of the Correlation between the Three Parts of the Questionnaire . . 7.1 Correlation between SES and Non-native Language Performance . . 7.2 Correlation between SES and Language Attitudes 7.3 Correlation between Non-native Language Performance and Language Attitudes 7.4 Discussion of Results

60 60 60 61 61

Conclusion

63

8

References Appendices:

Index

I. II. III. IV.

Summary of Statistical Calculations Principal Statistical Formulae Supplementary Tables Questionnaire Samples

64 66 66 68 72 86

LIST O F TABLES

Table Table Table Table Table Table Table Table Table Table Table Table Table

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

Table Table Table Table Table Table Table Table Table Table Table Table

14 15 16 17 18 19 20 21 22 23 24 25

Distribution of Questionnaires Questionnaire Returns Classification of the Sample by Age Groups Classification of the Sample by Sex Native Languages of the Sample Distribution of the Sample According to Income Income Distribution of the Sample According to Language Distribution of the Sample According to Profession Distribution of the Sample According to Education Foreign Languages Studied in School Distribution of the Sample According to Native Language Average SES Ratings of the Sample Extent of Non-Native Language Use Indicated by the Self-Evaluation Scores Attitudes of the Sample toward Non-Native Languages Extent of Language Use in Interpersonal Communication Extent of Reading: Books Read per Year Professional Usefulness of Languages Ranking of Languages according to 'Beauty' Standard Languages and Dialects in Everyday Communication Association of Dialect Use with Lack of Formal Education Association of Standard Language Use with Conceit Willingness to Learn Foreign Language Dialects Is it Useless to Learn a Foreign Language Dialect? Which Dialects are Considered more Difficult to Learn? Languages Used in Religious Functions Appendix

III.

Table A Non-native Language Performance of the Sample : Oral Comprehension Table Β Non-native Language Performance of the Sample : Speaking Table C Non-native Language Performance of the Sample : Reading

10 Table Table Table Table

LIST OF TABLES

D Non-native Language Performance of the Sample : Writing E Attitudes of the Sample toward Non-native Languages and Dialects F Reading of Periodicals G Distribution Curve for Normality of Sampling

INTRODUCTION

The Canton of Ticino, situated at the southern end of Switzerland, is the only area in that country in which Italian is spoken by the majority of the population. For many centuries this canton has been in contact both with the rest of the country and Italy. It has been closely associated politically and economically with the former and culturally with the latter. It has long been a meeting ground for people speaking languages other than Italian. It is therefore ideally suited for a sociolinguistic investigation of multilingualism. During the last twenty years there has been a considerable increase in the number of people who have come from other parts of Switzerland or from abroad to take up residence in the Canton of Ticino. This movement may be attributed to a number of reasons, the most important of which seems to be the mild climate of the area combined with the economic stability and favorable tax system of Switzerland. Whatever the reasons for the influx of large numbers of non-Italian speakers may have been, the fact remains that today the extent of multilingualism is still increasing. The virtual absence of any social friction between the various segments of the multilingual population make the Canton of Ticino even more interesting from a sociolinguistic point of view. The total population of the Canton of Ticino in 1967 was 230,7s!. 1 The population of the three major cities is as follows: Lugano: 52,356; Locarno: 25,949; Bellinzona: 22,840. About half the population of the canton lives in these three cities. It is interesting to note that between 1960 and 1967 the population increased by 18.0 percent. Between the years 1860 and 1967 the population almost doubled from 115,985 to 230,751. For administrative purposes the canton is divided into eight 'Districts' which are then subdivided into thirty-nine 'Circles' or 'Agglomerates'. These are further divided into 253 'Communes', villages which have their own local administration. Of the twenty-five Swiss cantons, Ticino ranks tenth in total population and nineteenth in population density. The economy of the canton is based largely on viticulture, where it ranks third, and on tourism, where it ranks fourth in the nation. 2 Commercial and industrial enterprises, on the other hand, are not welldeveloped and the Canton of Ticino ranks tenth nationally in these areas. According 1 Source : Ufficio Cantonale di Statistica : "La Popolazione del Cantone Ticino nel 1967" (Bellinzona, Switzerland, 1968). 2 Source: Ufficio Cantonale di Statistica: "Annuario Statistico del Cantone Ticino, 1966/67" (Bellinzona, Switzerland, 1968).

12

INTRODUCTION

to the census of 1960, 36,320 foreigners were living in the canton. The majority of them were German and Italian citizens. The same census which gives the total population as 195,566 also lists the following information under 'native language', without regard to citizenship: Italian: 166,361; German: 18,272; French: 2,796; other languages: 1,597. The highest concentration of speakers of these languages is found in the districts (not cities) of Lugano and Locarno.

1 OBJECTIVES

In the following pages the objectives of this study will be discussed. For reasons of clarity they are divided into 'General Objectives' and 'Specific Objectives'. 1.1

GENERAL OBJECTIVES

This study is a presentation of a sociolinguistic investigation, done in form of a survey of a geographically and politically defined entity, the Canton of Ticino in southern Switzerland. The survey was conducted during the summer of 1968. The primary data was collected by means of questionnaires. Secondary data was gathered through interviews with members of the population. The following three aspects were investigated : (a) the extent of multilingualism in the Canton of Ticino, especially in Lugano, Locarno and Bellinzona. (b) the attitudes of the population towards the standard languages spoken in the area, as well as towards the dialect varieties of each. (The standard languages involved are Italian and German, the dialects are Ticinese and Swiss German.) It was also attempted to determine the views speakers of one dialect have about speakers of a dialect in another language. (c) whether or not there is a correlation between the foreign language performance of the population as measured by the questionnaire and their language attitudes and to determine the extent of this correlation. An additional effort was made to determine whether the socioeconomic status of the population is correlated with their foreign language performance and to what extent these two factors are correlated. A third measure determined the extent of the correlation between socioeconomic status and language attitudes. 1.2

SPECIFIC OBJECTIVES

The following are the specific objectives of this study. 1.2.1

Socioeconomic

Status

In determining the socioeconomic status of a population the following indices were

14

OBJECTIVES

used, following the suggestions of such social scientists as G . D . H . Cole (1956), William L. Warner (1960), and Max Weber (1968): (a) Income: The annual income was rated on a six-point scale ranging from approximately two-thousand dollars (rating: six) to over six-thousand dollars (rating: one). The subject was asked to indicate his income by marking the appropriate place, (b) Occupation: This was also rated on a six-point scale. Ratings were based primarily on the prestige value of the occupation or profession of the respondent. The scale extended from 'Arts and Crafts' (rating: six), 'Officials', including local, state and federal officials (rating: five), 'Teachers', (rating: four), 'Hotels and Tourism' (rating: three), 'Industry and Commerce' (rating: two) to 'Professionals' (rating: one) such as lawyers, doctors and architects. The respondents were asked to mark that description which most closely corresponded to their professional activities, (c) Education: Educational attainment was again measured on a six-point scale ranging from 'Elementary school only' (rating: six) to 'Advanced academic degrees' (rating: one). These indices are taken to measure the socioeconomic status of a given population. The measures described above were then averaged out so that each respondent obtained an overall socioeconomic status (hereafter abbreviated SES) rating between 1.0 and 6.0. The correlation between any two of these measures was calculated and the results obtained will be discussed below. 1.2.2 Language Performance In this section of the questionnaire the extent of foreign language use by the sample population was determined. The respondents were asked to evaluate their own foreign language performance by rating oral comprehension, speaking ability, reading ability, and writing ability for each foreign language known by the respondent on a six-point scale. For each of these four 'skills' the ratings were described in a few short sentences and the respondent then marked that rating which he felt represented best his performance in each skill in a given foreign language. The self-evaluation scale devised for this study does not claim to measure with precision the foreign language performance of the respondent. It provides an approximate index which can be used for statistical calculations and which can be correlated with other measures of the questionnaire. The foreign language evaluation was preceded by a short section which determined the subject's exposure to foreign languages in the schools. No measurement or rating was taken. The subject indicated the number of years he had studied a particular foreign language in school. In addition, the native language and his familiarity with his native dialect, if any, was determined. 'Native language' was defined as the language which was used most frequently in the subject's home during the first five years of his life. He was also asked to indicate the dialect of his native language which was familiar to him. This could be done by listing the geographical area where he grew up. 1.2.3 Language Attitudes This section of the questionnaire determined the attitudes of the sample population

OBJECTIVES

15

toward standard German and standard Italian as well as toward Swiss German and Ticinese according to their native language. The respondents were asked to rate their attitudes on a six-point scale toward the above languages and dialects, excluding their native languages. Each rating was again described briefly. The respondents then marked their attitudes toward each language or dialect, not including their native language, on this scale. For example, if the respondent's native language was standard German, he would indicate his attitudes toward Swiss German, standard Italian and Ticinese. As an additional measure of language attitudes the respondents were asked to indicate which language or dialect is used in communication with acquaintances and friends and in communication of friends and acquaintances with the respondents. The extent of communication was measured on a three-point scale. The respondents were also asked to indicate the approximate number of books, papers and journals read in German, Italian, French, English and other languages. It was also determined (a) whether there are negative attitudes towards persons who speak mainly a dialect or who have incomplete mastery of their standard language; (b) whether there are negative attitudes towards persons who always use the standard language ; (c) the professional usefulness of the language involved ; (d) whether or not the sample population is receptive towards learning a dialect of a foreign language; (e) the language or languages used in religious services and private correspondence. None of these questions were rated on a scale. An attempt was also made to determine the relative difficulty in learning a foreign language dialect as regards speakers of other languages. The respondents were asked to express their feelings by selecting one of three possibilities.

2 SIGNIFICANCE O F THE STUDY

Over the last few years linguists have become increasingly interested in the various problems of describing the sociolinguistic aspects of a given geographical or political entity. The works of such scholars as Robert J. DiPietro (1968), Charles A. Ferguson (1959), Joshua A. Fishman (1966), John J. Gumperz (1964), Dell Hymes (1967), William Labov (1966) and Uriel Weinreich (1964) have greatly enriched the fields of multilingualism and sociolinguistics. Some writers, such as William A. Stewart (1962), have proposed sociolinguistic profile formulae of multilingual areas, considering primarily the extent of usage of the languages involved and their function in the community. Others, such as William Labov (see below), have dealt with certain dialect features and their presence or absence in given strata of the population. This study has a multiple purpose : It presents an overall view of the linguistic situation in the Canton of Ticino based on an investigation by means of questionnaires which determine some of the most prevalent sociolinguistic aspects, such as the attitudes of the population towards the dialects of their native language, the extent of multilingualism in urban areas, as well as the extent and degree of foreign language performance of the individual respondents. The sociolinguistic survey is based on the assumption that questionnaires can be constructed and distributed which are extensive enough to cover the most significant aspects of multilingualism and language attitudes, precise enough to permit a detailed statistical analysis of the findings and yet simple enough to be presented to a, linguistically speaking, unsophisticated population. This type of questionnaire might be applied in areas where there are not only two or more standard languages but also two or more dialects of these languages in contact. The Canton of Ticino provides a good example for this type of investigation, since there are at least two standard languages in contact : standard Italian and standard German, in addition to the dialect of the Canton of Ticino (an alpine-lombardic dialect which will be labelled henceforth with the term used in loco, namely 'Ticinese') and Swiss German. 1 Neither of these dialects will be further specified or classified within the framework of the present investigation. One must, however, be aware that there are 1

'Swiss German', as used in this study, includes all varieties of the German dialects spoken in the northern and central parts of Switzerland. For a further discussion on Swiss German dialects see Moulton (1962).

SIGNIFICANCE OF THE STUDY

17

a large number of different varieties of both dialects. Another feature of particular interest to the sociolinguist is the attitude with which speakers of one dialect regard their own standard language on the one hand and another standard language (or dialect) on the other hand. One asks oneself why it is that speakers of Ticinese appear to hold standard Italian and standard German in equally high esteem and yet consider Swiss German as the least desirable form of oral communication. The people of the Canton of Ticino often speak about the 'deficiencies' of the dialect and in general make few attempts to make it a 'socially acceptable' form of speech. This contrasts with the fact that the Swiss Germans have made their dialect acceptable on all social levels and see no particular advantages in using standard German for oral communication. If the questionnaires designed for this study can provide answers not only to questions regarding multilingualism but also to some of the questions about language attitudes within the framework outlined above, they may indeed constitute a useful tool for future investigations of a similar nature. 2.1

LITERATURE

To this date there have been few publications about the Canton of Ticino and its linguistic situation. Most of the available studies are descriptions of dialects or dialect varieties. The works of two men stand out clearly ; Oscar Keller and Silvio Sganzini have contributed since 1928 extensively to the study of Ticinese dialects. The former has, in a way, prepared the field with a number of books and articles, the most important of which are listed in the bibliography. The latter followed with the impressive Vocabolario which is based on the monumental work of Karl Jaberg and Jacob Jud. With the exception of the present study, no investigation of a sociolinguistic nature has been undertaken in the Canton of Ticino. 2.2

DEFINITIONS A N D DELIMITATIONS

The geographical area under investigation is limited to the three largest cities of the Canton of Ticino : Lugano, Locarno and Bellinzona. These cities are all located in the southern part of the canton and their combined total population is 101,145 (as of December 31, 1967), slightly less than half of the total population of the canton of 230,751. When studying the publications of the Cantonal Office of Statistics in Bellinzona one encounters conflicting figures in regard to the population of these cities. This results from the fact that the three cities also constitute administrative districts which extend considerably beyond their city limits. For this investigation only the population residing within the city limits was taken into consideration. From a dialectological point of view, Lugano, Locarno and Bellinzona are centers of what has been termed 'alpine-lombardic koinè',2· i.e. a standardized urban form of the dialect which is said to 2

According to Federico Spiess, in personal communication, the term 'alpine-lombardic koinè' was first used by Oscar Keller (see References).

SIGNIFICANCE OF THE STUDY

18

have originated in and around Milan and now is used with minor modifications in most urban areas of northern Lombardy and the Ticino. Although it is clear that all varieties of Ticinese have a stock of common features at all linguistic levels, there is little mutual intelligibility between varieties located at the extreme ends of the chain of dialects. The same would also hold on a larger scale for most Italian dialects. 2.2.1

Socioeconomic Status

(SES)

In determining the SES of the respondents three indices were used : (a) income, (b) profession and (c) education. These indices are assumed to provide adequate measures for the SES of a given population without placing undue emphasis on any one aspect of the area's social, cultural or economic structure. 2.2.2

Non-Native Language Performance

Using a self-evaluation scale, the respondents were asked to evaluate their performance in national and foreign languages, 3 excluding their native language, which they had studied. A six-point scale was used for each language skill.4 All non-native languages indicated by the respondents were tabulated but only French was used for the statistical calculations since it was the language most often listed by all groups of subjects. In descending order of frequency the following languages, excluding French, were tabulated: (a) Italian, (b) German, (c) English, and (d) Spanish. In general, native speakers of German indicated that they knew more English, whereas native speakers of Italian listed Spanish more often than English. The following languages were found only in isolated instances and were not considered for further analysis : Swedish, Polish, Yiddish, Arabic, Greek and Portuguese. As regards the ratings for German and Italian, the respondents were asked to evaluate only that language which for them constituted a non-native language, i.e. German for native speakers of Italian and Italian for native speakers of German. 2.2.3

Language

Attitudes

This part of the questionnaire investigated the attitudes of the sample population towards standard languages (German and Italian) and dialects (Swiss German and Ticinese). They were rated again on a six-point scale and a detailed analysis of these ratings is presented below. Other information reflecting the attitudes of the respondents toward languages and dialects are presented, but only partially included in the detailed analysis of the questionnaires. 3 The national languages here are German, French and Italian. The fourth, Romansh, was never listed. Foreign languages, for example, include English. 4 The four language skills considered here are: oral comprehension, speaking, reading and writing.

SIGNIFICANCE OF THE STUDY

2.3

19

SPECIFIC DEFINITIONS

Before proceeding, some of the terms used here must be defined, (a) Socioeconomic status. This is taken here as a measure of the rank obtained by the individual in the community where he lives. This rank is measurable with certain indices which provide the investigator with a set of tools to determine the standing of each respondent in relation to the rest of the community. In this investigation, income, profession and education were used as indices for the SES of each respondent, since these are often considered by the population as 'guidelines' along which the success or failure of an individual and his standing in the community can be measured. It is not claimed that these indices are the only ones possible in the study of this specific area; one might conceivably include the presence or absence of a television set in the household as an indication of socioeconomic standing. The indices mentioned above, however, are more readily available and are easier to establish. (b) Language performance. For the purposes of this study, language performance is taken to mean the actual use of non-native languages which the speaker is able to make in a concrete situation. It is limited by the fact that in our case it is the speaker who determines the extent of his performance by rating himself on a scale. He does not rate his overall performance in each of the languages, but only one 'skill' at a time, such as oral comprehension, speaking, reading or writing. (c) Language attitudes. These are taken within the framework of this study as those feelings which the individual has towards his native language or foreign languages. Language attitudes may be positive, indifferent or negative. They are measurable by the reactions of the subject when asked to rate or describe his feelings towards a language or dialect. In this study no attempts were made to determine the roots of certain language attitudes which may be found in cultural or political differences between groups of speakers. We take language attitudes here as the sum of individual feelings or prejudices which speakers tend to have towards certain forms of oral communication without being able to rationalize them.

3 METHODS A N D PROCEDURES

In this section we describe the procedures used to obtain the sociolinguistic picture of the Canton of Ticino. The methods for statistical evaluation of the data are presented and discussed. 3.1

THE QUESTIONNAIRE

To measure the SES, foreign (non-native) language performance and language attitudes of the population, a questionnaire was constructed and tested on the residents of the cities of Lugano, Locarno and Bellinzona in the Canton of Ticino. The respondents were selected by means of the local telephone directory. This method of selection is based on the assumption that the telephone as such does not constitute a status symbol. The telephone system of Switzerland is so highly developed that almost every house-hold is connected. The procedure employed was as follows: The three cities representing the highest population density were selected. The total population of each city was determined, based on information provided by the Cantonal Office of Statistics in Bellinzona. Then, one percent of the total population of the three cities was taken to be the size of the sample. The same source maintains that in these cities about one-fourth of the population speaks German. The sample was therefore divided in such a way that for every four Italian questionnaires there was one questionnaire in German. This ratio of four Italian speakers for every German speaker is generally accepted as being accurate for the Canton of Ticino and the findings of this study verify it. The distribution of questionnaires was as follows: Table 1.

Distribution of Questionnaires

City (Population)

Italian Quest.

German Quest.

Total

Lugano (52,356) Locarno (25,949) Bellinzona (22,840)

390 200 150

130 70 60

520 270 210

Total:

740

260

1000

101,145

METHODS AND PROCEDURES

21

In general, this method of selecting a sample proved effective. The questionnaires were then mailed to the respondents, providing each with a pre-addressed, stamped envelope. The respondents were also asked at the end of the questionnaires whether they would be willing to concede an interview to the investigator, so that some of the points touched in the questionnaires could be discussed in greater detail. If this request was answered positively, the investigator contacted the respondent by telephone and an interview was scheduled at a mutually agreeable time and place. Some 38 percent of the respondents agreed but only nineteen percent of the total sample were actually interviewed due to scheduling problems. The relatively few interviews that were conducted proved highly informative and helpful, and the interviewed persons usually showed great interest in the investigation in general and at times suggested ways in which the questionnaires might be improved. It was divided into four parts: Part I dealt with the person of the respondent, asking such questions as name, age, place of birth, length of residence in the Canton of Ticino and the reasons for selecting this part of the country. It also included rating scales for income and profession, the first two indices for establishing the SES rating. The third index, education, followed at the beginning of Part II, on the same page. The respondents were asked to indicate the level of their educational attainment by checking the appropriate description, such as 'Completed secondary education'. If the respondents had higher education they were asked to indicate the duration of their studies and the kind of degree obtained. They were asked to indicate the non-native language they had studied in school and the duration of these studies. The mother tongue was established and the native dialect specified. Part III contained the self-evaluation scales for non-native language performance in the following order: Oral Comprehension, Speaking, Reading and Writing. Each level of performance of the four 'skills' was described in a few sentences. The respondent then selected that level which he felt corresponded most closely to his performance in each of the non-native languages he knew. This part of the questionnaire was the longest and perhaps the most cumbersome since the respondent, before being able to answer, had to read the set of descriptions for each 'skill' and then determine which level of performance was suitable for each of the languages he wished to include. Part IV contained the attitudinal ratings for the two standard languages, Italian and German, and for the two dialects, Swiss German and Ticinese. Part V of the questionnaire was taken as an extension of the previous part using a different format. The first two questions dealt with interpersonal communication and language use. The respondents were also asked to indicate the approximate number of books, daily and weekly papers and technical journals read annually. If they answered positively, they were asked to specify the languages. Here, of course, only standard languages were considered. The respondents were then asked to list the languages which they considered most beautiful and most useful in their profession. The next four questions were designed to investigate the attitudinal relationship between standard languages and dialects.

22

METHODS AND PROCEDURES 3.2

PERCENTAGE OF RETURNS

Of the thousand questionnaires mailed, 7.6 percent were returned completed. The actual number of returns is higher but some had to be omitted from the statistical analysis since they were only partially completed. The above percentage represents the total number of questionnaires which were used for statistical calculations. The relatively low number of returns can be attributed to a number of variables which could not be controlled, (a) Time: The only possible time period during which the investigation could be conducted was the summer months June through August. During this time many people in the Ticino traditionally take their vacations and could not be reached, (b) Sampling: The telephone directory was being up-dated and the new edition had not yet been available. 1 Thus, a relatively large number of questionnaires probably never reached their destination, (c) Length: During the initial classifying and sorting process it became clear that some respondents found the questionnaires to be too long and too time-consuming, returning them only partially completed. These were eliminated from the final analysis. The completed returns may be broken down in the following way: Table 2. City

Questionnaire Returns

Italian Quest.

German Quest.

Total

Lugano Locarno Bellinzona

31 7 11

15 9 3

46 16 14

Total

49

27

76

As can be seen, the ratio of Italian and German questionnaires returned is more than 2:1. The ratio of questionnaires mailed was 4:1. This shows that about twice as many German speakers than Italian speakers were willing to return the completed questionnaires. In Locarno more German than Italian questionnaires were returned. In Bellinzona the pattern is reversed. Locarno has considerably more German speakers than the other two cities. It has been a resort town for a long time and attracts more people every year. Bellinzona, the cantonal capital, on the other hand, has never been a tourist center, and has never had large numbers of German speakers. 3.3

NORMALITY OF SAMPLING

The investigator attempted to obtain a representative sample of the population. For this purpose 'population' was taken to mean any group of individuals residing within the city limits of Lugano, Locarno and Bellinzona. A sample is defined as a subset of 1

In Switzerland telephone directories are updated every two years. A good number of questionnaires were returned with a stamp 'Deceased'.

METHODS AND PROCEDURES

23

a population (Weiss, 1968). A sample, in order to be workable, must be representative of the population under investigation. In order to determine the normality of sampling the SES indices were used. They were averaged out so that one SES rating was obtained for each respondent. Using a 'chi-square test' (Garrett, 1966), the frequency distribution of these ratings was calculated and their probability obtained. We define 'probability' simply as "the relative frequency over the long r u n " (Weiss, 1968). It means that there are 'x' chances in a hundred that the given differences between obtained and normal frequencies are due to chance or to fluctuations of the sample. The probability (P) was found to be between .50 and .60 (P = .58). Thus it is found that X 2 is significant and that there are fifty-eight chances in a hundred that the differences between obtained and normal frequencies are due to chance or fluctuations of the sample. The results obtained from the SES ratings of the sample population indicate a relatively normal frequency distribution. It follows, then, that the sample itself must be equally representative of the population under investigation.

3.4 RELIABILITY OF THE QUESTIONNAIRES

This part of the study determines the reliability of the entire questionnaire and its three major parts (SES of the sample, non-native language performance, language attitudes). Reliability is defined here as " . . . a general quality of stability of scores regardless of what the test measures" (Lado, 1961). This stability depends to a large extent upon the degree to which the scores obtained are free of chance errors. The 'split-half technique' to determine the reliability of the questionnaire was used. This method is regarded by many as the best way to obtain test reliability, although the 'rational equivalence method' has certain theoretical advantages. It is employed most frequently in those cases where it would not be possible to construct parallel forms of the test or administer it a second time. One of its advantages is that all pertinent data for computing the reliability coefficient are obtained on one occasion, thus eliminating variances between the two testing situations. A disadvantage of the method lies in the fact that chance errors affect the scores on both halves of the test in the same manner, because the test is administered only once. Increasing the length of the test makes the estimate of reliability more accurate. In the present case it was felt that the length of the questionnaire and its parts was sufficient to reduce the effect of chance errors on scores (ratings). 3.4.1

Reliability of Part I (SES)

Before determining the reliability of the entire questionnaire, the coefficient of reliability for each of its parts was calculated. To determine the reliability coefficient of the entire questionnaire we used the 'Spearman-Brown' prophecy formula. It estimates the reliability of the entire test from the correlations of the half-tests. The correlation between the half-tests was found through the 'product-moment' method. The invest-

24

METHODS AND PROCEDURES

igator was interested to determine the coefficient of correlation between the three sets of scores obtained from the SES ratings, income, profession and education. We computed the correlations between (a) income and education, (b) income and profession, and (c) profession and education. The correlations obtained from these three measures were then taken as partial indications of the coefficient of reliability. The 'SpearmanBrown' formula was then applied to determine the reliability of the SES ratings of the sample (Part I). The coefficient of reliability for Part I is (r) = .79.

3.4.2

Reliability of Part II (Non-Native Language

Performance)

For determining the reliability of Part II the procedure used was similar. Since this part of the questionnaire was designed to determine the non-native language performance of the sample with a self-evaluation scale, the ratings obtained for two 'skills', such as oral comprehension and speaking, were averaged out and correlated with another set of ratings from a different combination of 'skills', such as reading and writing. All ratings obtained here are based on French since that was the language most widely used. The correlations for the following pairs of 'skills' were calculated and the coefficient of reliability then established : (a) (b)

Oral comprehension and writing with speaking and reading. Oral comprehension and speaking with reading and writing.

It was found that for (a) the coefficient of reliability was : r„ = .91 and for (b) r«o = .87 These values were then used to establish the coefficient of reliability for Part II as described above. 2 (r) = .94 3.4.3

Reliability of Part III (Language

Attitudes)

Part III of the questionnaire (language attitudes) was divided into two parts, A and B. In Part IIIA we determined the language attitudes of the sample in relation to their native language, which was not rated. For each language or dialect an arbitrary numerical value was assigned in the following way: Italian: '4', German: '3', Ticinese: '2', Swiss German: Ί ' . Thus an average rating could be obtained for each individual. Part 2 The coefficient of both halves is always larger than that of each half.

25

METHODS AND PROCEDURES

HIB used the same numbers for the languages and dialects in the answers obtained to the remaining questions. These were divided into 'odd' and 'even' groups and then correlated by the 'product-moment' method. It was found that for Part IIIA the coefficient of reliability was : r u = .67 and for Part IIIB r 2 2 = .80 These values were then used to establish the reliability coefficient of Part III following the same procedures as before. (r) = .85 The reliability of the entire questionnaire was determined by using each of the three parts as one third in the 'Spearman-Brown' formula. Thus a coefficient of reliability (r) = .94 was established for the entire questionnaire.

3.5

VALIDITY OF THE QUESTIONNAIRE

The validity of a test or a questionnaire is dependent upon the fidelity with which it measures what it claims to measure. Reliability is concerned with the stability of test scores, validity is concerned with their evaluation in terms of independent criteria. We had hoped that the interviews conducted with a number of respondents could be used as the 'independent criteria' for establishing the validity of the questionnaire. To be valid a test must be reliable. Since the reliability of the questionnaire was found to be high, it was decided to use the index of reliability as a measure of validity. Within the framework of this study the index of reliability is taken as one possible measure of validity of the questionnaire. The following values were obtained using the formula Γοο=λ/Γΐ1· Part I. Part II. Part IIIA/B Entire Questionnaire

SES of the sample. Language Performance. Language Attitudes.

rœ rœ rœ rœ

= = = =

.88 .97 .92 .97

It is shown that the questionnaire is reliable and reasonably valid. It constitutes a sufficiently effective instrument for investigating the sociolinguistic structure of a multilingual area such as the Canton of Ticino. This consideration becomes particularly important and relevant when the correlations within the different parts of the questionnaire will be discussed.

4 RESULTS OF PART I O F THE QUESTIONNAIRE

In this chapter we discuss the correlations found between certain aspects of the different parts of the questionnaire as well as between the parts (I, II, and III) themselves. The methodological considerations will be discussed and the significance of the findings will be evaluated. Specific questions and objectives will be answered. We also present the socioeconomic structure of the area, the results of the self-evaluations of non-native language performance and of the language attitude ratings. Finally the sociolinguistic picture of the Canton of Ticino will be drawn.

4.1

CALCULATION OF THE COEFFICIENT OF CORRELATION

Correlations are often used to examine the relationship which is assumed to exist between two variables. One may wish to determine, for example, the extent to which one variable is dependent or independent from another variable. This correlation, if it is linear, can be expressed by the 'product-moment' coefficient of correlation. A relationship may be absolute, i.e. it never changes. Then the dependence between the two variables is also absolute. In this case the correlation is perfect and positive (r = 1.00). This kind of relationship exists between the diameter and the circumference of a circle. Since both dimensions are completely dependent upon each other, the correlation would be 1.00. If no relationship exists, the coefficient of correlation is said to be 0.00. Between these two limits (1.00 and 0.00) there is some degree of positive association, depending on the size of the coefficient. However, a negative or inverse relationship may also exist. It is said to be perfect when the coefficient is = — 1.00. Coefficients between 0.00 and —1.00 indicate an increasing degree of inverse relationship. One must keep in mind, however, that the coefficient of correlation is to be judged in terms of the conditions under which it was obtained as well as the objectives involved. In calculating the coefficient of correlation for a number of measures of the questionnaire the 'product-moment' method was used. The coefficient is calculated from a correlation table. It involves the construction of a 'scattergram', where the two variables to be correlated are arranged vertically on the left side and horizontally along the top of the diagram in predetermined intervals. The joint distribution of the two variables, also called bivariate distribution, is marked, first by tallies and then numer-

RESULTS OF PART I OF THE QUESTIONNAIRE

27

ically. Then a series of checks are executed to determine if the calculations are correct. These checks are then marked as shown in the correlation tables. Since the mean was assumed, it has to be corrected with the formulae : jy j y (N = size of the sample) Cx = — ; Cy = — (f = frequency of variable) After the correction (c) for (x) and (y) have been found, the standard deviation (sigma) for both is found by applying the formulae :

σχ

/ fx 2

= V "ñ "

Cx2 (i)CTy=

'-

/ fy 2

V Ñ" •- Cy2'-(i)

(i = step interval). The coefficient of correlation is then calculated with the formula : Σχ'γ' r =

Ν — cxcy xy

(Product-Moment)

4.2 DEMONSTRATION OF THE 'PRODUCT-MOMENT' METHOD

It may be convenient to demonstrate the application of the 'Product-Moment' method by discussing some of the correlations computed for the questionnaire and present some tables for visual demonstration. 4.2.1

SES

Indices

In Part I of the questionnaire the SES ratings for the sample were determined by using three indices, income, profession and education, rated on a six-point scale. They were then averaged out so that each individual had one overall SES rating. The scales used for the three indices are given below. A. Annual

Group Group Group Group Group Group

Income.

I II III IV V VI

(up to 8,000.00 sfr.) (8,000 to 12,000 sfr.) (12,000 to 16,000 sfr.) (16,000 to 20,000 sfr.) (20,000 to 24,000 sfr.) (over 24,000 sfr.)

Rating: Rating: Rating: Rating: Rating: Rating:

6 5 4 3 2 1

28

RESULTS OF PART I OF THE QUESTIONNAIRE

B. Profession (Ranked according to prestige). Group Group Group Group Group Group C.

I II III IV V VI

(crafts and agriculture) (officials, local and state) (teachers on all levels) (hotel and restaurant) (industry and commerce) (self-employed professionals)

Rating: Rating: Rating: Rating: Rating: Rating:

6 5 4 3 2 1

(elementary school completed) (secondary school completed) (maturity exams passed) (college, university attended) (college, university degree) (advanced degrees, doctorate, etc.)

Rating: Rating: Rating: Rating: Rating: Rating:

6 5 4 3 2 1

Education. Group Group Group Group Group Group

I II III IV V VI

It is clear from this kind of SES rating that a number of apparent discrepancies cannot be avoided. Several respondents, for example, did not find a suitable category for their professional activities and subsequently listed their occupation on the margin of the questionnaire. An attempt was then made to incorporate those cases as objectively as possible into one of the categories listed. The resulting rating was then used for the calculation of the average SES rating. (Respondents without income, such as students, members of religious orders and housewives, were assigned an income rating of six.) After having made these adjustments, we were especially interested to determine whether or not there is a correlation between these indices of SES and the extent of correlation. The following correlations were computed: (a) correlation between income and education. (b) correlation between income and profession. (c) correlation between profession and education. The following procedure was used : The questionnaires were arranged alphabetically and then divided into two halves, independent of the language of the questionnaire. This division was required for the calculation of the coefficient of reliability and validity; it is not needed for the coefficient of correlation. After numbering the questionnaires, the ratings of each respondent for the three SES indices was entered in its proper column. Then the total ratings and the average SES ratings were calculated. This gives us the raw data for the correlation tables. Then 'scattergrams' were constructed arranging the two variables, e.g. income and education, along the horizontal and vertical axes. Since the ratings were obtained from six-point scales, the step intervals (i) ranged in whole numbers from 1.0 to 6.0.

RESULTS OF PART I OF THE QUESTIONNAIRE

4.2.2.

29

Correlation: Income and Education

After making the corrections for the assumed mean and calculating the standard deviation, the coefficient of correlation (a) was found to be r a = .465, positively marked but not very substantial. The relatively low correlation between income and education should not come as a surprise. Sometimes a person with little formal education becomes highly successful in his profession, achieving a high income rate. That the reverse is also true need not be mentioned. We find, however, a higher frequency of occurrence in the upper income groups. This skewness of distribution may be contributed partially to the fact that the income ranges were estimated generally too low. The distribution of scores would have been more even had the income categories been adjusted upward by adding an additional category above, say, 30,000 sfr, and eliminating the lowest one. This would then have been more representative of the actual situation. Yet there is a positive correlation between income and education of the present sample population. It is not markedly high, but sufficiently indicative of the socioeconomic situation in the Canton of Ticino as represented by the major urban areas. 4.2.3

Correlation: Income and Profession

For the correlation between income and profession the same calculations were applied as for the previous correlation. The raw data were again arranged in a scattergram. Step interval and size of the sample remained unchanged. Correlation b, between income and profession, is positively marked and substantial, r b = .595. There is a substantially high correlation between the prestige of a profession and its financial rewards, that is, the more income to be derived from a given occupation, the higher its prestige in the community. It follows then that income is a better index of social prestige than educational attainment in the Canton of Ticino. 4.2.4

Correlation: Profession and Education

For the calculation of the coefficient of correlation between profession and education the 'product-moment' method was applied once more. The correlation c, between profession and education (rc = .597), is again positively marked and substantial. It is insignificantly higher than that between income and profession. There is a considerably high correlation between the three indices used to determine the SES ratings of the sample. The lowest coefficient was found to be between income and education (ra = .46). The coefficients between income and profession and between profession and education (rb = .595 and r c = .597, respectively) are considered the same, since they may be rounded up to r = .60. The figures presented here may not be very meaningful in themselves but they contribute to the overall picture of socioeconomic status in the Canton of Ticino. They also help to establish reliability and validity of the questionnaire and they show the

30

RESULTS OF PART I OF THE QUESTIONNAIRE

normal distribution of the sample. The correlations indicate that the three indices employed here are adequate for determining the SES ratings of a given population.

4.3 ADDITIONAL RESULTS OF PART I

Let us now look at some additional results from the first part of the questionnaire. We were interested in determining which age groups and which sex were most frequently represented. Dividing the sample in four age groups the following distribution was found: Table 3.

Classification of the Sample by Age Groups According to City and Language of Questionnaire

Age group

Lugano

Locarno

Bellinzona

Italian

German

Total

under 30 30—49 50—69 over 70

4.1% 27.1 % 21.9% 5-4%

1.3% 12.3% 6.8% 1.3%

4-1% 8.2% 5-4% 1-3%

8.2% 35.6% 16.4% 4.1%

1-3% 12.3% 17.8% 4.1%

9.5% 47.9% 34.2% 8.2%

The number of German speakers (regardless of nationality) is proportionately low for the first two age groups but increases for the two last groups. This confirms the observation made before that the Canton of Ticino attracts considerably large numbers of German speakers who have reached retirement age. It seems on the other hand that few young speakers of German are migrating into the area. On the whole, it becomes clear that mostly middle-aged people completed the questionnaire. The following table sums up in a similar fashion the distribution according to sex of the sample. The sex of the respondents was determined by the first name. Since this information was optional, the percentages do not always add up. Table 4.

Classification of the Sample by Sex According to City and Language of the Questionnaire

Sex

Lugano

Locarno

Bellinzona

Italian

German

Total

Male Female

40.2% 18.0%

22.2%

16.6% 2.7 %

54.1% 11.0%

24.9% 9.7%

79.0% 20.7%

—

31

RESULTS OF PART I OF THE QUESTIONNAIRE

As table 4 shows, there were about three times as many males in our sample than females. It is interesting to note the proportionately small number of Italian female respondents. The investigator found a considerable reluctance on the part of the female population to fill out this kind of questionnaire. As far as could be determined the proportion of single females (those having telephones listed in their names) is not unusually small. The low percentage of females (20.7%) is a further indication of a somewhat negative attitude toward outside intrusion (by means of questionnaires). It may be attributable in part to the strong cultural heritage of Italy in the Canton of Ticino, which, as in most southern European countries, limits the role of the female first of all to house and family. Another aspect concerns the native language of the respondents. They were asked to indicate that language or dialect which was used most frequently at home during the first five years of the respondents life. The following count was made. 1 Table 5.

Native Languages of the Sample According to City

City

Ticinese

St. Italian

Swiss German

St. German

Lugano Locarno Bellinzona

9.8% 4.2% 9.8%

28.1% 7.0% 5.6%

14.0% 5-6% 4-2%

8.4% 2.8%

23.8%

40.7%

23.8%

11.2%

Total

—

Again we find that the ratio between Italian and German speakers is about 2:1, counting the two dialects with the respective standard languages. As regards the relation between dialect and standard language, we find that in Lugano, for example, three times as many people indicated that their native language was standard Italian and not Ticinese. In Bellinzona, on the other hand, more persons indicated the dialect as the oral means of communication used at home. In all three cities the speakers who gave Swiss German as their native language outnumber those with standard German. The ratio is comparable in all three cities although no standard German was found in the Bellinzona sample. This is not surprising since it is not one of the major resorts in the Canton of Ticino. The distribution of the sample according to income category is presented in table 6.

1 Totals do not add up to 100 % because languages such as Swedish or Polish were listed. Since they were found only in isolated instances they were not considered for the calculations.

32

RESULTS OF PART I OF THE QUESTIONNAIRE

Table 6.

Distribution of the Sample According to Income

City

Cat. I

Cat. II

Cat. Ill

Cat. IV

Cat. V

Cat. VI

Lugano Locarno Bellinzona

9.5% 1-5% 1-5%

6.3% 1-5% 1-5%

6.3% 3.1% 4.7%

9.5% 7.9% 1-5%

3.1% 6.3% 3-1%

20.6% 4-7% 6.3%

12.5%

9.3%

14.1%

18.9%

12.5%

31.6%

Total

A number of observations can be made about this table. First, the distribution of the sample is skewed toward the two highest income categories (V and VI). This is another indication that the estimates of annual income, which were based on information for all of Switzerland, were too low for the Canton of Ticino. Second, there is a concentration of high income in Lugano. Fifteen respondents placed themselves in categories V and VI, as compared to seven in Locarno and six in Bellinzona. This was partially due to the size of the cities. For the two lowest categories the same observation may be made. Third, 17.2 % of the sample did not indicate their annual income, an indication that there is considerable reluctance on the part of the population to disclose their financial status. As could be ascertained during subsequent interviews, most respondents did not object so much against income categories as such but against defining each category in terms of certain amounts of Swiss francs. If the distribution of annual income is considered in terms of native language, Italian versus German, without regard to locality, the following table can be constructed: Table 7.

Income Distribution According to Language

Language

Cat. I

Cat. II

Cat. Ill

Cat. IV

Cat. V

Cat. VI

Italian German

3.1% 9.5%

6.3%

7.9% 6.3%

12.5% 6.3%

7.9% 4.7%

22.2% 9-5%

3-1%

The skewness towards the upper ranges remains but is more marked in the Italian sample. The German group is more evenly distributed along the scale. The relatively high number of German speakers in the lower categories is a further indication of the function of the Canton of Ticino as retirement haven for pensioners. They obtain their pensions (if they are Swiss citizens) from the government. 2 On the upper end of the income scale we find an equally high percentage of German speakers. This group consists promarily of businessmen and professionals, retired or not, who spend most of 2 Comparable to the Social Security system in the United States, the Swiss government has an obligatory pension plan (Alters- und Hinterbliebenen Versicherung, or AHV) for all its citizens.

33

RESULTS OF PART I OF THE QUESTIONNAIRE

their time in the Canton of Ticino, enjoying its climatic and geographical advantages. In this group one finds a large number of Germans who during the last twenty years have migrated into the area, attracted by its natural beauty and favorable tax system. This influx of wealthy Germans has taken on proportions which caused considerable concern on both the cantonal and federal levels of government. It resulted in a temporary building stop for most residential dwellings in the canton as well as additional restrictions and complications for those who wished to obtain permanent residence permits. This segment of the population may not have been very large numerically but it contributed considerably to the rapid economic development of the area during this period. Indications are that the 'boom' has levelled off. While it lasted, the Canton of Ticino profited economically although some people would be hesitant to admit this fact. The distribution of the sample according to profession is given in table 8. It is independent of the SES rating attributed to each category. Distribution according to the language of the questionnaire is included. Table 8. City Lugano Locarno Bellinzona

Distribution of the Sample According to Profession

Cat. I —

1.4% —

Cat. II

Cat. Ill

Cat. IV

Cat. V

Cat. VI

17.1% 4.2% 2.8%

2.8% 5-7% 5-7%

8.6%

11.4% 7.1% 5-7%

17.1% 4-2% 5.7%

— —

Italian German

1-4%

20.0% 4.2%

8.6% 5-7%

5-7% 2.8%

14.2% 10.0%

21-4% 5-7%

Total

1-4%

24.2%

14.3%

8.5%

24.2%

27.1%

—

Categories I and IV are not well represented. This may in part be due to the sampling process since there does not seem to be a shortage of persons working in arts and crafts. (Agriculture was also included in category I, but since the sampling was limited to the cities, it was later excluded from the results.) Some respondents considered this category so low in social prestige that they indicated their professional activities in a 'higher' category. A sculptor, for example, placed himself in category VI, which was intended for free professionals such as doctors, lawyers, etc. The apparent shortage of members of the restaurant and hotel branch is more difficult to explain. In many cases they might have considered themselves as belonging to a different category. The low number of German speakers in category II (officials on the local and cantonal level) is self-explanatory since officials are recruited from the local population, and foreign nationals are excluded. The small percentage represents Swiss Germans transferred to the Canton of Ticino.

34

RESULTS OF PART I OF THE QUESTIONNAIRE

Distribution of the sample according to educational attainment is presented in table 9. Again the distribution according to the language of the questionnaire is included. SES ratings are not taken into consideration. Table 9.

Distribution of the Sample According to Education

City

Cat. I

Cat. II

Cat. I l l

Cat. IV

Cat. V

Cat. VI

Lugano Locarno Bellinzona

5-2%

9.2%

3.9%

22.3% 10.5% 5-2%

7.9% 2.6% 3.9%

7.9% 2.6% 2.6%

7.9% 3.9% 2.6%

Italian German

2.6% 6.6%

22.3% 15.8%

6.6% 3.9%

11.8% 2.6%

9.2% 3.9%

11.8% 2.6%

Total

9.2%

38.1%

10.5%

14.4%

13.1%

14.4%

—

1.3% —

(In calculating the percentages, only the highest educational attainment was considered. Those who have attended university must necessarily have completed elementary school.) The table shows that there is a considerable break between categories II and III ; the former indicates the completion of secondary education, the latter possession of the certificate of maturity. In Switzerland as well as in the Canton of Ticino, school attendance is mandatory until the age of fourteen. 47.2 % of the sample has not continued its education beyond secondary school and 9.1 % has only elementary schooling. The 10.5 % which falls into category III have obtained their maturity certificates but did not attend university. Of those who did attend institutes of higher learning (none of which are located in the Canton of Ticino), 14.4% obtained higher degrees such as doctorates. Those who wish to pursue their education at universities must go to the German or French speaking parts of Switzerland or to Italy. According to counts made by the Statistical Yearbook (cf. bibliography) the former choice seems to be more popular since degrees from Italian universities are often not recognized in Switzerland. As regards the division of the sample according to the language of the questionnaires, the same proportion between Italian and German is found. The only exception occurs in Category I where more German speakers have attended only elementary school than Italian speakers. The division between mandatory and voluntary education falls between categories II and III.

4.4 LANGUAGES STUDIED IN SCHOOL

In addition to education as an index of socioeconomic status we attempted to deter-

35

RESULTS OF PART I OF THE QUESTIONNAIRE

mine which languages the respondents had studied in school and for how long. This did not play any part in the SES ratings but was listed as a separate item on the questionnaires. The results are presented numerically in table 10. Table 10. Foreign Languages Studied in School Italian Questionnaires / German Questionnaires Duration

German/ Italian

French

English

Spanish

Latin/ Greek

1 year 2 years 3 years 4 years 5 years 6 years 7 years over 7 years

1/3 5/3 8/2

V-

4/4 8/2 7/4 3/3 -/ι -/4 -/2

5/2

3/1 2/-

61-

11/3 3/i/2/1

1/1 9/2 3/2 6/1 4/2 5/6 16/4

-β - / -

i/-/2

-h -n 11-

-/I 2/10/2

More Italian speakers study German as a foreign language in school than speakers of German study Italian. Italian is not part of the regular curriculum in the schools of Germany. It is taught on a voluntary basis in the schools of the German speaking part of Switzerland. On the other hand, German is taught regularly in the schools of the Canton of Ticino. 75.5 % of all Italian speaking respondents have studied it for at least one year, but not more than five years. 34.6 % have studied it for more than five years. Only 14.8 % of the German speaking sample have studied Italian for more than five years, although 44.4 % have studied it for at least one year. This seems to corroborate the informal observations that, in general, native speakers of Italian are more willing to use German than vice versa. 95.9% of the Italian speaking group have studied French for at least one year, as compared to 66.6 % of the German speaking sample. In both Germany and Switzerland, French is an obligatory subject in the school system. Of the Italian sample 67.4% have studied the language five years or more, 32.6 % have studied French for more than seven years. This confirms the observation made before that French is generally used as the common means of communication between speakers of different languages in the Canton of Ticino. For English the situation is reversed. It is taught as an obligatory subject in most schools in Germany, but as an elective in most Swiss schools including those in the Canton of Ticino. We find that 74 % of the German group have studied English for at least one year, 25.9 % have studied it for five years or more. Comparing this with the Italian group, we find that 44.8 % have studied English for at least one year, but no one had studied the language more than five years. Spanish is studied only sporadically and usually on

36

RESULTS OF PART I OF THE QUESTIONNAIRE

a voluntary basis. One rarely finds this language in the regular curriculum in any school in Germany or Switzerland. It is most frequently taught in evening courses through adult education programs. 14.4 % of the German sample had studied Spanish, as compared to 12.2% of the Italian sample. The two classical languages, Latin and Greek, were grouped together since they are usually studied together. 26.5% of the Italian group had studied these languages for more than five years, as compared to 11.5 % of the German group, an indication that classical languages are still considered a necessary ingredient in the curriculum of the schools in the Canton of Ticino.

4.5 NATIVE LANGUAGES OF THE SAMPLE POPULATION

In determining the native languages of the sample the respondents were asked to indicate which language or dialect was most frequently used at home during the first five years of their life. The results are presented in table 11. Table 11. City

Distribution of the Sample According to Native Language

St. Italian

Ticinese

St. German Swiss German

Lugano Locarno Bellinzona

26.3% 6.6% 5-2%

9.2% 3.9% 9.2%

7.9% 2.6%

Total

38.1%

22.3%

10.5%

—

Others

14.4% 5-2% 3.9%

2.6% 2.6%

23.5%

5.2%

—

The languages listed under Others' were French, Swedish, Polish and Yiddish. It is doubtful that the high number of speakers of standard Italian corresponds to reality since respondents were often unaware of the differences between dialect and standard language. It is safe to assume that those who have standard Italian as their native language were equally exposed to Ticinese during the first five years of their lives. The dialect is still commonly used in informal settings (within the family, with friends and relatives). It is also true that the use of the standard language is increasing in those settings, especially in those cases where one parent speaks either a different language or a different regional variety of the dialect. On the other hand the figures indicate once more that Swiss Germans are very much aware that they do not speak the standard German of Germany. 4.6 SES AVERAGES OF THE SAMPLE POPULATION

The socioeconomic structure of the sample was determined by three indices, income,

RESULTS OF PART I OF THE QUESTIONNAIRE

37

profession and education. The average SES rating for the entire sample was found to be 3.6. Since all three scales went from one to six, the average should theoretically have been 3.0. The sample was skewed toward the lower end of the scale. It does not, however, invalidate the scales as such. If the sample is divided according to locality and language of the questionnaire, the following average SES ratings are obtained. Table 12.

Average SES Ratings of the Sample

Lugano : Overall rating : Locarno : Bellinzona: Overall rating: Italian: German:

3.6; Italian: 3.3; German: 4.0 3.3; Italian: 3.7; German: 3.0 3.8; Italian: 3.3; German: 4.4 3.4 3.8

Comparing the above figures it can be said that, on the whole, the German speaking sample ranks somewhat higher on the socioeconomic scale than the comparable Italian speaking sample. An exception is Locarno where the former group ranks a little higher. This city has a higher percentage of German speakers than the other two. For many years this has been where wealthy Germans and Swiss Germans spend their summers or retire. For Bellinzona the opposite is true. Most speakers of German are Swiss officials that have been transferred to this city. As regards the Italian group, Lugano and Bellinzona rank equally high on the socioeconomic scale, slightly above average. In the preceding pages we have discussed the socioeconomic structure of the sample in the three cities under investigation. We have used three indices to determine the SES rating of the respondents as well as of the entire sample. Some aspects which are pertinent to understand the sociolinguistic diversity of the Canton of Ticino have been described and analyzed. We have seen that the sample is quite evenly distributed in most parts of the questionnaire. In the following pages we will discuss some aspects of non-native language performance as manifested by the self-evaluation scores obtained from the respondents. Following the methods used above for the SES ratings, we will analyze some correlations between various language skills.

5 RESULTS OF PART II OF THE QUESTIONNAIRE

The second part of the questionnaire is concerned with the non-native language performance of the respondents. It was based on a self-evaluation scale, designed to help the individual establish a rating for each non-native language he knows in the four audiolingual skills. There were six different levels of performance, which were described briefly for each skill. The respondent then selected that level or rating which corresponded most closely to what he felt his performance was in each language by circling the appropriate number. For example, a respondent may select an oral comprehension rating of Ί ' (highest) for French and '6' (lowest) for English. The native language of the respondent was not rated. This shows not only the number of languages commonly spoken by the population but also the extent of their use. It serves as a measure of performance of all four audiolingual skills, and it highlights those aspects which dominate multilingual communication. Since the self-evaluation procedure is subjective in nature, the ratings obtained from each respondent cannot be taken as absolute values of non-native language performance in the sense of scores resulting from a standardized objective test. They must rather be taken as indications of that performance. (See Appendix for self-evaluation scales.) In developing the self-evaluation scales care was taken to describe each level of performance in such a way that it could be easily understood by linguistically unsophisticated subjects. This resulted at times in certain phrases which may sound odd to the language expert, especially in the English translation. The rating scales were structured so that they could accommodate the problems found with this particular population. If one were to make different investigations of multilingual situations, adjustments of the self-evaluation scales would be required. The self-evaluation scales proved to be an adequate means of determining the extent of multilingualism in the Canton of Ticino although some respondents felt that the descriptions of the various performance levels were too long and time-consuming, if one wanted to read them carefully.

5.1 SELF-EVALUATION SCALES FOR NON-NATIVE LANGUAGE PERFORMANCE

The results obtained from the self-evaluation scales are presented in the tables below. They were no longer classified according to locality since not much additional infor-

RESULTS OF PART II OF THE QUESTIONNAIRE

39

mation could be gained from that division. The following languages were mentioned in isolated instances and therefore not tabulated or statistically evaluated: Swedish, Polish, Arabic, Greek and Portuguese. For the sake of convenience the ratings obtained may be grouped in such a way that only three levels of performance are distinguished, so that (a) high level of performance comprises ratings Ί ' and '2', (b) medium level of performance comprises ratings '3' and '4' and (c) low level of performance comprises ratings '5' and '6'. The total percentages serve only as a quantitative index. (See Appendix III for Tables A and following.) 5.1.1

Oral Comprehension

We can say that 84.2 % of the entire sample have a high level of comprehension for French (89.3 % of the Italian speaking sample as compared to 75.8 % of the German speaking sample). Of the latter group, two-thirds indicated that their comprehension of Italian was on a high level; 53.0% of the Italian group chose that level for German as a foreign language. As regards English, the disparity between the two groups is even more marked ; only 8.1 % of the Italian sample evaluated their performance for that language on the high level, as compared to 59.2 % of the German speaking group. For Spanish the high level was selected by 6.1 % of the Italian sample but was not indicated by the German speaking group. On the medium level of performance we find the Italian sample represented with 28.5% for German as a foreign language. Of the corresponding German speaking group only 7.3 % indicated that level for Italian as a foreign language. Here the cantonal ratio of four Italian speakers for every German speaker is maintained. This is also true for the two groups as regards French. 12 % of the Italian sample chose the medium level for that language, as compared to 3.6% of the German sample. For English on this level, both groups rank about equally high in percentages (16.3 % vs. 11.1 %). The medium level for Spanish was selected by 11.1 % of the German group and by 8.1 % of the Italian counterpart. On the low level of performance, the percentages are small for the two primary languages, Italian and German, represented by 6.1 % of the Italian speaking sample and 3.6% of the German speaking group. Neither group indicated this level for French as a foreign language. For English, however, about twice as many Italian speakers indicated a low level of oral comprehension (30.6% vs. 14.7%). For Spanish it was low again, 8.1 % for the Italian group and 3.6% for the German group. It is clear that among the Italian speakers French ranks first in oral comprehension, followed by German and, to a lesser extent, English and Spanish. For the German speaking group French also ranks highest, but English and Italian follow closely. Spanish, on the other hand, is not well represented at all.

40 5.1.2

RESULTS OF PART II OF THE QUESTIONNAIRE

Speaking

The results obtained from the self-evaluation scales for speaking are summarized in a similar manner and presented in percentages in Appendix III, Table B. In comparing the results obtained from this rating with those for oral comprehension we can make a number of observations. First, there is little change for German as a foreign language, although some of the scores from the medium level of performance for comprehension have shifted to the low level for speaking. A similar shift is present for Italian as a foreign language where about 7 % of the high level for comprehension were redistributed on the medium and low levels for speaking. Second, for French we find that onefourth of the German speaking sample has selected the medium level for speaking. This constitutes an increase of some 20% over that level for oral comprehension, an indication that a larger percentage of the sample understands French much better than speaking it. Third, as regards English, we find no representation of the German sample on the medium level for oral comprehension but 25.9 % of this group indicated this level for speaking. This may be taken again as a downward shift in speaking. Most respondents, regardless of their native language, tend to evaluate their speaking performance much more conservatively than their comprehension. This is especially noticeable for the Italian speaking group as regards Spanish. 30.5 % indicated that they understood at least some Spanish, but only 14.2 % indicated that they spoke the language at all.

5.1.3

Reading

The results obtained from the self-evaluation scales for reading are again arranged in a table and presented in percentages (see Appendix III, Table C). On the high level of performance we find that about half of the Italian speaking sample reads German very well; the distribution of ratings Ί ' and '2' is about equal. 59.2% of the German speaking sample selected the high level as regards reading Italian as a foreign language. The percentages obtained for both groups for French are about the same (70.3% for the German sample, 69.4% for the Italian). The percentage of the German speaking group as regards the reading of French on the high level of performance has increased by 11.1% over the same group's speaking performance, but it is still lower (5.5 %) than the level indicated for oral comprehension. For the Italian speaking group we find a discrepancy of almost 20% between comprehension and reading of French, compared to a difference of 8.1 % between speaking and reading on the high level of performance. The differences between these three skills point towards a progressively more conservative evaluation. For English, this level is represented by 48.1 % of the German speaking group but by only 16.3 % of the Italian counterpart. Both percentages constitute a considerable increase over those obtained for speaking and, in the case of the Italian group, also over comprehension. On the medium level of performance we find the Italian sample represented by

RESULTS OF PART II OF THE QUESTIONNAIRE

41

24.4% for German and the German group by 14.7 % for Italian as a foreign language. As far as French is concerned, the differences between the two groups are less marked for reading than for speaking. The German sample decreased and the Italian sample increased its ratings. For English the ratings obtained from both groups are proportionately similar, showing an increase of about 4 % over speaking. This is somewhat striking because the medium level of comprehension was not represented by either group. For Spanish the medium level was selected for reading by 14.7 % of the German sample, indicating an increase of 11 % over speaking. The percentage of the Italian sample increased only by 1.3%. On the low level of performance we find many more Italian speakers as regards German than German speakers as regards Italian (18.3 % vs. 3.7 %). For French there is a proportionately similar rise of percentages in both groups as compared to speaking. For English, the Italian speaking group is again more heavily represented on the low level than its German speaking counterpart (20.4% vs. 7.4%). Both rate themselves lower in reading than in speaking. For Spanish we find 10.2% of the Italian sample on that level as compared to 3.7% of the German group. Comparing the total percentages obtained for reading with those for speaking, we see that in general more people read a given foreign language than speak it. This is especially important as regards French. More respondents of the German speaking sample read the language independent of the level of performance. On the other hand, more Italian speakers speak French. This indicates a more active use of this language by the Italian speaking segment of the population. For Spanish we find about twice as many German speakers reading the language than speaking it. The difference between speaking and reading Spanish of the Italian sample is slight (2 %). 5.1.4

Writing

The results obtained from the self-evaluation scales for writing are again arranged in a table and the results are summarized in Appendix III, Table D. On the high level of performance for writing we find that the percentages obtained for both language groups are comparable for German and Italian as foreign languages, the difference being less than 3 %. Comparing this with the previously tabulated language skill, reading, the situation is now reversed. German seems to be written more and better by the Italian speakers than Italian is by the German group. For French we find that the Italian speaking sample rates higher by about 8 % than the German speaking group. For English a much higher percentage of German speakers is represented on this level of performance (37.0 % vs. 4.3 % for the Italian group). For Spanish both samples were equally low. On the medium level of performance the percentages for German, Italian and French as foreign languages were proportionately similar to those on the previous level. The Italian sample ranks above the German speaking group in both languages. The situation is reversed for English and Spanish where the German speaking sample

42

RESULTS OF PART II OF THE QUESTIONNAIRE

is more strongly represented (33.3% vs. 22.4% for English and 18.5% vs. 4.3% for Spanish). On the low level of performance we find that 10.2% of the Italian speaking segment write little or no German as compared to 18.5% for the German speaking sample writing Italian. The same percentage of German speakers also writes little or no French. 11.1 % of the German group and 18.3 % of the Italian group write little or no English. On that level of performance we find that for Spanish the German group is not represented. The Italian sample is represented by 8.1 %. In comparing the results obtained for writing with those for reading one sees little or no difference between the self-evaluation ratings of these language skills for Italian and Spanish as foreign languages. For German, French and English as foreign languages both groups indicated somewhat lower percentages for writing than for reading. This means that fewer people write in the foreign language than read it, regardless of level of performance. For English we find a difference of 36% between the German and the Italian groups. If the four tables in Appendix III are combined, an overall ranking of non-native languages may be obtained. For this purpose the level of performance and the individual language skills are not considered. It simply serves as an indication of the extent of language use in a purely quantitative way. From this ranking it becomes clear that French most often serves as the common medium of communication in the Canton of Ticino among speakers of different languages. The German speaking sample showed an approximately equal knowledge of English and Italian (79.2% for English vs. 77.6% for Italian). The comparable Italian sample demonstrated a considerable difference between German and English (87.2% vs. 46.3 %). The fact that English is taught only as an elective during the last two years of high school accounts to some extent for the low percentage obtained by this group. French, on the other hand, is taught as the first foreign language in Switzerland and parts of Germany. In most German schools English is taught as either first or second foreign language. Italian and Spanish are generally not taught in Germany. Table 13 also shows that more than three-fourths of the German speaking sample is quadrilinguai to some extent (German, French, English and Italian) as compared to 90% of the Italian speaking sample which is at least trilingual (Italian, French, German). Table 13

Extent of Ν on-Native Language Use Indicated by the Self-Evaluation Scores

Italian speaking sample 0) (2)

(3) (4)

French : German : English : Spanish :

96.2% 87.2% 46.3% 19.5%

German speaking sample (1) (2)

(3) (4)

French : English : Italian : Spanish:

88.2% 79.2% 77.6% 15.0%

RESULTS OF PART II OF THE QUESTIONNAIRE

43

Considering the ratings obtained for the different language skills, we find that German as a foreign language is rated by about half of the sample on the high level of performance. For Italian as a foreign language the ratings are similarly distributed in all four language skills, and, with the exception of writing, all are on the high level of performance. We find that the low level is represented by less than ten percent of either sample for the respective languages. For French the distribution of ratings in all four language skills is, as indicated by the Italian speaking sample, comparable. The corresponding German speaking group indicated a very high percentage of oral comprehension on the high level of performance, but it is lower than that obtained for reading. In general the German speaking sample indicated higher ratings in the various language skills for English and the Italian speaking group for Spanish.

5.2 STATISTICAL EVALUATION OF PART II OF THE QUESTIONNAIRE

The statistical evaluation of the second part of the questionnaire, non-native language performance, follows closely the procedures used for the first part, SES rating of the sample population. The coefficients of reliability and validity were determined with the split-half method, using the 'Spearman-Brown' formula; the coefficient of correlation was calculated for two sets of language skills with the 'product-moment' method. Since French was found to be the non-native language most widely used in the Canton of Ticino, the ratings obtained from the sample for this language were employed for all calculations. Thus numerical stability in the sample is assured. Other non-native languages such as English or Spanish could have been used as well, but variations in the size of the sample would then make it difficult to compare the different parts of the questionnaire. The investigator was particularly interested to find out if there was a positive correlation between any two pairs of language skills and to determine the extent of this correlation. We decided to measure the correlation between : (a) (b)

oral comprehension and writing and speaking and reading. oral comprehension and speaking and reading and writing.

It is obvious that different combinations could have been selected, e.g. combining the two 'active' language skills, speaking and writing, with the two 'passive' ones, reading and oral comprehension. We doubt that major divergencies would have been found. It is important to point out that the findings obtained from these correlations cannot be generalized. They are limited not only to one language, French, but also to a specific geographical area. Any broader implications would have to be tested with different languages and preferably a larger sample. 5.2.1

Correlations of Two Sets of Skills and Reliability of Part II

For the statistical calculations of the two sets of skills (henceforth called correlation

44

RESULTS OF PART II OF THE QUESTIONNAIRE

(a) and correlation (b)) the raw data were arranged in four columns, one for each language skill. The ratings obtained from the self-evaluation scales for each respondent for French were entered in the appropriate column. The ratings were added and averaged out for the two pairs of language skills. The data were arranged in a scattergram, with the average ratings for one pair of language skills listed horizontally and the other vertically. Correction for the assumed mean, standard deviation and coefficient of correlation were calculated. Reliability and validity of each correlation (which constitutes a half-test) are determined. They are then combined to indicate reliability and validity for the entire part of the questionnaire. The calculations show that there is a high positive correlation between the two pairs of language skills, oral comprehension and speaking on the one hand and reading and writing on the other. The coefficient of correlation was r a = .77. The distribution of the ratings is fairly even with some concentration of scores on the high level of performance as might be expected for French. Similarly, after correcting for the assumed mean, calculating the standard deviation, the coefficient of correlation b was found to be rt> = .85. Taking the coefficients of reliability and validity for the two correlations (a) and (b) as half-measures for the second part of the questionnaire the 'Spearman-Brown' formula is applied. Reliability for correlation (a) r n a = . 87 Reliability for correlation (b) r U b = .91 Reliability r n ab = .94 Validity rcoau = .97 It is thus found that both indices are sufficiently high to make this part of the questionnaire a reasonably accurate measure of foreign language performance of the sample population of the Canton of Ticino. Since the coefficients of correlations were not only positive but also considerably high, other combinations would probably not differ greatly. Pairing 'active' versus 'passive' language skills might possibly show a higher correlation than what might be termed 'mixed' skills. No attempt was made to find that combination which would yield the highest correlation. Within the framework of this questionnaire, we only wanted to find out whether or not language skills can be said to be correlated. The findings show that a significant correlation exists between the language skills tested.

6 RESULTS O F PART III O F THE QUESTIONNAIRE

The third part of the questionnaire was designed to investigate some aspects of language attitudes of the sample population. We were interested to examine these attitudes not only for the standard languages but also for some of the dialects of the area. The respondents were asked first to rate their attitudes on a six-point scale towards the standard languages Italian and German and towards the dialects Ticinese and Swiss German. They were asked not to rate their native language but to consider it as a basis of evaluation in relation to the remaining three, and to be subjective in their ratings. The description of each rating was kept rather crude in order to make the individual think about his own position in relation to the languages and dialects around him. It may also seem that undue emphasis was placed on the professional aspect of language, but one must remember that most people tend to evaluate a foreign language in terms of its usefulness and not for its aesthetic value. 6.1

RATINGS OF LANGUAGE ATTITUDES

To obtain the ratings for each respondent, the scales were arranged according to the native language of each individual. The respondents checked their native language and then rated their attitudes towards the remaining three languages 1 below. This arrangement had the advantage of providing a second indication of the native language of the respondent. This brings to light some of the difficulties involved in selecting the native language. Frequently respondents would indicate "Ticinese" as their native language for the attitudinal evaluation after they had selected "standard Italian" in Part I of the questionnaire and vice versa. It seems unlikely that many respondents checked their earlier choice, much less corrected it. After having determined the native language, each respondent then indicated his attitude towards the other three languages by circling the appropriate number on the scale. The results obtained were examined and evaluated on a percentage basis and a prestige ranking for the various languages was established. The total size of the sample had diminished at this point to sixty-eight, as compared to seventy-six for the first two parts of the questionnaire. For the tabulation of the findings the rating scales were rearranged so 1

Here we include both standard and dialect varieties under the cover term 'language'. They are distinguished where necessary.

46

RESULTS OF PART III OF THE QUESTIONNAIRE

that rating Ί ' and '2' were taken to indicate a positive attitude on the part of the respondent, rating '3' and '4' an essentially neutral attitude, and ratings '5' and '6' a negative one. When the questionnaires were designed a six-point scale seemed preferable to give the respondents a wider choice of ratings. For the evaluation of the results a three-point scale was found to be sufficient. Table 14 presents the attitudes of the sample population towards non-native languages without regard to mother tongue. Table 14.

Attitudes of the Sample toward Non-Native Languages

St. Italian St. German Ticinese Swiss German

Positive

Neutral

93.4% 89.3% 24.4% 19.5%

4.4% 10.7% 43.9% 43.9%

Negative 2.2% —

31.7% 36.6%

The table shows not only the distribution of the percentages of the sample population but also gives the prestige rankings. The two standard languages rank very high and the two dialects relatively low on the scale. This may be taken as a confirmation of the observation that standard languages often are held in higher prestige by a given population than its dialects2 and that Swiss German is a low prestige dialect outside its own territory. The small percentage which indicated a negative attitude towards standard Italian is insignificant within the framework of multilingual communication in the Canton of Ticino. The sample was quite decisive in indicating its attitudes towards the languages and dialect with which it is in frequent contact. The results support the observation made on an informal basis that there is no social friction between the two principal language groups, German and Italian, since both hold the standard language of the other in very high esteem. The low regard for the respective dialects does not seem to modify the situation. Table 14 offers a summary of the results obtained from the attitudinal ratings of the sample population. Invariably, the standard languages are given higher ratings than the corresponding dialects. Ticinese is preferred by all groups over Swiss German, even though the latter is much more widely used throughout the country, giving it more significance as a means of everyday oral communication. (See Appendix III, Table E for details.) 6.2

L A N G U A G E USE IN INTERPERSONAL COMMUNICATION

The second question of this part of the questionnaire was designed to determine the 2

This situation is likely to be restricted to areas where the standard language is also an indigenous language. It probably won't apply to the same extent to standard languages based on a colonial language.

47

RESULTS OF PART III OF THE QUESTIONNAIRE

extent of use of the two standard languages and dialects in interpersonal communication (between the respondent and friends and acquaintances and vice versa). The respondents were asked to indicate the extent of usage on a three-point scale. They had an additional category at their disposition to accommodate those who used all languages and dialects listed. Since the ratings were kept global we found a substantial overlap between the categories. Multiple answers were also permitted so that the percentages rarely add up. After an initial evaluation it became clear that there was little or no difference in the direction of language use. It is therefore more convenient to present the results of both questions at the same time, by simply averaging the percentages. Table 15.

Extent of Language Use in Interpersonal Communication Rating:

St. Italian St. German Ticinese Swiss German Other Lang.

1 (few) 19.3% 46.6% 21.1% 25.9% 26.7%

2 (half) 32.2% 16.3% 9.9% 13.5% 4.0%

3 (most) 31.0% 14.3% 35.7% 19.0% -

Standard Italian is the language most widely used in the Canton of Ticino. For everyday communication the Italian speaking sample uses mostly Ticinese and standard Italian. The German speaking group uses Swiss German more than any other form of oral communication, Ticinese ranks second, standard German third and standard Italian fourth. The extensive use of Swiss German indicates that most speakers of German do not come from Germany but from Switzerland. A relatively high number of intermarriages between Ticinese speakers and speakers of Swiss German from other cantons account for the high percentage of Ticinese used by the German speaking sample. It is also interesting to point out that more than half of the Italian speaking sample uses languages other than those investigated here, most often French and English, at least occasionally. For the German speaking sample the percentage is only 5.5%. The relatively high percentage (35.1 %) for the limited use of standard German finds its corresponding slot in the table for both samples. The significance of this percentage is that many speakers of standard German use the language on a limited basis because their Italian speaking counterparts would rather use standard Italian. The German speaker, of course, is in the minority and thus he is expected to use the language of the majority more frequently than his own. The high regard which has been shown by the sample of both standard languages makes it clear that the terms 'minority' and 'majority' language can only have numerical implications.

48

RESULTS OF PART III OF THE QUESTIONNAIRE 6.3 READING AS AN INDICATION OF LANGUAGE ATTITUDES

The next questions concerned the reading habits of the sample population. The respondents were asked to indicate the approximate number of books read annually in the various languages by checking the appropriate rating for each language. Three quantitative distinctions were made: Rating Ί ' : up to five books annually in the language. Rating '2' : between five and ten books per year. Rating '3' : more than ten books per year. The results obtained from this question, together with those from the following ones, concerning the reading of news papers, magazines and journals, give an insight into the extent of reading in the languages under investigation. It is also assumed that the extent of reading in foreign languages is not only dictated by professional needs and the desire to be informed, but is to a considerable degree a manifestation of the attitudes towards the foreign languages in which one reads. Before discussing the results of these questions it should be pointed out that the people in the Canton of Ticino in general have free access to a large number of publications in many different languages. If one browses in any bookstore in Lugano, for example, one finds a very substantial selection of books in Italian, French, German, English and Spanish. This is also true for newspaper stands where large numbers of foreign language periodicals are available. Table 16 summarizes the results obtained for this question. Table 16.

Extent of Reading. Books Read per Year. Italian sample/German sample

German Italian French English Spanish

Rating: Τ 22.4%/29.6% 22.4%/29.6% 30.6%/25.9% 10.2%/22.3% 6 . 1 % / 7.4%

'2' 6-1%/ 7.4% 24.5%/11.1% 14.2%/11.1% -

/ 3.6%

—

/

—

'3' 6.1%/44.4% 38.7%/ 7.4% 8.1%/ / 11-4% / 3.6%

It seems from table 16 that both language groups read proportionately the same number of books annually in their respective native languages. Precise numerical indications are not possible since the three categories are not clearly limited. We can say that more than six percent of the Italian speaking sample reads more than ten books in German per year. We cannot say whether they read eleven books or twenty. Precise measures are not necessary since the investigator was primarily interested in determining a trend. As might be expected, the Italian speaking sample reads more French than its German speaking counterpart, which on the other hand reads more English books. Neither English nor Spanish is represented by the Italian sample on levels '2' and '3'. More than 80 % in each sample read at least five books in their native language per year. In addition to that the Italian sample is represented with 33% which read at least five books annually in German. More than half of the German

RESULTS OF PART III OF THE QUESTIONNAIRE

49

speaking sample reads that many Italian books over the same period of time. Disregarding languages and percentages, we can say that in general the German sample reads more than its Italian counterpart. The following questions were designed to show the extent to which daily newspapers are read by the sample population. The respondents indicated if they read a daily paper and, if so, in what language. The question did not require any rating, but only a 'yes'-'no' type of response. The results indicate that almost every Italian speaker reads at least one (Italian) paper a day. More than one-fourth of this group also reads a German and French paper. The German sample again favors English over French and more than half of this group reads two papers every day. As was pointed out before, the easy access to foreign language newspapers accounts in part for the high percentage of newspaper readers. The next question was designed to show the extent of reading weekly or monthly periodicals, especially illustrated magazines which are as popular in the Canton of Ticino as they are throughout Switzerland and the rest of Europe. We find a higher percentage of German speakers reading weekly or monthly magazines than daily newspapers. Both samples read more periodicals than dailies. The German sample reads somewhat more than the Italian counterpart (51.8% vs. 47.9%) in the three languages, Italian, German, French, and to alesser extent, English. In the last category, 'other languages', we find that Spanish is represented by some 6 % of the combined sample. The next question concerned the extent of reading technical or specialized periodicals. In this category we included any kind of periodical publication which is designed to provide its readers with specific information about a limited number of subjects. They do not have to appeal only to the educated reader. They may range from the stamp collector's journal to the bulletin of the Historical Society of the Canton of Ticino. We find that more than 50% of the combined sample reads a specialized journal. More than 40 % of the Italian sample reads them also in French and German. Of the German sample only about one-fourth reads them in Italian, but they read about three times as much in English. This is partially due to the fact that many more specialized periodicals are published in English than in Italian. As regards the results obtained from this series of questions, it is clear that the German sample in general reads more books and the Italian sample reads more papers, magazines or journals.

6.4

SUBJECTIVE ATTRIBUTES OF LANGUAGES

The next two questions are more directly related to language attitudes. The first concerns the professional usefulness of the languages and dialects involved, the second concerns the subjective 'beauty' of them. The term 'beauty' was left unspecified so that each respondent could answer according to his own interpretation. Multiple

50

RESULTS OF PART III OF THE QUESTIONNAIRE

answers were permitted and the results therefore do not add up to 100%. Table 17 summarizes the results of the first question concerning the professional usefulness of the various languages. To show the distribution more clearly, the combined sample was added. Table 17. Professional Usefulness of Languages

Italian sample German sample

Italian 61.2% 51.7%

German 83.6% 66.6%

French 73.4% 55.5%

English 42.8% 51.7%

Combined sample

56.4%

75.1%

64.4%

47.2%

German is generally considered to be the language which is professionally most important in the Canton of Ticino. The Italian sample ranks it above its own standard language, together with French. The same pattern is found for the German sample which ranks Italian on the same level as English. French is again confirmed as an intermediary form of communication between the various language groups in Switzerland. As regards German, the recent increases in the German speaking population in the Canton of Ticino and the resulting commercial expansion have undoubtedly contributed substantially to the high ranking of German. The next question provided the respondents with three spaces to indicate the languages which they considered to be most beautiful. Thus, multiple answers were permitted, but this option was not taken in many instances. In some cases respondents indicated that they considered all national languages of Switzerland to be equally beautiful. Table 18 summarizes the results. Table 18. Ranking of Languages According to 'Beauty' Italian sample German sample

Italian 54.1% 64.0%

German 37.5% 16.0%

French 75.0% 32.0%

English 20.8% 20.0%

Spanish 6.2% 8.0%

Combined sample

59.0%

26.7%

53.5%

20.4%

7.1%

One observation stands out very clearly : neither of the two groups considers its own language to be the most beautiful. The Italian speaking group selected French and the German speaking sample chose Italian. For the combined sample, however, Italian ranks first, followed by French, German and English. The percentages for the first two are as close as those obtained for the last two, about six percent in either case. Since the German speaking sample consisted of a large number of Swiss Germans, the preference given to French by this group is not surprising. The Swiss Germans, as a group, have traditionally been more inclined towards French and Italian than

51

RESULTS OF PART III OF THE QUESTIONNAIRE

towards standard German. Those that have come to the Canton of Ticino naturally will prefer Italian over French. It would be interesting to investigate the aversion which the Swiss Germans sometimes feel towards standard German within the framework of a sociolinguistic study. This aversion may, at times, result in almost hostile reactions on the part of speakers of Swiss German. Since this is not directly related to the present study, a few observations will have to suffice. It was observed in a number of instances that speakers of Swiss German would become evasive and claim not to understand when they were addressed in standard German, or a central or northern German dialiect. This was never encountered when the German speaker had a noticeable Austro-Bavarian (or Allemanic) accent. This has led some acquaintances of this investigator to use French or English rather than standard German when they are in the German speaking part of Switzerland. There is perhaps another aspect to this problem, namely that the speaker of Swiss German, when he is confronted with the need to communicate with a speaker of standard German, realizes the 'inadequacy' of his own brand of German. This might make him adhere to his native dialect to the extent that he becomes unwilling to even listen to the standard language, or he might wish to identify his own brand of German as being Swiss and therefore not be nationalistically involved with Germany. At times he fears that he might be ridiculed because of his 'inadequate' command of the more prominent phonological features of standard German, such as some suprasegmental features and some consonants. This fear is, of course, more often than not an imaginary one.

6.5 THE RELATION BETWEEN STANDARD LANGUAGE AND DIALECT

The next five questions concerned the relations between the standard languages and dialects involved and the extent of dialect use. In the first question of this group the respondents were asked to indicate whether they use more standard language or more dialect in everyday and professional communication, and to list the names of the dialects or standard languages. 'Everyday' communication is taken to be somewhat more informal than 'professional' communication, but both were equally inclusive. The results obtained from this question are presented in table 19. Table 19. Standard Languages and Dialects in Everyday Communication Extent of Use Standard Languages Italian German French

Dialects Swiss German Ticinese

Italian sample German sample

44.5% 29.6%

4.0% 33.3%

16.3% —

6.1% 40.7%

53.0% 18.5%

Combined sample

37.0%

18.6%

8.1%

23.4%

35.7%

52

RESULTS OF PART III OF THE QUESTIONNAIRE

We found that about 20 % of the entire sample gave multiple answers, that is, they could not decide whether they use more standard language or more dialect. The table shows that the combined sample uses both standard Italian and Ticinese to approximately the same extent. The Italian sample, of course, uses more its own dialect than its standard language. This is also true for the German speaking sample where the dialect is used more frequently. It was also found that the Italian speaking sample used more Swiss German than standard German (6.1 % vs. 4.0 %). The German sample, however, used more standard Italian than Ticinese (29.6% vs. 18.5%). There are 12.4% more German speakers who use Ticinese for everyday communication than there are Italian speakers using Swiss German. The latter group also indicated that they use about four times more French than German. Approximately 30% of the German sample use primarily the two standard languages, Italian and German. French was not represented for this sample. The relatively large percentage of German speakers who use Ticinese (18.5 %) is in part explained by the considerable number of intermarriages between speakers of Swiss German and Ticinese. 3 In many cases the family lives in the Canton of Ticino and the Swiss German partner naturally inclines towards the Ticinese dialect within the family. On the whole we found little difference between standard languages and dialects as regards the extent of usage. We cannot say, that one form of oral communication clearly dominates the other. If one were to establish a ranking for frequency of use, standard Italian would rank first, followed by Ticinese, Swiss German, standard German and French. 6.5.1

Complementary

Questions

The next two questions are complementary in their structure. The first determines the extent to which the predominant use of the dialect and at times incorrect use of the standard language are associated with a lack of formal education. The questions required a 'yes' or 'no' answer. This minimal choice placed the respondents in a position where they had only two alternatives, either one being somewhat extreme. The respondents might not be completely for or against either solution, but because the choice was limited they were forced to make a decision about their attitudes. It is doubtful that this approach is always desirable but if too many alternatives are permitted for this type of question, respondents will frequently hesitate to make a clear decision and then select a more vaguely phrased answer. This, in turn, may lead to misinterpretation of the results. A number of respondents added a category which may be labelled 'it depends'. To show the complete results of this question, this category is also presented. The percentages are too low to have any great effect on the overall results which are given in table 20.

' This statement is based largely on personal observation. The statistical sources consulted do not give specific information about intermarriages between members of different speech communities in Switzerland.

53

RESULTS OF PART III OF THE QUESTIONNAIRE

Table 20.

Are predominant use of the dialect and incorrect use of the standard language associated with a lack of formal education? 'Yes'

'No'

'it depends'

Italian sample German sample

61.5% 59.3%

32.4% 29.6%

6-1% 11.1%

Combined sample

60.4%

31.0%

8.6%

We find about two thirds of the combined sample consider persons who use mainly the dialect and who have incomplete control of the standard language as being less educated. Both subsamples indicated comparable percentages for either category. The German speaking group selected the third category more frequently than its Italian counterpart. Throughout the educational system of the Canton of Ticino the use of the standard language is greatly stressed. Students are not allowed to use the dialect in the classroom. Since the standard language is so strongly emphasized, it is, of course, associated with education and the use of the dialect, in many instances, with a lack of it. As was mentioned before, the use of the dialect is limited primarily to informal situations ; it is almost never used in formal situations such as the courtroom, the church, or even in business meetings. The predominant use of the dialect, especially when combined with insufficient control of the standard language, indicates clearly a lack of education in the eyes of the sample population. The next question served as a control question. It determined whether or not the respondents consider persons who use mostly the standard language, even in informal communication with close friends or relatives, as being conceited or snobbish. One would expect that those who answered the previous question positively, would answer this one negatively and vice versa. The results confirm this assumption. The Italian sample indicated its attitudes in this question somewhat more decisively than in the previous one. Both sub-samples selected the third category less frequently than in the previous question. We found that about one fourth of the German speaking sample indicated that they consider persons who speak mainly the standard language to be somewhat conceited. It is difficult to establish precise percentages, but we assume that the majority of this group are speakers of Swiss German. Table 21 summarizes the results. Table 21.

Is predominant use of the standard language associated with being conceited or snobbish?

Italian sample German sample Combined sample

'Yes'

'No'

'it depends'

16.3% 25.9% 31.2%

77.6% 66.6% 72.1%

6.1% 7.5% 6.8%

54

RESULTS OF PART III OF THE QUESTIONNAIRE

As Charles A. Ferguson (1959) has pointed out, in a diglossie situation such as the one in the German speaking part of Switzerland, one frequently finds that the substandard variety, in this case Swiss German, is given values which are normally characteristic for the standard variety. This eliminates to some extent the need of a standard language for oral communication, which in turn makes those who do use the standard form of the language look conceited, as wanting 'to put on airs'. It is one of the criteria of a diglossie situation that the substandard variety of the language is used even in those formal situations which in a 'straight' dialect situation would require the use of the standard language. This constitutes one of the most important differences between the sociolinguistic structures of the Canton of Ticino and the German speaking part of Switzerland. Swiss German is used in most formal situations (with the exception of newscasts and stage productions) whereas Ticinese is used only in those social situations which are clearly marked as to their informality. The difference in attitudes between the two sets of speakers were convincingly shown by the results of the last two questions. The social situation was, of course, limited to informal communication, where the Swiss German speaker would almost never use the standard variety and the Ticinese speaker would be unlikely to use standard Italian but would rarely object to its use by others. 6.5.2 Learning Non-native Language Dialects The next two questions determined the willingness of the individual to learn a dialect of a non-native language, if it would facilitate his being accepted in a new environment. even though it might be an additional learning burden. The following questions were again control questions. It asked the respondents whether or not they consider the learning of foreign language dialects as an essentially useless endeavor. Again either an affirmative or a negative answer was expected but a third category was added for the tabulation of the results. The percentages are presented in table 22. Table 22.

Would you be willing to learn a foreign language dialect?

Italian sample German sample Combined sample

'Yes'

'No'

'it depends'

63.2% 71.4% 67.3%

34.7% 23.8% 29.3%

2.1% 4.8% 3-4%

We find that the German speaking sample indicated its willingness to a somewhat greater degree than the Italian speaking group. More than two thirds of the combined sample answered this question affirmatively. The control question was designed to either confirm or reject this trend. This question was left unanswered by about 6.5% of the sample population. Table 23 presents the results.

RESULTS OF PART III OF THE QUESTIONNAIRE

Table 23.

55

Do you consider the learning of a foreign language as useless?

Italian sample German sample Combined sample

'Yes' 37.0% 32.0% 34.5%

'No' 63.0% 64.0% 63.5%

'it depends' —

4.0% 2.0%

The trend found in the results of the previous question is clearly confirmed by the percentages in this table. Almost two thirds of the combined sample do not consider the learning of foreign language dialects as useless. The German speaking sample indicated its positive attitudes somewhat more decisively. It is evident that both language samples consider the dialect as a useful and important part of language which facilitates communication on certain levels. This mutual respect for the dialect of the other language group is in part an indication of the lack of social friction between the various linguistic segments of the Canton of Ticino. It can be seen now that willingness of the individual to accept the dialects of his environment in a foreign language situation on the same level as his native dialect and not as an inferior manifestation of language are very much part of positive language attitudes. This may well be a precondition for the stable coexistence of different language groups in a given sociocultural environment. This precondition is filled in the Canton of Ticino. 6.5.3

Relative

'Difficulty'

Associated

with

Dialects

The following question is related to the two previous ones. It determines which dialect the respondents consider to be more difficult to learn, Swiss German for speakers of standard German or Ticinese for speakers of standard Italian. The questionnaire again allowed only two choices. After initial analysis it was found that many respondents had indicated two additional categories : (a) both dialects are about equally difficult to learn, (b) they were undecided. These two categories were then incorporated in the results, which are presented in table 24. Table 24. Which dialect is considered more difficult to learn by speakers of the standard language? Swiss German for German speakers

Ticinese for Italian speakers

Both about equal

Italian sample German sample

65.1% 66.6%

16.3% 8.3%

6.8% 25.0%

11.8%

Combined sample

65.9%

12.3%

15.9%

5.9%

Undecided

—

56

RESULTS OF PART III OF THE QUESTIONNAIRE

As can be seen the majority of the sample population considers Swiss German to be more difficult to learn for speakers of standard German than Ticinese is for speakers of standard Italian. Both subsamples responded with comparable percentages but one fourth of the German speaking group indicated that the two dialects were equally difficult to learn. About twice as many Italian speakers than German speakers indicated that Ticinese is the more difficult dialect to learn. The results of this question may be taken as an indication of the relative 'distance' between the dialect and the corresponding standard language as it is interpreted by the linguistically unsophisticated population. The present sample finds less linguistic relation between standard German and Swiss German than between Ticinese and standard Italian. Although there are as yet no studies relating the two dialects with their standard languages on a formal basis, it is this investigator's impression that both are about equidistant from their respective standard languages.

6.6 STANDARD LANGUAGES IN RESTRICTED SITUATIONS

The remaining two questions determined the extent of use of the standard languages in two restricted situations: (a) religious functions and (b) private correspondence. Both are considered formal situations. One involves mainly oral use of the language, the other its written use. The results are presented in table 25. Multiple answers were not excluded and therefore the percentages do not add up to 100%. Table 25.

Languages Used in Religious Functions

Italian

German

French

Latin

Italian sample German sample

55.1% 33.3%

6-1% 55.5%

8-1% 11.1%

28.5%

Combined sample

44.2%

30.8%

9.6%

14.2%

—

Hebrew —

Other Languages —

3.6%

11.1%

1.8%

5.5%

Even though the Canton of Ticino, like the rest of Switzerland, does not have an 'official' state religion, the native population is predominantly Catholic. The German speaking sample is mostly Protestant. As the results indicate, one third of this group attends religious services conducted in Italian, only six percent of the Italian speaking group attends services in German. The majority of either sample attends services which are conducted in their respective native languages. The small percentage found under 'Other languages' consists mostly of English speakers. (There is one Anglican church in Lugano and a Methodist chapel in Locarno.) For the last question the respondents were asked to indicate if they always wrote in the standard language, if they used some dialect words in their private correspondence,

RESULTS OF PART III OF THE QUESTIONNAIRE

57

if they wrote at times in the dialect or if they always used their dialect. Since there is not much written material to be found in either the Swiss German dialects or Ticinese, we assumed that neither was used much in private correspondence. The results bear this out. The two categories, indicating occasional and frequent use of the dialects, were not found at all. It is safe to say that the use of either dialect in correspondence is negligible. The small percentage (5.1 %) which does employ some dialect words presumably does so with a special connotation intended only for the receiver of the letter. We can exclude this form of communication from any sociolinguistic considerations in the Canton of Ticino.

6.7 STATISTICAL EVALUATION OF PART III OF THE QUESTIONNAIRE

The statistical evaluation of the final part of the questionnaire follows closely the methods employed for the two previous parts, socioeconomic status and non-native language performance. This assures not only consistency of the results but also provides the best basis for comparing the three major parts of the questionnaire. The coefficient of reliability and validity was determined by the 'split-half' method using the 'Spearman-Brown' formula. The coefficient of correlation was calculated with the 'product-moment' method. For this part of the questionnaire, language attitudes, two separate measures for reliability, validity and correlation were taken. First, it was attempted to determine if there was a correlation between the attitudinal ratings of the two standard languages and their corresponding dialects and the first nine questions which followed (extent of oral communication, extent of reading, professional importance of languages and dialects as well as the 'beauty' of the languages involved). If one wishes to label these two aspects one might call the first 'pure' attitudinal and the second 'applied' attitudinal aspects. The latter are taken as overt manifestations of the former. We decided to use only the first nine questions because the remaining ones required a 'yes' or 'no' response which does not lend itself easily to the statistical analysis selected. To obtain the raw data for this correlation (which we will label 'Correlation A') the following steps were taken: Since the 'pure' attitudinal ratings were obtained from a six-point scale, on which the respondents evaluated their attitudes towards the three language forms, excluding his native language, we arbitrarily assigned a numerical value to each of them, so that '4' corresponded to standard Italian, '3' to standard German, '2' to Ticinese and Ί ' to Swiss German. Using the same numerical values the average for the nine questions of the 'applied' attitudinal aspects was calculated. This resulted in two sets of scores and constituted the raw data for Correlation A. The second correlation, labelled 'Correlation B', is essentially a check of the internal consistency of 'applied' attitudinal part. In both cases, numbers are assigned to languages or dialects. When the averages did not result in even numbers, e.g. a rating average of 3.5 for the 'pure' attitudinal part, it

58

RESULTS OF PART III OF THE QUESTIONNAIRE

means that this particular respondent rated his attitudes as being equally divided between standard German ('3') and standard Italian ('4') without giving explicit preference to either language. The following illustrates the problem: For Correlation A we find that No. 29A has a score of 3.5 in the column 'Item' which means that he gave equally high ratings to both standard languages. In the column 'Check of Item' we find a score of 3.40. This indicates that this respondent uses mainly Italian but considers German to be the moie important language, as well as the most beautiful one. These numerical values are best taken as indications of a trend and not so much as precise measures. The raw data obtained in this manner were again arranged in a scattergram. The values for 'Check of Item' are listed on the horizontal axis, for 'Item' on the vertical axis. (See Appendix for calculations.) There is a positive and significant correlation between the 'pure' and 'applied' aspects of language attitudes in the Canton of Ticino. The coefficient of correlation is (Γ)Α = .67. The measures for reliability and validity are indicative only for this half of the part under investigation. The distribution of the data indicates the extensive use of standard Italian by both subsamples. A large portion of the scores is found between values 3.00 and 4.00. The trend to use the standard language is clearly also manifested by the distribution of the scores. The second correlation, Correlation B, is taken as a measure of the internal consistency of the 'applied' attitudinal ratings of the last part of the questionnaire. As was pointed out above, only the first nine questions were considered for these calculations, since the last six questions were structured differently, which would involve a different statistical analysis. After assigning numerical values to the languages and dialects, the scores were arranged into 'odd' and 'even' groups and subsequently averaged out, so that two sets of scores remained. These constitute the raw data for 'Correlation B'. The coefficient of correlation for the 'odd' and 'even' items of the 'applied' attitudinal aspects was found to be positive and substantially marked (ΓΒ = .51), although lower than that of Correlation A. The reliability and validity for this half-measure were also found to be somewhat lower. The distribution of the data is, compared to the previous arrangement, less even but it clearly indicates the trend towards standard Italian. The coefficient of correlation between the two sets of scores was lower than that of the previous correlation, but still significant. In order to obtain higher values for this part, it would have to be restructured and, perhaps, lengthened, so that there would be more specific questions about the extent of interpersonal communication and a clearer separation between this and the extent of reading of the sample population. Additional subjective aspects of language attitudes such as 'beauty' and 'professional importance' of languages and dialects might be amplified by more detailed questions. The reliability and validity of the attitudinal part of the questionnaire is calculated with the 'Spearman-Brown' formula. The coefficients of reliability and validity of Correlations A and Β function as half-measures.

RESULTS OF PART III OF THE QUESTIONNAIRE

59

Reliability for Correlation A: r u = .80 Reliability for Correlation Β : r22 = .67 Reliability for Attitudinal Part of Questionnaire (A and Β) : Γ]/2 I/2 = .85 Validity for Attitudinal Part of Questionnaire (A and B) : r œ = .92 The calculations presented indicate that the third part of the questionnaire, language attitudes of the sample population, is a reasonably accurate testing device. The last two questions regarding the use of languages and dialects in religious services and private correspondence were found to be irrelevant to the sociolinguistic structure of the Canton of Ticino. We can conclude that the questionnaire, although it is far from flawless, did show the most significant aspects of the sociolinguistic structure of the area under investigation.

7 RESULTS OF THE CORRELATION BETWEEN THE THREE PARTS OF THE QUESTIONNAIRE

As a final measure, it was attempted to show whether there was a correlation between the three major parts of the questionnaire and the extent of this correlation. The questions to be answered are : (a) Is there any correlation between the socioeconomic status and the non-native language performance of a multilingual population, and, if so, what is it? (b) Is there any correlation between the socioeconomic status of the population and its language attitudes, and what is its extent? (c) Is there any correlation between non-native language performance and language attitudes in our multilingual population and what is its extent? The calculations were based on the same methodological considerations as for the previous correlations. 7.1

CORRELATION BETWEEN SES AND NON-NATIVE LANGUAGE PERFORMANCE

For the correlation between SES and non-native language performance the raw data were taken from the overall SES ratings obtained for each individual respondent and the ratings of the self-evaluation scales for French, with all audio-lingual skills averaged out. In this way, one SES score and one non-native language performance score was obtained for each respondent. It was found that there is a positively marked correlation between socioeconomic status and non-native language performance of the sample. The coefficient, r = .34, however, is not substantial. It is safe to say that there is no significant relationship between the two aspects under investigation. 7.2 CORRELATION BETWEEN SES AND LANGUAGE ATTITUDES

For the correlation between SES and language attitudes of the sample population the raw data were obtained first from the average SES rating of each respondent and second from the average of the combined 'pure' and 'applied' ratings of the attitudinal part of the questionnaire.

RESULTS OF THE CORRELATION BETWEEN THE THREE PARTS OF THE QUESTIONNAIRE

61

The computations indicate that there is a positive correlation between SES and language attitudes of the sample population (r = .04). Although it is positive, the correlation is so slight as to be negligible. We can say, therefore, that there is no correlation between the two sets of data. The distribution of the scores in the scattergram (see Appendix) shows considerable concentration in the upper third, ranging through most of the spectrum of SES ratings. This is a further indication of the importance assigned by the sample population to the two standard languages, Italian and German, which were given the numerical values of '4' and '3' respectively. Ticinese is also quite evenly distributed along the scale with little or no concentration in any specific SES group. Swiss German, on the other hand, is hardly represented at all. The largely negative attitudes of the sample are indicated in the scattergram by the absence of scores below 1.76; Swiss German was given a numerical value of Ί ' .

7.3 CORRELATION BETWEEN NON-NATIVE LANGUAGE PERFORMANCE AND LANGUAGE ATTITUDES

As a final measure the correlation between non-native language performance and language attitudes was calculated. The raw data were obtained in the same manner as for the two previous correlations. For non-native language performance the average ratings of the audio-lingual skill for French were arranged horizontally and for language attitudes the average for 'pure' and 'applied' ratings were arranged vertically in a scattergram. Since there was a difference in the size of the sample, the smaller number (N = 68) was taken. We find that there is a positive correlation between non-native language performance and language attitudes for the sample (r = .10). The coefficient, however, is so small that the correlation becomes insignificant. It is, thus, safe to say that there is no relationship between the various aspects of language performance and attitudes. As in the previous two correlations, there is a concentration of scores in the upper third of the scattergram indicating a strong emphasis on the two standard languages. It extends about equally throughout the performance ratings which do not go lower than 4.50. The attitudinal ratings do not go lower than 1.76, indicating that Swiss German is represented only occasionally.

7.4 DISCUSSION OF RESULTS

From the discussion above it becomes clear that there is no significant correlation between SES, non-native language performance and language attitudes of the sample population. Neither of these aspects is, thus, dependent on the other. Within the framework of this experiment we cannot say, for example, that persons ranking high on the SES scale would rank equally high on the self-evaluation scales for non-native

6 2 RESULTS OF THE CORRELATION BETWEEN THE THREE PARTS OF THE QUESTIONNAIRE

languages, i.e. have a better command of various non-native languages. Comparing the three correlations, we find that the first one, between SES and language performance, has the highest coefficient (r = .34). This is taken to mean that there is some degree of relationship between these two aspects, even if it is small. Those persons who had ranked themselves high on the educational index for the SES ratings were, of course, more exposed to non-native languages in the schools and, thus, had a higher rating of their performance than those who had less formal education. We may also assume that those respondents, through their high standing in the community, are probably more often in contact with speakers of other languages. The virtual absence of a correlation between language performance and language attitudes speaks to some extent for the linguistic sophistication of the sample population. We found that the attitudes towards the language of one's environment were independent of income, education or profession as well as of the ability to use these languages. We also found that the use of the dialects, Ticinese as well as Swiss German, is in no way limited to certain socioeconomic strata. This is in opposition to one of the findings reported by Labov (1966) in his widely acclaimed study of the social stratification of English in New York City.

8

CONCLUSION

It was found that the three areas of sociolinguistic investigation, socioeconomic status, non-native language performance and language attitudes of the sample population, were not significantly correlated. They are largely independent of each other and each one, by itself, constitutes a valuable and reasonably efficient means to determine the sociolinguistic structure of a multilingual population. A set of data was obtained from which the sociolinguistic picture of the area could be shown conclusively. The investigator believes that the questionnaire developed for this limited study is essentially a very useful testing device. Improvements to be made are mostly concerned with the 'external' aspects of the questionnaire, such as mode of distribution, arranging and structuring of questions, as well as gathering and collating of the data. We have shown that the questionnaire is internally valid and reliable even though there is no significant correlation between its major constituent parts. There is no doubt that a wider application of this type of questionnaire is not only feasible but also highly desirable for the study of multilingualism and its attitudinal manifestations. A condensed format could be designed which would be applicable to the study of any given multilingual area which is clearly definable.

REFERENCES

Chomsky, Noam, Aspects of the Theory of Syntax (M.I.T. Press, Cambridge, Mass., 1965). Cohen, Marcel, Pour une Sociologie du langage (Albin Michel, Paris, 1956). Cole, G. D. H., Essays in Social Theory (Macmillan, London, 1956). Denison, Norman, "Some Observations on Language Variety and Plurilingualism", A.S.A Conference, Sussex, England (1969, mimeo). DiPietro, R. J., "Multilingualism in St. Croix" (1968, mimeo). —, "Contrastive Analysis and the Notion of Deep and Surface Grammar", Georgetown University Series on Languages and Linguistics 21 (1968). Ferguson, Charles Α., "Diglossia", Word 15, 325-241 (1959). —, "The Problems of Language Teaching with Diglossia", Georgetown University Series on Languages and Linguistics 15 (1962). —, "The Language Factor in National Development", in: F. A. Rice, ed., Study of the Role of Second Language Learning in Asia, Africa and Latin America (C.A.L., Washington, 1962). Fishman, Joshua Α., Language Loyalty in the United States; the Maintenance and Perpetuation of Non-English Mothertongues by American Ethnic and Religous groups (Mouton, The Hague, 1966). —, ed., Readings in the Sociology of Language (Mouton, The Hague, 1968). Garrett, Henry E., Statistics in Psychology and Education (D. McKay, New York, 1966). Garvin, Paul L., "The Standard Language Problem", in Garvin, On Linguistic Method (Mouton, The Hague, 1964). Gumperz, J. J. and Dell Hymes, The Ethnography of Communication (American Anthropological Association, Menasha, Wise., 1964). Hymes, Dell, Studies in Southwestern Ethnolinguistics (= Studies in General Anthropology) (Mouton, The Hague, 1967). Jaberg, Karl and Jakob Jud, Sprach- und Sachenatlas Italiens und der Südschweiz (Zofingen, Switzerland, 1928-1940). Keller, Oscar, "Die Mundarten des Sottoceneri (Tessin), dargestellt an Hand von Paralleltexten", Revue de Linguistique Romane X (1934). —, Beiträge zur Tessines Dialektologie, Vol. I, Die Mundart von Rovio (Lugano), Vol. II, Wörterbuch der Mundart von Val Verzasca (Librairie E. Droz, Paris, 1937). —, "Dialekttexte aus dem Sopraceneri (Tessin)", Zeitschrift für Romanische Philologie 61, 256-318 (1941); and ZRPH 63, 23-122 (1943). —, Die präalpinen Mundarten des Altoluganese (Winterthur, 1943). Kloss, Heinz, Die Entwicklung neuer germanischer Kultursprachen von 1800 bis 1905 (Pohl Verlag, München, 1952). Labov, William, The Social Stratification of English in New York City (Center for Applied Linguistics, Washington, 1966). Lado, Robert, Language Testing (McGraw-Hill, New York, 1961). LePage, Robert, The National Language Question (Oxford Univ. Press, New York, 1966). Macnamara, John, ed., "Problems of Bilingualism", Journal of Social Issues 23 (2) 1967. Moulton, William, "What Standard for Diglossia? The Case of German Switzerland", Georgetown University Series on Languages and Linguistics 15 (1962). Peal, Elizabeth and Wallace Lambert, "The Relation of Bilingualism to Intelligence", Psychological Monographs 6 (27) (1962) (whole No. 546).

REFERENCES

65

Rubin, Joan, "Bilingualism in Paraguay", Anthrop. Linguistics IV (1) (1962). Schönbach, Peter, Sprache und Attitüden (Verlag Hans Huber, Bern, 1970). Sganzini, Silvio, Elio Ghirlanda, and Federico Spiess, Vocabolario dei dialetti della Svizzera italiana, fase. 1-6 (1952-1961); fase. 7, 218-328 (1962) (Tipografia Commerciale, Lugano, Switzerland). Stewart, William, "An Outline of Linguistic Typology for Describing Multilingualism", in : F. A. Rice, ed., Study of the Role of Second Language Learning in Asia, Africa and Latin America (C.A.L. Washington, 1962). Ufficio Cantonale di Statistica, Annuario Statistico del Cantone Ticino, 1966/1967, and La Popolazione del Cantone Ticino nel 1967 (Bellinzona, Switzerland, 1968). Vildomec, Verboj, Multilingualism (A. W. Sijthoff, Leyden, 1963). Warner, William L., Social Class in America, a Manual of the Procedure of the Measurement of Social Status, with a New Essay 'Theory\ and a Method for the Comparative Study of Social Stratification (Harper, New York, 1960). Weber, Max, Economy and Society; an Outline of Interpretive Sociology, Giitnher Roth and Claus Wittich, eds. (Bedminster Press, New York, 1968). Weiss, Robert S., Statistics in Social Research (J. Wiley and Sons, New York, 1968). Weinrich, Uriel, Languages in Contact, 3rd ed. (Mouton, The Hague, 1964). Wölk, Wolfgang, "Language Attitude Tests; Speaker Reactions in a Bilingual Situation" (mimeo).

APPENDICES

I.

SUMMARY OF STATISTICAL CALCULATIONS

1

Reliability of the Entire Questionnaire: r n = .94 1.1 Reliability of Part I (SES of Sample) : r u = .79 1.2 Reliability of Part II (Non-Native Language Performance): r u = .94 1.3 Reliability of Part III (Language Attitudes): r n = .85

2

Validity of the entire Questionnaire: r œ = .97 2.1 Validity of Part I (SES of Sample) : r„ = .88 2.2 Validity of Part II (Non-native Language Performance) : r^ = .97 2.3 Validity of Part III (Language Attitudes): r œ = .92

3

Correlations 3.1 (Part I) Income and Education of Sample : r = .46 3.2 (Part I) Income and Profession of Sample: r = .59 3.3 (Part I) Profession and Education of Sample: r = .60 3.4 (Part II) Í Oral Comprehension 1 and j Speaking 1 _ ^ [Writing j [Reading J 3.5 (Part II) {Oral Comprehension! and {Reading ì f _ g^ {speaking J [Writing J 3.6 (Part III) 'Pure' and 'Applied' Attitude Ratings: r = .67 3.7 (Part III) Odd and Even 'Applied' Attitude Ratings: r = .51 (internal consistency check) 3.8 (Parts I and II) SES and Lang. Performance of Sample: r = .34 3.9 (Parts I and III) SES and Lang. Attitudes of Sample: r = .04 3.10 (Parts II and III) Lang. Performance and Lang. Attit. : r = .10 II. PRINCIPAL STATISTICAL FORMULAE Sfx

1. Mean: M = — Ν

APPENDICES

2. Standard Deviation: SD =

3. Chi-square: X2 =

(fo

—

f)2

/ Sfx2 " Ν , where f 0 = observed frequency of occurrence and f = expected frequency of occurrence

4. Reliability (Spearman-Brown Formula): r u =

Modified Spearman-Brown Formula: r l u

2rV«7i 1+r

WiïU l+^/.Va1/.

5. Validity: τχ = Vrn 6. Calculation for the Coefficient of Correlation fx' (a) Correction for the Assumed Mean : c x = — ; Ν

fy' cy = — . Ν

/fx 2 (b) Standard Deviation (sigma) : σ χ = \J — — c x 2 . (i) ;

/fy 2

(c) Product-Moment Formula (Coefficient of Correlation) :

r=

Sx'y' Ν

CxCy

GxGy

(1 —r 2 ) (d) Correction for Standard Errors : SEr = ——— V JN (for .95 and .99 confidence intervals).

68

APPENDICES III.

Table A.

SUPPLEMENTARY TABLES

Non-Native Language Performance of the Sample: Oral Comprehension

Italian Sample

FL German

FL French

FL English

FL Spanish

Performance Level High Medium Low

53.0% 28.5% 6.1%

89.3% 10.2%

8.1%

Total

87.6%

99.5%

38.7%

6.1% 16.3% 8-1% 30.5%

FL Italian

FL French

FL English

FL Spanish

High Medium Low

66.6% 7.3% 3.6%

75.8% 3.6%

59.2%

Total

77.5%

79.4%

German Sample

Table B. Italian Sample

—

—

—

30.6%

—

14.7%

11.1% 8.1%

73.9%

19.2%

—

Non-Native Language Performance of the Sample: Speaking FL German

FL French

FL English

FL Spanish

High Medium Low

53.0% 24.4% 10.2%

77.3% 18.3% 2.0%

8.1% 10.2% 32.6%

2.0% 2-0% 10.2%

Total

87.6%

97.6%

50.9%

14.2%

FL English

FL Spanish

37.0% 25.9% 14.7%

3-7% 3-7% 3-7%

77.6%

11.1%

Performance Level

German Sample

FL Italian

FL French

High Medium Low

59.2% 11.1% 7.4%

59.2% 25.9%

Total

77.7%

85.1%

—

APPENDICES

Table C.

69

Non-Native Language Performance of the Sample: Reading FL German

FL French

FL English

FL Spanish

High Medium Low

47.1% 24.4% 18.3%

69.4% 20.4% 6.1%

16.3% 14.2% 20.4%

4.3% 4.3% 8-1%

Total

89.8%

95.9%

50.9%

16.7%

FL Italian

FL French

FL English

FL Spanish

High Medium Low

59.2% 14.7% 3.7%

70.3% 22.2% 3.7%

48.1% 29.6% 7-4%

3-7% 4-7% 3.7%

Total

77.6%

96.2%

85.1%

12.1%

Italian Sample Performance Level

German Sample Performance Level

Table D.

Non-Native Language Performance of the Sample: Writing FL German

FL French

FL English

FL Spanish

High Medium Low

47.1% 26.5% 10.2%

63.2% 26.5% 2-1%

4.3% 22.4% 18.3%

4.3% 4.3% 8-1%

Total

83.8%

91.8%

45.0%

16.7%

FL Italian

FL French

FL English

FL Spanish

High Medium Low

44.4% 14.7% 18.5%

55.5% 18.5% 18.5%

37.0% 33.3% 11.1%

3-7% 18.5%

Total

77.6%

92.5%

81.4%

22.2%

Italian Sample Performance Level

German Sample

—

70

APPENDICES

Table E. Attitudes of the Sample toward Non-native Languages and Dialects ( expanded) Rating

1

2

3

4

5

6 (low)

Nat. Lang. Std. Italian Attitude toward Std. German Swiss German Ticinese

25.8% 6.3%

12.7% 4.2% 2.2%

2.2%

-

8.5% 12.7%

2.2% 4.2%

2.2% 8.5%

4.2% 4.2%

Nat. Lang. Std. German Attitude toward Std. Italian Swiss German Ticinese

23.8% 9.6% 9.6%

4.7% 4.7% -

— 9.6% -

4.7% 4.7% 9.6%

4.7%

4.7% 9.6%

Nat. Lang. Swiss German Attitude toward Std. Italian Std. German Ticinese

27.2% 18.2% 9.1%

2.3% 11.3% -

2.3% 9.1%

-

2.3%

-

9.1%

4.5%

4.5%

Nat. Lang. Ticinese Attitude toward Std. Italian Std. German Swiss German

29.0 10.2% 2.9%

2.9% 17.4% 1.4%

— 7.2% 10.2%

— — 2.9%

— — 8.7%

— — 7.2%

Table F. Reading of Periodicals (ia) Daily Newspaper

in

Italian

German

French

English Other Lang.

Italian Sample German Sample

95.9% 62.9%

28.5% 74.0%

26.5% 11.1%

2.0% 14.8%

11.1%

(b) Weekly Papers Italian Sample German Sample

93.8% 55.5%

36.7% 85.1%

55.1% 29.6%

6.1% 37.0%

2.2% 11.1%

(c) Specialized Periodicals Italian Sample 55.1% German Sample 25.9%

40.8% 51.8%

44.9% 11.1%

6.1% 18.5%

2.0%

—

—

APPENDICES

Table G.

Distribution Curve for Normality of Sampling

71

IV QUESTIONNAIRE SAMPLES GERMAN AND ITALIAN

FRAGEBOGEN ZUR MEHRSPRACHIGKEIT IM KANTON TESSIN I.

AUSKUNFT ÜBER IHRE PERSON (wird vertraulich behandelt)

Name : Alter : Jahre Adresse : Geburtsort : Land : Wielange wohnen Sie im Tessin: Jahre. 1. Welches war der Hauptgrund, weshalb Sie ins Tessin kamen? (Bitte zutreffendes ankreuzen.) a. berufliche Gründe: b. finanzielle Gründe: c. Steuervorteile: e. andere Gründe: d. Klima und Landschaft: 2. In welche der folgenden Gruppen würden Sie sich auf Grund Ihres durchschnittlichen Jahreseinkommens einstufen? (Bitte eines ankreuzen.) a. Gruppe I (bis 8000 sfr.) d. Gruppe IV (bis 20000 sfr.) e. Gruppe V (bis 24000 sfr.) b. G r u p p e l l (bis 12000 sfr.) c. Gruppe III (bis 16000 sfr.) f. Gruppe VI (über 24000 sfr.) 3. Welche der folgenden Berufsgruppen trifft für Sie am besten zu? a. Landwirtschaft (auch Viehzucht oder Weinbau): b. Hotel- und Gaststättengewerbe: c. Handwerkliche Berufe: d. Beamten (kantonale und staatliche): e. Lehrberufe: f. Industriearbeiter : g. Freie, selbständige Berufe : II. SCHULBILDUNG (Bitte zutreffendes ankreuzen oder ausfüllen)

1. Abgeschlossene Grundschulbildung? Ja: Nein: 2. Abgeschlossene Sekundär (Oberschulbildung? Ja: Nein: 3. Matura, Abitur oder ähnlicher Abschluss? Ja: Nein: Wenn Matura, welcher Typ? A: B: C: H:

APPENDICES

73

4. Hochschulbindung? Ja: Nein: W e n n ' J a ' , a. Universität: Wieviele Semester: b. Techn. Hochschule: Wieviele Semester: c. Handelshochschule: Wieviele Semester: d. Andere Hochschule: Wieviele Semester: 5. Hochschulabschluss? Ja: Nein: W e n n ' J a ' , a. Doktor: b. Diplom: c. Staatsexamen: d. Andere: 6. Welche Fremdsprachen haben Sie während Ihrer Schulzeit gelernt? (Bitte ankreuzen.) a. Klassische Sprachen (Latein, Griechisch) : Wieviele Jahre : b. Deutsch: Wieviele Jahre: c. Englisch: Wieviele Jahre : d. Französisch: Wieviele Jahre : e. Italienisch: Wieviele Jahre : f. Spanisch: Wieviele Jahre : g. AndereSprachen: Welche: Wieviele Jahre : Wieviele Jahre : 7. Welche Sprache wurde während der ersten fünf Jahre Ihres Lebens in Ihrem Elternhaus am häufigsten gesprochen? 8. Welcher Dialekt Ihrer Muttersprache ist Ihnen am geläufigsten? (In welcher Gegend sind Sie aufgewachsen, z.B. Kanton Bern, Sizilien, Oberbayern) III. BEWERTUNG IHRER SPRACHKENNTNISSE

Bewerten Sie bitte Ihre Sprachkenntnisse mit Hilfe der folgenden Skalen. Lesen Sie bitte die einzelnen Beschreibungen sorgfältig durch. Tragen Sie dann in den einzelnen Tabellen die Ihnen bekannten Fremdsprachen mit der zutreffenden Bewertung ein, z.B. ' Verstehen'. Fremdsprache: English. Bewertung: '5'. A. VERSTEHEN

1. Sie haben keine, oder nur geringe Schwierigkeiten die Sprache zu verstehen, auch wenn sie rasch gesprochen wird. Vorträge oder künstlerische Darbietungen (Filme, Theater) stellen keine sprachlichen Probleme dar. Sie fühlen sich mit der Sprache vertraut. 2. Sie verstehen den grössten Teil des Gesprochenen. Bei raschen Sprechern verlieren Sie manchmal den Faden und bitten um Wiederholung. Ebenso bei schwierigen Anweisungen und Erklärungen. Dialekte der Fremdsprache verstehen Sie nicht oder nur sehr wenig. 3. Sie verstehen etwa die Hälfte des Gesagten bei normalen alltäglichen Unterhaltungen, können aber Erklärungen und Anweisungen nur mit Schwierigkeiten folgen, auch wenn der Sprecher sich langsam und deutlich ausdrückt.

74

APPENDICES

4. Sie verstehen weniger als die Hälfte im normalen Gebrauch. Sie müssen sich sehr auf das Gesagte konzentrieren. Gespräche über Ihnen wenig bekannte Themen bleiben Ihnen meist unverständlich. 5. Sie verstehen nur einen Bruchteil des Gesagten, meist nur einzelne Wörter und Ausdrücke. Die Sprache ist Ihnen recht fremd. 6. Sie verstehen fast garnichts und suchen die fremde Sprache wenn möglich zu vermeiden. TABELLE A. VERSTEHEN (Bitte tragen Sie die Ihnen bekannten Fremdsprachen ein und kreuzen Sie die zutreffende Bewertungszahl an)

Fremdsprache: Fremdsprache :.. Fremdsprache :. Fremdsprache :.. Fremdsprache :..

4 4 4 4 4 B. SPRECHEN (Anweisungen wie oben)

1. Sie beherrschen die Aussprache der fremden Sprache völlig, Sie werden deshalb gelobt. Man versteht Sie leicht. Sie fühlen sich in der Sprache ganz zuhause. 2. Sie haben manchmal bei langen und komplizierten Wörtern Ausspracheschwierigkeiten. Der Akzent Ihrer Muttersprache ist wahrnehmbar. 3. Manchmal fehlt Ihnen in der fremden Sprache ein Wort, an das Sie nur in Ihrer Muttersprache denken können. Das Aussprechen von bestimmten Lauten (Vokalen und Konsonanten) macht Ihnen regelmässig Schwierigkeiten. Sie versuchen längere Sätze zu vermeiden. 4. An alltäglichen Unterhaltungen nehmen Sie teil, können sich aber nicht immer leicht verständlich machen. Sie werden auf Grund Ihres Akzentes schwer verstanden. 5. Sie verwenden nur kurze einfache Sätze. Sie finden die Aussprache der fremden Sprache kompliziert und fallen deshalb oft auf die Aussprache Ihrer Muttersprache zurück. Sie versuchen, wenn immer möglich, Ihre Muttersprache zu verwenden. 6. Sie beherrschen die Fremdsprache nur in einzelnen Wörtern und Ausdrücken. Sie fühlen sich darin fremd und vermeiden sie nach Möglichkeit.

75

APPENDICES TABELLE Β. SPRECHEN (Bitte tragen Sie die Ihnen bekannten Fremdsprachen ein und kreuzen Sie die zutreffende Bewertungszahl an)

Fremdsprache :.. Fremdsprache :.. Fremdsprache:. Fremdsprache : Fremdsprache:

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

C. LESEN (Anweisungen wie oben)

1. Sie lesen selbst schwierige Texte mühelos und rasch, auch wenn sie nicht in Ihr Interessengebiet fallen. Witze und Anekdoten begreifen Sie sofort, auch wenn es dabei um Feinheiten der Sprache geht. Kreuzworträtsel können Sie vom sprachlichen Standpunkt lösen. 2. Sie lesen Material von durchschnittlicher Schwierigkeit das in Ihr Interessengebiet fällt. Bei Gedichten und stylistisch kompliziertem Material benötigen Sie machmal eine Erklärung oder Übersetzung. 3. Sie lesen meist nur Zeitschriften etc. schrecken aber vor Leitartikeln und schwierigen Abschnitten nicht zurück. Manchmal fragen Sie oder verwenden ein Wörterbuch, wenn Sie ein Wort nicht kennen. Sie lesen recht langsam. 4. Sie lesen nur langsam und zögernd. Sie müssten häufig ein Wörterbuch verwenden, machen sich aber nur selten die Mühe. Von langen Artikeln und Büchern nehmen Sie Abstand. Das Ausfüllen von Formularen macht gelegentlich Schwierigkeiten. 5. Sie lesen nur wenig in der fremden Sprache. Sie ziehen Texte mit Bildern vor, da Sie so die Zusammenhänge leichter verstehen. Sie versuchen, sich das Material, wenn möglich, übersetzen zu lassen. 6. Sie lesen so wenig wie möglich in der fremden Sprache. Texte lesen Sie Wort für Wort, zögernd und langsam. Strassen- und Reklameschilder bleiben Ihnen oft unverständlich. TABELLE C. LESEN (Bitte tragen Sie die Ihnen bekannten Fremdsprachen ein und kreuzen Sie die zutreffende Bewertungszahl an)

Fremdsprache :.. Fremdsprache: Fremdsprache: Fremdsprache:. Fremdsprache:.

2 2 2 2 2

4 4 4 4 4

76

APPENDICES D. SCHREIBEN (Anweisungen wie oben)

1. Sie schreiben mühelos in der fremden Sprache über Ihnen bekannte Themen. Sie können stylistische Feinheiten ausnützen und z.B. Humor wiedergeben. 2. Sie schreiben nur beschränkt in der Fremdsprache, meist Privat- oder Geschäftskorrespondenz, die Sie allerdings mühelos meistern. Sie beherrschen die gängigen Formulierungen und machen keine Grammatik- oder Rechtschreibfehler. 3. Das Schreiben in der Fremsprache macht Ihnen Schwierigkeiten. Grammatikfehler kommen vor. Sie müssen öfters nachsehen, wie ein Wort geschrieben wird und was es bedeutet. 4. Das Schreiben in der Sprache macht Ihnen recht viel Kopfzerbrechen. Sie setzen Briefe erst in Ihrer Muttersprache auf und übersetzen sie dann in die Fremdsprache. Selbst dann machen Sie noch viele Fehler. 5. Es fällt Ihnen schwer in der Fremdsprache selbst einfache Briefe zu schreiben. Sie überlassen das am liebsten anderen. 6. Sie schreiben überhaupt nichts in der fremden Sprache. Sie haben es nie gelernt und haben auch kein Bedürfnis es zu erlernen. TABELLE D. SCHREIBEN (Bitte tragen Sie die Ihnen bekannten Fremdsprachen ein und kreuzen Sie die zutreffende Bewertungszahlen)

2 2 2 2 2

Fremdsprache:. Fremdsprache:. Fremdsprache: Fremdsprache: Fremdsprache : IV.

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

IHRE BEZIEHUNG Z U SPRACHEN

Bewerten Sie bitte mit Hilfe der folgenden Skala Ihre Beziehung, Ihr Verhältnis zu den folgenden Sprachen (oder Dialekten), nicht aber zu Ihrer Muttersprache. Sie können dabei ganz subjektiv sein. Bitte kreuzen Sie die zutreffende Zahl an. Skala: Ί' '2' '3' '4' '5' '6'

= = = = = =

Sehr gut. Gut. Mittelmässig. Eher gleichgültig. Recht schlecht. Sehr schlecht.

Die Sprache ist nützlich, schön und wertvoll. Die Sprache ist beruflich wichtig und hat literarischen Wert. Die Sprache ist hauptsächlich beruflich wichtig. Die Sprache ist weder sehr nützlich, noch besonders schön. Ein notwendiges Übel, ohne nennenswerten Vorteile. Ich würde die Sprache nicht lernen. Finde sie hässlich.

APPENDICES

77

Α. Ihre Muttersprache ist Hochdeutsch Ihr Verhältnis zu a. Schweizerdeutsch ist b. Italienisch ist c. Tessiner Dialekt ist

1 1 1

2 2 2

3 3 3

4 4 4

5 5 5

6 6 6

B. Ihre Muttersprache ist Schweizerdeutsch Ihr Verhältnis zu a. Hochdeutsch ist b. Italienisch ist c. Tessiner Dialekt ist

1 1 1

2 2 2

3 3 3

4 4 4

5 5

6 6 6

C. Ihre Muttersprache ist Italienisch Ihr Verhältnis zu a. Hochdeutsch ist b. Schweizerdeutsch ist c. Tessiner Dialekt ist

1 1 1

2 2 2

3 3 3

4 4 4

5 5 5

6 6 6

D. Ihre Muttersprache ist Tessiner Dialekt Ihr Verhältnis zu a. Hochdeutsch ist b. Schweizerdeutsch ist c. Italienisch ist

1 1 1

2 2 2

3 3 3

4 4 4

5 5 5

6 6 6

V. Bitte verwenden Sie für die ersten drei der folgenden Fragen die vorgeschlagenen Skalen. Für die restlichen Fragen tragen Sie bitte das Zutreffende ein oder kreuzen Sie es an.

1. Wieviele Ihrer Bekannten und Freunde sprechen mit Ihnen ? (Skale: Ί ' = wenige. '2' = etwa die Hälfte. '3' = die meisten). a. Hochdeutsch: 1 2 3 b . Schweizerdeutsch: 1 2 3 c. Italienisch : 1 2 3 c . Tessiner Dialekt : 1 2 3 d. Andere : 1 2 3 2. Im Umgang mit Bekannten und Freunden sprechen Sie? (Skala wie oben), a. Hochdeutsch: 1 2 3 b . Schweizerdeutsch: 1 2 3 c. Italienisch: 1 2 3 d. Tessiner Dialekt : 1 2 3 d. Andere: 1 2

3

3. Wieviele Bücher lesen Sie im Jahr auf? (Skala: Ί ' = keine. '2' = etwa fünf. '3' = etwa zehn. '4' = mehr als zehn), a. Deutsch: 1 2 3 4 b. Italienisch: 1 2 3 4 c. Englisch: 1 2 3 4 d. Französisch: 1 2 3 4 e. Spanisch: 1 2 3 4 f. Andere: 1 2 3 4 4. Lesen Sie eine Tageszeitung? W e n n ' J a ' a u f a. Deutsch: d. Englisch:

Ja:

Nein: b. Italienisch: e. Spanisch:

c. Französisch: f. Andere:

5. Lesen Sie eine oder mehrere Wochen- oder Monatszeitschrift? Ja: Nein: W e n n ' J a ' a u f a. Deutsch: b. Italienisch: c. Französisch: d. Englisch: e. Spanisch: f. Andere:

78

APPENDICES

6. Lesen Sie eine oder mehrere Fachzeitschriften? Ja: W e n n ' J a ' a u f a. Deutsch: b. Italienisch: d. Englisch: e. Spanisch:

Nein: c. Französisch: f. Andere:

7. Welche Sprachen sind Ihrer Ansicht nach in Ihrem Beruf am wichtigsten? a b c. 8. Welche der Ihnen bekannten Sprachen empfinden Sie als die Schönsten? a. b. c. 9. Sprechen Sie in Ihrem Berufs- oder Alltagsleben mehr Dialekt oder mehr Hochsprache? a. Mehr Dialekt: Welchen: b. Mehr Hochsprache : Welchc : 10 Halten Sie Menschen, die ihre Hochsprache nur unvollkommen beherrschen, für weniger gebildet ? Ja : Nein : 11. Halten Sie Menschen, die selbst im Umgang mit Freunden immer die Hochsprache sprechen, für eingebildet oder hochnäsig? Ja: Nein: 12. Angenommen Sie sind Ausländer, würden Sie sich bemühen den Dialekt Ihrer neuen Umgebung zu erlernen um leichter akzeptiert zu werden und Anschluss zu finden, auch wenn dies für Sie sehr schwierig sein sollte? Ja: Nein: 13. Halten Sie das Erlernen fremdsprachlicher Dialekte grundsätzlich für Ja: Nein :

unnötig?

14. Welcher Dialekt ist Ihrer Ansicht nach schwieriger zu erlernen ? a. Schweizerdeutsch für einen hochdeutsch Sprechenden: b. Der Tessiner Dialekt für einen italienisch Sprechenden: 15. Welche Sprache(n) werden bei der Ausübung Ihrer Religion gesprochen? a. b. c. 16. In Ihrer Privatkorrespondenz verwenden Sie? a. immer die Hochsprache: b. meistens Dialekt: c. wenig Dialekt: d. nur einzelne Dialektausdrücke : P.S. Würden Sie sich gegebenenfalls zu einer kurzen Besprechung bereitfinden, bei der auf die einzelnen Fragen genauer eingegangen werden könnte? Wenn ja, geben Sie bitte an dieser Stelle Ort und Zeit (vor dem 21. Juli) an, die für Sie am angenehmsten sind. Mit bestem Dank für Ihre Mitarbeit.

Name, Adresse, Telefon: Tel.: Datum und Zeit : Ort:

Uhr

79

APPENDICES

QUESTIONARIO SUL PLURILINGUISMO NEL CANTON TICINO I. INFORMAZIONI PERSONALI (Saranno tenute in stretta confidenza)

Nome: Indirizzo: Luogo di nascitá: Professione: Da quanti anni abita nel Ticino:

Età:

anni.

Stato: anni.

1. Quale é stata la ragione principale della Sua venuta nel Ticino? (Scelga una voce sola). a. ragioni professionali : b. ragioni finanziarie : c. vantaggi fiscali : d. clima, paessaggio : e. altre ragioni : 2. A quale dei gruppi seguenti Lei pensa di appartenere in base al Suo reddito annuale? a. b. c. d. e. f. 3. Di a. b. c. d. e. f. g.

Categoria I Categoriali Categoria III Categoria IV Caterogia V Categoria VI

(fino a (fino a (fino a (fino a (fino a (più di

8000 sfr.): 12000 sfr.): 16000 sfr.) : 20000 sfr.): 24000 sfr.) : 24000 sfr.) :

quale dei seguenti gruppi professionali fa parte? agricoltura (compresi allevamento e viticoltura): attività alberghiere : artigianato: funzionario (statale o cantonale): insegnamento: industria: libera professione : II.

ISTRUZIONE

1. Istruzione elementare completa? Si: No: 2. Istruzione secondaria completa? Si: No: 3. Maturità? Si: No: In caso di risposta affermativa, quale tipo? A: Β 4. Istruzione universitaria? Si: No: In caso di risposta affermativa a. Università: b. Politecnico: c. Handelshochschule: d. Altri:

C: Quanti Quanti Quanti Quanti

H: anni: anni: anni : anni:

80

APPENDICES

5. Laurea? Si: No: Quale? a. Dottorato:

b. Diploma:

c. Altre:

6. Corsi di lingue: Quali lingue estere ha studiato in scuola? Per quanto tempo? Quanti anni? a. Lingue classische (latino, greco) : b. Tedesco : Quanti anni? c. Inglese: Quanti anni? d. Francese: Quanti anni? e. Italiano: Quanti anni? f. Spagnolo : Quanti anni? g. Altre lingue : Quanti anni? Quanti anni? 7. Quale é stata la lingua pili usata in famiglia durante i Suoi primi cinque anni?

8. Quale dialetto della Sua lingua materna conosce meglio, cioè in quale regione é cresciuto, p.es. Sicilia, Canton Ticino?

IH. AUTOVALUTAZIONE DELLE SUE CONOSCENZE LINGUISTICHE

Favorisca valutare la Sua conoscenza delle lingue per mezzo della scala seguente. Legga con cura ogni descrizione e per ogni lingua estera contrassegni il numero corrispondente in ogni delle quattro tavole, p.es. 'Comprensione'. Lingua estera: Inglese. Voto: 3. A.

COMPRENSIONE

1. Ha pochissima o nessuna difficoltà nel comprendere la lingua estera anche se parlata rapidamente. Conferenze, rappresentazioni (cinema, teatro) non offrono difficoltà linguistiche. 2. Comprende la maggior parte di ció che Le viene detto. Ascoltando una conversazione rapida, qualche volta Lei perde il filo e deve chiedere che Le si ripete qualcosa. Ció Le capita anche quando riceve istruzioni e spiegazioni difficili. Comprende poco o nulla dei dialetti delle lingue straniere considerate. 3. Comprende circa la metà di una normale conversazione, ma riesce solo con difficoltà a comprendere istruzioni e spiegazioni anche se chi parla, si esprime lentamente e chiaramente. 4. Comprende meno della metà di una normale conversazione. Deve concentrarsi molto. Conversazioni su soggetti con i quali Lei non é familiare, Le riescono linguisticamente incomprensibili.

81

APPENDICES

5. Comprende solo una frazione di ció che si dice, per lo più solo singole espressioni e parole. Si sente piuttosto estraneo alla lingua. 6. Comprende quasi niente e cerca di evitare questa lingua. TAVOLA A. COMPRENSIONE (Scriva il nome di ogni lingua e contrassegni il numero corrispondente)

Lingua Lingua Lingua Lingua Lingua

estera : estera : estera: estera : estera:

1 1 1 1 1

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

B. LINGUA PARLATA (intruzioni come sopra)

1. Lei é facilmente compreso e lodato per la Sua pronuncia. Si trova completamente a Suo agio in questa lingua. 2. A volte ha qualche difficoltà nel pronunciare parole lunghe e complicate. L'accento della Sua lingua materna é evidente. 3. Qualche volta Le manca una parola nella lingua straniera e può pensarla solo nella propria lingua. Ha regolarmente difficoltà nel pronunciare certi suoni (consonanti e vocali). Cerca sempre di evitare lunghe frasi. 4. Prende parte a normali conversazioni ma non sempre può comunicare con facilità. A causa del Suo accento ha qualche difficoltà nel farsi capire. 5. Usa solo frasi corte e semplici. Trova difficile la pronuncia della lingua straniera e perciò ritorna alla pronuncia della Sua lingua materna cercando di usarla il più possibile. 6. Conosce solo singole espressioni e parole nella lingua straniera. Si sente a disagio in essa e la evita se possibile. TAVOLA B. LINGUA PARLATA (Scriva il nome di ogni lingua estera e contrassegni il numero corrispondente)

Lingua Lingua Lingua Lingua Lingua

estera : estera: estera: estera : estera :

1 1 1 1 1

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

c . LETTURA (istruzioni come sopra)

1. Legge facilmente e rapidamente testi difficili anche se non sono nella area della Sua specialità. Non ha difficoltà nel comprendere barzelette e aneddoti anche se ciò

82

APPENDICES

richiede la conoscenza di certe sottigliezze del linguaggio. Da un punto di vista linguistico é capace di resolvere parole incrociate ecc. 2. Legge testi di difficoltà media nell'area dei Suoi interessi. Nel campo della poesia e di materiale linguisticamente complicato Le occorre una spiegazione o una traduzione. 3. Legge soprattutto giornali e riviste, ma non evita articoli di fondo e periodi difficili. Talvolta usa un dizionario o chiede ad altri il significato di parole che non conosce. Legge piuttosto lentamente. 4. Legge testi difficili lentamente e con esitazione. Dovrebbe fare frequente uso di un dizionario, ma lo usa raramente. Evita lunghi articoli e libri. Trova difficoltà nel compilare moduli e questionari nella lingua estera. 5. Legge poco nella lingua estera. Preferisce testi con illustrazioni che ne facilitino la comprensione. Cerca di far tradurre il testo. 6. Legge il meno possibile nella lingua straniera. Legge parola per parola e molto lentamente. Cartelli stradali e pubblicitari Le sono spesso incomprensibili. TAVOLA C. LETTURA (Scriva il nome di ogni lingua estera e contrassegni il numero corrispondente)

Lingua Lingua Lingua Lingua Lingua

estera : estera : estera: estera : estera :

1 1 1 1 1

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

D. LINGUA SCRITTA (Istruzioni come sopra)

1. Scrive con facilità nella lingua straniera su soggetti che Le sono familiari. Sa impiegare raffinatezze stilistiche e far uso di umorismo. 2. Il Suo uso della lingua straniera scritta é limitato principalmente alla corrispondenza privata o d'affari, con la quale non ha difficoltà. É padrone delle comuni formule epistolari e non commette errori di grammatica o di ortografia. 3. Ha qualche difficoltà nello scrivere la lingua straniera. Commette errori di grammatica. Deve spesso controllare l'ortografia e il significato delle parole. 4. Scrivere nella lingua straniera é per Lei un problema. Di solito scrive prima una brutta copia nella Sua lingua materna e poi la traduce nella lingua estera. Anche procedendo cosi, restano molti errori. 5. Lei scrive il meno possibile nella lingua straniera. Non ha mai imparato scrivere in essa e non lo vuole imparare. 6. Lei non scrive affatto nella lingua straniera.

83

APPENDICES TAVOLA D. LINGUA SCRITTA (Scriva il nome di ogni lingua estera e contrassegni il numero corrispondente)

Lingua Lingua Lingua Lingua Lingua

estera: estera : estera: estera: estera :

1 1 1 1 1

2 2

3 3 3 3 3

2 2 2

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

IV. ATTEGGIAMENTO VERSO LE L I N G U E

Favorisca valutare il Suo atteggiamento verso le seguenti lingue (o dialetti) o le Sue relazioni con esse - ad esclusione della Sua lingua materna - mediante la scala seguente. Può essere completamente soggettivo. Contrassegni il numero che Le pare corrispondere meglio alla Sua valutazione. Scala : Ί ' = ottima. La lingua é utile e bella. Ha valore letterario. '2' = buona. Professionalmente importante. Una apprezzabile letteratura. '3' = media. Importanza soprattutto professionale. '4' = piuttosto indifferente. Non é specialmente utile né bella. '5' = piuttosto cattiva. Senza particolari vantaggi. '6' = pessima. La lingua non é bella, non Le importa di impararla. A. La Sua lingua materna e il tedesco Il Suo atteggiamento verso a. il tedesco svizzero é b. l'italiano é c. il ticinese é B. La Sua lingua materna é il tedesco svizzero Il Suo atteggiamento verso a. il tedesco é b. l'italiano é c. il ticinese é C. La Sua lingua materna é Vitaliano Il Suo atteggiamento verso a. il tedesco é b. il tedesco svizzero é c. il ticinese é D. La Sua lingua materna é il ticinese Il Suo atteggiamento verso a. il tedesco é b. il tedesco svizzero é c. l'italiano é

1

2 1 2 1 2

3 3 3

4 4 4

5 5 5

6 6 6

1 2 1 2 1 2

3 3 3

4 4 4

5 5 5

6 6 6

1

1 2 2 1 2

3 3 3

4 4 4

5 5 5

6 6 6

1

1 2 2 1 2

3 3 3

4 4 4

5 5 5

6 6 6

84

APPENDICES

V. Favorisca rispondere alle prime tre domande mediante le scale proposte.

1. Quanti dei Suoi amici e conoscenti parlano con Lei? (Scala: Ί ' = pochi. '2' = circa metà. '3' = la maggior parte), a. tedesco: 1 2 3 b. tedesco svizzero : 1 2 3 c. italiano: 1 2 3 c. ticinese: 12 3 d. altre lingue: 1 2 3 2. Con amici e conoscenti parla Lei? a. tedesco : 1 2 3 b. tedesco svizzero : 1 2 3 c. italiano: 1 2 3 c. ticinese: 12 3

d. altre lingue: 1 2

3

3. Quanti libri legge all'anno? (Scala: Ί ' = nessuno. ' 2 ' = circa cinque. ' 3 ' = circa dieci. '4' = più di dieci), a. intedesco: 1 2 3 4 b. in italiano: 12 3 4 c. in francese: 1 2 3 4 d. in inglese: 12 3 4 e. in altre lingue : 1 2 3 4 d. in spagnolo : 1 2 3 4 4. Legge un giornale quotidiano? Si: In caso affermativo, in a. tedesco: d. inglese:

No: b. italiano: e. spagnolo:

c. francese: f. altre lingue:

5. Legge una o più riviste settimanali o mensili? Si: In caso affermativo, in a. tedesco: b. italiano: d. inglese: e. spagnolo:

No: c. francese: f. altre lingue:

6. Legge una o piti riviste specializzate? In caso affermativo, in a. tedesco : d. inglese:

Si:

No: b. italiano : e. spagnolo:

c. francese : f. altre lingue:

7. Quali lingue considera più importanti nella Sua professione? a. b c. 8. Tra le lingue che conosce quali considera le più belle? a. b.

c.

9. Nella Sua attività professionale o nella vita d'ogni giorno parla più in dialetto o più in lingua? a. più in dialetto : Quale: b. più in lingua: Quale: 10. Considera meno istruite le persone che hanno imparato la propria lingua non completamente e parlano soprattutto in dialetto ? Si: No:

APPENDICES

85

11. Considera superbi o snob coloro che parlano sempre in lingua anche con amici intimi? Si: No: 12. Se Lei fosse uno straniero, cercherebbe di imparare il dialetto del Suo nuovo ambiente, per essere più facilmente accettato e per stabilire contatti, anche se ciò costituirebbe un addizionale problema linguistico ? Si: No: 13. Considera essenzialente inutile imparare i dialetti delle lingue straniere? Si: No: 14. Che dialetto pensa Lei che sia più difficile da imparare? a. tedesco svizzero per un tedesco della Germania: b. ticinese per un italiano : 15. Che lingua (lingue) usa Lei nelle funzioni a. b. e d

religiose? c. f.

16. Nella Sua corrispondenzprivata scrive? a. sempre in lingua : b. sempre in dialetto : c. talvolta in dialetto : c. con singole espressioni o parole in dialetto :

P.S. Potrebbe forse concedermi una breve intervista nella quale domande come le precedenti potrebbero essere discusse in maggior dettaglio. In caso positivo favorisca scrivere la data e il luogo pili conveniente per Lei (prima del 21 luglio 1968). La ringrazio moltissimo per la Sua ccoperazione.

Nome, Indirizzo, Numero telefonico: Tel.: Giorno e Ore: Luogo:

INDEX

Age groups 30 Allemanic 51 Alpine-lombardic dialects 16, 17 Applied attitudinal ratings 57 Audiolingual skills 21, 39, 40, 41, 42 Austro-Bavarian dialects 51

Jaberg, Karl 17 Jud,Jacob 17

Beauty of Languages 50 Bellinzona 13, 17, 37 Bivariate Distribution 26

Keller, Oscar Koinè 17

Catholics 56 Chi-square test 23 Coefficient of correlation 26, 57, 58, 60, 61 Cole, G . D . H . 14 Complementary Questions 52 Correlation of socioeconimic status 28, 29, 43 Diglossia 54 DiPietro, Robert J. 16 Distribution of Sex and Age 30 Distribution of Questionnaires 20 Education 14, 28, 34 English 35,37 Extent of non-native language use

42, 43

Ferguson, Charles A. 16, 54 Fishman, Joshua A. 16 Foreign languages 18, 36, 39 French 12, 35, 37, 60 Frequency of language use 52 Garrett, Edward 23 German, standard 12, 17, 24, 31, 35, 37, 45, 47, 52, 57 Gumperz, John J. 16 Half-measures 58 Hymes, Dell 16 Income

14, 27, 32

Intermarriages 52 Interpersonal communication, language in 46, 47 Italian, standard 12, 17, 24, 31, 35, 37, 45, 47, 52, 57

17

Labov, William 16, 62 Lado, Robert 23 Language Attitudes 14, 18, 19, 21, 45, 46, 47, 49, 57, 60, 61 Language Attitude Ratings 45, 46 Language in formal situations 54 Languages in the schools 35, 36, 42 Learning of non-native language dialects 54, 55 Locarno 13, 17, 37 Lugano 13, 17, 37 Methodology 20 ff. Minority language 47 Moulton, William G. 16, fn. Multilingualism 38, 42, 46, 63 Multiple answers 47 National languages 18, 36 Native languages 14, 60, 61 Non-native language performance 14, 18, 19, 60, 63 Normality of Sampling 22, 23, 71 (Curve) Objectives, general and specific 13 Oral comprehension 21, 39, 40, 41, 42 Performance levels 39, 40, 41 Population 11,20,21 Prestige of languages 46 Procedures 20

87

INDEX Product-Moment method 27, 43, 57 Profession 14, 28, 33 Protestants 56 'Pure' attitudinal ratings 57 Questionnaires

20, 21, 63, 72 (Appendix)

Reading 21, 39, 40, 41, 42, 48, 49, 57 Reliability 23, 24, 44, 57, 59 Returns 22 Sample 22, 23 Sampling Procedures 20, 45 Scattergrams 28, 57 Schoolsystem 34,42 Self-evaluation scales 38, 61 Sganzini, Silvio 17 Significance 16 Skewness 32 Socioeconomic status 13, 18, 19, 24, 27, 29, 37, 60, 63 Sociolinguistic survey 16, 59 Speaking 21, 39, 40, 41, 42 Spearman-Brown formula 23,24,43, 57, 58, 67 Spiess, Federico 17, fn.

Split-Half Method 43, 57 Standard Languages 31 (cf. also: German, Italian) Standard Languages restricted use 56 Standard Languages in correspondence 56 Standard Languages in religion 56 Standard Language vs. Dialect 53 Stewart, William 16 Subdivisions, cantonal 11 Swiss German 13, 15, 16, 17, 24, 31, 45, 47, 52, 57 Ticinese 13, 15, 16, 17, 24, 31, 45, 47, 52 Ticino, Canton of 11, 20, 21, 26, 32, 33, 37 Usefulness of Languages Validity

25, 44, 57, 59

Warner, William 14 Weber, Max 14 Weinreich, Uriel 16 Weiss, Robert 23 Writing 2 1 , 3 9 , 4 0 , 4 1 , 4 2

50

janua linguarum Series Practica 141 Gething, T.W.: Aspects of Meaning in Thai Nomináis 142 Sala, M.: Phonétique et phonologie du Judéo89 Espagnol de Bucarest 146 Bouma, L.: The Semantics of the Modal Auxiliaries in Contemporary German 147 Jochnowitz, G.: Dialect Boundaries and the 77 Question of Franco-Provençal 148 Moessner, L.: Automatische syntaktische Analyse englischer nominaler Gruppen 149 Sciarone, A.G.: La Place de l'adjectif en 46 Italien moderne 150 F u m é e , E.J.: Die wichtigsten konsonantischen Erscheinungen des Vorgriechischen 153 Hemphill, G.: A Mathematical Grammar of English 154 Haugen, E. and T.L.Markey: The Scandinavian Languages 155 Plotkin, V.Y.: The Dynamics of the English Phonological System 158 Edmonson, M.S. (ed.): Meaning in Mayan Languages 160 Omar, M.K.: The Acquisition of Egyptian Arabic as a Native Language 162 Chayen, M.J.: The Phonetics of Modern Hebrew 163 Hodge, C.T. (ed.): Afroasiatic 74 164 Lupas, L.: Phonologie du Grec attigue 165 Miko, F.: The Generative Structure of the Slovak Sentence 166 Hensey, F.G.:'The Sociolinguistics of the Brazilian-Uruguayan Border 167 Wolf, P.de: The Noun-Class System of ProtoBenue-Congo 168 Atkinson, J.C.: The T w o Forms of Subject In37 version in Modern French 170 Detering, K.: Automatische Erzeugung englischer Sätze 173 Schroten, J.: Concerning the Deep Structures of Spanish Reflexive Sentences 182 Jensen, F. and Τ.A.Lathrop: The Syntax o f the O l d Spanish Subjunctive 184 Iliescu, M.: Le Frioulan 129 185 Defromont, H.J.: Les Constructions perfec65 tives d u verbe anglais contemporain 189 Hayon, Y.: Relativization in Hebrew 194 Donaldson, W.D.: French Reflexive Verbs 58 195 Vasiliu, E. and S.Golopentia-Eretescu: The Transformational Syntax of Romanian 196 Wells, R.A.: Dictionaries and the Authoritarian Tradition 204 Gertner, M.H.: The Morphology of the Modern French .Verb 68 211 Householder, F.W. and G.Nagy: Greek 221 Rensch, K.H.M.: Nordkalabrischer Sprachatlas anhand der Parabel v o m verlorenen Sohn 227 Moody, M.D.: A Classification and Analysis of "Noun + De + Noun." Constructions in French 98

Dfl. 32,FF/ 58,36,FF/ 50,30,FF/ 30,96,24,48,20,76,64,16,25,FF/ 48,38,34,58,FF/ 24,32,36,28,F F / 80,FF/ 42,72,FF/ 38,58,32,FF/ 44,30,60,FF/ 64,-