14R? .3311 State *7 ROM USE ONLY ABSTRACT CORRELATES OF LANGUAGE ATTITUDES OF STUDENTS FROM INDIA By Marvel June Allard This dissertation is concerned with attitudes to- ward language. It is.an attempt to examine.attitudes by ex- ploring methods for measuring such attitudes and by studying selected attitudinal and socio-cultural correlates of lan- guage attitudes. The.research was eXploratory and correlational in nature. No specific theoretical framework.guided the study. Sociological,.psychological.and.linguistic.literature were reviewed and language concepts selected for study. The major language concepts selected were prestige, affection, liberal- ism, Specialization and interaction patterns. .Questions and indices were devised to measure these language concepts. Attitudinal and socio-cultural variables suggested in the literature and from informal sources were selected and questions and scales written to measure them. Attitudinal variables included change proneness, approval of.current social change in India and attitudes concerning were govern- ment power should reside. Socio-cultural variables included rural-urban residence, travel, social class, family tradi- tion of foreign study and Nerth-South residence in India. Marvel June Allard Relationships between the language concepts.and the attitu- dinal and socio-cultural variables were studied through a series of low order hypotheses. The questions written as measures of the language, attitudinal and socio-cultural variables as well as a number of questions concerned with general language back- ground were combined into a questionnaire. The,qugatiennaire was administered to 96 students from India, studying in the United States at Midhigan State University and the University of.MiChigan. The results showed the language concepts :prestige, affection, liberalism, specialization and interaction, to be both useful and measureable. All of the hypotheses_received some degree of support. Socio-cultural correlates were found for language prestige, affection and liberalism. Travel and social.classemerged as the primary socio—cultural correlates of the.language measures. The perceived language liberalism of the respondent's father, and to a lesser extent, the perceived language S liberalism of his mother, subcaste, social class and region, were correlates of the respondent's own language liberalism. In addition, the perceived language liberalism of these reference groups was generally associated strongly with both the language specialization and interaction patterns of the respondent. Marvel June Allard The respondents change proneness, approval of India‘s current social change and attitudes concerning where political power should reside all related to his language liberalism. His Change proneness and approval of India's social change related to his language specialization: his approval of India‘s social change and attitudes concerning where political power should reside related to his inter- action patterns. . Einally,_socioecultural.cerrelates.uere.found for the attitudes concerned with change and government power. Travel was found to be the most important correlate of change proneness, approval of India's social change and attitudes concerning when political power should reside. OORRELATES OF LANGUAGE ATTITUDBS OF STUDENTS FROM INDIA By Marvel June Allard A THESIS Submitted to Michigan State University -in partial fulfillment of the requirements fer the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1965 <19 capynght by MARVEL JUNE ALLARD 1966 “(NOMRDGMEN TS The anther wishes ,1 to ”express ...her ...appreciatien to her thesis adviser, Dr. NEugene..Jaoebenn,._end to the other members of her committee, Dr. Jack Berry, Dr. Erwin Bettinghaue. Dr. Wilbur Breekover, Dr. John Gullahern. Dr. Hideya Kumata..and Dr. Frederic Wickertfer their help in the planning of the research and preparation of the manuscript. 11 ,CHAPTER I II III IV V .TAELEMOEWOONTENTS .INTRODUCTION Language and Culture .India and.Language Language and Personality Culture and.Pereonality was sonny or ms meanness or LANGUAGE arrxmnss Language Attitudes Correlates of Language.Attitudes Language and Cultureflypetheses .Language and.Personality Hypotheses Culture and Personality Hypotheses DESCRIPTION OF THE SAMPLE Population and Sample Travel Family Tradition of Foreign Study Summary ' LANGUAGE BACKGROUND OF THE INDIAN STUDst General Language Background Prestige and.Affection Associated withrLanguages Language.Preferences EnglishrLanguageeBadkground Attitudes.about Bilingualism Attitudes about Indiats.Language '.Preblems Summary CONSTRUCTION OF INDICES .Languagewhiberalism LangUAge~Specializatien Interaction Change Orientation A Approval of Social Change Attitudes Cenoerning where Government Power Should.Reeide ‘ iii Page omen '133 TABLE or CONTENTS (Continued) eeee CHAPTER Social Class VI RESULTS AND DISCUSSION VII SUMMARY REFERENCES APPENDICES iv Page 147 156 274 .295 300 TABLE 'q as Ln b u» h) F“ O 9. 10. ll. 12. 13. 14. 15. 16. 17. 18. 19.- LIST OF TABLES Size of City Where Most of Life Lived Home State or Territory Number.of Places Lived in India Travel in.Indiaa Travel Outside India Vicarious Culture Contact ,Generations of Family Members with Education Outside-India Foreign Study of Brothers and Sisters ‘Pereign Study of Parents, Grandparents, Uncles and Aunts Languages Known Native Tongue Home State or.Territery.and Corresponding NativehTengue- Fluency in Hindi Knowledge of Classical Indian Languages Number.of Languages, Dialects, Sub- languages and Vernaculars Known by Parents Number of.Languagee-Learned-fnem.Parents Prestige Accorded Languages by.3ntire Indian Group Prestige Recorded Native-Tengue Differences between Prestige.Ratings Given Languages by Entire Indian Group and Native Tongue Speak ere Page 51 52 53« 54 54 56 57 58 59 62 63 64 65 66 67 6.8 69 7O 71 LIST or TABLES (Continued) .... TABLE 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 39. 40. 41. 42. Affection Expressed for Native Tongue .Affectien Expressed feerin i and.English by Entire Indian.Group Meet Comfortable Language . Language(s) Most Useful Professionally Language(s) Meet Useful Socially Languagecs) Preferred Languageh) Liked Language(s).Net Appealing Language(s) Used Unwillingly Number of Languages Wbuld Like to Learn Languages- Weuld Like to Learn Value in Studying.Fereign Languages Age at which English Language Study Began Years of English Language Study Nationality of.English Language Instructors Similarity of English Learned at Home to thatMUsed-in”United-States. Adequacy ef.3ngliehnLanguage Instruction Desirability of- Using. Mere. than- One Language Prestige.kttaehed te Bilingualism“ .Appreval ef Inteneixing Languages Actual.Intermixing.ef,Languages Concern-for-India’s.Language-Preh1em Willingness to Accept Hindi as India's Official Language vi Page 72 72 73 74 75 76 77 79 80 81 82 83 83 84 85 86 86 87 89 9o 90 ‘ 91 92 LIST or TABLES (Continued) ... .TABLE 43. 44. 45. 46. 47. 48. 49. 50. 51 .... 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. First Choice for India. II Official Language Second Choice. for India'. 8 Official Language Strength of Conviction- aboutLanguage Choice Approval of Practice of Drawing State Lines According. to Language- Boundaries. Languag e. Liberalism . Indices. Correlations hong Language Liberalism Indices- . .lntergroup Cerrelations Anong "Languages Not Appealing" Measures Intergroup Correlations Among “Languages Used Unwillingly' Measures. Intergmup Correlations Among " Intermixing APproval! Measures. . Intergreup Correlations Among ' Actual Inter- mixing“ Measures. Intergroup Correlations Among ' Prestige Attached. to Bilingualism” Measures Intergroup Correlations Among ‘ Acceptance of Hindi!I Measures. Language. Specialization. by . People Language. Specialization- by Place Language Specialization- by. Occasion Nehru. Discussi.en...Indices. Confidence- Scores- . Self. Percept, I Scares. Motivation. Scares. . Interoorrelat ions... onengd Metivatien Measures Orientation Toward Change Indices vii Page 93 94 95 96 108 110 113 114 115 116 117 118 121 122 122 A 124 126 128 131 132 133 LIST or TABLES (Continued) . . . TABLE 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. Intercorrelations Among Orientation Toward Change Measures.. Approval of Social Change Speed Approval of Social Change Direction Approval of Social Change Scope Index of Social Change Approval Intercorrelations.&meng Approval of Social Change7Measures. Power Accorded Central, State, and Local Governments. DifferenceePower.Measures. Outcomes When Government Levels Disagree. Power Indices--. Interoorrelations Among Measures Concerning Where.Government-Power.5heuld.Reside Per Capita Income- Number of-Beehs-in Family Library Number.of Servants in.?amily Educational.Level.ef Fathers and Mothers Father’s Occupatien. Ratio of Brothers.and Sisters Attending College to.Bretherswand.Sisters.ef.Cellege.Age Social Class Indices- Intercerrelatienslhmeng Secial-Class.Measures Relationships between Affection Felt for I Languages and Rural-Urban Residence viii Page 134 135 136 137 137 139 141 143 144 145 148 149 150 151 151 152 153 154 155 159 LIST or TABLES (Continued) TABLE 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. Relationships between Prestige Attached to Languages and Travel Relationships between Affection Felt for Languages and Travel Relationships between Prestige Attached to Languages and Social Class *‘ r“ Relationships between Affection Felt for Languages and Social Class Relationships between Prestige Attached to Languages and Family Tradition Of Foreign Study , --..-- - Relationships between Affection Felt for Languages and Family Tradition of Foreign Study srf‘ Relationships between Prestige Attached to Languages and North-South Residence in India- re . Relationships between Affection Felt for Languages and North-South Residence in India. Relationships between Language Liberalism and Social Class ~ ‘ Correlations between the Self Language Liberalism Index and the Language Liberalism Indices of the Father, Mother, Subcaste, Social Class, and Region a, -11 . Relationships between Items in the Self Language Liberalism Index and CorreSponding Items in the Language Liberalism Indices of the Father; Mother, subcaste, Social Class, and Region Relationships between Language Specializa- tion and Diversity of Language Preferences Cut-off Points for Dividing Language Liberalism Indices of Reference Groups into Liberal and NonéLiberal Categories .' ;.. ,Mean Language Specialization Measures for Homogeneous and Heterogeneous Reference Groups . is Page 162 164 167 171 172 T73 174 183 ’187 199 LIST OF TABLES (Continued) TABLE 98. 99. 100. 101. 102. 103. 104. 106. 107. 108. Roan Language Specialization Measures for Non-Liberal, Liberal, and Heterogeneous' Reference Groups Correlations between Language Specializa-_ ation and Range of Language Liberalism Among Reference Groups Relationships between Language Specializa- tion and Importance of Family. subcaste, Social Class, and Region Mean Proportions of Contact in Native Tongue Groups Mean Preportions of Friendships in Native Tongue Groups. . Relationships between Heterogeneity of Language Liberalism and Discussion about Nehru ' s Death The Relationship between Discussion About Nehru' 8 Death and Importance of Reference Groups 7 Relationships between Language Liberalism and Change Orientatior 4. Relationships between Language Liberalism and Approval of Social Change in India Relationships between Language Liberalism and Attitudes Concerning Where Government Power Should Reside Relationships between Language Specializa- tion and Change Orientation Page 200 201 204 211‘ 211 214 216 218 221 109. Relationships between Language Specializa- tion and Approval of Social Change in India . 223 110. Relationships between Language Specializa- tion and Attitudes Concerning Where Political Power Should Reside 227 lll. Relationships between Discussion About Nehru 8 Death and Approval of Social Channe in India - 231 .,_- ’ LIST or TABLES (Continued) TABLE Page 112. Relationships between Discussion About Nehru‘s death and Attitudes Concerning Where Govern- ment Power Should Reside 232 113. Relationships between Change Orientation and Travel 237 114. Relationships between Change Orientation and Social Class 239 115. Relationships between Approval of Social Change in India and Rural-Urban Residence 241 116. Relationships between Approval of Social Change in India and Travel 243 117. Relationships between Attitudes Concerning Where Government Power Should Reside and Travel 247 118. Relationships between Attitudes Concerning Where Government Power Should Reside and Social Class 251 119. Relations between Language Liberalism of Reference Groups and Change Orientation 259 120. Mean Change Orientation Measures for Non- Liberal, Liberal and Heterogeneous Groups 261 121. Importance Attached to Liberal and.Non- Liberal Reference Groups for Reapondents with High and Low'Orientation Toward Change Indices 262 122. Relationships between the Language Liberalism of Reference Groups and Approval of Social Change in India 263 123. Importance Attached to Liberal and Non- Liberal Reference Groups for Reapondents with High, Medium and Low Indices of Social Change Approval. 265 124. Relationships between Attitudes Concerning Where Government Power Should Reside and the Language Liberalism of Reference Groups 268 xi LIST OF APPENDICES APPENDIX Page A. QUESTIONNAIRE ON LANGUAGE 301 E. AGE ' A ' 333 C, ACADEMIC OBJECTIVES 334 D. ACADmIC SPECIALTIES 335 E. NUMBER OF MONTHS SPENT IN THE U.S. 336 E, RELIGIOUS AFFILIATION 337 G, EMPLOYMENT BEFORE AND AFTER SOJOURN 338 H. NUMBER OF YEARS SPENT OUTSIDE INDIA 339 x, DISTRIBUTIONS OF RESPONSES TO ITEMS COMPRISING 340 THE LANGUAGE LIBERALISM INDICES. J. INTEROORRELATIONS AMONG ITEMS IN RESPONDENT S LANGUAGE LIBERALISM INDEX 348 K. CONFIDENCE SCALE 354 L. KOSSOPF SELF PERCENT SCALE 357 M. DISTRIBUTIONS OF RESPONSES TO ITEMS COMPRISING MOTIVATION SCALE 359 N. IMPORTANCE OF REFERENCE GROUPS 365 xii CHAPTER I INTRODUCTION Language is an area of vital concern to developing countries today. In these countries, progress in unitication and.aoonomicmdeneLOpmentnis_seriouslynhamperedmby«language.prOb- lens. The language problems stem from lack of education and absence of a dominant language. Sub-cultural groups, eadh with a language and culture of.its own, fight for recognition of these languages and stoutly refuse to accept any other as a nationalglanguage. It_isnthenpnrposewo£.thisithenis“tonexamineulanguage. attitudesiatwtheuindizidualmleuelWandntowatndyIthemattitudinal and.socio=cultnral-correlatesnofntheseIlangnage«attitudes. Thejnhgects in .the enndyene.nhnaenu-1mm_1na1e, currently studying at Michigan State University and the Univer- sitynofidnichigan. TheynarehnotCamrepresentatiye«orOSSesection of Indian enniety, inhey me highly -eeIeet, highiyedneeted group. Some of them will have leadershiproles in their country.and_for,this.reason“theirIlanguagehand«social“attitudes are important. Their attitudes are important, .too, .hecnuee Observers have singled out the university students as exhibit- ing the most yieihlehnneet. In thecultureconflict situation in India. Modern writers pointeout.that.themedncatedflfiindunis, no longerwcompletely Hinducand-certainly not completely Whatern. 1 PI IF. .0; If 2 The language and socialnattitudes of_thesewetudents.from India represent the balancing of Eastern and Western cultures on the individualEleyelnandwsuggeet,won a.broaderulevel,.the_form.of “social thinkingmamong thenyoung educated generationuin India. To understand better the language attitudes of these Indian students,.an understanding Of_the.social situation in whiCh the attitudes_exist-is_necessary. An examination of general sociO-cultural factors. language background factors, and personality characteristics is made in this research to suggest some of.theasocial,and.personal factors associated with particular language attitudes. Manyneszeloping Asian and.African..countries_haye-been overwhelmed.byflthenproblcm_of a_multiplicity-cinindigenous ‘ languages compoundedwbytheeimpositionmot a Nbstern tongue. The .nznhienuennne hhmaaingmmmemmahmnm It is.a socialnandecultural.mattermas well. Thereais a uddeSpreadn adknowiedgement that_language is a.ksy toacultural.identity and unity. It isaa. hedgent. group mbarship_and has even .heen regarded as the ”Mary criterion ..for defining nationhood” (spencer,.1963: Hayes,_1960). As auCh it acquires deep emotion- a1 significance In: thembars..o£..its1ociety, 1.8hentcn. 4933). a: ..nhe name-..t.1.me_it isnaPpaLentuin.,-dmLelcping Airican and‘Asian.countries.todayathat_language_is.a.diuiaiueafactor flux.that nations areIraralymhomogeneous;cultural groups. In 1mnia,as~ineothar deuelOpingncountries,«manynsubcultures.exist. language serves.to unite eacb.subcultureaand.distinguish.it 3 from other subcultures. From a national perSpective, however, language does not serve to unite. Subcultural languages separate and divide a nationlinto subqnations. The newly deveIOping countries were -not unitied culturally crllinguistically before independence. Local jealousies overlanguage More usuallycogreatthat the language of the colonizerhad tote used to unite the country enough to tight for independence. Iheachieyment offlindependence only complicated the language problan. The tidal wave of nationalistic sentinent that swept “eachcountry brought with_it .thestrong .desire to be rid .of .all vestiges ofacolonialism-eincludinglanguage. What language, then, should be used as the national language? Every indigenous language suggested . antagonized the speakers Of every other indigmousnlanguage. Using ..a hone-indigenous language offended national pride. Morethan that, .it meant thehannernrurenlnrlnuerulruralherlrage. ,nnethecther hand, .unificationand glorification of cultural .,heritage were not enough. They nereonlythelbeginning. -Ir‘rhelrerhere of posterty andignorance meretonbe rammed, greatrandcrapid economic “and technological advances were necessary. For this 11: uaeneeeeeary tohave‘hookslllearningnand atechnical Runw- ledge, .and such things werelnot _toube.1ound.in.indigmous languages. And so .,the..prohl.elns,multipli.ed.,. .. . India and Language India has found just these problems. The multiplicity Of Indianianguages isastounding. According to -the -1951 census .atotaloi' B45 languages or dialects exist. Rourteen Of than are recognized. by the Indian Constitution. About .91 -percent of. the IpOpulationIspeakone .orinore .of. theseltourteendanguages. .Thecatiue speakerscf each of these languages .generallyiivein geographic proximity to each other and. are considered to constitute linguistic-cultural regions. “Linguistic regionalism“ sometimes called “linguistic communalism" is widely recognized .as a divisive .factOr in Indian society. It is the disharmony among these iinguistic cultural units andthe assertion by each of its individuality and independence trom the Others which poses amaJOr threat to Indianunity. mueain.usnl: Nanauatiand mu. .1951) Superimposedulpon this pattern ntlenguage diversity , is English. English is not so widely known as other languages. It is, however. thalanguage of the educated. It is themeans by smichtheeducatedcommunicateamongthemseluesand with the Western world and until recently it was .mlanguage of government. The constitution adapted in .1949 namedHindi as the .officialianguage of the Union to replace English by ._19.65. The Hindi complex. inrdu. sinduetani-.andhindi).ia._nnehen by .ahnut 42 percent of the population. It was intended that Hindi he- come the language of. all-Union matters such as courts, post Offices, customs, communicatiohshetween states...etc., nithths 5 states free to make their own language choices for intra-state affairs. The designation Of Hindi for use in national activities engendered.resentment from many quarters. There were many who sawpit_as.anrelevation which_o1ershadowed and degraded their ounqnative_tongue. There were.others who decried the selectioncf Hindicn .the heels of its poverty .-in .-uneahulazy and literature. They pointed to the rich-literary traditions of other Indian languages and the poor showing of Hinditin com- parison. Still others felt that if India were to move ahead economically. she must-adopt English: neither Hindi.nor the other Indian languages were.suitable vehicles for‘western tedh- nological.knowledge or‘Western governmental structures. What- ever the -criticism given. language battles seem to be the vehicles of expression for underlying fear and friction among dissenting subcultures. I . . . the ever-present fissiparous tendencies that wererhound-to.increase.with.the.deveIOpment.of political consciousness among the.masses.began to .manifest themselves_after the Eirst world war. They.had.heen-mainly.ggligigug before Eartition. and had Opposed Muslims to Sikhs and Hindus. In independent India, they becameligggigtigIL (de Riencourt, 1960, p. 352). Nearly euery writer comments.onithe.emotiOnal.inuolve- ment of Indians.on.the.languageeissue. .Murphy‘summarizes it very well: ' . . . and comparing various south Indian groups with one another, one realizes the.enormous intensity and profound personal.meaning of the language group as a mother group. a normative group, a .center and axis of self-sufficient life. (Murphy, 1957, p. 339). 6 In terms of national unity the seriousness Of the problem created by such strong involvement with linguistic regions became apparent to the Linguistic Provinces Commission. In.1948.they.reported, "Some of the ahlest.men in.the country came before.ue.and.confidently.and.emphatically stated that language in this country stood for and represented culture, race, history, I individuality, and finally a sub-nation.” (quoted in Harrison, 1955-56, p.621). ngggage ggd ngsgnglity .It_is_apparent thatnlanguage is a.many-faceted vari- able. Itis ..social: -.it-.is.-.cultural: it is psychological. Language is not only atmeans Of communicating.and of binding groups together: .it_is the_key to the.individual's own definition of.who he is. It is the means by which the indi- vidual establishes a sense of identity and security. ,The notional importance of aianguage lies _in the fact that it contains the voices of .one 's.mother, .father, .hrothers and sisters,. .and..one1s.dearest.friends. Our deepest ‘ emotions and most intimate memories are tightly bound up with our native tongue. (Armstrong, 1963, p. 69). nyrhnlngiateghauednng recognized theralatinnehiphe- tweenrlanguage.andmpersonality. .The deveIOpmentIOftlanguage I goes hand inhend uith..the detainee-at .otlpmonalityIand die- . Attitudes seem to .causemuchcf theianguage problen in .India. It.-is..the.refusal toiearn_other indigenous .lenqn- ages,.tomlearn.lnglishtor-Hindi.and/orntomaccept.Hindi.as.the Official tongue which.hlocks.progress.on.the.prohlem. The 7 identification of the individual with his sub-cultural groups results in his adeption-of their provincial_language attitudes. A nationwhereidentitioation _is .strongerct .the euheoultural level than at the national level is not a united nation. The educated elite in India.have been introduced to larger worlds than.those.known.by.the.simple.villager. Hope? fully some'have found their horizons broadened enough to become identified.at_larger-levels-enationally.and even.internationally. Even more hOpefully, this broadening of parapective is re- flected in attitudes toward cultural problems such as language. It becomes of interest then to explore the personality domains associated with “enlightened“ {or at least broadened) attitudes on issues such as language. In.particular.it.isgof interest to examineto what extent.positive.-feelin.gs.cbout-change at therindividual.and.social.leyel.relate.totflenlightened“.langu— age attitudes. SilEE££—IEQ.Z££§QEAL¥E¥ The young educated generation in developing countries is.exposed.to.two.culturesa'.a.very.modernLculture.and.a.tradi- tional. culture. To some extent this ..is true Linevery country, for.no generation sees the world-in exactly the sameulight as.its parent. generation. In the Asian..and Africancountries, . however,.the differenceretween the two traditions is often startling. The young.person in these.cultures.feels the pullaof.two different naysooftlife. .Ontoneohandtis.the . modern‘uestern.culturetand.on.the«other.istthettraditional 8 Eastern or African culture. Individuals cOpe with this problem in.several ways, ranging from outright rejection of the traditional culture on one hand, to outright rejection of the new'culture on the other. As might be expected, most solutions seen to be part wayebetween these two extremes. Some values from each tradition are accepted. The results of cultural intersection, however, are not always pleasing. The system of education devised by Macaulay lacked in wisdom and provision. It has created young educated people out of harmony with their environment. It has created misfits, young men who aspire to live like EurOpeans but without working like Euro eans. Chand and Kapoor, 1957, p.43). The results of-cultural.intersectionrmay be examined in several ways. .The.extent to which an_indimidual is willing to cast aside his provincialism and idmtify at national and international.levels.mayvwell.be‘reflectedrin“hishattitudes toward.language and toward social change and toward govern- ment in general. The question becomes one of what personal attitudinal factors and what socio—cultural factors.influence an.individual to accomodate-cultural intersection in-one mannerrratherothan.another. .No 912111.11:an language ethihndes-ie.enmpiete without a consideration errhepeyehningieei end enciningieeireehnre involved. By the same token the picture isrmore.complete when.the psychological and sociological”factorsrarenin turn related.to-eaeh_other. 9 Although.language problems may be most clearly seen in develoPing countries such as India, they are by no means foreign to countries with a single accepted language. Perhaps the best illustration of the importance of considering psycho- logical, sociological, and linguistic factors in combination is from‘warner and arole's The Sepia; Systems of émegicgn W: As the P1.child°s personality develops toward greater orientation to the external relations, a reaction sets in against the ethnic language, provoking antagonism in the parents. (warner and Srole, 1945, p.221). CHAPTER II THE STUDY OF THE CORRELATIS OF LANGUAGE ATTITUDES From psychological literature, from sociological liter- ature.--.from linguistic literature. .and from -informal conversa- tions, a number of ideas and concepts about language were gathered. Most of these ideas and concepts were vague and poorly. defined. Eewhad been operationalized. Some, however, seemed particularly relevant to the study of individuals in language situations such as those existing _in India. It was decided to examine a few of these ideas and concepts _in detail to determine their ..usefulness. forstudying. . language attitudes. W The ..1irst concept ”selected was prestige. Throughout the literature references are nedeto social statuser prestige in._connection withlanguage. Sometimss.it4s._W with which high cornlow prestigeisrassociated. Income countries visiting ..aliens speaking many languages. enJoy high prestige as ..do thenative born citizens who haveecquired foreign tongues .as secondary ..languages. Sometimes Atris with particular .W that highmornlow prestige is associated. ' In most countries -smne languages hold more “status than others and _in developing countries there erelocal dialects «that 10 ll educated citizens arerreluctant to admit they speak. Sometimes it is with_the.gnltu;31_grguprrepresented by a language that the.bigheorhlow prestige is associated. .Immigrant groups are said to experience embarrassment.ouer accents and alien .mother tongues. Whether the prestige associated with languages is ,high.or~loquee s to.depend_on.circumstances. Inrany.event a dimension.of prestige seems to be associated wuthrlanguages.‘ Another theme in the literature is that of the signifi- cance of affect. Affect takes many forms. Immigrants teach the old country language to their children in order that it null not.die.outAin4the.new.country. .Tribal.andrregional groups«refuseWtorlearnrotherelanguages.and.are gealous when therlanguage.ofmanotherlgroupeis_chosenwfor-gowernment«use. Nationsrrarely seriously.considermgiring-up-theirnnative tongues in favoreof the conveniences of a.world.language. Individuals,nhoweuer._profess“affectionate.feelings_for_langu- ages other than the native tongue. They express affection for languages for a variety of reasons. Sometimes it is an appreciation of the,literary or tonal qualiti smof a language. Sometimes affect is the result of memories of happy_or sad experiences associated with.arlanguage. .Thenmanyefeelings andrinfoemalwexperieneesereported.euggestnthat.some.sort.of ~positiue.orrnegatixeeaffect;exiets.forelanguages. “In.the research reported.here,uaffect-was examined in the form of affectionrfelt.for.languages. The concern was.not.with a .clinicalediagnosis of affection.or-its.hases. .The.concern was simply in determining if an.indiuidual could express the.degree 12 of affection he felt for various languages and if socio- cultural factors were correlated with the affection expressed for,arlanguage. The third language concept examined was a composite of ideas and attitudes grouped as evidence_ofulanguage_liberalimm. A dimension of-liberalismeConservatism seemed to pervade a numbereof specific opinions and attitudes. «Liberalism was defined.as.a.realization-andJacceptance.ofrother,2iewpoints non.language and an.appreciation.of the value.of.otherrlanguages.’ In.one.sense_liberalism.is a question.of perspective. Amper- spectiverlimited to the confines.ofLansinglenlanguage.or .regional group is.arnarrow.perspectiverlikely to be_reflected in narrow attitudes. A broader perspective with an appreci-' ation.of where arregion,or_language gnoup.fitsninto enlarger picture seems moreulikely toube.expressed.in.hroader, more liberal attitudes. In another sense,nliberalism concerning language attitudes can be viewed as an indication.of.identi- fication. Strong.identification.at.the_regional-or«language group level will.prohahly be reflected in appreciation of only.one.point,of.wiew~ethat_of.the.regionqor.languageegroup. Conversely, when-identification is strong.at~theenational level. and weaker at..the.regiona1, ..then..the iengnegeehurnm resulting seem morelikely ..to heiihereiin earn: es .they are basedmon a genuine.concern-for what.willnhe.best for.the nation-rather than.for a.particular-subgroup_inithe.natian. Eromneither.approach.rit-seemed.toemake.an.intuitiueekindof sense to view-the-language..attitudes .as..liheral.ror _non-Jihsral 13 indices of an individual's approach to the whole language question in India. Another approachntoianguage behavior was _made via the language usage concept -of specialization. This was_the .fourth languageconcept studied in this research. Language specialization refers .to differences iniangu- age behavior due to differencesin social situation, 9.9.. topic, audience, occasion, place, -etc. -In_one way, _language specialization can .he thought of ..as social sensitivity. It seaned reasonable to assume that some individuals would be more sensitive than,.others..to..such-social situations. .Indivi- dual differences might. be .eXpected, too, because -not all ..indi- viduals live in situationscalling for the same.amount of specialization. More .homogenous situations donotrrequire, or even give opportunity for, asmuch specialization as heter- ogeneous situations. Specialization appeared to bee mseful concept .for investigating ianguageusage ofth ..a personal- ity and a situational perspective. The «fifth ianguage concept investigated was also a language usageconcept. Thisconcept was.interaction. In this research, two .hinds ,of._interaction were..._examined. The first .kind concerned friendshipand contact patterns .among Indianetndente. .saeieaiiytheeonoern was..one_of who.inter- acted with whom within the .-Indian community. rheeeeond .kihd of interaction was -one .of amountof interaction .in the Merican community relative to that .in the "Indian community. Language differences existed both within the Indian community and .14 between the American and Indian communities- It wasnfelt that if language were- extremely important to -indimiduals, then .it would be “reflected in the .interaction patterns of the .-indi- viduals. ngrgates .gf Laggage Attitudes An investigation .ofthe correlates of these language concepts was made. Several different kinds of correlates seemed both relevant .and interesting. First werethe socio- cultural correlates. Thesocio-cultural variables examined were “rural-urban residence, amount of travel, social class,. family. tradition of foreign “study .and North-South..residence in India. Second were the attitudesthetespondentsperceived theirreferencegroups to hold. .Ihe referencetgroups selected for study were the family, subcaste, socialclass and region. The perceived attitudes included .hoth measures cf_..language liberalism and of, language preferencBB. Another. class of .correlates -inszestigated..concerned the relationships of the student to his reference groups. He was asked to indicatenhownjmportant the groups mere .to-..him, both individually and .in.relationship to -each _other. The fourth correlate was concerned with the Indian communityronthejiichigan Statecampus. Ihetatinetongiie compositiontofctheccommunity was {examined torcdetemine the extent to which the. student 's. interaction patternsmfollowed native tongue lines in the ..Indianmmmunity. 15 The last set of correlates of language attitudes studied were attitudes toward change, broadly defined. These attitudes included general responses to change, attitudes toward social change in India and.attitudes toward centralization of politi- cal control in India. ,These.attitudescareclabelled: "attitudes toward change and political control. . A.large amount.of language background data was . gatheredethrougb-questionstabout“languagescknown.touthetindi- vidual,.languages known by his parents, English.languageuback- ground,_language‘preferences-and.related data. ‘Measurestused for.obtaining.this.information included: LANGUAGE BACKGROUND DEER ,LenguagescKnowncto.the,Indiuidual Number.ofulanguagestknown Specific languages known Reading, writing, speaking and understanding of speech.in each.language known Specifictlanguages_learned from parents Reasons‘fornlearning-otherAlanguages Knowledge of classical Indian languages Years of study of classical Indian languages Languages Known to the Parents of the Individual Numberiofhlanguages{known $pecific.languagestknown Reading, writing, speaking and understanding of speech in each language known English.language,Background Age at which study started Years oftormal study * Nationality-of.instructors ”Similarity of English learned in India and English used in UeseAe 16 Language Preferences Most comfortable language Language most useful professionally Language most useful socially This information, loosely knitted together, formed a background against which theclanguagewconcepts.and their correlates might be examined. The language concepts and.their correlates were studied through a series of hypotheses. These hypotheses were.of.lowcorder. They were not deductions.from any theory. Indeed, verywlittle relevant theory seemed to exist. In like manner, these_hypotheses.were not the outgrowth of a large body.of.relevant research. Relevant_research was sparse. The_hypotheses were exploratorycin-nature: .they came from.informal observations and from.observations in thecliter- ature. Many were quite frankly speculative in nature. This chapter presents the hypotheses. When they were suggested -by ,fomal observation or research, the literature is cited. When they weretnot, the speculation is presented. They were.arhitrarily“grouped.into.threeccategories: ,language and culture, language and personality,.and.culture and person- ality. ”Hypotheses.involving.language“attitudes,andtlanguage usage on the part ofctherrespondent werecclassified as Wlanguage“.hypotheses.f-Hypothesescinvolving.socioecultural factors,.perceived_attitudes.of_referencecgroups,.feelings aboutcreference.groups.and.the Indiantcommunitytonnthetuichigan State campus were labelled ”culture”;hypotheaes. “Hypotheses involving attitudes toward change and politicalficontrol.wsre l7 titled ”personality” hypotheses. The variables used for test- ing hypotheses in this research were: LANGUAGE VARIABLES Language.Attitudes Prestige Affection Liberalism Language Use Specialization Interaction CORRELATES OF LANGUAGE VARIABLES Socio-cultural Factors Rural-urban'residence Amount of travel Social class Family tradition of foreign study North-South residence in India Perceived Language Attitudes of Reference Groups Language.liheralism Language preference Reference Group Importance Importance of reference groups Relative importance of reference groups Native Tongue Composition of Campus Indian Community Attitudes toward Change and Political Control Change,orientation Approval of social change in India Attitudes concerning where government power should reside In d C t H t 8 Language and culture studies have been mainly the province of linguists. Probably the most famous and intriguing 18 of.all hypotheses about language and culture is the Sapir- Wharf hypothesis. In very general terms, the Sapir-Whorf hypothesis sought to establish relationships between the way people.of different cultures_looked at the world.andctheelan- guages they spoke.The hypothesis was not clearly stated, howb ever, and numerous interpretations of itcresulted (fiblker, Jenkins, and Sebeok, 1954). Sociologists assert that language and culture go hand in hand, yet they have been slow to deal with the two in combi- nation: The sociological literature as a whole reveals “but rudimentary.concepts.for the analysis of .language as a social system, for comparative studies,.andnforgstudyflof_the_rolecoftlanguage wina total society. Theretseems,tocbe,fairly widespread recognition that languageeis import- ant, but no one has been quite sure what to do about language as a general social pattern. (Useem, 1963, p. 30). Recently,“psychologistschave_been.examining_language and culture. Osgood and colleagues £1962).have heen.concerned with whether speakers of differentclanguages use the samel dimensions to describe the world. In some of this researdh nearly.identical dimensions emerge when.differentclanguages are compared. In other studies the dimensions aremnot so similar. -Lambert“and“associateschave.been.studyingcthe_effects A ofmaccents“andclanguagestontthe«perceptioneofcother.people.' Changesrintthenperceptionuofmantindividualhs«personality"as the_individualflchanges.accentstor_languages.are.interpreted A astreflections.of”attitude_toward“theeculturetrepresented.hy theaccenter language iLamhert, Anisfield, is. Xeni—Komshian, 19 1963: Anisfield, Bogo, & Lambert, 1962: Lambert, Hodgson, Gardner & Fillenbaum, 1960). It is from the linguists, however, that the.impetus for the first hypotheses came. In addition to describing languages, linguists talk about the changes in languages and the.influence of one language upon another in bilingual areas. When the discussion turns to the factors contributing to these influences and changes, sociological oonsiderations.enter the picture. It has been suggested that language.might.be used as an index of social change or acculturation. Samora and Deane (1955-56), for example,.propose using language preference as an acculturation measure. Casagrande £1955) states that the English fluency of a Comanche is an index.of lhis relative acculturation to American culture patterns.“ In discussing linguistic borrowing, Gumperz Ll961).notes that . . . studies on the spread of modern urban Hindi in rural areas, for example, might form a basis for scalar indices for the measurement of social -change and of the relative integration of rural areas in modern Indian national lifeo (campers, 1961, p. 982:). Qne.ofothe«factorshinvolvedcin“linguistic"change.and interference is prestige. .Casagrande.il955)”cites.the-trela- tive prestige of competingclanguages“.as a variable affecting the ratenofclinguistic change. ‘Weinreich.Llafil).listscprestige ascanmexampletof.thecstereotyped”attitudestofthilingualcgroups which affect.linguistic interference. JHodson £1936) suggested that social status of.a_mother-tongue group_is a determining tactcrintherateerhiiingnaiism. nieheid assumes the relative status of the two languages as one of the more 20 significant factors inJanguage contact. Haugen (1953) notes that for immigrants to the United States, ”Very little prestige has been attached to the having.of bilingual capacities. ”_ Another attitude .ofparticularconcern .is affection. Although not directly suggested in the literature, affection appears as an undercurrent in discussions of attitudes and. feelings about language ,_.in the scientific literature as well as in informal reports. Armstrong £1963) talks about the motional importance of language to the individual. Shenton (1933), states that nations have “deep sentimental attachments to their,.o.m.languages.” Bossard ~(.1945) talks about...anbarrass- ment over. bilingual. backgrounds. Haugen (1953) describes escape from.bilingualism as ,a very common reaction ..of bilinguals in the United States. The motional involvement of residents of India .on the “language issue has been -noted many times. It is theorized that .a culturemight well accord a language high prestige, but. feelrelativelywlittleaffection for it. English, in India, is said to be accorded high -pres- tige--—higher ,than. many of thetregional languages. Yet it is not advocated for -the officialulanguage _by everyone-£43 not evenlisted in the Indian constitution as an -official language. Many Indians are .tooemotionally involved .nith theirbm tongues -to.‘.consider accepting ..any _other _Indian .01: non-Indian. language as the “official language. this same phenomenon occurs -on an international level, when, periodically, the issue ofhaving a single international language -israised. Nationalistic sentiment isotoo‘strong-enohcountry wants to 21 accept another's language at the expense of its own, no matter what the prestige of the other language. There.are.many.factors which might be expected to affect attitudes such as prestige and affection. One set of factors“concernsrthe.rangemof-culturencontact.inuone's_back- groundq The.first.such.factor is_rura1=urban.residence. The effect of residence should be clearly seen in India where the descriptions of rural areas with highly formalized and stable.social-caste.systems_and{descriptionswofilarge.cities with a fair degree of westernization and diverse cultural groups present a stark contrast. Singer (1959) describes the urban characteristics found in Madras as “. . ..a heterogeneity of linguistic, religious and ethnic and social groups.” Another avenue.of culture.contact is travel. Indiridnals who have.trauelled widely inside.and/or outside India will have come into contact with many other languages as well as other cultures. If.indeed.there is truth.in the old adage. “travel is.broadening“, then thawindiyidual whorhas,trauelledmm19ht be expected to differ from-hisvuntravelled.countrymen.in the prestige and affection he attributes to variousulanguagea. Social class is a third factor which may be eXpected to relate to language attitudes. Social.class.differences inrlanguage-form.and_usagemhave.heenrobseruediin«manyicountries (Eerguson.& Gumperz.41950). 12o: example, Englishiusagerand mastery-in Hawaii.(&mith. 1939),-Bnglishtprestigemin the Qudan (surname-953) andin Nigeria LBW, 4953). .and English usage. pronuncifiionAeximnchoices “in .nngiand‘laamtainueez) .1... I\ U. 0;- '90. t H " UV I llo‘ . 'ah I.‘ . a‘ 'ee "no . i" o N. ' In .0. It, 5'. 3‘... t a... J;- (Iv (if 22 and degree of bilingualism in the U.S. (Arsenian, 1937) were all found to relate to social class and/or education. In dis- cussing immigrants to the U.S., Haugen (1953) notes that the highly educated visitors from Norway deplore the corruption of their native tongue by the borrowing of English words on the part of the Norwegian immigrants. In the case of India, it is particularly interesting to consider the effects of social class on attitudes concerning English and Hindi. English and Hindi are the languages of the educated young: they are both “imposed“ languages in many areas, and they are the disputed and controversial languages of govern- ment. ,. The effects of a family tradition of foreign study on attitudes. toward languages have not been discuss ed .in the liter- ature to the author's knowledge. In India a family with a tradition of foreign education is a family with more education and a family where non-indigenous__languages are known. One or both of these factors might well be expected to relate to feel- ings about languages. ‘ I Finally, a”, last and most obvious factor in feelings 1’ about language: North-South residence. The South's resentment of Northern dominance and the imposition of the Northernlangu- .1 age, Hindi, as the official language .are well known (Murphy, l953). It ,is expected that all of these socio-cultural factors-erural—zurhan background, amount of travel, «social class, family tradition .offoreign .study andMNorth or South .residence will have some effect on the prestige an individual attaches to A“ 23 various languages and on the affection he feels for them. Stated in hypothesis form these eXpectations reads Hypothesis I Language prestige and affection are related to socio-cultural factors: a) Language prestige and affection are related to rural-urban residence. b) Language prestige and affection are related to amount of travel. c) Language prestige and affection are related to social class. d) Language prestige and affection are related to family tradition of foreign study. d) Language prestige and affection are related to .North-Southcresidence-in.1ndia. Thewthird.attitude.consideredrin-thisrresearchiisclan- guage liberalism. Because it is an abstraction from a number of Specific attitudes, there is little research to be cited in connection with it.- Two of the Specific attitudes suggesting a dimension of liberalism.onatheiindividualilevel.came.from Whinreich's list.L1951) of extraelinguistic group factors,affemtingilingu- istic interference. The attitudes were: “attitudes toward bilingualism as.such*land.“tolerance oriintolerance:withiregard to.mixing.languages.andfltoiincorrectwspeechAin-eachtlanguage.“ This last attitude has ,.been.noted on the. part of educated Norwegian visitorsctowardwthe intermixingcofuNorwegian.and _ hEnglishwhyuuorwegian“immigrants“toflthe-nnited,States.LHaugen, 1953). It has also.been noted on the.part of-elder Comanches. towardetheucontaminationiof.theirinative.tongue.by.their English-speak ing children 1 Casagrande, .1955) . D 5” up I, 24 Other attitudes suggesting a general liberalism in- cluded acceptance of Hindi as India's official language and the degree to which other languages were rejected.or used un- willingly. It.is expected that the range of culture contact experienced by the individual will affect his general liberal- ness regarding language Just as.it is expected that range of culture contact would affect the prestige and affection he associated.with.particular languages. Consistent with these expectations,ithe_hypotheeis.is_made.that.ruralzurbaniresidence,7 amount of travel, social class,.family tradition of foreign study and North-South residence are correlates of language liberalism. Hypothesis.II Languageiliheralismoisirelatedlto“socioecultural.factors. a) Language liberalism is related to rural-urban residence. b) Language liberalism is related to amount-of travel. c) Language liberalism is related to social class. d) Language liberalism is related to family tradition of foreign study. e) Language liberalism is related to North-South .residence in India. .Theisocioeculturalifactors.listediahouerareinot.the .only correlates ofmconcern. “Earhapsctheteingle.most.important _iculturalrfactor.affectingianilndianis"attitudes.are.themattitudes of.his.referencergrnnpl.VcIndiaiiscsaid.toiheiaulandiof.groups -and.ailand in which.everyrgroup_is a minorityigroup. (fimith, 1962). It is membership inithese.groups-that.is,hypothesised. to give the .Indian his identity. In.this_..r.eaearch..thelanguage 25 attitudes of four groups have been selected. for study. These groups are : family, subcaste, social class and region. Every group is not a reference group on every issue or for every set of attitudes. If a group is a reference group for -an individual for a set of attitudes .or behavior, than the individual. may be expected to exhibit ..a .certain amount of con- _ fonnitytothe .group.norms concerned ariththis .set_.of .-attitudes oLbehavior. Thergroupn approves .of conforming behavior and , group approval is rewarding whenthe, group ,is important to the individual. The first group to beexamined is .theifamily. Armstrong (1963) states that theenotional importanceof _a languagels due to its holding the voices of oner parents. Ineiculture where family is as important as it is in India, .family attitudes .might -certainlybe expected. to influence an individ’ual'e ..ettitedee. On the other hand, however, in Merican immigrant and native ‘ Indian groups, the first. or younger generationpappears tobave obtained .a more external orientation with a .reaction-against the native tongue appearing (Casagrande, 1955: Warner a arole, 1945). In the present study, newcomers to higher education are immigrants in a sense. Indians who, on the one hand, are from tradition-erientediamilies ghereieducation _islow, .but far as the 3.8. , may~eellireiect the traditional langmgenatti- _ tudes .of. .theirujmilies. The ..secondgmupof interestisihe euhcastebr Jati. In India today the 1ati-is-the-.basicw,unitn-of.the...caste systan (Whiting, 1963). It is difficult to assess the influence of 26 the caste system. “Modern.educated Indians often deny-their belief in_it, yet the caste.system pervades so.much of their background, it is hard to believe that it could have little or no effect on their attitudes and behavior. . . . Second in the.pyramid of cohesive structures is the caste, bound by lineage, by occupation, by customs and by the respon- sibility of mutual support, all interlocking in respect to the maintenance of the lineage, that is, through intermarriage. (Gardner Murphy, 1957, pp. 337-8). The third group of interest is the social class. The literature concerning social class has already been cited in the discussion preceding Hypothesis I. In Hypotheses I and II, social class was treated as a socio-cultural factor and the.intent“wasrto1measure.thereffects,of differencesiin social class upon.language attitudes. In the present discussion, however, social class is examined as a reference group. Neither measurement of social class per se nor differences among classes are of interest. It is how the individual perceives his social class as well as how he perceives each of his other reference groups to feel about language that is of concern. The fourth and last group is the linguistic region. The .existence..of..linguistic regions ..in . India is common know- ledge. Inaconnection with linguistic issues, Harrison states that, ”Themcontest between.theicentral poweriand.the forces ofregionalism willlong .remain. the bedrock issue .inundia'. (1955-6). is an interesting one. .The linguistic boundarymis.essentially the caste boundary. In the words of Selig Harrison, 27 “. . . the caste structure is regionally self-contained.“ (1955- 56, p. 626). Caste and language are not identical, however. . . . the caste subdivision and the language subdivision are usually “at right angles,“ that is, that there are all castes present in a language group and all language groups of_a certain region present within a certain caste, . . . (Gardner Murphy, 1953, p. 75). It was not possible in this research to study the family, social.class,.auhcastecandiregioncof each Indian student.directly. There waano way.to_collect.information on the language attitudes of these groups from the groups themselves so that correspondence between each of them and the respondentsicould.be independently determined. Information about group norms had to.come.from the student. Comparing the students' attitudes to those he attri- buted to his groups_produced.an indication of the individual's perception of his conformity to his group norms. In sudh a situation, the question of whether or not the group in question actually holds the attitudes the student has attributed to it is not of primary concern. The concern is with how the individual.seesnhis“owniattitudes.in1relation.toithose.of.the 91039. It was the intent of this study to.collect.ailarge amount-of-descriptive.materialmregardingotheilanguagethack- ground, attitudes,.andcusage of_the Indian students and to assess change proneness and other social attitudes.of the students. It was not.feasihle to ask theirespondent.to answer for his family,rsocialcclass, subcaste and region, .28 every question asked of himself. It was decided to select a small core of questions concerning language liberalism and language preference and.ask themindividual to indicate how each of his four reference groups.felt about them as well as howuhe personally felt. .Qorreapondence.hetweentthetlanguage attitudes of .a-theiindividual andbisgroups wouldlieian11n- directnmeasureqoftconformitymtongrouptnorms. It was expected that the amount of influence of‘a ,reference,group.would dependionthe_importance of the group to the individual. The amount of influence of a group could be assessed by how far the individual deviated from the group in the_1iberalness of his attitudes. In hypothesis form these expectations read: Hypothesis III Language liberalism is related to the.perceived_langu- age liberalism of reference groups and to feelings about the reference groups. a) Language liberalism is*rerated-to~the perceived language liberalism of family, subcaste, social w_classend..region. b) Deviation from the perceived language liberalism of the family, subcaste, social class~end region is related to the importanee of each of these jhrnups to1the.individual. , Eurther”consequencestnfethetinfluence.oftreference titoups.may.;hetexplored. .Hhen_referencetgroupatarecperceiwed m‘51have.strong.1.14ang1.1ag:e1..preiiezmni'ies...and....an_indixszitxlnal attaches inHeartancettotheaegroups..heimayihe_expected.tothonorthe'ir Preferences by using the appropriate 1anguage(s) when dealing 29 with them. If all the reference groups have the same prefer- ences the task is easy. When there is diversity among the groups regarding the.language(s) in which theyrprefer to communicate, “the _indiyidual must me different languages with different groupsrif.hemiseto.pleasemthemwall. .Itewould appear that the amount of specialization an individual employs would depend on how diverse his groups are in their prefer- ences and how'much.importance he attaches to each of the groups. In like manner, an individual whose groups are heter- ogeneous, i.e., some of the groups have.liberalelanguage attitudes and some have.noneliberal language attitudes, would probably find it necessary to specialize if he is to please all his groups. Thus he may have to avoid using certain langu- ages around certain groups while he is free to use those langu- ages around others of his reference groups. Again it seems reasonable to.predict that therimportance_of the groups.may wellrrelatehto_the.amount ofspecialization-practicedrby the individual. Hypothesis IV Language specializationris_related to the perceived attitudes of reference groups and to.feelings about the reference groups. a) Language specialization is related to the perceived diversity of language preference among family, subcaste, social.class.and_region. b) Language specialization is related to the perceived heterOgeneity.of language liberalism among family, subcaste, social class and region. c) Language specialization is related to the import— ance of family, subcaste, social class and region. 30 Still another aSpect of reference group influence in the language realm is the influence reference groups exercise over the individual when he lives in a situation where his culture and another intersect. Indian students.on.an American campus are in contact with two cultures, American and Indian. They may choose to interact entirely within the Indian com— munity, entirely within the American community or within both. Lambert and Bressler (1956) have examined the life styles of Indian students studying in the U.S. They found, for Indian students who had been in the U.S. for one to two years, an "alternation between tworopposing trends: the tendency to be- come more American than the Americans while simultaneously re- affirming ties with the home culture.“ This alternation was intensified in the behavior patterns of those who had been in the U.S. over two years. For most of the Indian students, however, (14 out of 19), the majority of associations were within the Indian community. In the research described here the language attitudes Indian students ascribe to their reference groups at home are examined in relationship to the individuals with whom they choose to interact in the U.S. Ifilanguage attitudes are re- lated to interaction, then.the attitudes of reference groups and the importance attached to reference.groups would pre- sumably have a bearing on the.interaction.patterns.of the individual.‘ It was expected that in the case of Indianrstudents whose groups all had liberal language attitudes, the students 31 would interact with Indians Speaking other than their mother tongue and would interact with the American community as wmdl as with Indians speaking their mother tongue. The same expect- ation held for Indians with heterogeneous reference groups, i.e., Indians who perceived some of their reference groups to be liberal and some to be conservative regarding language attitudes. For Indians perceiving all their reference groups to be conservative, the expectation was that most interaction within the Indian community would occur with Speakers of the same native tongue. In addition, a comparison of interaction within the American community to that within the Indian com- munity would show a high preponderance of the.interaction to occur within the Indian community. These expectations are affected by the importance with which the individual regards his reference groups. Groups having much importance will affect interaction patterns more than groups having little importance for the individual, i.e., the expectations are stronger when the groups are important and weaker when they are unimportant. Hypothesis V Interactionois related to native tongue, perceived.language liberalism of reference groups and to feelings about the reference groups. a) Interaction is related to native tongue within the Indian community on.the.Michigan State campus. b) Interaction is related to the heterogeneity of language liberalism among family, subcaste, social class and.region. 32 c) Interaction is related to the importance of family, subcaste, social class and region. Lgpggage and Persgnality Hypotheses In general, the research on bilingualism and personal- ity seems to have followed two lines. The first line of research has been concerned with learning a second language. Lambert and associates have conducted research which supports the position that an individual's attitudes toward a lingu- istic culture are related to his progress in.learning that language (Anisfield and Lambert, 1961: Gardner and Lambert, 1959). Another line of research has focused on describing the bilingual personality. Much study has revolved around the question of the relative intelligence of bilinguals and monolinguals (Peal and Lambert, 1962: Arsenian, 1937). Some personality investigation has sought to Character- ize the bilingual in terms of his social adjustment. Bossard (1945) examined case histories of U.S. bilinguals to eXplore effects of having a minority group tongue on personality deve10pment. Social setting and attitudes toward language emerged as important factors. He found further that the child who rejected.the parental language also rejected.the parent who Spoke it. Embarrassment over parental language and its resultant “foreign accent“ in the child led the child to employ protective devices: a) restrained speech, b) incon- spicuous behavior, c) home avoidance, and/or d) meticulous English. 33 When personality and social characterizations of bilinguals are made, the theme of embarrassment and shame runs through the writings on bilinguals in the U.S. Perhaps it is best stated by Haugen in his summary of years of obser- vation of Norwegian immigrants in the U.S.: As for the bilinguals themselves, some have accepted their lot with enthusiasm, others with resignation, but many more have sought to evade it by becoming anglicized as soon as possible. However prominent bilingualism may be as a trait of American life, it is eclipsed by the continual flight of the bilinguals themselves from bilingualism . . .. (Hanger, 1953, p. 2)° The most recurrent theme through the studies of bilinguals done by psychologists,.linguists and sociologists is that of attitude. The attitude of the individual toward language or the culture represented by the language relates to his language usage, progress in learning a language, per- sonality and social adjustment. It is apparent, too, that the bilingual's social and cultural setting has some influence on these attitudes. The.situation in the.developing countries does not allow for wholesale generalization from American observation and research. In most, if not all, develOping countries, there is no dominant language. The language cleavages are generally so numerous and so large that every language is a minority language. It may be that in such a situation, social conditions and social pressures do not.dictate inferior- ity feelings and negative language attitudes quite so strongly as in the 0.8. In such a situation other.personality factors may come into play more. 34 It is the complaint of the linguists that psycho- logists had done little work on the bilingual personality. . . . American psychologists so far have apparently refrained from formulating theories on bilingualism, as no experiments in this field seem to have been undertaken. (Weinreich, 1953, p. 72). Although psychologists have begun to study bilingual- ism, the interest has not so much been on personality as on environmental factors relating to the development of one or more kinds of language systems in the bilingual (cf. Lambert, Havelka and Crosby, 1958: Ervin and Osgood, 1954). Ervin and Osgood (1954) have made some theoretical interpretations of bilingualism. Their work was extended by Lambert, Havelkaand Crosby (1958). Both .setsnnof research are concerned with the effects of the context of acquiring the language on bilingualism. They examine the_language systems of the bilingual and attempt to relate environmental factors to the develOpment of these systems. This research is not concerned with the degree or kind of bilingualism. The term bilingual has been defined variously in terms of degree of fluency in the languages of concern and in terms of the context of.learning.of,the.langu- ages. The term bilingual in this research refers merely to the fact that the.individuals involved have knowledge of more than one language. The concern is not with the kind-or degree of bilingualism, but with the attitudes held by the bilingual. Relationships between language attitudes and background factors have already been posited. Relationships between 35 personality and language attitudes are also of concern. The possession of a liberal outlook on India's langu- age problems is suggestive of liberal attitudes in other areas. If an individual is able to reject the traditional language provincialism so dominant in India, it may mean that he is able to reject other traditional attitudes. On a broad level he may be positively oriented toward change, i.e., his per- sonality may be such that he is inclined to look for new perspectives and new ways to doing things. On a more specific level, “modern“ liberal language attitudes may be accompanied by 3 general approval of the social.change occurring in India and by modern liberal.attitudes about where political power should reside. In hypothesis form this is: Hypothesis VI Language liberalism is related to attitudes toward change and political control. a) Language liberalism is related to change orient- ation. . b) Language liberalism is related to approval of social change in India. c) Language liberalism is related to attitudes concerning where government power should reside. If language-attitudes.are related to personality and other social attitudes, then language usage may also relate to personality and socialnattitudes. .Language usage in this instance refers to specialization and interaction. Several writers have suggested that many “bilinguals have a specialization of language by topic.or audience. 36 Ervin's observation is that no bilingual has exactly equivalent experiences in the two languages. There is a “specialization of functions“ for each language for a given individual (ditto, no date). Language specialization would seem to be one method of adjustment in the personality realm since bilinguals report they change as social beings when they.change languages (Haugen, 1961). Hypothesis VII Language specialization is related to attitudes toward change and political control. a) Language specialization is related to change orientation. ' b) Language specialization is related to approval of social change in India. c) Language specialization is related to attitudes .concerning where government power should reside. If the language issue is as important in India as we are led to believe, then the Indian's interaction patterns should also be consistent with his socio-cultural attitudes. More explicitly, the Indian with conservative individual attitudes,.i.e.. negative change orientation, negative atti- tude toward India‘s.social change and traditional attitudes concerning where government power should reside would be ex- pected to have conservative language interaction behavior. His interaction would be. primarily confined to the Indian community and he would.be expected to.have relatively little interest in interacting with non—Indians. 37 Hypothesis VIII Interaction in related to attitudes toward change and politi- cal control. a) Interaction is related to change orientation. b) Interaction is related to approval of social change in India. c) Interaction is related to attitudes concerning where government power should reside. C tu a d Perso ‘ t H otheses Culture and personality studies have traditionally been made by anthropologists. The many attempts to describe national character and typical.personality have been examined and reviewed rather' critically recently (Singer, 1961: Duijker & Erijda, 1960: Inkeles and Levinson,.1954).‘ Many theorists no longer feel that any national character can be defined for modern nations. There have been numerous attempts to relate personal- ity and culture change in terms of the effects of culture change upon personality. AnthrOpologists usually engage in culture and person- ality studies involving the administration of projective techniques to members of non—literate societies which.are in the process of adopting ”characteristics of the dominant society in their immediate enviromment” (Kennedy, p. 419, 1961). Psychologists have recently conducted personality and culture change research by means of student sojourn studies designed to measure attitude change resulting from 38 the cross—cultural experience. Selltiz and Cook (1962) dis- cuss a number of these studies. Generally few significant personality or attitudinal changes appear (McGuigan, 1959, 1958: Smith, 1955). Psychologists have also.examined individuals whose‘Western contact comes in their own countries. Such studies are usually attitude studies which find acceptance of some Westerniattitudes: the acceptance being.accompanied by partial rejection of traditional values (Jahoda, 1961: Lystad, 1960: Doob, 1960). Studies have been conducted on early adOpters of new agricultural. family or community practices. Generally such studies have sought to characterize adopters in terms of reading habits, social status, education etc. (Putney and Putney, 1962: Rogers, 1962: Adams and Masuoka, 1961: Freeman, 1961: Lionberger, 1960: Back, 1958). Some of these studies, however, have begun to ask questions about the attitudes and personality characteristics of change prone Decals- Examples.of-theiattitudesuand_personalitywcharacter- istics found to relate.to adoption of different practices are a scientific and non-traditiona1.outlook and a business attitude toward farming in India (Bose._1959)-and a proneness to change in general in Puerto Rico (Back, 1958). Studying American students abroad, Smith found no relationship.betwaen attitude.change and an introverted.or extroverted personality.but found that 39 Knowledge of a person's preexisting attitudes appears to be a better predictor of his response to a heterogeneous inter-cultural experience than is information about the intercultural experience itself. (Smith, 1955, p. 477). In general, one's personality and outlook on society, like one's language attitudes, may be expected to relate to socio—cultural factors reflecting the range of culture con- tact in one's background. .An.individualis_orientation toward change.should be more positive if his residence is urban. At the same time, urban residents might also be eXpected to view the ongoing social change in India as not rapid enough or wide-spread. enough.while.theirdrural.counterparts find it threatening to thestableaociety they know. For rural residents, social change.might well be viewed as proceeding at far too.fast a pace. A third area in which rural and urban residents might well differ involves attitudes concerning power at various government levels. This appears to be particularly relevant for India where regional loyalties are often extreme and resentment of the.national government not at all unusual. It would seem that to rural folks with a focus on themselves and their way of doing things, change would be rather frightening- a disturbance of the status 93g. On the other hand, people from more‘Weaternizedrurban areas.are moreelikalyuto welcome change and to value a relatively~strong.central government to unite the diverse Indian papulation and hasten change. Travel may also be expected to affect -the individual's interest in change and his view of the current .social change 40 in India. For the urbanite, travel to rural areas may under- score the need for change. For his rural counterpart, travel to urban areas may whet an appetite for changes in his own surrounds. Another socio-cultural factor related to social atti- tudes is social class. In India, the upper classes were the primary agents of culture contact. Their contact with British administration and their education and travel Opened doors to Western culture for them that other groups did not have. Singer (1959) raised the question of ... whether the Brahman literati in Madras are changing their social role, giving up their traditional role as cultivators of the Great Tradition (and agents of Sanskritization, as M.N. Srinivas would say) to become intelligentsia, i.e., agents of Westernization and modernization. (Singer, 1959, p. 144). Lamb (1963) was in agreement with Singer's observations. In all imitation, there tends to be cultural time lag, and this is particularly significant in India today. While the lower castes have been imitating traditional Brahman customs, a considerable percent? age of the Brahmans themselves have been imitating the customs of the West. ... This time lag helps to explain the existence among the great bulk of the middle and lower castes of India of a traditionalism often greatly underestimated in the West. (Lamb, 1963, p. 149). The Useems (1955) found differences in readiness to change among classes and castes also. They felt, however, that it was the ”lower castes and lower middle classes” whidh had a greater receptivity to change (p. 27). In view of the difference of Opinion on who it is that changes, an examina- tion of.how social class is related to change orientation is in order. 41 In the Useem study of‘Western-educated Indians, the authors reported that one of the factors associated with change proneness was family tradition in Western education. They found that if the individual was the first in his family to study abroad, he was more apt to have a positive orient- ation toward change than if his family traditionally sent its young abroad. Apparently being the first in one's family to study abroad_ig,a breaking of tradition. (This.poses an interesting problem. Following a hypothesis of broadmculture.contact resulting in.broad liberal.attitudes, it is the individuals whose families have traditionally studied in the Western world who might be expected to have the positive orientation toward change, who.might wish to see.social change in India increase and who might wish to unite India under a strong central government. The Useems'.finding contradicts the hypothesis about change orientation andaraises the question about the individual who breaks family tradition to study abroad. Does this individual whom the Useems found to have a positive orientation toward change also view.socia1 change as.not proceeding rapidly enough--at least more so than his fellow sojourners coming with a family.traditionfiof.foreign education? .Does this,same individual also go so far.in.break- ing tradition as to wish to see the central government stronger than local or regional governments? A final_socio-cultural factor that may alsolbe ex- pected to influence antindividua19s outlook onilife is 42 Northern versus Southern geographic residence. This factor is important because the central government in India is located in the North and the North-South conflict is strong. What does this mean in terms of change orientation, attitudes toward the current social change in India,_and attitudes about the power various government levels should have? Much social change is being attempted by the central government and probably.much that isn't is being,attributed to the central government. Since the central government is located in the North, northerners may feel a strong central govern- ment to be a decided advantage and may be more willing to accept social change, particularly as it appears to emanate from the government. .On the other side is the South whose resentment of Northern domination is no secret. Southerners are likely to oppose a strong central government and social change for no other reason but that it comes from the North. Five socioecultural factors have been discussed as influences on personality and social attitudes. In hypo- thesis form,.the discussion may be summarized as: Hypothesis IX Attitudes toward change and political control are related.to socio-cultural factors. a) Change orientation is related to rural-urban .residence. b) Change orientation is related to amount of travel. c) Change orientation is related to social class. d) Change orientation is related to family tradition ofnforeign.study. e) Change orientation is related to North-South residence in India. 43 f) Approval of social change in India is related to rural-urban residence. 9) Approval of social change in India is related to amount of travel. h) Approval of social change in India is related to social class. 1) Approval of social change in India is related to family tradition of foreign study. 1) Approval of social change in India is related to North—South residence in India. k) Attitudes concerning where government power should reside are related to rural-urban residence. 1) Attitudes concerning where government power should reside are related to amount of travel. m) Attitudes concerning where government power should reside are related to social class. n) Attitudes concerning where government power should reside are related to family tradition of foreign study. 0) Attitudes concerning where government power should reside are related to North-South residence in India. As in the case of.language liberalism and.language usage, socio-cultural factors are not the only factors posited to be related to personality and attitudes toward social issues. A second group of factors to be considered.is the individual's reference groups. To assess the influence ofmreference groups on the individual's change orientation, attitudes.toward social change.in India, and attitudes concerning where government power should reside, it would.have been.best to have.independ- ent measures of these attitudes from each of the groups.and from the individual. This was not possible. .Because of the nature of the measures, it was not.feasible to ask the_respond- ent how.each of his groups would have answered eachlof the measures, either. An -.indirect indication of the influence of reference groups on the student's attitudes.toward change 44 and political control would be suggested, however,.by demon- strating that a respondent's language attitudes and attitudes toward change and political control were related to each other and that both were related to the perceived language attitudes of the.reference groups. If such.relationships were found, then there would be reason to,.expect that the attitudes of the respondent and those ofhis reference groups were related. While such evidence would be indirect and in- conclusive, it would be strengthened by taking into consider- ation the importance attached to the reference groups by the individual. Hypothesis VI predicted that individuals withrpositive attitudes toward change and political control would also have liberal language attitudes. .Hypothesis III predicted that the individual's language attitudes would be related to those of his reference groups. Hypothesis x predicts that indivi- duals who perceive their reference groups to have liberal language attitudes would be expected to have a positive change orientation, to approve of social change in India and have attitudes favoring greater concentration of power at higher government levels. Conversely, the individual who perceived his reference groups to have noneliberal language attitudes would he expected to have conservative or negative attitudes toward change "andpolitical control. The individual whose groups have heterogeneous attitudes is.caught in the.middle. It is expected that he will probably take a middle position, i.e.,-have-attitudes that.are neither very positive nor very negative. These expectations regarding attitudes toward 45 change and political control and language attitudes on the part of the reference groups assume an attitude consistency within the individual and agreement between the individual and his reference groups. Finally, it is expected that the importance theeindi- vidual attaches to his groups and his own attitudes will be related. Individuals with positive attitudes toward change and political control will attach greater importance to reference groups with liberal language attitudes and.lesser importance to those with.non=liberal attitudes. Therconverse holds for individuals with relatively negative attitudes to- ward change and political control, i-e., lesser importance is associated with groups having liberal language attitudes and greater importance to those with_non21iberal language attitudes. Hypothesis X Attitudes toward change and politicel control are related to the perceived language liberalism.of reference groups and to feelings.about.the reference groups. a) Change. orientation is related to'the language liberalism of family, subcaste, social class and “region. b) Change orientation.is related tO‘thB heterogeneity of language.liberalism.among family, subcaste, .social class and_region c) Change orientation is related to the importance and liberalism of family, subcaste, social class .Mand region. d) Approval of social change in India is related to the language liberalism of family, subcaste, social class and region. e) Approval of.social-change.in India is related to the heterogeneity of language liberalism among family, subcaste, social class and region. 46 f) Approval of social change in India is related to the importance and language liberalism of family, subcaste, social class and region. 9) Attitudes concerning where government power should reside are related to the language liberalism of family, subcaste, social class and region. h) Attitudes concerning where government power should reside are related to the heterogeneity of langu- age liberalism among family, subcaste, social classvandnregion. Overview. Ten hypotheses have been made.in this chapter: five language and culture hypotheses, three language and personality hypotheses and two personality and culture hypotheses. These hypotheses are complex: they involve a number of language variables and a number of language correlate variables. A diagram is presented below to give an overview of the hypotheses and the variables with which they are con- cerned. The roman numerals indicate the hypothesis positing a relationship between the two sets of variables. LANGUAGE [greetige ‘12— '1 \ 7 [Affection :L 1 a Liberal ism is i) LLiber em I ,iEt; lipecialization Ii \ 7 I Interaction jg LANGUAGE [_Liberal ism IL 31W. 7 (:gpecialization§ /’ [_Interaction _L & > PERSONALITY Attit des towa d e nd 29; it i cal Cont £91 ,. Change orientation Approval of social change in India Attitudes concerning where government power should 45V res i. d... e. - Attitudes toward Change and- Pglgtical Control Change orientation India Attitudes concerning where government power should }e__ reside * Approval of social change in 1Used only in Hypothesis IV 47 CULTURE ‘SOCio-cultural Factogs Rural-urban residence Amount of travel Social class Family tradition of forei _ study North-South residence in .____India_i P rceived Lan ua Attitudes Reference G u 8 Language liberalim Language preferencel Reference Grogp Importance Importance of reference groups Relative importance of reference groups Native Tongue Compgsitiog of C Indian un t PERSONALITY Attitudes toward Change and P litical C nt Change orientation Approval of social change in India Attitudes concerning where government power should reside CULTURE Socio- tu a Fa rs Ruralaurban’residence Amount of traVel Social class Family tradition of foreign study North~South residence in India ‘Perceived.Language Attitudes W Language liberalism Referggce Grggp Importange Importance of re erence groups 2Not used in Hypothesis V ot used with ”Attitudes concerning where government power 8110111 I] rnei a a” CHAPTER III DESCRIPTION OF THE SAMPLE Rgpglgtigg.and Sample The pOpulation selected for study was composed of the students from India studying at Michigan State University and the University of Michigan during the summer of 1964. Participation of the approximately 180 Indians at the two universities was requested by letter followed by phone calls. The process of letters and phone calls was repeated several times at both universities. The sample finally obtained consisted of 96 Indians, 55 from Michigan State University and 41 from the University of Michigan. Participants appeared in small groups, usually five or six at a time to fill out the questionnaire. (The question- naire appears in Appendix A). The only other person present while the students completed the questionnaires was the lanthor. Respondents signed their names to the questionnaire with the promise by the author that all information would be kept confidential. Approximately one-half.the population at each univer- sity participated inthestudyo _There.is nonasis.for speculating.about the.students who did.not respond to.the request for participation. eNo information was available on .what.native tongue they spoke and nothing was known about 48 49 their language attitudes. The data from Michigan State were collected during the summer at a time when many had gone on vacation and some had just graduated from the,university and were already en route to India. The data from the University of Michigan were collected in September and October. The data from the two campuses were compared and found not to differ in any significant reapect. The data were combined in the analyses and discussions which follow. Ninety-two of the respondents were_male,.four were female. The range in age was from 19 to 46 years with a mean of 27.5 and a median of 26.4 years. The distribution of ages appears in Appendix B. Most of the Indian visitors were studying at.the universities for academic degrees. The heaviest concentra- tion was at the graduate level with over half the sample (52 percent) studying for the doctoral degree and 27 percent for the master's degree. Seventeen percent were studying for a bachelor‘s degree and four percent gave miscellaneous acadunic objectives. The distribution of academic objectives is presented in Appendix C. Engineering was given as.the academic.specialty of nearly 45 percent of the students. The only other fields of study.for.five.or more students were biology, pharmacology and mathematics.and statistics. The academic specialties are presented in Appendix D. The sample.included individuals who had been in the United States a fewmmonths and ranged all the way toithose 50 who had been in residence well over four years. Approximately one-third of the students had been in the U.S. one year or less: another third had been here between 13 and 30 months, and the last third had been in residence over two-and-a-half years. The mean time spent in the U.S. was slightly over two years. Appendix B shows the distribution of time in the U.S. The majority (some 61.5 percent) of the Indians were Hindus. .Another 17.7 percent listed Jain, Islam, Sikh or Zoroastrian as their religion: ten percent gave a Christian religion, five percent claimed atheism or no.religion,.and an- other five percent declined to answer. The distribution of religions is found in Appendix F. Two-thirds of the students had been working before they came to the U.S. Only onemthird had definite jobs to go back to in India. See Appendix G. The majority of the Indians had spent most of their lives in large towns or cities. Very few claimed a back- ground that might be considered rural in nature. Table 1 shows the size.of.city where.students had lived most of their_lives. The lack of students from villages and.small towns is unfortunate in that it.renders investigation of rural-urban differences very difficult. Most of the Indians come from Northern statesqalthough several Northern states are not represented in this-sample: .Assam, Himalch Pradesh, JammusKashmir,,and,Madhya.Pradesh. Table 2 shows the distribution of home states. The.states 51 listed under Dravidian Language Areas are identified here as Southern states: those listed under IndomAryan.Language Areas are considered to be Northern States. TABLB_1 Size of City Where Most of Life Lived Number in Percent of 5.3.921. 539212 Large city (300, 000 or more population) 51 53.1 Large town (100, 000 to 300, 000 pOpulation) 14 14.6 Medium town (20 000 to 100 000 ion) 14 14.6 Small town (1,000 to 20, 000 pOpulation) 13 13.5 Village (1,000 or under population) 1 1,0 No.response 3 3.1 Tbtal 96 99.9 grayel Bight different measures of travel were included on the questionnaire. ‘Three.dealt with travel in India, three withtravel outside India andatwo with vicarious travel. ’ The three.questions dealing with travel in India asked: a) the number of different states in which the respondent had lived for 6 months or more: b) the number of cities,.towns or villages in which he had lived for.over 6 months,-and.c) how'much.traveling he.had done in India generally. 52 TABLE 2 Home State or Territory m Number in Percent of 5 my; 9 8 mg; g Dravidian Language Areas Andhra Pradesh 5 5.2 Kerala 3 3.1 Madras 9 9.4 Mysore 2 2.1 Indo-Aryan Language Areas Bengal 3 3.1 Bihar 4 4.2 Delhi 6 6.3 Goa 1 1.0 Gujarat 9 9.4 Maharastra 29 30.2 Orissa 3 3.1 Punjab 7 7.3 Rajasthan. 2 2.1 Uttar Pradesh 1 2 l 2. 5 No response 1 1.0 Total 96 100.0 _fi This_lastwmeasure had five alternativesjranging from “very much“ to ”none“. The Indiansappeartohaveheen fairly mobile in India. A11.but 10 percent hadclivedin.more.than one.city as shown in Table 3. On the average theyfihad lived.in four cities located in two states. They did not change state quite sorreadily.as.city: .overwBS percent had lived.in-three .or fewer states. .Table 3.shows.the number of.states in which respondents had lived. 53 TABLE 3 Number of Places Lived in India Number Number Percent Number Number Percent of in of of in of Cities Sample Sample States Sample Sample 1 10 10.4 1 35 36.5 2 15 15.6 2 29 30.2 3 20 20.8 3 18 18.8 4 18 18.8 4 5 5.2 5 8 8.3 5 2 2.1 6 6 6.3 6 2 2.1 7 4 4.2 7 l 1.0 8 4 4.2 8 0 0.0 9 3 3.1 9 0 0.0 10 6 6.3 10 0 0.0 11 0 0.0 12 1 1.0 No response 2 2.1 No response 3 3.1 Total 96 100.1 Total 96 100.0 Mean Cities 4.2 Mean States 2.2 Std.Dev. 2.5 Std.Dev. 1.6 Over half the respondents indicates a background of ”much“ or “very much" travel in India. Table 4 presents the responses to the general question on travel in India. Only five people said they had.travelled 1itt1e.or.not at all in India. The question of travel abroad yields rathercdifferent , results. .Nearly 80 percent (161respondents) had not been :outside of India before.their.trip.to the United States. , Eighteen. respondents had travelled. outside India. Table 5 slabows the number of.trips taken outside India exclusive of the sojourn to the 0.8. 54 TABLE 4 Travel in India m General Amount Number in Percent of ..5L___¥sL_f Tra _sml—‘B s _msl-e... Very much 22 22.9 Much 29 30.2 Some 39 ' 40 . 6 'Very little 4 4.2 None 1 .1.0 No response 1 1.0 Total 96 99.9 TABLE 5 Travel Outside India T s t Ind Cgugtrigs Visited Number Number Percent Number Number Percent of in of ‘ of in of Times .Sample r. Sample Cogntrieg ggpgig ngple 0 76 79.2 0 72 75.0 1 14 14.6 1 7 7.3 2 1 1.0 2 3 3.1 3 0 0.0 3 1 1.0 4 1 1.0 4 l 1.0 5 1 1.0 5 0 0.0 6 1 1.0 6 l 1.0 No response 2 72.1 No response 11 11.5 9 Total. 96 99.9 Total 96 99. ,0f.tbee18mrespondents withtforeignetravel,etbirteen indicated,they.badespentlover,ammohthuintateleast“onewforeign 1::untry: _seven.had livedtinmonetcountry, three.in two 55 countries, one in three countries. one in four and one.in six countries. Table 5 shows the number of countries in which the respondents had visited for a month or longer., Respondents were asked.to indicate howumany years they had spent outside India. Some confusion appears to have resulted from thianuestion. The question_askedirespondents about strayel- outside. India, notcountmg .the 11.5. -trip. Fortyefour.individuals said.they had been outside India one year or.longers Previously, however. only 18 students indi- catedntheylhadmeverumade.any.trips outside India. inf. Table 5) The discrepancy is probably due to some students counting the months spent in the United.8tates when they answered the question. In view of the confusion.about this question, it was not used.in testing‘hypotheses.about travel. The distri- bution of responses appears in Appendix.H. Vicarious contact with other countries was assessed by two questions. The first asked the respondent howwmuch reading he.had done about foreigncountries. Table 6 Shows the responses- .EXactly half said they had done isome”.read- ing. Over 37 percent had done 9much“ or ”verquuch“ and ’ -‘slightly.over_lompercent-hadmdone “very-little”sor inane”. In.general the students.indicated theymhadedone quitenambit ofimreading“aboutwforeignmcountries. ,Indianwstudents.downothenerallyfleome”fremwfamilies . haying extensive contact with nonslndians. See Table 6. ..Qver.20.percentwsaid.their.familieamhadmnotsuchwcontact,and -28 percent said there was “very little“.eontact. .iny.15nper- secentmcmmewtnom families where there was fimufih“ or “very mudh“ of such contact; 56 TABLE 6 Vicarious Culture Contact Reading About Foreign Countries Family Gogtact with non- I d'" s - Number Percent Number Percent in of , in of mm 1321212 Easels am: Very much 11 11.5 Very much 9 9.4 Much 25 26.0 much 6 6.3 Some 48 50.0 Some 32 33.3 Very little 9 9.4 Very little 27 28.1 None 1 1.0 None 20 20.8 No Responses 2 2.1 No Responses 2 2.1 Total 96 100.0 Total 96 100.0 Family Traditionhgg Foreign Study For.relativelymfew"Indianslinwthemsampleawasmthere awfamily tradition of -foreignstudy. Table 7 shows the number of generations in each family with educationroutside India. Sixty percent of the students were the first in their families tostudy abroad. Twenty—eight percent indi- cated.thatVoneror;more,familywmembers.firoma.single.qenera- tion.had_such study. Nearly always.thesemfamily.members' were,£rom.the«respondents'towngeneration.-e.g..ubrothers, sisters or.cousins: “sometimes they were from a.different gmeration, e.g.,_.s£ather, uncles. or aunts. ‘ .Most.of the students did not.have brothers.or sisters with past or current study abroad. .,.Twenty-five had one or ,more brothers withtfioreign study andnflourehadwsisters with 57 TABLE 7 Generations of Family Members with Education Outside India Number Percent in of Easels _iERL—S 9 None 58 60.4 One generation 27 28.1 Two generations 6 6.3 Three generations 1 1.0 Four generations 1 1.0 Five generations 0 0.0 Six generations_ 1 1.0 No response 2 2.1 99.9 Total 96 such education. Sixteen students had one brother with foreign education, seven had two brothers._one had three brothers and one had four.brothers. Two students had one sister with.study abroad. one had two.sisters and one had five sisters with.such study. Table.8 shows the countries in which brothers.and sisters hadstudied. The country in swhiCh brothers had most often studied was the United States. The 0.8. was.followed closely by England. For.sisters.the most frequentecountryenamed was.the audan. This is.because _all.five.sisters ofnonewstudent studied in.the Sudan. The foreign study of parents, grandparents. uncles and.aunts.is shown in Table 9. Six_indiuidualswhadwfathers with foreign study. three had grandfathers.M18.had uncles and.six had aunts. ‘Eor same.ofuthewindiuidualsmmore than one.uncle.or.auntflhad.studied.ahroad. 58 TABLE 8 Foreign Study of Brothers and Sisters r "_.' Number Percent in of ' Total Sampl e Sampl e Brothers: 34 United States 14 14.6 England 11 11.5 Germany and 0.8. 4 4.2 England and 0.8. 1 1.0 Germany 1 1.0 Burma 1 1.0 Sudan 1 1.0 Aden 1 1.0 Sisters: 9 Sudan 5 5-2 England 2 2.1 United States 1 1.0 Aden l 1 0 W The most frequent country of study was England. Twenty-five of the 43 sojourningjrelatives.studied.there. Thirteen studied in the U.S.. three in Germany, two in Switzerland. one in Australia and one in Belgium. 59 TABLE 9 Foreign Study of Parents, Grandparents, Uncles and Aunts m Number Percent in of Total Easel: .flmL.5 1° ' Father: 6 England 3 3.1 United States 1 1.0 England and U.S. 1 1.0 England and Germany 1 1.0 Grandfathers: 3 England 2 2.1 Germany 1 1.0 Uncles: 27 England 15 15.6 United States 9 9.4 Germany 1 1.0 Australia 1 1.0 Switzerland 1 1.0 Aunts: 7 England 3 3.1 Belgium' 1 1.0 United States 2 2.1 Switzerland 1 1.0 5mm The students averaged.27.and a half years in age. About half wereistudying tor.the doctorate and themother half for lower degrees. .Nearlynhalf were“majoringwinrengineering with theremainder scattered over other academiciields. On the average they had been in.the U.S. tun years. Over 60 per cent were Hindu. Tweethirds had workedheforeooming .to the U.S.: one-third had jobs to gowback to. duost‘hadwlived.in large cities.in India: almost.none had apent.tbe_ma}ority.of 60 their lives in very rural areas. ‘As defined by language area. about 20 percent were from the South and 80 percent from the North of India. The students had lived in four cities and in two states in India on the average. Theyconsidered themselves to have done much general travel in India. Only one-firth of them had been outside Indiaand onlyasi'x .nr .theae..individuals had .lived in more than one other country for a month orilonger. ‘ .Vicarious contact with other.cultures took place far more through reading than through.family.eontact with non-Indians. Over. so percentnofathe individuals were the first .in atheir family to study abroad. When.there were other family mhmbers.with.foreign study.they.usually belonged to the indi- vidual's own generation. When.the family members with foreign Astudy were not from the same generation as the student they were generally.his father. grandfather. uncles or aunts. sojourning family members from all generations.most often had travelled for study to England or to the United States. CHAPTER IV LANGUAGE BACKGROUND OF THE INDIAN STUDENTS General Language Baghgpgggd The respondents claimed knowledge of a large number of languages. ‘When asked to list the languages, subclanguages, vernaculars and dialects they knew. the 96 students in this .sample listed some 515 languages.. This averaged out to 5.36 languages per person. ‘Table 10 summarizes the languages H listed by the Indians. It isninteresting.to note that 8 students forgot to list the language they were using -- English. By far.the best known.Indian language was Hindi._listed by 88 of the respondents. laindustani is given.here.as_a.langu- age separate from Hindi because some of the respondents listed it that way. Hindustani_is a combination of Hindi.and Urdu. The next most frequently.named languages were Sanskrit, Gujarati. Urdu and Marathi spoken by 40. 38. 38 and 37 indivi- dualswrespectively. I 1 Hindi was given.as the native tongue of 20 of the respondents. Gujarati. Punjabi.and.Tamil.werenthe.only other .languages.given as thenative tongue by.tsn.or more.respond- ants. .The distribution.of native tonguesnis given in Table.ll. 61 Number Percent in of Lamas: {Souls male Assamese 1 1.0 Bengali 20 20.8 Gujarati 38 39 . 6 Hindi 88 91.7 Hindustani 4 4.2 Kannada 7 7.3 Kashmiri 0 0.0 Malayalam 11 11.5 Marathi 37 38.5 Oriya 4 4.2 Punjabi .28 29.2 Sanskrit 40 41.7 Tamil .18 «18.8 Telugu 10 .10.4 Urdu -38 39.6 English 88 91.7 .52 TABLE 10 Languages Known nAlthough India has states“organized.along«language lines.aancomparisoniofitheihome.statenandWnativemtonguemgiven by the.students.showed.more.than onenlanguage listed as the native tongue for residents of halfthe states.. Table.12 . shows .thehome state given by .-.each..st.udent._and.,his-_reportedv native tongue.. The state of Maharastramhad.by far the great- estndiversitylofmlanguage. .The explanation forathis is not immediately apparent. .Whethernthe.multiplnwlangnigfifioPor . state. reflect theactual intermixture. ariangnagainina .states or indicates a great,.deal [of.ntravel on..the pm‘ of ins students is not clear. 63 TABLE 11 Native Tongue Number Percent Native in of 222199.? Easels seam Hindi ’ 20 20.8 Gujarati 15 15.6 Punjabi 11 11.5 Tamil 11 11.5 ‘Marathi 7 47.3 Telugu S 5.2 Bengali 4 4.2 Oriya 3 3.11 Konkan i* 3 3 . l Kannada 3 3.1 Malayalam 3 3.1 English* 3 3.1 Hindustani 2 2.1 Sindhi*‘ 2 2.1 Rajasthani* 1 1.0 Kannies* 1 1.0 Kutchi* l 1.0 Marwari* l 1.0 No response 1 1.0 Total 97+ ‘— W * Not an official language _ + Total is,97 instead of 96 because one student named two languages: (Kannada and Tamil). A rather intriguing aspect of the language data is that several students listed as.their native tongue a.langu- age other than that spoken with their families. VOften a student listed two or three languages as being used with his family and then gave one of these as his.native tongue. For ten students, however. there was no.overlap between thenlangu- ages used with the family and the.1anguage.listed as the native tongue. Five of these ten individuals gave Hindi as 64 soscsnx use needs one: eosmcou o>ausc smog} acoosomnou one now czocx no: ensue 03mm. .suaaocn sea osmcou obsess ones? endogenous use new caosx nos euuuu.osom emm bin me NMN NN c-h-lr-Iu-l '1‘ 'Il' .‘.|‘.l! \ .l\ L I]! . i .suvnsusnsz as: ensue OED; ones? unaccomnou one new caosx nos moose» e>uusz . w i 4 NH N b m m : m M50) :40 as. a a snowman sen» MMfl as means use sauna -uunns Inna uneuo saunas sauce I... ,III ill lll.}, IIIIIIIIIII‘ I‘llllll " I III- ‘slll huouuuuea no ensue 050m cameos o>ausz menosommouuou one “huouwuuou Nov ousum meow .euseocomeou sous» Mom euoaasoocu ceaussuousne. H n v m N m m m . eases” . H . . mosh—cum M «asses: assassin «ansunonnsw. «apnea «asunsoaemw . easement m suasmnaszo .... sosccsx m sundae «request a assess a1 m fineness” Issuance. N m «Guam, mum 14m. mmm once. mam mam mmmmsmm mmmmmw lean tam abwuflz us: sonz_uuos nuances . . . i. ' 1% ‘A ‘I., NH flnmdfi row—... ..~— (Aww'~"rm-.-' 65 their native tongue. One individual remarked to the author, ”I am answering this question as an Indian, and even though I do.not like Hindi or speak it well,nas.an Indian responding to an American. my.native tongue is Hindi!“ Most of the Indians considered themselves tonhave good command.ofeHindi. «On.thewauerage,-two~thirds:ofeallnthe:Hindi speakers rated themselves in the Weanellent“~or igood“mcete- gories. See Table 13. ‘Mean fluencytratings were_calculated by assigning ”excellent“ a value of 4, “good“ a value-of 3, lfair“ a 2,-and “poor? all. Thenmeans_indicate.that-the.best performance in Hindi.was.in understanding the speech of others. This wasfollowed by.reading.f1uency. Third was speaking fluency and last was writing fluency. TABLE 13 Fluency in Hindi Excel- Mean ‘ Std. Lil—st _LGOd Isis 29.2; __sl'1‘ot 2.1.2295): .siDv Reading Fluency 38 30 13 7 88 3.13 .94 writing , Fluency .26 32 20 9 87 .2.86 .86 Speaking Fluency ”30 33 18 7 -88 .2.98 .93 Understanding Speech .38 36 7 5 86 .3.24 -83 Half of the sample (48irespondentslnclaimed”some.know~ ledge.of Sanskrit. “Earnfewer“knewtthenethereclassical”langu- ages, Pali and Prakrit. Pali was known by-only one person: 66 Prakrit by four. On the average Sanskrit was studied over three and a half years. Table 14 presents these figures. TABLE 14 Knowledge of Classical Indian Languages Number in Percent Mean Years Language Sample of Sample of Study Sanskrit 48 50.0 3.67 Pali l 1.0 2.0 Prakrit 4 ° 4.2 1.0 The figures in Table 14 conflict with those in Table 10 where only 40 Indians claimed knowledge of Sanskrit and no one claimed to know Pali or Prakrit. The figures in Table 10 came from the first item on the questionnaire requesting the individuals to list all the languages they knew. Apparently eight individuals forgot to list Sanskrit, one forgot Pali and four forgot to list Prakrit just as eight forgot to give the language they were using, English. Data for the table above, 14, came from the third item on the questionnaire which asked: "Can you read any of the following?" and then listed Sanskrit, Pali and Prakrit. The Indian students in this sample claim knowledge of more languages and dialects for themselves than they claim for their parents. They listed an average of 5.4 languages known by themselves (Table 10), 3.9 known by their fathers and 2.8 by their mothers. Table 15 shows the languages reported to be spoken by parents. 67 TABLE 15 Number of Languages, Dialects, Sub-languages and Vernaculars ’ Known by Parents ‘ ' m Number of Father Mother Both Parents* Languages Number Percent Number Percent Number Percent 1 6 6.3 22 -22.9 6 6.3 2 12 12.5 23 24.0 11 11.5 3 21 21.9 20 20.8 19 19.8 4 22 22.9 15 .15.6 22 22.9 S 15 .15.6 8 8.3 16 16.7 6 11 11.5 5 5.2 12 12.5 7 7 7.3 . 1 1.0 9 9.4 No Response 2 2.1 2 2.1 l 1.0 Total 96 100.1 96 99.9 96 100.1 Mean Languages 3.9.5 12.82 4.08 Std. Dev. 1.62 1.52 1.70 fir *Number of different languages known between the parents. The discrepancy in number of.languages.known by the generations together with the finding that the majority of students learned only one or tworlanguages.£rom their parents suggests asomewbat broader.and.perhaps less provincial back- ground on the part of_the younger generation. _ The reasons for learning 1anguages.beyondtbose learned from parents fell.into two classes: compulsory anhool subjects Lparticularly English .and.Hindi)uand speaking With friends. uneasianally.parents_or grandparents were lovers of English.literature and this prompted the Child to learn the language. The majority of Indians learned_only.onemlanguage from their parents, although a few learned.as many as four. 68 Table 16 shows the number of languages learned in this manner. TABLE 16 Number of Languages Learned from Parents Number of Number in Percent assesses. __2____5am 19 W 9 e 1 60 62.5 2 23 24.0 3 9 9.4 4 2 2.1 No Response 2 2.1 Total 96 100.1 Mean Languages 1.50 Std. Dev. .76 t d f 1! As ci ted es The students were asked to rate each of the fourteen official languages and English on how much prestige they accorded the language. Prestige categories used. for the rat- ings ...... ,s =-1-ligh prestige, 4 =-. Fairly high prestige, 3 a Medium prestige, .2 = Eairly low prestige. .-l = Lowprestige. Table 17 shows .the .mean prestige rating given each language by.the entire sample .nf students. .TheJangnages .are listed in.,nrder by prestigewithdanguages .assignedthe high- est prestige .ratings, appearingat .-the..tep of Atharlist. sanakrit. Englishwand Hindi .1ead.-the-1ist4n,.prestig.e. The small standard deviatinn indicates , agreenent mg the students with regard to the .high prestige “accorded... these .languages. 69 .TAELE 17 Prestige Accorded Languages by Entire Indian Group Language. Haitians: M...“ _shD Sanskrit 90 4.83 .37 English 84 4.62 .67 Hindi 83 4.39 .88 Hindustani 75 3.87 1.27 Bengali 79 3-56 1.12 Urdu 78 ‘3.33 1.24 Gujarati 75 ‘3.17 .1.23 Marathi 73 2.97 11.23 Tamil 75 2.88 1.29 Punjabi 75 2.79 1.23 Telugu 72 2.79 1.25 Kannada -72 2.56 1.15 Malayalam 71 2.42 1.16 Kashmiri 71 2.38 1.08 Assamese .67 .2.28 1.13 Oriya 69 2.26 1.18 Number Mean Prestige The ordering of the list-changes when.the mean pres- tige rating.forarlanguageiis.ealcnlated.only.from,the ratings made by individuals listing that language as their native tongue. Table 18 shows the mean prestige rating for eadh language when.it is rated only by speakers listing it as their native.tongue. .As might be expected,itherprestigemratings«accorded a language byrits native speakers were always higher_than those given the -eame-lang1.1age by the Indian group -as a :whole. Table lflflsbowswthe discrepancies between prestige ratingegmade.by the.entire..group and thesemadeby .native,.epealters. The languages for whichwthemgreatest.discrepancy.in ratings was notedare.Telugu,.Kannada. Bengali, Oriya and Punjabi. .Thetlanguages wherecthe,least discrepancy.was.found 70 TABLE 18 Prestige Accorded Native Tongue Number of - Mean Prestige ‘ t . . Native Tongue Ratings Rating Dev. Bengal 1 4 S. 00 . 00 English 3 5.00 .00 Hindustani 2 5. 00 . 00 Hindi 20 4. 80 . 37 Telugu . 5 4. 60 . 55 Gujarati 13 4. 38 .77 Punjabi 7 4.14 . 90 Tamil 7 4.14 . 90 Kannada 2 4. 00 1 . 41 Marathi 7 3. 71 l . 49 Oriya ‘ 3 3. 67 2. 31 Malayalam 3 2.67 1.16 are Malayalam. English and Hindi. Malayalam receivedrather low prestige. ratings by its native speakers: no other langu- age was lower. English and Hindi .on the other hand, received very high ratings by their native speakers, an opinion shared by the group as a whole as indicated by the small standard deviations foundfor the Hindi and English ratings in Table 17. Each respondent was asked -how much affection he felt for each of the languages he knew. Affection was espressed in five categories: 5 = Very much affection, 4 = Much affec- tion, 3 = Some affection. .82 = Little affection, l = No affection. As expected, the affection expressed by an individual for. his native tongue was generally very high. Eorrseven languages, Bengali, Hindustani, Kannada, Oriya. English, Konkani. and Rajasthani. every native speaker expressed the very highest affection. No language received a mean affection TABLE 19 Differences between Prestige Ratings Given Languages by Entire Indian Group and Native Tongue Speakers , Native Tongue Difference Language* Group: Mean speakerssMean Between Wines Lia—meat . Ra s W Telugu 2.79 4.60 21.81 Kannada 2.56 4.00 21.44 Bengali. 3.56 5.00 21.44 Oriya 2.26 3.67 -l.41 Punjabi’ 2.79 4.14 -l.35 Tamil 2.88 4.14 21.26 Gujarati 3-17 .4.38 21.21 Hindustani 3.87 5.00 -l.13 Marathi 2.97 3.71 -.74 Hindi 4.39 4.80 -.41 English 4.62 5.00 -.38 Entire'Indian Malayalam. 2,42 2.67 -.25 *Urdu. Kashmiri and Assamese do not appear on the list be— cause they were not given as a native tongue by any of the {individuals in the sample. rating less than 4.0 (“much“ affection). Table 20 shows the mean affection ratings. The affection expressed.for Hindi and for English were examined.in greater detail. Table 21.shows-the.affection felt for Hindi and English by all the speakers of these languages. The affection felt forHHindi was moderately high for the group as a whole. The.affection expressed.for English was even.higher Lthemean is 4.28 compared to.3.69 for Hindi). The difference between.the two means is not significant, however. (t test). 72 TABLE 20 Affection Expressed for Native Tongue Native Number of Mean Affection Std. Tongue Ratings Rating Dev. Bengali 4 5.0 .00 Gujarati 14 4.2 .1.16 Hindi 17 4.3 1.31 Hindustani _l .5.0 .00 Kannada 1 5.0 .00 Malayalam 3 4.3 1.16 Marathi 6 4.3 .82 Oriya 2 5.0 .00 Punjabi 10 4.4 .70 Tamil 10 4.7 .67 Telugu 5 4.8 .45 English 3 5.0 .00 Konkani 3 5.0 .00 Kanries l 4.0 .00 Sindhi 2 4.5 .00 Rajasthani l 5.0 .00 Kutchi l 4.0 .00 TABLE 21 Affection EXpressed for Hindi and English by Entire Indian ‘ Group Hindi English Number in Percent Number in Percent . ML etasnrls sample... was 5 Very mudh affection 23 24.0 39 40.6 4 Much affection 24 25.0 23 24.0 3 Some affection 25 26.0 5 5.2 2 Little affection 4 4.2 .l .1.0 ‘1 No affedtion 5 5.2 4 4.2 No Response , 15 15.6 24 25.0 Total - 96 100.0 96 100.0 Mean affection. 3. 69 4. 28 Std. Dev. 1.12. 1.04 73 Language Preferences English was listed by over half the sample as the most comfortablefilanguage for them. Nearly everyone gave more than one response to this question which asked: ”In which language are you most comfortable?“ Table122 shows that for 34 individuals English was the.first choice although.it was listed as a native tongue by only three individuals. .Hindi, the native tongue of 20.people,.was a.comfortable language.for 29 student 3. TABLE 22 Most Comfortable Language We. Eat Ems-1 bier 6822.291 2253.1 English 34 l 2 53‘ Hindi 18 l 29 Gujarati Punjabi Marathi‘ Urdu Tamil Bengali Hindustani Konkani Oriya Telugu Sindhi Malayalam Kannada l _L Niobium use-bum“? PP wwwhwmw [oi-4 .... H Nh . P Pwuwwwwquwo English.wasnby.far”theilanguageiconaidersdamostMuse- fnl professionally. Table 23fisbows.the,response to.the question. ”which language do you think will be most useful to you.in your profession in India?“ English alone was named by 65 of the 96 respondents with another 24'respondents 74 naming it in conjunction with another language. Altogether, 89 of the 96 individuals mentioned English as the most useful language professionally. Hindi was the secondmost frequently named language. Twenty-six.people_listed it alone or jointly with another tongue. When Hindustani was lumped together with Hindi. this total rose to 28. Four other languages were named: Sanskrit, Gujarati, Malayalam. and Pun jabi, with the latter given twice. TABLE 23 Language(s) Most Useful Professionally. Number in Percent of Language Simple SQEBla English - 65 67.7 English and Hindi 18 18.8 Hindi * 6 6.3 English, Hindi and Sanskrit l 1.0 English and Punjabi l 1.0 English and Gujarati l 1.0 English, Hindi and Punjabi l 1.0 English and Malayalam l 1.0 English and Hindustani l 1.0 Hindustani l 1.0 99.8 Total 96 . When the question was one of what_language was most useful gggiglly,.Hindi and English traded places. The.ques- tion read. "Which language will.be.the mostiuseful socially lunch-professional situations in India?" In.answer to this question, Hindi was mentioned 54 times.and Hindustani 5 while English was mentioned only 29 times. These two.languages were far ahead ofiany.others. Table 24.shows these figures. 75 TABLE 24 Language(s) Most Useful Socially Percent of __§smsle_ Number in Language _ésmnls__ Hindi 35 English 16 Pu English and Hindi Gujarati Hindustani Punjabi Hindi and Telugu Bengali Oriya Malayalam .Hindi and Urdu Punjabi and English Hindi and Hindustani Hindi and Punjabi Gujarati, English and Greek Marathi, Hindi and Gujarati Hindi, Urdu and Punjabi Hindi, Punjabi and English Hindi, Gujarati and English Regional language Regional language and Hindi Regional language and English Metropolis: English and Hindi. otherwise state language Unclear None No Response Total Hum HHPHHHI—‘HI—‘HHHwamq P‘ ta P‘ r4 96 OH” OOOOOOOOOOOOPI—‘waqm O O I 0 Hum HHPPHHHHPHHHprmqmm H O 1.0 1.0 1.0 99.4 English and Hindi also,led the list.of preferred languages. Table425_showsJthe.responeesato.tbe.question, “What 1anguage(s) do you.prefer?“ English dominated.Hindi: slightly fewer than two-thirds of.the respondents named English.and slightly under one-half named Hindi. 76 Two of the three miscellaneous responses in Table 25 belonged to students who declined to name a specific_language saying that it depended on the purpose for which the language was to be used. The third miscellaneous response was a ”No preference“ with the added note, ”. . .-but I certainly like my native tongue for sentimental reasons.” TABLE 25 Language(s) Preferred Qrder.in which Language Listed We 4...? rat “Second 2229. .M 29.91 English 16 Hindi ll Gujarati Marathi Tamil Punjabi Hindustani Urdu Bengali Konkani Sanskrit Oriya Telugu ah H P‘Fflbcse g.- H #5 F‘NF‘HBJNUUNF‘NUO$ H" huoh' hJétnhdn P‘N ...- b 'Nhowcobdnuun~Jq No Response to U Miscellaneous Thailanguagesrnamed withttheinext\highestifrequency on the listiof preferences and the.languages given inuresponse to “. . .,are there otherilanguages that.youilike?.Lspecify)“ were primarily North.1ndian.languages. .Tahle-26.shows the languages listed in response to theiquestion. inis.is not surprising in viewiofmthe fact ..that the sample .consisted of approximately 80 percent North Indians and 20 percent.Soutb Indians. 77 TABLE 26 Language(s) Liked W Order in which language listed Language First. Seggnd Third Fourth Total Bengali ll 14 Urdu 14 Hindi 13 Sanskrit French Gujarati Marathi German Punjabi Telugu Tamil Sindhi Latin English Russian Bhojpuri Assamese l Kannada 1 Portuguese 1 Marthali l Konkani l Rajasthani l Porvi 1 Memoni 1 All languages All languages-in India Many like Kashmiri and Sinohi HHMwamqqmm HP HHP PH ...» of: ...; l-‘ NHHmNPHHHHNfimN Hra nah» HNNHHHHHHHHMNMwwabmmq A question asking what languages were.notiappealing and one asking what languages were used.unwillingly were also on the questionnaire. Tables 27 and 28 show the responses to these items. English was never listed as a language whidh was not appealing, but three individuals said they used it unwillingly. Hindi ledithe list of the languages used unwillingly. Fifteen peOple gave this response. Hindi was listed only four times as not appealing, however. 78 The two languages most frequently considered as not appealing were Tamil and Telugu, both Southern languages. The third most frequently named as an unappealingilanguage was Marathi, a Northern language. Part of the reason for Southern languages leading this negative list may be because of the Northern dominance in the sample. It is doubtful that this is the entire reason however, because it was North Indian languages which formed the largest share of languages used unwillingly. Another reason for the little appeal of Tamil and Telugu is their harshness to Indian ears. When asked if there were any particular characteristics they associated with particular languages, most respondents answered in terms of sound, and in terms of literary achievements. Languages were described in feminine terms: sweet, soft, musical, sentimental, poetic and rich. The words ”sweet“ and "rich“ appeared again and again in the descriptions. Both words were used to describe the sound of the language and to des- cribe its literature. Rarely was a language described in negative terms. The languages most often described in un- flattering terms however, were Tamil, Telugu and Marathi. The complaint against all three was one of harsh and jarring sounds. One individual went so far as to describe Marathi as ”Like a horse talking”. In the case of Telugu however, an equal number of peOple considered its sounds as sweet and melodic. 79 As already mentioned, Hindi was used unwillingly by more individuals than any other language. The reasons for this are probably partly political and social since no one said anything unflattering about Hindi. It was most often described as known by a large number of.people,.easy to learn, sweet-sounding and.“the national language of the country“. Althoughrespondents were asked “when and why?" in connection with the languages they used unwillingly, only 40 responded. Very few made any reference to Hindi. Those that did said they didn't.like to use it because they didn't know it well. One answer reads ”Learnt because of compulsion, which is bad technique for.learning languages.” TABLE 27 Language(s) Not Appealing Order in which Language Listed Language i t Second Third Fggrth Total F rs Tamil 6 Telugu Marathi 6 Sanskrit 4 4 2 2 2 .... O Punjabi Gujarati Hindi Bengali 1 Urdu' 1 German Kannada l Assamese l Rajasthani 1 Chinese 1 Dagri 2 French 1 Russian 1 Konkani ? HPNPPw mm HP Hum HHHNHHHHNNweemddm 80 TABLE 28 Language(s) Used Unwillingly Number in Percent of Language We ——EL§-Sa‘“ Hindi 15 15.6 Marathi 8 8.3 Bengali 4 4.2 Gujarati 4 4.2 Urdu 4 4.2 English 3 3.1 Punjabi 2 2.1 Telugu 2 2.1 Konkani 2 2.1 French 2 2.1 Sanskrit l 1.0 Kannada l 1.0 Malayalam l 1.0 Tamil 1 1 .0 Miscellaneous* 7 7.3 *Six respondents said that all but two or three languages were used unwillingly. The seventh indiVidual answered, "Regional languages.” On a more positive note,.most Indian students.indi- cated that there were languages they would like to learn. Table 29 shows there were only five people who said there were none. Most people indicated one or two languages, a few wanted to learn all languages. What languages did they want to learn? Table 30 gives . the answer to this. Four languages were named by 10.people or more. Three of the four were non-Indian languages: French, German and Russian. The reasons given for learning these.languages were primarily those of: professional and scientific purposes, Ph.D. requirements, to be able to.read the literature in the language, and to have better communica- tion with the native speakers of these languages, For Bengali, 81 one reason predominated. Nearly everyone wanted to take advantage of its rich literature. TABLE 29 . Number of Languages‘flbuld Like to Learn ‘ W Number of Number in Percent of Lgnggages Sample SEEEL° None 5 5.2 One 22 22.9 Two 23 24.0 Three 11 11.5 Four 7 7.3 Five 1 1.0 Miscellaneous 5 5.2 South Indian languages l .1.0 Foreign languages 3 3.1 All or many languages 5 5.2 No Response 13 13.5 Total 96 99.9 The interest in foreign languages.expressed by the desire to learn them was corroborated by the answers given to the question, “Is there any value in.studying foreign .languages. Why?” The responses to the first part of the question are tallied in Table 31. Nearly 90 percent of the _respondents answered in the affirmative: eight percent in the negative and two percent did not respond. The largest majority gave broadening of outlook and understanding of.the culture of.others as their principal reasons. For some.the references to.better communications and understanding of other cultures were presented in terms of India. Apparently foreign languages to them.were other 82 TABLE 30 Languages Would.Like to Learn .A‘ _F_——... ___ __._ _A”..- -..--.- ”we: n -,__ M jPerfienwfmflk images: .3321.“ e _mnl—S a French 31 32.3 German 26 27.1 Bengali 22 22.9 Russian 1? Tamil Urdu Marathi Telugu Sanskrit Hindi Gujarati Sindhi Spanish Chinese Kannada Malayalam Arabic Assamese Greek Italian Latin Swedish Kanarise Persian Maharashtrian . H HHPHHHHHMMprwprmmqoq O oooooooowwwpwwwwuwwweq HHHPHHHHNMNNwwwkbmmqm Indian languages. A few people talkedin terms of access to theiliterature in other languages and_one said the ualue was in social advancement. Most of the individuals who sawino valueiin.learning foreign languages gave rather self-centered reasons. .They felt they knew enough to get by.and-that they would only.for- get languages_if they didn't have to use them. One individual replied that there was just “no end to.languages." 83 TABLE 31 Value in Studying Foreign Languages Is there any value Number in Percent of in studying foreign Sample Sample ...—lesseesssl... Yes 86 89.6 No 8 8.3 No Response 2 2.1 Total 96 100.0 Egglish_Langugge Background- Badkground in English was.assessed by means of five questions. Responses to the first question are.presented in Table 32. This table shows that the students began learning English between infancy and 14 years of age. The.sverage age for beginning English study was about seven.and one-half years. TABLE 32 Age at whidh English Language Study Began W Age in .Number in Percent of Ygars Sample Sample ‘ l 2 2.1 2 2 2.1 3 4 4.2 4 8 8.3 5 17 17.7 6 9 9.4 7 5 5.2 8 10 10.4 9 11 11.5 10 12 12.5 11 3 3.1 12 9 9.4 l 3 3 3.1 14 l 1.0 84 The Indian students claimed everything from no formal study of English to twenty-two years of such study. See Table 33. It seems likely that individuals claiming more than 16 years of formal study misunderstood the question and interpreted formal years of study to mean years in which they studied via thehmediumbf English. If their answers are assumed to be accurate, the mean years of formal English instruction *was 10.2 TMfiE33 Years of English.Language Study M Years of Number in Percent of Study Sample' Samplg 0 1 1.0 l 0 0.0 2 0 0.0 3 0 0.0 4 5 5.2 5 2 2.1 6 6 6.3 7 7 7.3 8 10 10.4 9 11 11.5 10 15 15.6 11 5 5.2 12 8 8.3 13 5 5.2 14 5 5.2 15 l 1.0 16 3 3.1 17 5 5.2 18 0 0.0 19 0 0.0 20 2 2.1 21 0 0.0 22 l 1.0 No Response. 4 4.2 Total 96 99.9 85 Instruction in English was given primarily by teachers of Indian origin. This is shown in Table 34. A number of individuals studied under British and/or American teachers as well. TABLE 34 Nationality of English Language Instructors Number in Percent of Natiogality Sample ngplg Indian 66 68.8 Indian and British a 12 12.5 Indian, British and American 5 5.2 Indian and American 4 4.2 American 1 1.0 American and British 1 1.0 Indian and Anglo-Indian 1 1.0 Indian and Scotch l 1.0 Indian and Spanish l .1.0 Indian, British and Spanish 1 1.0 Indian, Spanish and Italian 1 1.0 Indian, Belgian and Australian l 1.0 Indian, British, French, Irish and American 1 1.0 Total 96 99.7 Respondents were asked how similar they felt the spoken English in the U.S. was to the English they learned in India. Most individuals considered the English they learned at home to be somewhat or very similar to that used in the U98. See Table 35. Only one person went so far.as to pronounce it identical. Eight people considered it very dis- simil ar. 86 TABLE 35 Similarity of English Learned at Home to that Used in United States Spoken English in the Number in Percent of tgg gggntgies is: Sam e ngnlg _ Identical l 1.0 Very_Similar 37 38.5 Somewhat Similar 27 -28.1 Somewhat Dissimilar 17 17.7 Very Dissimilar 8 8.3 Other 4 4.2 No Response 2 2.1 Total 96 99.9 Respondents were a3ked.to indicate.how adequate they felt their English instruction had been. Everyone labelled his English instruction as fairly or completely adequate. Table 36 shows that no one checked any of the igadequate alternatives. TABLE 36 Adequacy of English Language Instruction How adequate do you feel Number in Percent of your English instruction Sample Sample was? ‘fiv Completely adequate 37 38.5 Fairly adequate 52 54.2 Somewhat inadequate 6 6.3 Fairly inadequate 0 0.0 Completely inadequate 0 0.0 No Response 1 1.0 Total 96 3100-0 —7 87 Attitudes about Bilinggalism Five questions were asked about bilingualism. In reaponse to the first two, Indian students showed more in- clination to endorse the necessity for.individuals to use more than one language than for a nation to have to use more than_one language. Table 37 shows.the.responses made to these two questions. The number to the.left of each endorse- ment category represents the value assigned.that category in calculating the mean.enddrsement. The means showrthat on the average there was a slight positive feeling for the necessity of an individual knowing more than one language and a slight negative feeling when it is necessary for a nation to do so. TABLE 37 Desirability of Using More Than One Language Pg; Eh! IDQIXIQHEJ For thg Ngfiiog Using more than one a Number' Percent Number Percent lagguage is a in Sample 9f ngple .in Samplg g§_§gmplg 5 Very good thing 21 21.9 14 14.6 4 Good thing 15 15.6 10 10.4 3 Neither good nor bad 35 36.5 17 17.7 2 Bad thing 21 21.9 43 44.8 1 Very bad thing 1 1.0 9 9.4 No Response 3 3.1 3 3.1 Total 96 100.0 96 100.0 Mean Desirability 3.37 2.75 Std. Dev. 1.10 1.22 88 Feelings about using more than one language were ex- plored further by the question: “How'much prestige do you attach to being bilingual?“ The_response categories were: 5 = Very much, 4 = Much, 3 = Some, 2 = Little, 1 = None. ReSponses to this question were.rather evenly distributed over all categories. See Table 38. “On the average, the response was isome“ whiCh tends to indicate a.rather non- committal feeling. Yet it was with great pride that most Indians announced the number of languages they knew to the researcher collecting the data. Americans, who know few by comparison, are generally quite impressed when informed that such-and—such an individual speaks five languages. In deal- ing with Indian students on an informal level one finds them most anxious to impress their American hosts by a recital of the languages they speak. Why then soilittle prestige given bilingualism? One possible reason is a false modesty in answering the question, i.e., playing down an accomplishment which is known to impress Americans.r Another possible explana- tion.is.that in Indiarlittle prestige is.attached.to bilingual- ism.because nearly everyone.there speaks more than one lan- guage.Consequently the student really does not attadh great prestige to this accomplishment even though he finds his host country impressed by it. ‘With the.language.issue such a_burning one in India and on the campus the question arose as tohowuindividu-als' felt about intermixing languages, i.e., using words and phrases from one language when communicating in.another.‘>- 89 If everyone felt rather strongly that some one language should be the language of India, then strong opposition to the inter- mixing of languages might be expected. TABLE 38 Prestige Attached to Bilingualism M Amount of Prestige Number in Percent of Attached to Bilingualism Sample Sample 5 Very much 17 17.7 4 Much 17 17.7 3 Some -26 .27.1 2 Little -18 18.8 1 None 15 15.6 No Response 3 3.1 Total. 96 100.0 Mean Prestige 3.03 Std, Dev. 1.32 Such was.not the case, however, The attitudes were primarily.those of indifference or approval. See Table 39. It would appear that part of the passion.aroused over national language is a vying for.the honor of having.the title. Part, too, is probably a strong fear that if one's language is not.recognized in this manner it wdll.fa11 into disuse and eventually die out and with it will die the culture that it represents. In any event the students indicated by their answers that for them, at least, the issue was not one of “keeping the language pure.” In keeping with their approval of intermixing.langu- ages Indian students generally admitted to ”some“ actual 90 TABLE 39 Approval of Intermixing Languages . m Approval of Inter— ' Number in Percent of mixing Languages SQEELG ngplg ' Approve highly 14 14.6 Approve somewhat 25- 27.1 Indifferent or neutral 28 29.2 Disapprove somewhat .16 16.7 Disapprove highly 7 7.3 Other 3 ,3.1 No Response 2 2.1 Total 96 100.1 TABLE 40 Actual Intenmixing of Languages Ahurr» . __ . .. Actual Intermixing of Number in Percent of Laggggges Practiged 53mple Sample Very much 1 3 1 3. 5 Much 14 14.6 Some 39 40.6 Little 21 21.9 None 3 3.1 Other 2 2.1 No Response 4 4.2 Total 96 100.0 intermixing of languages. Table 40 presents their ratings of how'much intermixing.they felt'they actually practice. 91 Attitudes about India's Langgage Problggs The questionnaire contained a series of items per- taining to the issue of an official language for India. As expected the students eXpressed considerable concern about India's_language problem. Only 7 felt.little.or.no concern while well over half were ”very“.or “extremely” concerned. Table 41 summarizes the degree of concern.expressed for.the language.problem. Even at Michigan State University at the time Hindi became the official language of India, feelings of regional- ism rose to the point where the students argued among them- selves and some of the language groups in the India Student Association broke off and threatened.to,form.small groups on the.basis of language or region. TABLE 41 Concern.for India's Language Problem Respondents were: Number in Percent of Sggple Samplg Extremely concerned .25 26.0 Very concerned 28 29.2 Somewhat concerned 35 36. 5 Not very concerned 4 4.2 Unconcerned 3 3.1 No Response 1 1.0 Total 96 100.0 92 Students expressed a high degree of willingness to accept Hindi as India's official language. See Table 42. Over 60 percent said they were “completely” or ”somewhat“ willing while 30 percent said, “completely“ or “somewhat unwilling! The.remaining.-10 percent ..fell in the neutral and no response categories. The willingness expressed by North Indians was com- pared to that of South Indians. The difference was signifi- cant.at the .05 level (t test).with North Indians more willing to accept Hindi than South Indians. TABLE 42 ‘Willingness to Accept Hindi as India's Official Language Number in Percent of W! _nle..5am ...SERLW Completely willing 45 46.9 Somewhat willing .13 13.5 Neutral or indifferent 5 5.2 Somewhat unwilling 10 10.4 ' Completely unwilling 19 19.8 Other 2 2.1 No Response 2 2.1 Total 96 100.0 When.asked what they would choose to be Indials official.language, it was.Hindi and English which led the list, Over half wanted Hindi alone._and. about a third wanted English alone. .Another five percent.of the respondents wanted. both Hindi and English and six percent wanted Hindustani. 93 Only one person named anything else -- Sanskrit. Table 43 shows these figures. TABLE 43 First Choice for India's Official Language m Number in Percent of Language Samplg ngELE——- Hindi 52 54.2 English 31 32.3 Hindustani 6 6.3 Hindi and English 5 5.2 Sanskrit l 1.0 No Response 1 1.0 Total 96 100.0 Hindi and English were also the leading second choices for India's official language. Over one-third of the sample named one or the other of these two languages., Hindustani was given twice: Sanskrit, Tamil, and Punjabi were each mentioned onoe..Tab1e 44 presents the second_choices. It is readily apparent thatnforntbisrhighly.educated sample the choice of.official.languagenfor India boils down to two alternatives: Hindi or.English. JNearly everyone in the sample gave one as his first choice and the other as his. second choice. The questionnaire went one step_further and-asked each respondent to give reasons for his choices. For Hindi the reason was almost always that of-oommunication: “far.more Indians understand it than any other language. Occasional references were made to preservation of cultural heritage 94 (Hindi is Sanskritic in origin) or to nationalism (an Indian language for Indians). There were five major reasons given for choosing English tofbe Indials official language. First and most frequent was international.understanding. To com- municate and deal with the rest of the world, many respondents felt English to berth; language. A second reason given.for the choice of English was that of scientific and technical progress. Another reason concerned unification of India without favoring any one regional language. Still another reason was that of equal job opportunities.for individuals from the North and South. The fifth reason was that of avoidance of the considerable disruption in government in- volved in changing from English (now in use) to a different language. TABLE 44 Second Choice for India's Official Language .4 . . ‘ Number in Percent of _Langnage Samplg ng212 English 23 24.0 Hindi 11 11.5 Sanskrit 2 2.1 Hindustani l 1.0 Punjabi. l 1.0 Hindi and Hindustani 1 71.0 English.and Tamil 1 1.0 No Response 56 58.3 9 TotaLm 96 99. 95 The concern about the language problem mentioned earlier is further reflected in the strength of conviction about the respondent's choice for India's.official.1anguage. Table 45 gives the distribution of responses. The question asked was: How strongly do you feel about your answer to question 31 (a) above? (i.e., how strongly do you feel that the language you named should be the official language?) As strongly as possible Veryustrongly' Rather strongly Somewhat strongly Mildly ‘Only six percent feltcmildly abontrthe language prob- lem. Everyone else expressed stronger.feelings with half the sample saying they felt as strongly as possible. .Feelings were intense even among the most highly educated. TABLE 45 gggpggdentg felt: Sample ngple As strongly as possible 45 46.9 Very,strongly 20 20.8 Rather strongly 18 18.8 Somewhat strongly 6 6.3 Mildly 6 6.3 No Response 1 1.0 Total 96 100.1 7* The last question directly .concerned with ..the langu- age issue asked how much thenin'dividual. approved .,of the practice of drawing state lines according to language bound- aries. As Table 46 shows, this item.produced quite a range.of responses. 95 Over 57 percent of the sample expressed disapproval. Most of these respondents did so because they felt such a practice worked against the unity of India. At least 43 of the 57-said that provincialism, communalism, prejudice, bigotry, etc. were the result of drawing statelines by language boundary and the one thing India needed most of all was unity. While the disapprovers focused on the national scene, the approvers adjusted their sights to.the state.scene. The approvers were concerned with what would help a state function internally. In general they felt that drawing state.lines.by language boundary would greatly facilitate communication, education and government business within the state. A few said they approved because the cultural heritage, customs or interests of people in each language area are the same. TABLE 46 Approval of Practice of Drawing State Lines According to Language Boundaries “M “ ‘cen of * Deggee gf Appgoval ngple Sample Approve highly 9 9.4 Approve somewhat 1 2 l 2. 5 Indifferent or neutral 17 17.7 Disapprove somewhat .14 14.6 Disapprove highly 41 42.7 No Response 3 3.1 Total 96 100.0 97 mean ngera; Langggge Background! Most individuals spoke an average of 5.36 languages. The best known language was English, followed by Hindi, Sanskrit, Gujarati and Marathi. The most frequently.named as native tongues were.Hindi, Gujarati, Punjabi and Tamil. Most individuals considered themselves to.have.good command of Hindi. Half the sample had.studied Sanskrit while very few had studied Pali or Prakrit. The students claimed knowledge of more languages than their parents. Fathers were reported to.know approximately three.1anguages on the average. Students reported that they learned only one or two languages from their.parents, generally. Prgs_t_. gigs and Affection Assogiated with ngggages. Both the entire sample and native speakers alone assigned English and North Indian languages higher prestige than they did southern languages. North.and South Indiansalike accorded English the same high prestige. South Indians gave Hindi lower prestige than did North Indians. Individuals expressed very high affection for their native tongues. .Higher affection was expressed.for English than for Hindi. The North and the South did not differ in the affection they felt for Hindi or.for.English. Languggg Prefggggcgs. English was more oftennnamed as the most comfortable language. English was considered' the most useful professional.language: Hindi was the.second choice. Hindi was the most useful social language: .English was the second choice. 98 The most preferred language was English: Hindi was the second choice. North Indian languages were most fre- quently named as the “other languages liked.“ Southern languages led the list of unappealing languages. The sound of the language and the wealth of its literature appeared to be an important basis for making language evaluations. Hindi led the list of languages used unwillingly. Here the reasons appeared to be political.and social since evaluation of Hindi's sound was positive as were all other evaluations. The most frequently named languages the students said they would like to learn or wished they knew were.Erench, German, Bengali and Russian. The reasons given for desiring to learn the three non-Indian.languages were those of professional and scientific communication and.for better personal communication with the native speakers of these languages. For Bengali the attraction was its rich literature. Nearly 90 percent of the Indians felt there was value in studying foreign languages. Broadening of horizons and understanding of other cultures were the values of such study. Often the “understanding of other cultures“ referred to.other Indian language groups. “Foreign” languages meant other Indian languages as well as, or rather than, non-Indian langu- ages. Those who saw no value in the study of foreignilanguages responded in terms of immediate.personal gain. WW- 111- Indian-imam“ started the study of English between the.agesgof.one and 14. The average age was seven and one-half years. In general 99 they had 10 years of formal class-room study principally under Indian instructors. Most felt the English they learned in India to be somewhat similar to that spoken in the U.S. and they felt their English instruction to have been fairly or completely adequate. Attitudes about Bilinggglism. There was a slight tendency for the respondents to consider it desirable for an individual to have to use more than one language. There was* a slight tendency to consider it undesirable for a nation tO' have to use more than one language. In general respondents accorded bilingualism only “some“ prestige although opinions were widely split on this. There was indifference or approval of the intermixing of languages and a fair amount of actual intermixing of languages done by the students. Attitudes ghgut India's Lgnggage Prghlgms. A great deal of concern for India's language problem was eXpressed by the group. About 60 percent of the sample expressed some degree of willingness to.accept Hindi as the official langu- age and 30 percent eXpressed some or total upwillingness to accept it. Hindi was the first choice for India's official.langu- age by 54 percent of the students. Thirty-two percent.wanted English and five percent wanted both. Second choice was English for 24 percent andeindi.for 11.5 percent. The stu- dents felt quite strongly about their.language choices with 47 percent saying they felt “as.strongly as possible.“ _Forty— three percent indicated strong disapproval of drawing state 100 lines according to language boundaries. The remainder of the answers were widely split ranging from some disapproval to indifference to strong approval. If theipresent sample is any indication, Hindi.is becoming as widely known among.the educated classes as English. (Ninety-two percent of the.samplehclaimed_knowledge of it.) Hindi.received.rather high prestige ratings from the students with only Sanskrit and English more highly rated. Hindi was also.considered to be a comfortable language, second only to English in the number of times it was listed as the individual's most comfortable.language. In spite of such votes of.oonfidence, Hindi led the list of languages used unwillingly. Even among the most highly educated, Hindi, seems to be an extremely controversial issue. Slightly fewer than one half of the students were completely willing to accept Hindi as India’s official language. All degrees of acceptance were expressed ranging from complete willingness on the part of 47.percent of the sample to complete unwilling- ness on the part of 20 percent of the respondents. The picture painted by the answers to.the questions about choice for India's language is far too simple. The answers suggest that at least for educated groupsrlike this one. the naming of both Hindi and English wouldmplease.almost everyone. On an.intellectual level perhapsmthis is true,, but not on an emotional level. The same people who responded _to these questions beforequndi became the official language,' 101 bitterly argued with each other afterward and at last report had split into language factions on the Michigan State campus. Answers to the question asking approval of the draw» ing of state lines by language region showed that at least 40 percent of this group did not disapprove.and about 22 per- cent actually approved. The approvers.may.haveselected Hindi and/or English as.officialelanguages,.but their primary focus was still on the state or regional level. CHAPTER V Construction of Indices Many of the language and correlate concepts explored in this thesis were being.,_.investi.gated.quantitatively. for the first time. ,There were few existing studies.to guide the con- struction of data-producing.instruments. Because of this, a number.of questions were often used to produce evidence about a single.concept. Sometimes these questions were analyzed singly....sometimes.-indices were .formed.from,them. ._It hishthe purposeeofhthis”chaptereto”describehsomerof.thewmore.unusual measures...used and to describe , the...construction-‘.of...indices. No assumptions about homogeneity or dimensionality of. the concepts were made. when indices were formed..they were intended as explorations of the dimensionality of concepts and as aids in examining relationships among concepts. The indices-ars;quite.frankly first attempts._ Ho effort washmade to refine them. .It...is..hopsd...that.subsequent research will (do this. The.measures and indices are.presented here to.providee guidesto.defining.theflconcepts-andwto.improvingwoperational procedures.used.infiexamining the concepts. Inrthislchapter measures.of-thetfollowingrconcspts willwbe.describeds 102 1.03 L r i . To assess the degree of liberal- ism-provincialism of language attitudes, Language Liberalism Indices were constructed. Six.Language Liberalism Indices were made for each person (one for himself..and one each for his father, mother, subcaste, social class and region). Lnnsnagg.§n§s$nliza§ign. To examine the amount and kind of specialized usage of languages in differing situations Language Specialization measures were used. . Inhalafitign. Another concern of this research was with the.interaction patterns of thesrespondents.. One of the measurements of .this‘was aNehru Discussion Index which was intended to measure the amount of discussion an individual had within the Indian community (in the U.S.) relative to the.amount of discussion he had with Americans. W193. The fourth concept explored was change pnoneness. It wasexplored via.a personality measure called an Orientation. Toward Change Index. The Orientation Toward Change Index was composed of Confidence. Self-Percept and.Motivation scales. ‘Apn;gygl_gg_§ggjgl_ghgggg‘ Fifth is an Index of Social Change Approval. Thiseindex.measured.satisfaction with.on-going.social change in India. ' d I 533;gg. Sixth is a.series ofemeasures“concerning“attitudes about where government power should;reside. The purpose of these measures was .to‘determine.to whatsextent.thesrespondsnt felt power should be.vestedein local and/or regional govern- ments and to what extent he felt it should be vested at the 104 national level in the Central government. .§93131_£1§§g. The seventh and final measurement des- cribed in this Chapter is the Social Class Index. This socio- economic measure was based on per capita income, books in family.library, family.servants, father's education. mother's education. £ather!e.occnpation.and brothers and sisters attending college. u ' . ' ' ’ Discussions in the literature of the Indian.language problem and.informalconversations with Indian students ' suggested a number of issuestahout-which attitudes seemed to reflect.a liberal or conservative position. .Questions con? cerning these issues were combined to.form a Language Liberalism Index. Eirst was the issue at.tolerance regarding other languages. 'Presumably, the Indian with a.liberal outlook is tolerant of other languages. i.e., his emotional involvement with.his own tongue. and his_fears for its prestige.and sur- vival are not so great as to blind him to the values of other tongues. .The liberal Indian does notrbdindly dislike scores of other languages nor does he use other languages grudgingly. A “second...issue waslthatuofemixing mm: andn-.phrases from one language into another or switching hack.and forth among imam. The liberal .-indiszidual does ..not worry that his .own .-.langnage will .be -.contaminated “by. switching back and torth among languages or.by-horrosing words and phrases.£rom other languages. .105 Indeed, the Indian who is most liberal considers it an accomplishment to know’more than one language. This seems to be another aspect.of liberalness concerning language. The. individual.so~complete1y tied up.in his own language that.he finds-it difficult.to,see.muchvalueinother_languages would not be expected to attach much prestige to being bilingual.“ Conversely,.theuliberalflindividual would be expected to recog- nize the merit in other languges and be proud to know'more than one. The last indicationaof liberalism is probably the most obvious one for India-attitudes toward Hindi. To the extent the individual is willing to put aside his own person- al desires and, for the sake of Indian unity, accept Hindi as the official languagerto that extent be is a.very liberal man in India. These attitudes.of liberalismaconservatism.were.ex- amined by means of six items on.the questionnaire. The items. were: What languages are not appealing? What.languages do you use unwillingly? When peOple know and use several languages, there is a possibility that they will use words and phrases from one language when using another language. How much do you.approve of using words and phrases from one language when communicating in another language? Approve highly Approve somewhat Indifferent or neutral Disapprove somewhat Disapprove highly How'much do you think that you do use words and phrases from one language when communicating in an- other language? 106 Very much uch Some. ittle None a: AAAAA 0 w much prestige do you attach to being bilingual? Very much Much Some Little None EEEEE ll I'll How willing are you to accept Hindi as the official language in India? Completely willing Somewhat willing .Neutral or indifferent Somewhat unwilling Completely unwilling cher,($pecify) The numbers in.parentheses at the left of the response categories did not appear on the questionnaire but are given above toiindicate thescoring of the responses to each item. Thus a response of “Approve highly" to item 27 was given a score of 5 while ”Disapprove_highly“ was given a score of.l. The direction of scoring for the last four items above is such that the higher the score, the more liberal the language attitude is considered to be.. For the first.two.items,.languages which are not appealing and languages-used unwillingly, the scoring was reversed. .fiincetanhigh.score,-i.e..:manyalanguages_named, , would not indicate..-a..lib.eral-.attitude. A but -.;an-.-intolerant .one iinsteadlwthemitemsiwere”scored bytcounting the,numberxof languages listed and subtractingwthettotal from 6. The .resultlofwthistconxersion warmthatmthese.twoiitems were then e.escoredx.in..thc-.samemanner -as...the-last.four: a highsco're indicated liberalness and a low score, non-liberalness. ”107 Responses toall six items were summed to form a Language Liberalism Index for the respondent. In addition to the items asked about his own language attitudes, each reapondent was asked the same six items about his parents,. his subcaste, his social class and his region. Distributions of responses to all six items for each group appear in Appendix I. The items were scored in the same manner as for the.respondent and summed to form indices. In all, six .separate_indices“weretcalculatedtforteachtrespondent:nonetfor- rhtmself and one each for his.father,.mother, subcaste, social claSs and region. Indices of_language.liheraliem for.individual Indians (“Self" indices) varied from 9 extremelytconservative to 34 ,extremely liberal with a mean of 20.3. Indices for.fathers' varied from 9,to 29 with a mean of 19.6 and those for mothers from 15 to 22 with a mean of 17.9. ,Subcaste indices ranged from 11 to 25 (mean =.16.6),.social.class.indices ranged from 9 to 23 (mean =,.16.7). and. regionalindicesurangedhfrom 14 to' 25 (mean = .18. 4). Table 47 shows the. distribution .of..indices. for each group. he Table 47.shows,wtheretis yery.littleydifferenCe among the groupetin mean liberalness. “There is quite a difference in the variability, however. This isteuidenced . by hath. the standard, deviations .and the. tange. Most..notice-. -tahly differenttare the indices for.the mothers whichthavesana ~extremelytrestricted.range. «This is-phobably partly.due to thetsmall .sampl.etsize... .‘mta ,for ..all. -items-in the Index were 108 TABLE 47 Language.Liberalism Indices m.. Index Self Father Mother Subcaste Social Region .Clgss .... N [0 MN HNNwNH Hmnwawa H wP mum H NMuH HH HHN H HPwHwhNMN H N N H NHwbwuwmqmwQQQbmh HHHH 34 Tdtal 80 24 ll 19 16 22 Mean Index 20.33 19.58 17.91 16.63 16.69 18.36 Std. Dev. 4. 75 4.94 2. 39 4. 36 4. 32 2.99 available.for only ll.indiuiduals. Thetresponsemfor.mothers was often "don't know” or fishetdoesnlt care“. The impressions Againedtfrom,thetresponses,totquestionstconcerningtmother's .attitudes istthat.Indian women haveilittlettaste.for social and political.issues.or that sons discount.any attitudes. .their mothers might have to the point of not being able even to recount what thetattitudestare.- 109 Thetindices for the individual,..,his father, mother, subcaste, social class and region were intercorrelated to see how much similarity individuals perceived. among their groups. Table 48 shows "these correlations. Thetnumberstin parentheses are the samplesizes. Only three of. the,..fifteenfcorrelations were signi- ficant. This indicates that some -students did perceive these groups _to have, general attitudes about language which were definable and to some extent independent of each Other. The attitudes attributed by the individual to his groups were not merely projections -of. .his-own..or..those of his family. “In. fairness it .must .be pointed out, however, .that “thevxcorrelaq tions werenbasedton. very small “samples. Except for correla- tions with Self, thesample was...generall.y,tunder .10. Correla- tions with Self were basedwonll‘to -22 observations. The samples were small because Language Liberalism Indices were calculatedonly on completedata. Nearly every person answered some questions about .all his groups, .-but unless all questions about any one group were answered, no index was calculated for that group. Correlating further compounded the missing data problen because two complete indices are necessary for correlating. The-correlations that .mre.-ei:gnificant were those between the individual, and his. father, between the subcaste — and. the social.class. andhetween the social “class and the region. The correlation between the liberalism of the student and his father .is meaningful. Repeatedly in -later analyses 110 the relationships found between an individual's attitudes and correlate measures were paralleled by relationships between the father's attitudes and the-same correlate measures. The significant relationships between.subcaste and social class and between region and social class suggested at first glance that the individual might have lumped all these groups into ,one and ascribed a single set of.attitudes to them.. There are.reasons to.reject this possibility, however. [Kirst, the correlationmbetween«subcastewanderegion was extremelymlow and1 insignificant (r=.08). secondly, the large amount of.missing‘ data,argueuagainstethat.interpretation-.”Studentsmwhowdidrnot answer.the.series1of questions about a.group generally wrote,- “Don't know.” Sometimes this wasrfollowed by a lengthy ex- planation of why it was not possible to generalize about one or more of his particular groups. TABLE 48 Correlations Among Language Liberalism Indices Self Father Mother subcaste Social Region - Qiass ' 7 Self -— ’ Father .60**(22) .- Mathar ,.15 (11) .52(7) _- subcaste .07 (17) -11 (s) .54 (6) -- Social Class.46 (16)-.24 (6) -.35 (5) .se**(7) -- Region .04 1(21)2.42 (9) .17 (a) .08 (9) .67*(11) A—r —' 3’ * significant at the .osmlevel ** Significant at the .01 level. 111 The number of languages listed as unappealing to the Self was correlated in turn with the number of languages listed as unappealing to each of the reference groups. Responses to the same_item were intercorrelated among refer- enze groups. In like.manner- the responses to each of the. other five items forming theJLanguage Liberalism Indices were intercorrelated across groups. .The item by item intercorrela- tions for the groups appear in Tables 49 to 54. The number in parentheses by each correlation coefficient refers to the size of the sample on which it was computed. There were only four significant correlations.out of 15 among the responses to the question concerning.the number of languages not appeal- ing. Positive and significant correlations were found between attitudes ascribed.tos ,subcaste and self: -subcaste and- father: social class and father: and social class and sub- caste. When the item was the one concerning the number of languages used unwillingly, only three.intercorrelations were significant. Perceived attitudes-of social class and sub- caste correlated significantly: and perceived attitudes of the region.intercorrelated significantly with those of both the subcaete and social class. There was high.agreement perceived.amongtthe groupsy i. n tem.of. their. Approval-..ef-.intermixing .-lenguages. Twelve. of the fifteen relationships wereeeignificant. sThe only non- significant relationships were between subcaste andself: subcaste and father: and region and mother.; 112 A.fair number of.relationships were found among the reference groups in terms of the actual intermixing perceived to be practiced. fEight of the 15 correlations were signifi- cant.. Significant relationships were found between: Self and father: self.and social class: father and mother: father and.social class: .mother and subcaste: subcaste and social class: subcaste and.region: and social class and region. I All the correlations for both the prestige attached to bilingualism and for the acceptance of Hindi were signifi-. cant. On these two.ieaues, students saw high correspondence among.their.reference groups.and between themselves and their groups. The intercorrelation of Language Liberalism Indices among reference groups indicated far less correspondence among their perceived language attitudes than did the item by item intercorrelations. How'much of the correspondence reflects accurate perception of reference group attitudes by the student and how much reflects projection.of the studentls own attitudes is not clear. It is relevant to note, however, that the sig- nificant correlations.between.groups for individual.items and between groups.for.thelLanguage.Liberalism Indices were not always the correlations between the Self and reference groups.. Often they were.between reference groups. In these cases, .correspondencemindicateseeither.homogeneity of attitude A(based on accurate.perception) or a ladk.of clarity about the .groups and.the ascriptien of a.general.set of attitudes to them. .ao>oa no. on» an nanuaueamaa . 113 Amy me. avv ehmo. wee mm.u «my oo. not an. mv MAQ‘B we o oo. in . coo.a me e .8. .oav mm. “Hat om. acumen Ah v .Mh. condo deacon meat .He. cyanogen «h v mm. uenual em V mm. unease “a“. ”fw*< A.-. c " “~P ‘a. " ...“ ..—- r—-. . u Ho>mH do. 030 an ”CGOHUHGDHQ em A“- ‘. 114 I.’ ’~e-—-.‘—-§-' ~W-- 1"...” aa>oa mo. «nu an ucuusmecmem . nu ANHV..¢F. ANHV .oe. we~ ~¢.u as V oo. “oHV Hm. acemom II AHHV scum. “my mm.) A0 y 00. m0 v who momdu deduom II AmV mN.I Am V 00. “HHV aw. ouueunom n- 83 8. SC 8. .8502. .... E V an... autumn _ ..- «in . mama «flaw. «flawed «and Hum M .l‘ri ||( 1‘ on Henna” Mflmflfi enhance: : Hmcaaaa3co can: nomssmcsas moosd escapedouuoo .soumueucH Ho>oa Hoo. one as ucuuaueamem.wm Ho>oa do. on» an unoueuacmem .m Ho>oa mo. an» an uaoueuenmem . 5 1 1.. Anhveeeom. Ammv emu. II Aooveeewo. 1-. I ) i'l“ ll. \l‘ll Illll Achy NH. «moveemm. Avaeehm. I‘CII. Hm flflmda imoV .«om. AeeV...mv. Ahmveeehv. fidmv..-MN. nonveeedh. Manse: :Hebouem¢ mcdxaeueucHs wooed oceaueaeuuoo msoumuoucH «VBV emu. “may no. imee.emm. «beveemm. codmem «undo deacon cannonsm ensues Moscow “Hem “1§~*~— H¢>oa doc. or» an unaudueamewcww Ho>oa H0. or» an unavauaamem cm Ho>oa no. on» an unsceusemem . 116 g Admvee¢Mo Amhveeemfio IOGHU HGdUflm Amhveeevmo ummmmmmm momw vH. mmhw--NN. adbveedm. mm flamdfi mmhw mo. .5 Ammv . Hm. ‘ .mNo Avaeeemw. [I menseeez.=mo«xaeueucn H05904: ocoe< unedueaeuuoo mmoummowom m¢m~ ..HN. “why eevm. mmeV mo. «Mme .hH. Avaeeehm. Godmam neeau Heauom eunounom H.300: Hannah “Hem -?--—-..~' --———.~H HO>¢H H00. 0&0 an Ho>oa Ho. 050 um ucooeuecmemcww HGGUHNHGmfim It | ' ' - .l I. | ‘ a: .omV...m¢. Ame...me. ievewwve. imeVmwamw. w~m~.««~m. convex m II Nomveeemmo :hvwwnfim. “Ohvwaemo. mauveeemm. moeao Hewoom II Ammveeemm. «Hoymwwwo. “moveawoo. unmounfim II Avaecceh. wthaawmw. menace II Ambveeeomo Hunudh nu meow 3mm... 3 3 «Adam undue «am Ill Illlv."|| . enhance: --=Eu«aesmcewam OB oecoeuu< omaumeum= wooed mcofiueaouuoo esoumu.oucH mm mamdfi Cu- Ho>ma goo. any so unaueueamem... 8 ll Ahhveeemm. «Obveeemmo «mmvwwwboo Adbyelwmmo “Hmvccmbm. 60%?“ M II Aboveeehm. Ammywwwmn. whmvwwwmm. wmhveeeuh. condo anduom I: AHoVacahn. «Hovwwwmo. «HhVaeemo. eueeonsm ll 3K. veeedm. «MN. Veemom. H0304 In “mchccmh. . uenuem ......) “Ham nadmlm. duddmldddudm. «Huuuflflm munmqm udfluflm “Adm (I, E. monsoon: geccqm no oucougouue. mace: ucoeuuaouuoo esoumuouca. .vm Nan. -N._. ... .‘ ‘ .r“‘< WW“- . ‘o- ”‘4 ...—.-. -.. v ”fit. .ll9 The Language Liberalism Indices were examined from one further point of view. Inter-item correlations were calculated within each group. The intereitem correlations for the individual as well as tho.se..for-his.-father. mother. subcaste, social class and region are presented in AppendixJJ. Briefly these intercorrelations show the following. -Eor the individual. (Self) , two. significant inter—itemcorrelations were found and all items correlated with his Index. For the father, two.significantminterqitem correlations were found and all-but.one.item.correlated with his Index. tor the mother,.twoesignificant.intereitemecorrelations were found and only two items.correlated with her Index. Eor-the-sub- caste five significant inter-item correlations were.found and three items correlated with the Index. For the social class three significant inter-item correlations were found and two items correlated with the Index. For.the region two significant inter—item correlations were found and two items correlated with the Index. All.in.all, the.intereitem correlations indicate very little homogeneity within the Indices. For every group there was a significant correlation-between approval of inter- mixing words.and phrases among languages.and the actual.inter- mixing of words and phrases practiced by the group. The prestige attachedcto bilingualism often.correlatedcsignificant- ly with either approval-or.practice of.intermixing. ‘ 120 Language Specialization Another way of dealing with language behavior is to ex- amine its usage by occasion and by audience. .At.the simplest level for a bilingual this involves using one language rather thanlanother when communicating with.certain groups of people or when.communicating on certain topics. To measure such specialization of usage in India, three questions were asked: “with what particular people in India do you use what languages? Please explain.“ ”In what particular places in India do you use what languages? Please eXplain.“ ' ”On what particular.occasions in India do you use what languages? Please explain." In response to the first question, some individuals listed different languages for different peoples, e.g., Hindi to servants, Marathi to family and English in school. Other respondents listed only a few languages, e.g., Hindi and English and._followed.each.language .witha detailed 1131: of groups of people with whom it was used. .cher Indians listed three or four groups of.people and indicated they used two or three particular languages withmeach. .Because of the diver- .sity.of-response.patterns.tbe.number.of.different.languages listed by each.individual was counted.and in addition the .number.of particular people listed was counted. Table 55.gives the frequency distribution of.people.and languages for question 10: "With_what particular.peoplein India do you use what languages?“ . 121 TABLE 55 Language Specialization by People m Number of Number Percent Number of Number Percent Languages in of People in ' of Mentioned l 4 4.2 1 9 9.4 2 32 33.3 2 33 34.4 3 34 35.4 3 27 28.1 4 14 14.6 4 9 9.4 5 5 5.2 5 8 8.3 6 l 1.0 6 2 2.1 7 1 1.0 8 l 1.0 No Response 6 6.3 No Response 6 6.3 Total 96 100.0 Total 96 100.0 The second item used to measure specialization requested the places in India wheremone.language rather than. another was used. .Here.as.incthempreceding.item.some.respond- ents gave the same number of languages as places, others gave many places with few languages and still others gave few places and many languages. As many.as.eight.languages.were given and as many as seven places. Both.languages-and places. were tallied. Table 56.shows the distributions of languages and places given in.answer.to.the.questiona ”In what parti- cular places in India-do you use what languages?" ‘ The.third item concerning language specialization requested the.languages used for certain occasions. ”Ae‘in thelpreviousequestions.both"the.languages,named.and.the occasions given wereecounted. Table 57.shows the number of .122 TABLE 56 Language Specialization by Place Number of Number Percent Number of Number Percent Languages in of Places in of WWMWMW 1 5 5.2 1 7 7.3 2 23 24.0 2 41 42.7 3 26 27.1 3 17 17.7 4 19 19.8 4 11 11.5 5 4 4.2 5 3 3.1 6 2 2.1 6 2 2.1 7 2 2.1 7 1 1.0 8 1 1.0 No Response 14 14.6 No Response 14 14.6 Total 96 100.1- Total 96 100.0 languageswand.occasionsngivenein.answer.to.the.questionc "On what particular occasions in India do youmuse what.langu- ages?" TABLE 57 Language Specialization by Occasion M Number of Number Percent Number of Number Percent Languages in of Occasions am Easels .W camels female 1 7 7.3 1 9 9.4 2 35 36.5 2 34 35.4 3 20 20 8 3 .16 16.7 4 6 6.3 4 9 9.4 5 3 3.1 5 1 1 O 6 6 - 2 2 1 No Response. 25 26.0 No Response 25 26.0 Total. 96 100.0 Total 96 100.0 123 Interactien Another aspect of behavior of concern in the research is interaction patterns. One of the measurements of inter- action was an index constructed to measure the proportion .of people with whom the student communicated who were in the Indian community and the proportion who were outside this community. Thisindex was, called the Nehru Discussion Index. Jawaharial Nehru died shortly before this study was undertaken. The event was one of considerable concern to the studentsand was discussed by all of them. The situation pro- vided. -an- opportunity .to obtain ..a...specific. measure .of . Indian- American interaction. Withthisin mind, the students were asked: ”With how many Indian friends .did you discuss the event?" “With how many American. friends did you-.discuss the event? “ To construct the ..index,...the-nmnber .of Indians with whmihe event was discussed ..was ..divided ..by the totalnumher , of, individuals with whom. discussion ..occurreCL Theresulting ratio varied from 0 to 1.00 with 0 indicating ..no discussion with Indians and.l.00-indicatingediscussion to be entirely with Indians. Table .58 shows -the .-distrihution _of...the Nehru Discussion Indices. 124 TABLE 58 Nehru Discussion Indices W Index Nugber in Pegcent of .9000 - up 14 14.6 .7500 - .8999 8 8.3 .6000 - .7499 2 2.1 .4500 - .5999 6 6.3 .3000 - .4499 0 0.0 .1500 - .2999 11 11.5 .0000 - .1499 41 42.7 No Response .14 14.6 Shanss_9riss£stiss An index of change proneness or change orientation was created by combining the; Qonfidence.Self Percept and Motivation scales. Since the scales.as well as the index derived from them were used in thefianalyses of the data, each scale will be.described in detail. A description of the Qrientation Toward Change Index derived from them follows the description of.the scales. Enniigsncfl. One.component.of change proneness was theorized to be self.confidence. The individual who lacks self confidence.is not likely to venture into new fields and try newuthings. For him, failure is too.costlyoto self esteem. He ...cannot attord to risk tenure-especially since. he is al- ready convinced failure will come. 125 At the other extreme are the individuals possessing a high degree of self-confidence. More Optimistic, they do not eXpect failure to occur constantly nor do they fear the con- sequences of failure should it come. It isnot so damaging to self esteem. The degree of self-confidence for each individual was measured by a Confidence scale previously used by the researcher on American students. Thescale was composed of 18 items. Each item was followed byna continuum on which the ,respondent.placedoa check.tomindicate.the extent to which he felt the item.described.his behavior. Sample.items are: In your present business or school gnoup,~howvmuch do you think others will look down on you if you .express a.deviant opinion on an important issue? _— Very much inot atfiall Do you usually mull over decisions you've made and worry about whether or not you've made‘the right decision? Elways ' ' never The items concern the extent to which the.individua1 is willing.to.express a deviant opinion. howrmuch he.enjoys interacting with others..the degreewof confidence he feels he has..thenamountrof-support”he.feelsnhe.getsflfrommothersaand the extent to.which heworriesmahoutwthe course of.action.he. takes. The.,items.comprising -the..,scale.-appear. in Appendix AK. Responses techs. items were? scored ..hy ”use. of ”an over- lay whichhmarked.eff the continuumofor each item into seven equal intervals. Items worded in a negative manner and con- tinua labelled.in a negative direction were.reversed.so that 126 every item was scored in the same manner: the higher the score, the greater the self confidence indicated. Items were summed to form a Confidence score for each individual. Higher scores indicated higher self confidence. Scores for the Indian sample ranged from 59 to 119.7 Actual possible range was from 18 to 126. The mean Confidence scorecbtained was 84.02 with a standard deviation _ofJ. 3. 7. Table 59 shows the.distribution of Confidence scores. Split- half reliability for the sample was .69 (corrected by the Spearman-Brown formula). TABLE 59 Confidence Scores Confidence Number Percent Confidence Number 'Percent Scores in of Scores in of magma ___* -melsimsls. 115 - 121 1 1.0 80 — 86 12 12.5 108 - 114 3 3.1 73 - 79 15 15.6 101 - 107 4 4.2 66 - 72 8 ' 8.3 94 - 100 10 10.4 59 - 65 5 5.2 87 - 93 10 10.4 No Response 28 ‘ 29.2 §§Qj_£3;gep§. Another component of change proneness was hypothesized to be theoindividualis picture of himself in terms.of.howrmuch.he.likes to explore new worlds. Does-he see himself as.an individual who likes.to try newnthings,or.as someone who prefers _the “tried andtrue." Even .theindivi-g .dual with great.selchonfidence.may.be a great.traditionalist. 127 He may not like modern ways. He may have .made a very successful adjustment to the traditional world and be very confident and self assured in coping with it: he may see no reason to change his world by trying new things. At the other extreme are the individuals who see themselves as "young moderns.” To be a “young modern“ in a changing society implies that the individual must be ready to change with society or even to initiate change in the society. Such an individual should be change prone if his self image is accurate. The degree to whichan individual pictured himself as an adventurous person who liked to try.nsw things was measured by the Kossoff Self Percept Scale. 0n.this.scale Ahigh.scores indicate an openness and.positive1approach.to experMmenting with new ideas: low scores indicate a cautious approach and unwillingness to explore the novel. i-e., Low scores indicate preference for old and accepted ways of doing things. The scale consists of 16 statements to.which the respondentqindicatesMhow’stnongly_heuagreesxor“disagreesflby means of the following system: .5. agree very much: .4. agree somewhat: 3. neutral: 2. disagree somewhat: 1, disagree very much. The.scale appears in Appendix L. Scorescfor the.lndianstudents on the Self Reroept scale rangedfrom 37 to 74 with a mean of 5.2-5 and standard deviationof 6.. 3. feasible range on -this scale .was‘from :16 to 80. Scores.for-theflsamp1e were normally distributed as 128 can be seen from Table 60. Split-half reliability for this data was .65 (corrected by the Spearman-Brown formula.) TABLE 60 Self Percept Scores Self Percept Number Percent Self Percept ..Number Percent Scores in of Scores in of _mL33 16 Easels * Easels Easels 69 - 74 1 1.0 49 - 52 28 29.2 65 - 68 2 2.1 45 - 48 12 12.5 61 - 64 4 4.2 41 - 44 5 5.2 57 - 60 10 10.4 37 - 40 3 3.1 52 - 56 17 17.7 No Response 14 14.6 flntixaiign. The third and last scale in the Orienta- tion Toward Change was a Motivation scale. The possession of great self-confidence and a willingness to try new things by themselves do not guarantee that one will bother to seek out new'methods and.ideas or adoptdthem or even advocate their adoption by.others. Naome impetus.tofibehavior.is needed. The‘ true innovator must see the value of newmideas-and-processes and mustflsee the need_for,their1tria1 and adoption. He must he motivated.to.seek out, to.try, and to accept advancements. Combining. these three. elements produces a descriptionof in- novatorsrandmhighly.changesprone people as individuals with great self confidence who see themselves.as interested in new ideas and who are motivated to try out and adOpt those new ideas. The Motivation scale was constructed from five ques- tions. The questions used were: 129 As people grow older sometimes their level of drive and motivation changes. As for yourself, would you say that your drive and motivation is: Much higher now than it used to be somewhat higher now than it used to be _____About the same now as it used to be somewhat lower now than it used to be 'Much lower now than it used to be Some people are said to be strongly motivated and to have a lot of ambition and drive. Others are said to be less motivated and to have less ambition and drive. How'much motivation would you say that you have? Very much ____;Much Some ‘Little None As compared with the rest of your family would you say that you are more or less ambitious than they are? Much more ambitious somewhat more ambitious _____About the same in ambition Somewhat less ambitious Much less ambitious As far as you.are concerned, how important would it be to have more drive and ambition than you now have? Extremely important Very important Somewhat important Not very important Unimportant Some people are always trying to get ahead and improve themselves and others seem to be satisfied with their present condition. As far as you yourself are concerned would you say that you are: Much more ambitious than most people . Somewhatmore ambitious than most people About the same in ambition as most people Somewhat less ambitious than most people Much.less ambitious than most people The distributions oeresponses to each item are pre— sented in Appending. The1responses indicate that in general‘ individuals_felt they.had.much motivation and that they were: more ambitious than most.people. There was a slight tendenCy 130 to consider themselves more ambitious than their families. Overwhelmingly they felt their drive and motivation to be greater than in the past. There.was considerable variation in how important it was to them to have more drive and ambition. In general, however. it was felt to be important to have greater drive and ambition. When the students compared themselves to their families, about 59 percent rated themselves as having more ambition and a third saw little difference between themselves and their families. Explanations offered for responses to the family and Self comparisongindicated that those who saw no difference between themselves and their families considered their families as very ambitious. Those who saw themselves as having greater ambition than their families saw travel as the key. Some_used.trave1.as the 9proof“. saying they,were‘ the only ones in their family to travel abroad for.education and this was evidence of their higher motivation. Others saw travel as the motivation. ‘In travelling they sawuother cul- tures..other ways of doing things and this made them want to do something great for.their.ewn-country or to achieve some personal goal. I Responses-to,the.five-questionswaboutmotivationflwere .summed to form a Motivation.score. High.scores-indicatedfa high degree of. motivation. .lowscores -.a.,1ow- degree. A The.range of scores.possible was from 5 to 25.' Scores for the.sample-ranged.from 12 to 25. The.mean was 20.1.andrtbe.standard deviation, 2.7. Table:61.shows.the distribution of Motivation scores. .131 TABLE 61 Motivation Scores Motivation Number Percent Motivation Number .Percent Scores in of Scores in of £2221: .éemele . Easels ‘éamnls 25 - 27 6 6.3 16 - 18 18 18.8 22 - 24 17 17.7 13 - 15 1 1.0 19 - 21 43 44.8 10 - 12 1 1.0 No Response 10 10.4 It wasinot feasible to calculate oddeeven.or split- half reliability on the five items comprising the Motivation scale so the items were intercorrelated. Table 62 shows the intercorrelations among the items- The general motivation item seems to correlate best with the other items; the import- ance of having more motivation correlates the least well. All.items.oorrelate significantly at the.-OOl level with the Motivation score. .Not.a11 items correlate with each other. however. Ihree items: -general-.motivation. motivation- compared to others. and motivation compared.to.familyminter— relate significantly. Two items: change in motivetion.and importance of haying.more:motivationiirelated to only one‘ ,other item- If the last itemieimportance,ofhaving.more motivation,.werereliminated,frometheaMotivationesoore.nthen the general motivation item would interoorrelate.significantly‘ with all other items.and.mightwbe used in.plaoe of the Motivation score. 2 3 1.. once madman we eocuuuemEH Ammveeemm. «maveeemv. Ammo «a. cmuue>dmum ca emseno 0"." neuoneez.seaue>«uoz.mo084 macaueaeuuooueusH Aomveeemm. “HGVOMNo Ammv NN. Naneum an veummaoo Hosea Hoe. on» no uaoueuaamau.em a¢>oa do. one an unouauacmem .w Hebea no. es» we unmouuaomwm a Ammvaeahv. ammveeewm. Ava 00. “may..00o mam“ .-fiH. .Aomveewm. Avaiiemfio mmmvCC§Hmo II Avmveeedm. aflqflmfl awesome enunum 0» vehemeoo l‘I'll'r.‘l"‘ll'q fiAmauez ones msa>sn we euseuueaEH seaue>auez ca emceno maueeu on teammaeo «Menus on vehemeeu coauo>euoz_auuoaoo -W~,---.——" e v.1““ 133 Orientatigg nggrd Change. scores on the Confidence, Self Percept and Motivation scales were combined to form the Orientation Toward Change Inder. Data on all three component scales were complete for only 58 of the Indians. Orientation Toward Change Indices ranged from 123 to 187 with a mean of 156.0 and.standard deviation of 15.0.. Tahle.63 presents the distributionaof Orientation Toward Change.Indioes. TABLE 63 Orientation Toward Change Indices ==========aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa: Orientation Toward Number in Percent of _éamnle__ . 181 - 190 3 3.1 171 - 180 8 8.3 161 - 170 10 10.4 151-160 13 13.5 141 - 150 18 18.8 131 - 140 4 4.2 No Response 38 39.6 Table.64 presents therintercorrelations.among the major components.of the Qrientation.1oward Change Index.- Confidence, Motivation and self Bercept correlate significantly wdth,0rientation.Ioward Change, but not with each other. The .low insignificant correlations among scales show that they do not overlap: Jeach measures somethinquifferent. wAnnrezal_e£_§esiel_2haase. I D theological step.beyond the assessment of an indivi-' rduallselanguage.attitudes,andwthemeffectnofichangeeproneness ”134 TABLE 64 Intercorrelations Among Components of Orientation Toward Change Confidence Motivation self i figment Confidence - Motivation .18‘(64 u self Percept f -.02 (61 .08 (76) Orientation Toward.Change Index -88***(58)-34**158) -40**(58) * Significant at the .05 level ** significent at the .01 leVOl ***Signiflcant at the .001 level on them is.to examine how the individua1.feels.about the change currently taking place in India. Language decisions are only one part of the,change.occurring,as.Indiamattempts toeuniteuand,advance technically. It was theorizedithat.the individual who was most progressive and.most desirous.of.see-4 ing”Indiaradvance”technically”andneconomically‘wouldwfesli .thatuchangeawasnoccurring.toowslowly,Norwifnnotwtoonslowly,2 . at least not too.qui:kly. nubile this.individual might wish- to hastenwsocialichange, he.wouldmin general approve of the directionmof”socialmchange,mbut”wishetoflseemit.occuruon'a; ”broader.p1ane. In line with these expectations..the following .questions were usedmtonaesess”theietndentle‘generaliapproval of the changes occurring.in his.country. India is changing in many ways. How do you personally .feel,about the changes that are being made? The changei-eregggggg made: "-wMuch~tee~qntCKiy' fiemewhat tee quickly __+_About right 135 somewhat too slowly Much too slowly Among the changes being made: All are in the right direction Most are in the right direction About half are in the right direction and half in the wrong direction. Most are in the wrong direction Allare in the wrong direction The changes being made are: Much too large in scope (size) Somewhat too large in scope (size) About the right scope (size) Somewhat too small in scope (size) _____Much too small in scope (size) Table 65 shows the.distribution of responses to the first question which waseconcerned.withethe speed of social change in India. Responses were concentrated in two cate- gories. The majority of individuals felt the changes were occurring somewhat too.elowly or were about right in speed. No one felt they were taking place much too quiCKly. TABLE 65 Approval of_Socia1 Chang. SPdi ‘ Number of Percent of ' The changes are being made: 3 r 5 ‘MuCh too quiCKly 0 0.0 4 Somewhat.too quickly 6 6.3 3 About right 35 -36.5 2 Somewhat too slowly 40 41.7 1 Much...too..,elaowly .-s .s...3‘ No Response 7 7.3 Total 9.6 ...ioo..1 136 The.direction of social change was pleasing to a majority of Indians in the sample. See Table 66. Nearly two-thirds of the respondents checked the alternative: “Most (ZEhanges7 are in the right direction“ and another 24 percent replied that they liked about half the changes and didn't like the other half. No one felt that all the changes were , in the wrong direction. TABLE 66 Approval of Social Change Direction ' Number of Percent o Among.theichangesiheingumadel ._Batinga_ ‘_§gnglgp_. 5 All are in the right direction. 5 ..5.2 4 Mostare in the right direction 61 63.5 3 About half are in the right direction and half are in the wrong direction 23 24.0 2 Most are in the wrong direction 4 4.2 1 All are in the wrong direction 0 0.0 No Response 3 3.1 Total 96 100.0 The distribution of responeee.,for-the question con- cerning ..the scopeof India's social-changeuis‘givenin Table 67- The...distribution was- fairly normal with .a majority of responses in. the middlecategory -- social .. changes ..are about right in scope. Only one-person went so far-as .toneay that social changes were. too small in scope. Responsesctowthe .threeprecedingitens were ._summed .to . create an Index of. emu Change Approval. .Theuscoring for the first it» (.speedofehange) anduthe third .item (scope of 137 TABLE 67 Approval of Social Change 5C°P° Number of Farcent of The changes being made are: 3:21098 Sample 5 Much too large in sc0pe (size) 7 7.3 4 Somewhat .too large in scope( size) 12 12.5 3 About the right scape (size) 44 45.8 2 Somewhat too small in scape(size) 27 28.1 1. Much too.small in scope (size) 1 1.0 No Response L 5 5.2' Total 96 9 change) maintained. .in WW Index. . The ,highest possible.lndex was 15 which could be achieved only by indicat-. ing that changes in India were occurring much too slowly and were too small in scope. but.warenin the right direction. The lowest possible Index. 3. wouldmindicate attitudes that changes were much too rapid and too 1arge-in scope and were all in the wrong.direction.. Actual Indicesnfound for the sample ranged.from 7 tolls with anmean of 9.1 and a standard deviation of 1.6. The Indices are shown in.Table,68.w .TABLE.68 Index of Social Change Approval , A Index of Number Percent . Index of Number APercent Social .in mof Social “in ‘tof Change Sample Sample Change Sample Sample W___._ ...... Assam __ *7 13 4 4.2 9 .13 13.5 12 10 10.4 8 8 8.3 11 24 25.0 7 1 1.0 10 30 31.3 (No Response 6 .6-3 138 Zhe approval of social changecmeasures.intercorrelated significantly with each other, but only two correlated signi- ficantly wdth the Index of Social Change Approval. See Table 69. Approval of.social.change.speed and scope corre-- lated significantly with the Index: napproval of social change direction did not. The approval of direction of social change contributed little to the Index in its present form and might well be eliminated. -: C... P933; measures. The tendency to feel that the* region or state is the most important government unit or conversely.the.tendency to feel that the Central government must dominate state and local units waslfelt.to be a reflec- tion of the respondent's identification. If the primary identification is as an Indian,.i.e., with the country as a whole, this should be reflected.in feelings that the Central government must have.more power.than the state governments. If. of course, identification is primarily at the state level. e.g., "I am a Bengali“,.thennthis«identification should appear in the form of a desire.to.accord.the state government the most power. Attitudes toward where political power should reside were.assesaed by questions,ooncerning the power.the_student . wished to accord ,variousegovernment levels and theimportan‘ce he attached to Specificwgronps. I The first question aSKed.thewindividual.howumuch power he felt the Central. state and.local governments ."shouldr 139 Hosea Hoo. on» an uaaoauacmew.mm Hmsma Ho. on» an unnoeuacmew cm Hosea mo. on» an unconnecmam c Ammveeevh. .omv .ma. tome saw. cacao aaumuuuua emseso Heaven we Huseummd ,omceno Heaven no He>ouma4 menswear— emceno Heaven {Ill lililril Ill Ahmvccavo. Hm>oumm4 «manna Heaven «0 xeocm Abmv yawn. omoum omcoso Heuuom no He>ou004 Ammv econ. coauuonan «mango Heaven no He>oum94 II.. 009mm omseno deacon no He>ouom¢ undue omcmso Heaven no He>oumm< , we ae>oume4 ms054 oceaueaewuoouousn mm fldmdfi “...... - _,., 140 each be accorded. The response categories were: 5: Absolute power, 4 2 Great power, 3 = some power, 2 = Little power, and 1 I No power. Table 70 shows the responses to this question. The Indian students accorded the Central government more power than the state government .and the state government. more power than the local government. W. In addition to using power ratings m g; to test, hypotheses, three derived measures were also used. The power accorded the state government was subtracted from that accorded the Central government to yield a. difference-power measure. .11: likenanner the power rating for the local government was subtracted from that for the state government to yield a second difference-power measure. The third difference-power measure was the difference between the power measures for the Central-.and.local governments. These difference-power measures provided a‘measure of the extent to which the student wished to vest differing amounts of power in two government levels. The distribution of these difference-power measuresare given in Table 71. A constant of 4 was, ”added to .each measureto eliminate negative signs. Thus a difference-power score above 4 indicates the individual gave. the higher government level more power than the loaver: a..differences-.power, score. of ..exactly 4 indicates ..the -same amount of power given eachlevel: and a difference-power . score of less than 4 indicates “that the “lower government level was awarded greater power .than the higher level- The mean difference—power shownnin Table 71 ..indicates that there-- om. No. on. .bin .Ouu mm.~ mv.m md.v ue3om one: H.ooa . om o.ooa om a.mm om House N.v e a.~ N a.~ m eesomnem oz m. 1.. o.a a 0.0 o 0.0 0 scene oz H ~.m~ mu H.~ N o.H H noses causes m o.om we 93 9. m .3 me ~38 osom. m m.aa as m.o¢ mm o.om we segue venue e ~.v e ~.¢ v m.mm «m noses assesses m madman ammuuom nausea museums. «queen ummmmmm sees a no useuuem we uenssz no useuuem no Henasz us aseouem no Honesz m - uceschebew Hausa usessue>so ensum uneasue>ew Heuucou us seamen eucesfiue>oo Hausa use .eusum .Heuuceu cenueuo¢.uesom Oh MAQGB A‘_.‘..._...-._o—.-~ ‘ 142 was a slight tendency to favor the higher government level. This tendency was strongest when the difference in power was between the Central and local governments. 'Disagreement among respondents, as indicated by the variability of difference; power scores, was also strongest here. W, A second question was asked to determinethepower awarded government levels relative to each other. The question asked wast- When two of the governments are in disagreement, what do you feel the outcome should be? (Check one alter- native for each column) (a) Central ' (A) shim. m Central V8. V8. WWW]... a.should always prevail over B A should usually prevail over 8 -_ A should“pravail _ about half the time over B A should rarely . 'prevail over B _r r_ a.should never , imprevail over B Threeescoresrresultedmfrom this question: “an.9utoome score for when the Central andestate.governments were in diss‘ agreement.eoneforrwhen-itvwas.the.state.andrlocangovern- mentsrdisagreeingrand“amthirdwoneeforedisagreement.betwean ,Qentral«andrlocalmgovernments.» T‘ ‘ .Table 72 shows.the.entcomes desiredehen tno.govmmn-* ment levels disagree. “ltlissteadilyrapparent“fromttbettable that it was generally”felt“that,thetbigbermofuthe,two“govern-a ment levels should usually prevail.. This corresponds to the power ratings given the govemnment levels in Table 70. va.a o~.n H.ooa om ~.¢ w u mAH S .l «.mm «m a.>~ ow a.ma ma ~.¢ « H.~ a «dam «mg no no Henson useouem useadue>oo Haven asses Heuuseu Hesemueuceueuueo o.ooa . ~.¢ o.H, N.¢ o.am vomm Now 33m “0 uceouem mv. ooow 0m 0v vM usesoue>ow Hausa noses ousum - Hesemleuoeueuuan H5 MAflQB a.mm H.~ o.a o.a m.» n.5m m.a~ m.m o.a. «aueum H0 useouem mm. Oh.¢ om mo HoAEsz usesoue>ew eunum moses Houuooo- . Hesemaoooouomufin eeuseeez.uesefi I euoeueuudn .>en .oum Mesomeuceueuuwn one: Hosea oeeoonem oz m Nm'd‘tfltDl‘. llllflflfluuddflllll nesomleuseuewuun mN 2n mm o 05 o o>GG .09m 144 mm .m 003v «a .w efiooufio one: A . co." mm 0. con mm m . mm m m Hosea v.0 m ¢.m m m .m m. encased 02 m.@ m . o.o o o.o o .m. nope Hen>oue uo>oc nausea .4V H m.aa Ha m.> h H.m m .Amv . . no.5 3.26am hashes masons 23 m v.m m m.ma . ma v.oH oH “no uo>o use» on» . . , «it uses» Segue 33am .3 m 0.3 «m 525. ov 0.2.. mm “my . Hobo 325.3 a." Hoses odsofi 2v v m.mm hm a.mm . hm m.mm . pm «my . - . Moos Hanson—m stanzas Canons 33 m no no no .. mo ucmuuom Hone—92 ucuuuom Huflfiaz Hausa a“ momma “m. Oa> . On> u - . HuHudOU 2: 0.99.5 «(V Hnuucou 2: . , . shunned «Housed useaoueoeo cog nesoouso N h Manda. m 1.541; b ‘ ‘M— T Fauna..- “- ‘4“..— 145 W. A final measure of power was calculated. This waste single measure, calculated from the Outcome measures, indicating to what extent the student felt that. _theultimate power should reside in the higher levels of . government. Higher .Qut come measures favored [the higher of. the governaient levels, and.lower Outcome-..measures favored lower government levels. For..the Powerlndex the three Out- come measures were summed. Theresulting index was ..such that individuals with high scores favored the higher government levelsin cases of disagreement between levels: individuals with” lower scores iavored....thelower government level. Table 73 shows the Power Indices. ME 73 Power Indices -——-—-——— Power A ..II umber - Percent . Power 4. Number Percent Indices in of Indices in n§emnls. :fiannls i§amnle .flannls l S 20 320.8 -10 5 5. 2 l4 4 4.2 9 8 8. 3 l 3 l 2 12.5 8 3 3.1 1 2 20 20 . 8 7 ' 2 2. 1 ll 11 4.11.5 6 ,l a 1.0 .-No Response _.lO ..lo. 4. government level, ,.Difference—.Poner..measnres, Qutoomemaaures for mangovemment levels disagreeandrbeimr Index were _. intercorrelated. The Antercorrelatione Appeariin Table 74. ample sizes were large. ranging from 84 to 94. I .146 The correlations were as might be expected. The Difference-Power measures correlated with the government Power Measures from which they were created. The one excep- tion to thieiatne -correlation between the state government. power measure and the Difference-Powerhmeasure'for the state and local governments. Thepositive correlationsjound be— tween the Difiference-Power...measures -and the ..pomr accorded the Central government indicate thatnindividuals according the Central government muchpower. tendedto. {accord higher government levelsrelatively. more “power than .lowerlevel’s. The..negative.....corr.elatinns,.hetween. power accorded the state or the. local. government. and the. ..Qifferenceei’owergmeasures indi- -~ cated that individuals according highpower to state. and local levels tended to grant .the higher government levels still more power than these levels. The Outcome measures. all interrelated significantly and ’all correlated highly with the Power Indexcreated from . All of -themeamesof government power werederived from ..two.items on, the questionnaire. lathe lightest-.1113, ‘ it isninterestingeto ask -to matezrtent .thsneasuresof power overlap, i.e., _to _what.\.extent..ethey.measure thesame thing. Some,...-hut not template. overlap reassure. The Outcome, measure for Central vs stategovernmentlrel ated -.to ..both the Power meaame.,..for .the Central and .state governmentand. to the Difference-Power, measure for Central ”andustate. governments. » The. Qutcomemeasnre for state vs- local governments. related to ‘ the state government Power measure and to the Difference-Power 147 measure for state minus local government. It did not relate to the local government Power measure. The Outcome measure for Central vs local government did not relate to either the Central government or the local government Power measure. Neither did it relate to the Difference-Power measurelcalcu- lated between the.two government levels. The Power Index related to Central government power and to Difference-Power measures calculated between state and local and between Central and local governments.. The general impression gained from studying the government Power.ratings.and their derivativeSand the Out- come measures and their derivative is.that the two kinds of measures related most when it is the Central government which was the topic and least when it was the local government. W .An index of social class was constructed similar to that of Mangalam et.a1 (1960). As.in.Mangalam!s research. seven items were used: per.capita_income.wnumber.of.books in the family library. number of servants.,fatheris.education. mother's education. father!s.occupation and brothers and sisters with.college education. Inwthis“researchutbenlast uneasure.was modified and.themproportion.of.brothers and sisters «over lB.w1th college.education.was used in thewindex. Theufirsttindicator”ofnsocialiclassmused.wasiincome.‘ .Most of the Indians fall in.thewupper”incomenlevels. It-is misleading toeassnme the majority of.theasample to be well-ho- do based .on yearlyincomerigures. however. .sinceusome Indians CA -u¢>¢~ ucc- um ucmtvtuccuu chosen «no use .ucmoumacman use seduuaouuoo+w+ .He>ou do. us unsouuqcmun . - --- endows are usp .uneouuucmae use scuuoaeuuou w+ He>oa Hoo. Gnu us unsoauasmumemm .Ho>oa no. as unmouuacmau 4 Ho>oa Ho. onu no uaouewucmam .m nausea «nu use .uaauauuamuu you codenaouuou +w Ho>oa no. on» an unuoeuacmem . O'Come {CCHhe £5.60. . .mNe . itfifle fifle HNel. .MHe . CNN. ,KOUGH H0500“ Aemv Avmv .vmv “Nov Awm. Amm. Ammv Ammv Ammv I... eOiFMe thNo @Ho .hHo mo. bHol‘ *HO. OH. ou.>0mu Ammo .emv heme «emu Ammo Ava. “mm. Ammo Hausa .u> Houucoo nu ...nm. ma. ....oe. ~o.s 50... ..am. ma. .u.>oo Ahmv Ammo «may Ammo immv Ammo Acme Huuoq .-> ounum II emu. . AN. 1.3... . ma..- +++ma. eeNm. .u.>00 .mm. .mmv .eme .mmc lama .eme «yuan .m» enhance . -.. mummmmqmflqeammum II eeeho. .eewdm. .ee.m>.| mwbw.l. eeeab. .u.>00 ANmV Away “New Au “may H0004 esc«E.HeHuceu 8 . e . e. . e . 4 u- ea on n ma .oe. .u >90 .l «mag “we“ Away .«mu Hausa uncus.ououm nu . .bm.v . m.u. .cmn. .u >00 Ammq Awmwm Avmq eueuu asses amuuseu e ..u: t“ . “a z m “Hun II eeVm. ado! Raga - “now “Now useeaue>eo Hoooq In. .Msd. ue3om Aemv uneasue>oo eunum In Hesse uceesueboo «nausea .u.>00 o .a {hiya ou.>fl0 ourflflo .u.>flmlu “Can a “gum a; Hausa Haven neueum Haven Hausa eusum .u.>em .u.>oo .u.>oo .e> .e> .e> ensue ensue ensue Haven eunum Heuucoo Hmuuseo eusum Heuuceu Heuuceo eueum Heuuseo ufl,J,.!flI,, 4%JWMy?!fm ..-- liil§§i311iiliiifnfiilt?i5}{l§- m0.‘..: u we- wx ,.u,u‘-=. ‘ . .-u... as #5 aflmdn. v’-~‘ MW~W . "_~,- _’-' '-—_- 149 consider family income as that earned by ”family" in the Western sense and others interpret family income as that of the entire extended family. To correct for differences in definition of family, the students were asked to indicate how many. ..peopleshared theiincome. An average er._per .capita incomewwasmthenmcalculatedwby.dixidingithernumhernofiindivi-' vdualemsharingithemincomemintomthe median valuemof.therincome category checked. Table 75.3hows the per capita incomes for the.sample. TABLE 75 Per capita Income m Yearly Income Number in Percent of in Rupees S§NDJ§ Sang]. Under 1000 18 18.8 1001 - 2000 30 31.3 2001 - 3000 6 6.3 3001 - 4000 20 20.8 4001 - 5000 2 2.1 5001 - 6000 0 0.0 6001 - 7000 0 0.0 7001 - 8000 0 0.0 8001 — 9000 O 0.0 9001 -10000 1 1.0 No Response 19 19.8 Total 96 100.1 “Ameecondmindicationmof.sacialmclassmemployedmwas”the number of.hooks in the familymlihrary. ,The averageinumher of hooks estimated was 528. This average had a standard devi- ation of 800.7. Table 76 shows the distribution to be very skewed with the heaviest concentration offlresponses in the smaller categories. .150 TABLE 76 Number of Books in Family Library ...— .___ “W.-.“ __ _.—.——— _.._._—_.__ - —~ _ ,7 -__~ ..-1-.__— , .— -.._._ Number of books Ngmber in Pgrcent of 100 and under 27 28.1 101 - 200 17 17.7 201 - 300 8 8.3 301 - 400 1 1.0 401 - 500 7 7.3 501 - 600 1 1.0 601 - 700 0 0.0 701 - 800 0 0.0 801 - 900 0 0.0 901 - 1000 9 9.4 1001- 1100 1 1.0 1101- 1200 1 1.0 over 1200 7 7.3 No Response 17 17.7 TOtal 96 99.8 J ‘—y _ —— ‘— The third componentmofrthe.$ocial Class Index was.the -number of household servants. Most.Indiansuin.therpresent sample reported one to three servants. See Table 77. Very few individUals reported more than four servants. 1 The fourth andwfifth.indices.of social class.were the educationallevel ‘of‘the father and the educational .1evel ef the mother. Table 7a shows the educational.levels of the parents. Itiis.apparentmthatmthemeducational“level of the.fathers exceededthat ofmthemothers. Themean. ..educationa111evels ' Shownthat,“anatheraverage.“fathersrhadwcompleted.highwschool while mothers hadwabout nine years ofweducation. ATheudis-' .tributionof mother!s education is extremely skewed, however,x and.nearly twoathirds.of the mothers had eight years or less education. 151 TABLE 77 Number of Servants in Family W Number of Servants Ngmbef in Parcent 9f ' 0 9 9.4 1 27 jfihl 2 23 24.0 3 17 17.7 4 7 7.3 5 2 2.1 6 O 0.0 7 1 1.0 8 1 1.0 9 O 0.0 10 1 1.0 No Response 8 8.3 Total 96 99.9 TABLE 78 Educational Level of Fathers and Mothers ; . '“ ather ‘ ther Highest Educational Number 'Percent Number ’Percent Level Attained in of in of’ 1 . Easels Easels Easels .Eamnle 6 Doctorate 6 6.3, 0 0.0 5 {Master's degree or equivalent 11 11.5 1 1.0 4 Bachelor's degree or equivalent 26 27.1 5 5.2 3 Intemediate (11 to 12.years) ' 13 13.5 2 2.1 2 Matriculationilo years) ' 20 20.8 22 22.9 1 ,Belownhigh school (8 years or less) 18 18.8 62 64.6 No Response 2 2.1 4 4.2 Total A 96 100.1 90 100.0 Mean Educational Level ”3.11 1.49 Std. Dev. 1.32 .37 152 The sixth component of socio-economic status was the occupation of the father. Table 79 shows the Occupational distribution of the.smmple.. The distribution was_quite .skewed with most fathers in the higher occupations. ”Only. one student reported his fatherts occupation to fallrin,any of the bottom fourcategories.. The valuesmrangingmfrem onto 8 appearing to the left of the occupational categories represent thb values assigned each.category. ,These values were used in the calculation of the Social Class Index. TABLE 79 Father's Occupation ’_ ‘ Number Farcent Occupation in of k l £20013 Emile 8 Owner or manager of a large business: government official . 23. 24.0 7 Professional (lawyer, medical doctor, etc.) ' ' " 23, -24.0 6 owneror manager of a large rural estate or of a small business 25 26.0 .5 Office worker, government clerk, or ,_business clerk 4 4.2 Owner.of a small or.average farm 11 .1115 3 Skilled craftsman (carpenter, tailor, mete.) . 0 0 .0 2 Agricultural worker renting land or ”sharescropping «l -1.0 l Skilledgnrbanmwnmker“ ' .0 -0.0- 0 Unskilled urban worker: agricultural workerinot1owmingmland _ 0 0.0 .No Response 9 9.4. Total 96 1100.1 11153 The final componentof the Social Class Index concerned the number of brothers and sisters who had attended or were attending college. ,This.component was constructed by making- a ratio of the number.of brothers and sisters attending (or having attended) college. to the number of brothers and. sisters of college age or.older. .Table 80.shows these ratios. -The.higher the ratio.1themhigher the.proportion of brothers and sisters who attended college. TABLE 80 Ratio of Wand Wan College to Brothers and Sisters of College Age 1 a Number in ‘ Percent of BELLE. .ésmsls__. _ésmnls__. .81 - 1.00 31 . 32.3 .61 - .80 21 21.9 .41 - .50 14 14.6 e21 "' .40 9 9.4 .01 - .20 7 7.3 None 8 8.3 No Response . 6 6.3 Total 96 100.1 To obtain his Social Class lndex,theuindividual‘s ”responses to each of the components were transformed.into z scores and summed. The.reau1ting Social Class Indices. ranged from 14-80 to Q6-50 wdth.a meanwof.2.88 and a standard.devi-. ation of.2.56., Table alwshows the distributionrfor.the.56 students for whom data were complete. 154 TABLE 81 Social Class Indices W j f‘ __ social Class Number Percent Social Class Number Percent Index in of Index in of 1 Easels Easels. ___ mm -6.00 to -6.99 l 1.0 0.00 to 0.99 7 7.3 -5.00 to -5.99 1 1.0 1.00 to 1.99 4 4.2 -4.00 to -4.99 1 1.0 2.00 to 2.99 4 4.2 -3.00 to -3.99 10 10.4 3.00 to 3.99 5 5.2 -ZaDO to -2.99 4 4.2 4.00 to 4.99 1 1.0 -1.00 to -1.99 5 5.2 5.00 to 5.99 3 3.1 -0.00 to -0.99 7 7.3 6.00 and above 3 3.1 No Response 40 41.7 —_._ The-interoorre1ations.among the components of social class are presented in Table 82- In general the components. did not interrelate well- The items interrelating most often with other items seem to be father's education and mother's education. All components, however, correlated significantly with the Secial Class Index. .Hflbfld doc. us ucoususcmam sheave ago usn .usmodudcmam uos cowuoaeuheu+++ .Ho>oa no. no ucuuawusmam sheave .30 usn .ussuauusman uoc sowusaeuuooi. eeo¢o eeemm. Ammv Ammv II NH. “0mg a. 5 II .1 . 1 ca eueuedm NWQBHQHQ.;}IZ eeemm. eeemmo Aomv Ammv .00. no. Aomv amwv .eHN. eeemm. Ammv Awmv II eeemm. RE. 1005000 newuuosvm deuuuoavm euse>uem _ 28 Enemy»? 1.50:. 3.35; eeemm. eeemfi. 33 is... HH.I. mo. 32 :5 - mdcl. Ho.I $2 a: . ++¥mOe h§NMe Ava Amhv emo.l . oo. :3 a: _ ll nihNo .mnv husunun no masses. ueusoz ca exoem am>os Hoo. as» us nanosuacmum... H¢>ma do. «nu um unsceuanmam .. Ho>oa mo. on» us unnueuanmam . .eeemm. Awmv ,oa. “Nev OEGUGH euamuo won XOUGH mumHU Hawoom emeaaeo nu eweueum one mneuuoum nOauemsouo m.uonuem soaueosvu m.we£uoz sewueusbn e.uenuew museeuem no wenssz humunua Museum cu «soon «seven euammu Hem assesses eenao Huaoem maesd unedueaouuooueusH Nm mflmfia -,__.,.1 v '- _(.-, e V > - __,.~_——.... «.77; - o CHAPTER VI RESULTS AND DISCUSSION In this chapter, data relevant to the ten major hypo- theses will be presented. Each of the major hypotheses is composed of'sub-hypotheses, e.g., Hypothesis I is composed of sUb-hypotheses Ia, Ib, Ic, Id and Is. The statistical find- ings for each subehypothesis will be presented, followed'by a discussion of the findings for the entire hypothesis. This pattern is repeated for each of the ten major hypotheses. HYPOTHESIS I. LANGUAGE PRESTIGE AND AEFECTION1ARE RELATED TO SOCIO-CULTURAL FACTORS The prestige accorded languages was measured by means of the following instructions: Please rate g1; the languages below on how'much prestige you accord them. 5 a High prestige 2 = Fairly low prestige .4 = Fairly high prestige l = Low prestige 3 = Medium prestige* The instructions were followed by a list of languages: Assamese113engali,.Gujarati,”Hindi,-§industani.-Kannada, Kuhhiri. Malayalam, Marathi, 9:117:91, Punjabi, Tamil, Telugu, Urdu,1and.3nglish. Affection.was measured by asking.the.respondents to rate. each - of «themlanguages- they. knew -in. terms of...1the...affec- tion they felt for it. .The following categoriesrofraffection 156 157 were specified: 5 = Very much affection, 4 = Much affection, 3 = Some affection, 2 = Little affection, and l = No affection. Affection ratings were made by five or more respondents for each of the following languages: Bengali, Gujarati, Hindi, Malayalam. Narathi. Punjabi. Tamil, Telugu, Urdu, English and Sanskrit. Other languages such as Assamese, Hindustani, Kashmiri, Oriya and.Kannada did not.receive enough ratings to. be included in analyses involving affection. Hypothesis’I a) L d ff o r let to rural—ggbgn residence. The measurement of rural—urban residence was madexby asking the individual in what site.town he had spent most of his life. Answers were confined to specified categories: large city (300.000 o: more), large town (100,000 - 300.000), medium town (20.000 - 100,000), small town (1.000- 20,000) or village (1,000 and under). Unfortunately, very few people in the sample had spent a majority of their lives in rural areaS. (see Table 1, Chapter III). gzggtigg. Correlations were calculated between the prestige ratings and the ruralqurban measure for each of the _14.official languages and for English and.Hindustani. In . addition.-contingency tables were made.and chi square computed between ..prestige ..and, rural-urban «residence. . None of the statistics was-significant. I 4§;;§;t;gg. Correlations and chi squares were calcu- lated between the affection ratings and rural—urban residences for the 11.1anguages. Three of the correlations were 158 significant. Table 83 shows the significant results. In this table, as in all tables reporting significant relation- ships for data on which both.correlations and chi squares were calculated, all.significant correlations.are reported. Significant chi squares are.shown only in those instances. where the corresponding correlation.coefficient is not signi- ficant. Significant.chi squares and insignificant correspond- ing correlations occur.when two variables.relate to each other in a non-linear manner. Correlation coefficients were based on samples vary- ing from.5 to 78. The negative.correlation for_Gujarati (calculated on.arsample of 35) and the negative correlation for Telugu (calculated on a sample of 7) indicate that rural residents expressed.higher.affection forthese languages than did urban residents. The positive correlation for Urdu indicates the reverse. Rural.residents expressed less affection for this language than did urban residents. .Ihe correlation was calculated on a sample of 33 ratings. 8 t e is I “Y” 15 t and ffection a r t d t gmgggt of tggvgl. Egngfiigg‘ Correlations.and chi squares were.calcu- lated between prestige ratings and semen trayelmeasures. , Significant results are shown in Table 84. The first three travel measures concerned travel in.IndiaI a) number.of cities, towns andvillagesnlimedmin, b).number.ofistates lived in, and c) general.travel_in India. 159 TABLE 8 3 Relationships between Affection Felt for Languages and Rural-Urban Residence 3engali Gujarati r Hindi Malayalam Marathi . Punjabi Tamil Telugu r Urdu_ r English Sanskrit ll 1 0 .k N i -.ss* .41* *Significant at the .05 level. For Gujarati,~ there weremtwotsignificant and negative correlations. Individuals_who.have.lived.in relatively few ; cities, towns, villages or states considered Gujarati to have higher prestige than did individuals who have moved around more. The correlations were calculated on samples of 74 and 73. The situation was muCh the same for Hindi and Hindustani. Indians who had lived in relatively few cities, towns, and.villages.considered Hindi and Hindustani to have ”higher prestige.than Indians who had moved around more. Thesewcorrelations were based onsamples of 82 and,74 -respectigely. . . The significant chi squarerfound between.the prestige .accorded. armband. the...~number-.of states lived .in was computed .on 76 individuals. Thetrelationshiptappears.to be curvilinear. -The prestige of Urdu was high for those individuals who had -160 lived in either relatively few'or relatively many states. The prestige of Urdu was lower for individuals who had lived in a moderate number of states. No relationships . were ..found between any. . of the prestige ratings and the.amount of general travel in India. Two questions.dealt with travel outside India. The .first.question.asked the_individual howrmany trips he had taken outside India, and the.secondtaskedmhow many countries he had visited. Neither.of.these-two.travelrmeasures-showed any relationship to the language prestige ratings. . I _ The last two measures of travel were concerned with .vicarious culture.contact. .The.first measure asked about .the amount of reading the individual.did regarding other cultures and the second measure concerned the amount of con- tact his family had with non-Indians. A number of significant results were found between language prestige ratings and the amount of reading done about other cultures. .3 significant chi.square.was found between Assamese prestigerand reading done aboutrother,cultures. “Examination of the chi.square.tabletshows.the relationship to be.a.posi- tiveone...The.greateruthe,amonnt.ofwreadingMaurindividnal , hadrdone.about othertcultures, thesgreater the prestige be attached to Assamese.. $ample size for the.chi.square was 58. .The-significant-correlationrbetweenrthe,prestige _-attached.to.Bengali.and-amount.of.reading-about.otherrcultures indicates.thatmindividuals.who-hadmreadmmoreraboutmother -.cultures attachedmhigher.prestige to Bengali.than did indivi- -wduals who.hadiread less. A sample of 78 was used.in.calculation of the correlation. 161 The relationship between Kashmiri prestige and read- ing was curvilinear. Indians who accorded.Kashmiri the highest prestige had done a moderate amount of reading about other cultures. Those who.accorded.Kashmiri.the_least prestige read verymuch or very little about other cultures. The sample size was 70. The relationships.between reading.about other cultures and Oriya prestige, Urdu.prestige. and English prestige were the same as the relationship.between Kashmiri.prestige-and .reading. -The greatest amount of prestige was attached to each.of these languages.by Indians who had done.moderate reading. Very much or very littlewreading.about other cul- tures was associated.with lower prestige.attached to the languages. The statistics were calculated on samples of 68, 77 and 83 respectively. .Affsstien. The measureaof travel were also examined in relation to the affection-expressed for.languages. Corre- lations and.chi squares wereicalculated. Table.85 shows the significant.results. Only .one. of the. measures, concerning ..travel .-inside . Indiarelatedto any .of..the.affection..rating.s. There was a -eignificanttand positive.correlation.betweentaffection:sx- .pressed for Punjabi.and,numberof.statestlived in. The sample size.was 23. Individuals who had lived.inmrelatively .xnorewstates.expressed more.affection.for Runjabi than did individuals who had lived in relatively fewer-states. A significant chi square was found between the affection .Hosoa Ho. on» so unuoamacmam .. .Ho>os mo. on» so unuosuacmam . an 162 .N” ..Nx 'ONO N .H .Nx mmwacmHFmOQ noumwaso ImMMHMd> canon Sue: eunucoo Henna usog4 moauucsoo Masses msfivsom munaH. opamuso on Hm>eu9 ca 00>«A nomeaaa> use one: enema assess mqumqawlmuqmammlumaauuuam.:uuuem «manned auwuum an H when: iNx aammol u u aamm.l n H avN.l u H amN.l n M an Qflflmd have» @559. £0.35" > NE UHHXDCflm nuaamcm sou: fibfidflfi Hesse fineness swauo «Sussex awash.» He: «Hassmex opossum assumsocam «ceam «sausage aaemcem unmanned l ommsmsen "‘- 'Q-hc—F nflflmmImImmmmIIIImIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIlllllflflflflflflflflflflflflflflflflfln. Ho>sua use condemned on ponoeuu4 sesamoum guesses emanatedueamm . 163 expressed for English and the number of states lived in. The chi square was based on a sample of 70. The relationship appeared to be curvilinear with individuals who had lived in either.the fewest number or the.greatest number of states expressing the least affection for.English. Indians who had .lived.in a moderate number of states indieated.the most . affection for English. There were two sets of significant results for chi .squares and correlations calculated between affection rat- ings and the.measures of travel outside India. Both measures of travel outside Indiawrelatedttoraffection.for Malayalam. and for.Urdu. .In the case of Malayalam,tthe relationship was negative. . The greater .the..number ..of. trips outside India and.the.greaterrthe number of.countries visited,.the less the affection expressed for Malayalam. For Urdu the relationship was reversed. The greater the number ofltrips outside India and the greater the number of countries visited, -the greater the affection for Urdu. .The statistics for .Malayalamweremcalculated.on.samples of 9:. those.forurdu .on.samplesrof.34.and135. Correlations”and.chi.squares.calculated.between.the . affectionrratings.for.the languages and the two vicarious .chlturercontact.measur s yielded no significant results. Hypothesis I . ~ . ac) Liflflflififl pggggigg ggd affectign are rglateg to .sosialaslass. We. The Social Class ..Index am..its-mmmts (discussed in Chapter V).were-correlated with the.prestige -7--- .. .... ~. -77- ... - ...“-.- a. - .Hmsoa so. on» um scmuauacmem a. .ao>ua no. or» an uuuuauaumam . usuxmsem. ..Nx gunman l 4. iWMs H .H hMVa H .H ”UN: 6 SOOHOH. .. 11 Adams .Hm. n u sneezed. “sunset; chm.u u u ano.t u u snaeueanC HUGHm, «usumnswl «Hemcom.. IMdflHUdmudud.IluuHQMdnwl.IflflMflMflNl .llllMdudmlll MHMGMIQAI. odluufldmm IIIIQAIquHAMI «unwinddaf nae: uosucoo sense wooed meduussou shampoo 0on2 ca H0>MMB seamen nomeaaw> one haesum mcaomsm moans Academe ncaoa .mefiuwo muumummnumqmaumluququuuaw. nuauum_udauuma auwuua . «anew «manna Hummus III" I] ‘ I 1"] il'll‘ll' --rlill' I, He>eua.psu nomesmcmq won wash ceauoouud consume messagedueaom mm HQQflB 165 ratings given the languages.' Chisquares wsrssalse calculated between-social class and prestige. Table 86 shows the significant results. The Social ClassIndex_related to only one language - English. The correlation-ibased on 50 cases) was_positive.- The higher the Indian's-social class, the.greater.the prestige he accorded the English language. No.significant.relationships-were,foundrbetween-per capita income and language prestige. A significant correlation was found, however, between the size of the family.library.and.the prestige attached to Bengali. .Eor the 68 Indians on whom_data were available. a relatively large family library was associated with.greater prestige accorded Bengali. No significant relationships were found between the number of servants in the family and the prestige attached to languages. .The education of the father correlated negatively with the prestige attached to“ Gujarati (N = 73). The higher the.education.of.hisrfather,.the_less“themprestigewassigned Gujarati by. the respondent. ' Mother's educationrrelated.in-opposite ways to.Qriya andwto,EngliShwprestigewratings.M“Therstudent:whoseumother ewes,highly.educatedrattaohed-low_prestigeuto Oriyamandmhigh prestige to English. samplejises were 55‘ and -80..-.nespeat- ' ively. 166 In like manner the occupation of the father related in Opposite ways to the prestige~of Punjabt and*English. Respondents whose fathers had a relatively high” education accorded low prestige to Punjabi and high prestige to English._ Statistics were based on samples of 70 and 78. Finally, the proportion of college-aged.brotherstand sisters attending college related positively to the prestige ratings given Hindi.and Marathi. The more brothers and sisters the respondent had who were in college, the greater the prestige he attached to Hindi and Marathi. Samples were 78 and 68 in size. .Aifegtign. The social class measures and.affection ratings for languages were examined by means of correlations and chi squares. The significant results are shown in Table 87. The Social Class Index.re1ated to affection felt for only one language - Hindi. For the sample of 46 students, the relationship was negative. The.respondent whose social class was high tended to feel less.affection for Hindi than did the respondent whose social class was not so high. The.measure of per capita income related toaaffection for..three..of.the languages. For -all...three , Bengali..liindi. and Marathi.ithe relationship was negative. Indianwstudeats _from families with high per capita.income felt little-affec- ..tion.for.these three languages. Sample sizesflfor the three .relationships were 13,.64 and 26,.respectively. amm.uu " '”IQ‘I" xmcaH a DQUHU M .5308 «mm. H cmN. "H J «muddum ad usouusm use uuosuoua u H cmNo " H iVNoI scaummouuo MO.HQ£umh .mNo " H CMN 0.!" H ammo!" cauuooovu noduoUsGu nuni>uom saruusuozx s.uosuos no sosssz s mumamlaaauum om mHmGB .H¢>oa mo. on» as assasussmeu s chm. "H MHflundQ %Hdfiflha H OXOOQ oeoocu madman Hem Hauxuaou soaamsu sous amfldflB Adana «sundae «wauo unused: eunuwnadz «naenmax uuwccafi «couusocdm «beam «assumes Samson unmanned «manuaum 1.,1..~ - ‘ ---‘..—._-,—.. z . 4 a“- -.-.-‘ , . —~w—- TET w" 1-.-..- .7... "- ‘ :rv‘fiv ...—...“-.- __:_-'...‘._. ~,7_~‘__‘H‘ '_ ‘ - 168 The number of books in the family library also related negatively to the affection felt for Marathi. When the family library was large, affection felt for Marathi was low. The sample size was 29. No significant.relationships appeared between number of family servants and any of the.language affection ratings. The education of.therespondent's father related significantly to the affection eXpressed forulunjabi. The higher the education of the father. the greater the affec- tion the respondent felt forqunjabi. Twenty-four ratings were used ..in N.thea...,statistical_..calculatian.. Motherlsmeducation related in,a negative manner to the affection expressed for Hindi. .The more educated the mother, the .lesssthe affection "felt- by students ..for Hindi. samplesise was 78. f The occupation.of the father.related to feelings of affection for two of the languages. The relationships appear to be negative in both instances. .Students;whoee fathers had relatively high occupations indicated theysfelt relativelylittle affectionsfor Gujaratiand aindi. In. .ammple sizes were 34.for Guiaratiwand 73wforfifiindi.a » No...relationships were found..-betmen -the proportion _tof“collegeaagembrotherswandpainters.attendingrcellege.and Vlanguage"affectionmmeasures. 169 - ”fir “‘0‘." .-—_.x ~-r_---‘-' HO>0H Ho. $89 an Manuaudflmfimee so>os no. as» us assueussmsm . Udhlflfldm sesame“ sou: sesame Hesse _ :mm . nu Banana new... as cumin u «nudes: --w 5363: imm olflh Cgfiw cVN o Inn-H CmN a..." H fiUGflm s¢m.: us .- assssfisu s ..o>.uu s «Hanson nuanqusqm wg as ”MMMW numwma soausmono scaususvu scaususon no has—sh. @3003” w Heauom euenuoum m.uenus.m «rhesus: m.uo£ue.m .3852 3" 3.00m unease Hem _ IufldU HCHMGQ aesau Heaven use nemgmssa sou sash nodubouud segues meannessusdem e ussbuom banned hm mnmdfi 170 Hypothesis I d) La a on r la d family tgggition gf foreign study. To test Hypothesis 1 d. the students were divided I into two groups: those who were the.first in their families. to study abroad, and thosefromfamilies where at least one member had studied outside India. ‘ firm. The mean prestige attached to each langu- age was calculated for students who were the .firstin their families to study abroad. Meanpr'estige -ratings were also , calculated for the group of students who were .not the first in their families. .to..study. abroad. The two,_.mean prestige ratings for each language werecomparedby t test. The dashedline inthe t .'oolumn indicates that the difference between the two mean prestige ratings for Sanskrit was not tested for significance because of the small number of ratings. Table 88 shows the only significant difference in the prestige attached to the languages. This difference was for Hindustani. Individualsfrom families whose ..manbers had studied abroad attached higher prestige to Hindustani than. did ..individuals who werethe. first .in . their ..families to study ..abroad. 6 -W- The. ..mean ..affection expressed. for each _ language .was_..calculated.for .the.group,.of students who were the.first. in, their families .to...study __.abroad....and for ..th-e .11er Y of students who were .-not ..the first-manb.ers..of.,-their- families to study abroad. The.two.mean.._affection. ratingsfor each, language werecomparedhy t...test. .. Significancetests were TABLE 88 Relationships between Prestige Attached to Languages and Family Tradition of Foreign Study Prestige Ratings 1 W W W A ‘ W Assamese 2.37 1.26 41 2.16 .94 25 Bengali 3.50 1.24 48 3.63 .96 30 Gujarati 3.19 1.29 42 3.16 1.19 32 Hindi 4.40 .95 50 4.34 .79 32 Hindustani 3.59 1.30 44 4.23 1.17 30 Kannada 2.62 1.23 42 2.41 1.05 29 -2.21* Kashmiri 2.48 1.23 42 2.21 .83 28 *- Malayalam 2.56 1.22 43 2.22 1.09 27 Marathi 2.90 1.32 42 3.07 1.14 30 Oriya 2.40 .1.33 42 2.04 .92 26 Punjabi 3.00 1.30 46 2.46 1.07 28 Tamil 3.04 1.32 46 2.57 1.23 28 Telugu 2.93 1.30 43 2.57 1.20 28 Urdu 3.35 1.23 46 3.26 1.26 31 English 4.54 .73 50 4.73 .57 33 Sanskrit 5.00 .00 2 4.75 .50 --- 4 *Significant at the .05 level. not made for Malayalam and Telugu because of the small sample sizes. 2 Only one significant.difference was found between the mean.affection ratings.for the languages. Those who were thefirst in their families to study.ahroad felt more affection ..for 11m; than thoseirom families with a..tradi- tion.of.foreign-study. Hypothesis I Gnu-'13}; 0 8 10 nd 'v,_,fCt'—,"! - 3 . i1. ' 50“- O s ;- 0:.-9‘ ‘ o “90‘ >.. See Table.89. ’d _ Indian students were classified as to whetherutheir .residence was in a North or.a South Indian-atate.or.territory according to the following system: 172 TABLE 39 Relationships between Affection Felt for Languages and Family Tradition of Foreign Study .hffection Ratings _irst_ts.§tnd¥_ Eamil¥_2raditien Lansnses_. .JuxuuL. ' ..ssn..§.2..,.§ Ween .212. .E t Bengali 3. 50 1. 38 12 3. 67 1 . 5.1 6 Gujarati 3.67 1.35 15 3.05 1.31 19 Hindi 3. 79 1.18 48 3. 55 1. 06 31 Malayalam 3.40 1.67 5 2.50 1.29 4 --- Marathi 3.06 1.51 18 3.00 1.56 15 Punjabi' 3.65 1.00 17 3.00 1.91 7 Tamil 5.00 ..00 6 3.50 .l.60 8 2.659 Telugu 4.25 1.50 4 3.33 2.08 3 --- Urdu 3.23 1.11 22 3.75 .l.36 12 English 4.31 .97 45 4.20 1.22 25 Sanskrit 3.30 -l.34 20 3.00 1.52 14 *Significant at the .05 level. Wis W Bengal .Maharastra Andhra Pradesh Bihar Orissa Kerala Delhi Punjab Madras Goa Rajasthan. Mysore Gujarat Uttar Pradesh .zggggigg. .Thehmeanmprestige.rating“attachedwtotsedh -language was calculatedtfor north Indianssand-for-&outb rlndians.l Tbewtwo.maans for each-langnage were compared by-t test- Table 90.3howstthe.meanmratingswand-the.significant differences found. No significancestest was.made for Sanskrit .because of the small number.of.&outb Indian.ratings.- 173 TABLE 90 Relationships between Prestige Attached to Languages - and North-Seuth Residence in India Prestige Ratings fisrth_lndian Language Essiggngs 9 I R m ...s—a. 3 5.933 _s—o .- 5 Assamese 2.39 1.16 52 2.00 .96 13 Bengali 3.68 1-03 62 3.20 1.28 15 Gujarati 3.35 1.20 60 2.39 1.15 13 2.71** Hindi 4.56 .72 66 3.67 1.08 15 3.04" Hindustani 4.05 1.24 59 3.00 1.07 14 3.20** Kannada 2.45 1.16 56 3.00 .93 14 Kashmiri 2.39 1.13 56 2.31 .91 13 Malayalam 2. 46 1.25 55 2.43 .73 14 Marathi 3.02 1.24 58 2.85 1.23 13 Oriya 2.35 1.21 56 2.00 .1.00 12 Punjabi 2.92 1.24 60 .2.39 1.00 13 Tamil 2.72 1.27 58 3.53 1.09 15 -2.48* Telugu 2.58 1.18 57 3.85 .95 13 -4.15** Urdu 3.34 1.23 62 3.14 1.25 16 Engiish 4.61 .69 66 4.63 .60 16 4.80 .40 5 5.00 .00 1 Sanskrit —T_ “—7 . _._._,._._.. * Significant at ** Significant at the .05 level the .01 level As might be expected. North Indians accorded higher prestige to North Indian.1anguages {Gugarati..fiindi.and Hindustani).and.5outhIndiansmaccorded;higher prestige rat- ings to South Indian languages (Tamil and Telugu). W. Mean affection .ratingswfor -each..language 'were calculated separately for North Indians and South ludians. .for each language werettestedtfor“significance. affectionmratingsmfor.thenlanguages-andtthewsignificant differences found between Northern and douthern ratings .Differences between the.Northiand South.ratings Theumean: appear in Table 91. i174 Many of the.mean affection ratings were based on very small samples. An arbitrary criterion of five ratings was used to determine when significance tests would be made. Means calculated on a sample of less than five ratings for either Northern or Southern Indians were not used in testing' for significant differences. The dashes in the t column of Table 91 indicate when one or both of the mean ratings for a language were based on samples so small that no t test was performed. 1 Unfortunately there were not enough ratings available to test dufferences.between.North and southaffection for two.1anguagess Hindi and English. The t tests for.these two languages indicate.that North Indians felt greater affec- tion.for Hindi and.less affection for English than did South Indians. TABLE 91 Relationships between.Affection Felt for'Languages and North-South Residence in India W Affection Ratings . Ngzth Ingjan . Language ..ssidsnss Esnthuindisn_§ssisssss ideas..§.2. E, :5338 .512. .E t Bengali 3.71 1.23 17 1.00 .00 1 -—— Gujarati 3.38 1.29 32 3.50 1. 50 2 --- Hindi 3.81 1.11 67 3.08 1. 04 12 2.22* Malayalam 2.00 .00 .l 3.13 1.65 8 --- Marathi 2.97 1.40 31 5.00 3.00 2 -.- Punjabi 3.52 1.28 23 2.00 .00 1 --- English 4.16 1.11 57 4.69 .61 13 -2.36* Sanskrit 3.07 1.41 45 4.00 .71 4 -.- T * Significant at the .05 level **Significant at the .01 level. 175 DISCUSSION: symmesss In, 11;, lo, Id, 10 The same prestige stereotypes about language seemed to prevail in moderately sized urban communities as in the largest urban centers. The leek of differences found may reflect the universality of the prestige.stereotypes about languages in India. An.alternative explanation rests on the composition of the sample. Few respondents had lived most of their liveshin rural areas. .Itimaymhe.that.themfew who had lived in rural areas had also lived long enough in larger urban areas to adopt the prevailing sterebtypes. Ruraleurbanhdifferences.wereifnundrfor_thewaffection expressed for languages. .The affectionmratings.differed for ruralvand urban residents.for three of the.ll.1anguages. Respondents frem.more rural areas felt less affection for Urdu and more for Gujarati and Telugu. It is.easier to- understand possible reasons for.the Urdu differences than it- is for Gujarati and Telugu.. Urdu is considered a language rich in poetry. If the rural man is less well—educated— and well-read than his urban (counterpart, new-away Urdu mrghr-hrsnp-ctehrte—htimthan—that—ot-hia urban'ce‘un‘Eerpart: Why differences in-affectienseccur for Gujarati and Telugu rather than furfigbfiiefifigfiam: -335135; Zruraisurhan differences ~ . should -be. roundror affection ...and...no.t .-for- prestige. One answer is that there is awhasic.differencetbetween the hprestigerand.affection ratings. -Every respondent made pres- tigeiratings.on.each.ofuthe.fourteeniofficial«languages-and. also on English.andeindustani. Respondents indicated their‘ 176 affection, however, only for those languages which they were able to reads, write. speak and/or understand. Another answer is that the prestige ratings probably depend more on stereo- types than the affection ratings- The.affection ratings were intended to tap emotional involvement rather than stereotypes.. ' , Amount of travel was a correlate of both prestige and affection. Relatively more .trauelinside India was associated with.1ow prestige for Gujarati..§indi and. Hindustani and high prestige for Urdu. It was also associ- ated with iower..et£.sotion.._ror English. .and.higher..aff.ection for Eunjabi. Oneexplan'ationforuthe prestigefindingsris that.respondents.who-haverlived-in-many-citiestorwstates. came.into contact with many.1ndian languages. Qomparisons. among languages may have not favored Gujarati. Hindi.and Hindustani. Urdu.enjoyed high prestige among those who had lived in many or in few states. but low prestige among those having lived in a moderate number of states. One explanation is that in this-sample, residents of a moderate number of. states may have come intercontact mainly with those langu- ages overShadowing Urdu inprestige. Residents of many statesgmaynhave;had.atbrnader-sampleiof.1anguagesxwith-whiCh to compare Urdu. It.may.be too..that Urdu enjoyed;high‘ , prestige.among those who had lived in wary many.or very few statestbecauseethese.respondents.wereitheunuslimsHQr;Urdu ,.natise tongue speakers in the sample. .Thiswpossihility was not explored in this research. 177 Interestingly, the more states an individual had lived in, the greater his affection for Punjabi. The reasons for this are not clear.s ‘ Residence in a moderate number of states was associ- ated with high affection for English. Residence.in many Or few states was associated with less English affection. One interpretation for this isin terms of amount of language use. Living in.a small number of states may not.have neces- sitated the use of English. The native_tongue.or regional languageemay have sufficed with the result that the indivi- -dualis”affections”remainedwprimarily“withmhismnative-tongue ”or regional language. At the other.extreme..for individuals living in a large number of states. English may have been only one of many languages necessary and as such it shared only moderately in the speakers' affections. It was the indi- viduals who lived in a.moderate number of states who felt the most affection for English. In this case English may _have sufficed as the medium of communication in all these states. being used often and gaining in the affection of its speaker. ' Travel. outside inanimate: (associated «with prestige. but wasiasseciated withmaffectinn. “Mere travelioutside India qwasuassociated with greateriaffaction for Urdnrandwless affection for Malayalam.m It mayibewthat.travelflmademnhe- ....individual...aware. that -thehliterary prestige. of Urdu. extends , beyond IndiaPsmberdei-‘s w while Malayalam -isinot very wellknown. outside India. 178 Vicarious culture contact related to prestige, but not to affection. Much reading about other cultures was associated with high prestige.for Assamese and Bengali. Moderate reading about other cultures was paired with high. prestige for Kashmiri, Oriya. Urdu and English. Why high prestige for a language was related to moderate reading in some instances and much reading in others is not readily. mapparent. In the same manner why relationships were found for these particular six.languages and not for others is al- so not.apparent. All that isclear is that reading more ‘about,other cultures was associated.with attaching high prestige to languages—~up to a point at least. The.absence of relationships.between vicarious.cul- ture contact and.affection-is.not surprising if affection reflects language usage in part. Brestige,von-the other hand. may well involve a comparison of a language with others. Reading.about other cultures and presumably about other languages might induce comparisons which.cast particular languages in a good or bad comparative light. Social classrelated to both.prestige.and affection. High social class was paired with high prestige accorded English. Bengali, Hindi and.Marathi and low prestige accorded- Gujarati, Oriya. and Punjabi. Thebigh .prestige..,fornfinglish _ and Bengali ..isinot. surprising. English .is the language .of the educated andmis.informally,acknowledged as a prestige -symbol associated with high education and high-social status. .Bengali.isifamed.formits.richnliterarystradition-eawfeature 179 most likely to be appreciated by the educated. This inter- pretation is reinforced in these data because the component of social class relating positively to the Bengali prestige ratings was .the number of books .in the family library. The, high prestige for Hindi and.Marathi.was the result of positive relationships between prestige ratings for these languages and the prOportion of collegeaaged brothers and sisters with college education. Since Hindi iscnow compulsory in schools, it has become.amlanguage.of the.educated. As.such. it is a language not only of the individual but of those of his sib- lings with higher-.education. ‘ Whytthewcollege_education.of,the"respondentlswbrotbers and sisters should affect his own prestige.ratings of Marathi is not clear except that Marathi and.aindi.are closely re- lated languages. Social class was_strongly related to language affec- tion. High.social class was generally paired with low affection for Hindi. Occasionally this was true for Marathi, Bengali and.Gujarati. Only once was the relationship in reverse: whigh social class paired.with.high affection for Rungabi.. If the.language usage explanation-for the affection relationships.holds, then.the high prestige. but lowuaffece tion felt foriflindi.bymhigher”sociallclasses“iswmoreiunder- »standable.‘”Englishtisithewprestige“language,among.those,of high social class in India.and is probably usedias often as' possible (if.informaltreportswareetrue). .Hindi reportedly isnotwregarded in this manner. Its. -use..may .notbeflso great 180 as evidenced by the frequency with which it was listed as a "language used unwillingly“ (cf Table 28). The expected differences in affection for English did.not materialize: Indians of higher social class did not feel greater affec- tion for English than did Indians of lower social class.. If language usage is the key to affection. then the constant required use of English.by the students in the'sample would explain the lack of social class influence on English affec- tiOn.for these individuals. ' Family tradition of foreign study did not prove to be an important correlate oftlanguageprestige and.affection- Individuals who were the first in their families to study abroad accorded Hindi rather low prestige and expressed rather high affection for Tamil. It may be that those who were the first to venture abroad for education were ready to reject Hindi while those from families where others had studied abroad had grown up in a tradition of appreciating Hindi for its_merits-eit does not seem to be a language that is normally discussed in the U.S. The affection expressed :for Tamil by.those whose families.did notiinclude.other foreign stholars is puzzling. .It may beirelated:to the cohesiveness on campusinoted among the.students.for whom Tamil was.the nativewtongue. The”prestigelanduaffactionwrelatienshipsifoundcfor . North and fiouthiresidence.eetablish geographic residence as, a.significant correlate ..cf , languageattitude. For “the. .most part the relationships are selfeexplanatory. «North Indians H181 conferred high prestige upon the North Indian languages of Gujarati, Hindi and Hindustani. South Indians accorded high prestige to the South Indian languages of Tamil and Telugu. In like manner. North Indians expressed higher affection for the North Indian language Hindi, and South Indians expressed higher affection for English. The last finding correlates well with observers' reports that the hue and cry of ther South is "Equal status for English!" 182 THYPOTHESIS II. LANGUAGE LIBERALISM IS RELATED TO SOCIO- CULTURAL FACTORS The language liberalism of each respondent was measured by means of a.Language Liberalism Index. The construction *of this index is described in Chapter V. Hypothesis II a) L l s r ted t ru - r missus. Correlations and chi squares were calculated between the rural-urban residence measure and the respondents' Language Liberalism Indices. No significant.relationships were found for the 79 individuals fin.whom data for both measures were available. e Hypothesis II ’MLMWQW Correlations and chi squares were computed between Language Liberalism Indices and each of the seven travel measures. Sample sizes for the calculations varied between n78 and 80. None of the correlations or chi squares calculated between language liberalism and travel inside India were significant- .In like.manner none of the statistics.computedi hetwaen.languageliberalism.and.travel outside India.or bo- tweenulanguage.liberalism.andivicarious culture .contact were significant. Hypothesis II Language Liberal Indices were correlated with the Social Class Index and each of its components. 'Chi squares were also calculated between these measures. Table 92 shows 183 the relationships found. The Social Class Index did not relate to the Language Liberalism Index. When items that were combined into the Social Class Index were examined separately, relationships, with Language Liberalism were found. Negative correlations were found between Language_Liberalism.and the number of books in the family library-i Awnegative.correlation was.also found between Language Liberalism and the education of the mother. This means that respondents whose family libraries were relatively large and-whose mothers had relatively more educa- tion tended to have relatively conservative language attitudes. Sample sizes for the two relationships were 68 and 76 respectively. No other significant relationships between language liberalism and components of the Social Class Index were,found. ThBLE 92 R t s s betwe n L a e L beral s and Soc Cla s Per capita.income Books in family library r = -.24* Number of servants Father's education Mother'sieducation r Father's occupation Brothers and sisters in college ’0 34** Social.Class Index * significant at the .05 level. ** Significant at the .01 level. 184 Hypothesis II d) L u b ism s re at to famil tggdition of forgign stgdy. A mean Language Liberalism Index was calCulated for those Indian students who- were. .the. firsthin their families to study abroad. A mean Language Liberalism Index was also calculated for those from families where other members had foreign.education. The difference between the two means was tested for significance by means of a t test. The difference between the two means was not significant, i.e., whether or not an individual was from a family with a tradition of foreign study had no relationship to the liberalness of his language attitudes. Hypothesis II. 0’ ~WWMW W; A mean Language Liberalism.Index was computed for ’Indians from the North. A mean Language Liberalism Index was also computed for Indians from the South. The difference between the two means was examined for significance by a t test and found not to be significant. That is. there was no difference.between North Indians and South Indians in terms of liberalness of language attitudes. DISCUSSION: HYPOTHESES 11a, 11b, IIc, IId, IIe Neither rural-{urban residence. nor amount , of travel were correlates of languageiliberalism. In likeimanner. family tradition oquorei n stu y and NorthpSouth residence did not.correlate with.language liberalism. The.only 185 socio-cultural variable related to language liberalism was social class. The two components of social class which _correlated with language liberalism did so in a negative manner. ,Morenspecifically,.the higher the education of the mother and the greater the number of books in the family library, the less liberal the student's language attitudes. These relationships were opposite to those expected. The most obvious interpretation of the insignificant findings is that range of culture contact simply does not relate to liberalness of language attitudes. Another possible interpretation is thattthe measures of.culture contact used to test the hypothesis were.not.sensitive enough. This, interpretation does not seem too likely, however, because all of these measures of culture.contact related to the prestige _ and/or affection..ratings .examinedin Hypothesis I. is... is a third possibility, too, that the measures.were not appropriate, i.e., they were not really measures of range of culture contact. The significant findings in Hypothesis I do not establish the validity of the measures.. Still another possibility is that language liberalism .is a variable of a different order than prestige.andiaffec— tion. This is a rather strong.possibility. .Language ' .liberalism was a composite of a number of specific.attitudes: prestige.and affection were each specific attitudes. It may also be that the measures oficulture contact related significantly, but in different ways, to some.or all of the; components of.the Language Liberalism Index. ErasentwanAIyses did not explore this possibility. 186 The significant and negative findings for social class are puzzling. It is tempting to resort to the ”upper classes are more conservative" explanation. This is a 4 possibility. of course. Another possibility concerns the _ particular components of the Social Class Index: number of . books in family library and mother's education. It may.be. that these tmcomponents measure something .in addition to social class. -187 HYPOTHESIS III. LANGUAGE LIBERALISM IS RELATED TO THE PER- CEIVED LANGUAGE LIBERALISM OF REFERENCE GROUPS AND TO FEELINGS ABOUT THE REFERENCE GROUPS The measures of language liberalism used to test this hypothesis were the Language Liberalism Indices des- cribed in Chapter V. Hypothesis III a) L ua ib ralism is related to the er- ggivgd Lapguggg libgraligm of familyI subcaste, W This hypothesis was tested at two levels. At the broadest level, the Language Liberalism Index calculated for the Student (i.e., “Self“ Language Liberalism Index) was com- pared to those calculated for his father, mother, subcaste, social class, region. Comparisons were made by means of correlation coefficients., Table 93 shows the single signi- ficant result. The Self Language Liberalism Index correlated positively with the Language Liberalism Index of the father. The correlation was based on a sample of 79. TABLE 93 Correlations between the Self Language Liberalism Index and the Language Liberalism Indices of the Father, Mother, Subcaste, Social Class and Region Was Father Mother Subcaste Social Region , . Class Self Language Liberalism Index .60*** fir ***Significant at the .001 level. 188 The.hypothesis was tested on a more specific level by interrelating each of the six items comprising the student's index with their counterparts in the indices of his father, his mother. his subcaste, his social.class and his region. In all, 30 comparisons were made. Chi squares and correlation coefficients were used to make the comparisons. Eighteen of these comparisons were significant at the .05 level or be- yond. Table 94 shows the significant relationships found. Five of the six items comprising the.student's (Self) index were related to their respective counterparts in the indbx of at least one of his reference groups. The only item.in.the.student!s index failing to relate to its counter- .part in the.indices of the reference groups wasthe second _item: the number of languages used unwillingly. At the other extreme were the last two.items: _prestige.attached to .bilingualism and acceptance of Hindi as India's official language. These two items correlated in a positive manner atthe .001 level with their counterparts in thawindex of every reference group. Statistics for most items were based on samples varying in size from 71 to 83. The two excep~ tions to this were: the correlation.between Self and sub- caste for “Languages Not Appealing" and the chi square be- .tween Self and.region for "Intermixing Usage." The.samples .for these two statistics were ll and 12 respectively. Hypothesis III b) Essistiea_fr2m_the_2sresized.isnsuase_liberal- is m of the famil ngsubcaste social class and regi 9n is rglatgd to the Timpgrtance of each s .n s to“ e divid odObOH H00. 039 “fl UGMUflMHCUfiQCtt 0H0>0H HO. 0&0 H“ uGflUHHHC UHMiI oHG>OH m0. 03“ UH.»C§UHHHCD#W¥ t Aamveasbmonh Amhveaamhonu seamo. NH ‘aaeom.nu asamh. nu «poem Aawv «th Amhv mo ouceumeoo¢ AmmveaNMoflH Amhveaamm.uu eaaoo. "H ae¢m®.nu «seem. an MOHUmOHm Amen lose “out smeessmseeem Avmv aeNN. Amhvaevm.nu «chm. "H mcfixHE , Amwv tueucH Hesuod Ahhvsasmflouu Agbvemmouh Wammoflh teammo "H HG>OMQQ4 ;, Aw v Abby mcfixesueucH .9 m semeeeaeaso use: memmsmcsq Aaavsan. nu mcaaeedmd uoz wemsomcma L, i x H Emfiamuenan_emmsmcsq mama on» on semen cowmem.pce sumac deacon , . .eunsonom .uenuoz .uesuem on» no neoapsu Enaasuenaa emsomcmq_e£u ca msouH mew upcoomeuueu ram xech EmeHsHeQaA emssmcsq “Hem ecu cu useuH awesome mdancheumaem vm mumfla 190 Five separate language deviation measures were used to test this hypothesis. Deviation from the language liberal- ism of the father was calculated by subtracting the Language Liberalism Index of the student from the Language Liberalism Index of his father. Asimilar procedure was followed in the calculation of the deviations of the student‘s Language Liberalism Index from the Language Liberalism Indices of the mother, subcaste, social class, and region. These differ- ences or deviations were then correlated with the importance rating and with the relative importance rating the student ‘gave each group. The importance of each reference group to the indivi- dual was measured by means of a rating scale which ranged from 1 to 5. On this scale, 5 = extremely important, 4 = very important, 3 = somewhat important, 2 = not very impert- ant, and l = unimportant. Separate importance ratings were made for the region, social class, family, subcaste and country. .In addition,.relative.importance ratings were made. The respondents were asked torankin.egdor3thefivo'rderinctl groups in terms of their importance to him. The most import- ant group was assigned a value of 5 and the least important, a value of l. The distributions of the importance and relative importance ratings appear in Appendix N. Five correlations were made between the deviatiOn ,scores and their respective importance ratings and five correlations were made between the deviation scores and .“theirirespective.reisttvertmportance ratings. None of the lcherrelatiens waszsignificant. .191 DISCUSSION: HYPOflHESES IIIa, IIIb Hypothesis three predicted relationships between the individual's language attitudes and those of his father, mother, subcaste, social class and region. Taken as a whole, i.e., the Language Liberalism Index, the student's attitudes corresponded only with those he saw his father as having. This finding suggests that the respondp ent saw‘his own attitudes in closer correspondence with those of his father than with any other group. Put another way. it was the father who was the referent for these language atti- tudes of the individual. An.examination of the components of.theiLanguage Liberalism Indices shows some of these.specific attitudes (i-e., components) in the.respondent's Index to relate signi- ficantly with many of their counterparts in one or more of the reference group Indices. of particular note were the acceptance of Hindi as India's official language and the prestige attached to bilingualism - attitudes upon which the individual saw himself in agreement with every one of his groups. The.respondent's attitudes related significantly‘ with four of the perceived attitudes of the father, social eclass and regionrand three of the perceived attitudes of the mother and subcaste. In this more.spacificfanalysis. the .individualls attitudes did not appear.to be more.closely re—' ..lated to the father'..s-attitudes than .to those .of.-o..ther refer-v ance groups- From the specific analysis, it would be diffi- . cult to select any one of.the groups as being more of.a 192 reference group than the others. The lack of correspondence between the studentfs Language Liberalism Index and the indices of each of his reference groups wOuld suggest that the student was not,simp1y projecting all of his own attitudes to his reference groups. The.specific analysis, however, did not indicate the same thing. Many components of the respondentis Language Liberalism Index corresponded with those of his_reference groups. Here there_is strong reason to question whether therrespondent did project his attitudes to his reference groups.. The high degree of correspondence found between components of the Language Liberalism Indices was not consist- ent with the laCK of correspondence between the overall Indices. Presumably thisclack of consistency was due.to the presence of components in the.lndices which.did not relate to those in the respondent‘s Index and to.the.component,‘ which related in a non-linear.fashion (see chi.square statistic in Tahle.94). Feelings.on language questions ran rather high among. the students. An.illustration of this is found in Table 45 Where.strength ofconviction about personal choice for India's official language.is_presented. .Eeelings on issues such as this may have run so.high.as.towhave.obscured.anymohiective rappraisal of the position of the reference groups.§ In con-, .sequence, the respondent!s attitudes may have been.projected to.hisnreference.groups. The possibilityimramains, however,: thatfihisureference.groups may .-nearly .always...agree with the ' respondent on these issues and that the significant‘correlations .193 represent reality, i.e., accurate perceptions of reference group attitudes. In general, the.findings.suggestthat the parents, subcaste, social class and region may act as reference groups for the student on issues involving language. The validity of sudh an interpretation rests upon the accuracy of the student's perceptions of his reference groups. Although no relationships were found between the over-all attitudes of the student and most of his groups, it was expected that such relationships might appear if the importance attached to the.groups were taken into consider- ation. For example. if the region were, very important to some.students and.not to others. the extent of.a student's. deviation from the.attitudes of his region.might reflect the importance of the region for him. Importance.ratings had.no effect, however. on the correspondence between attitudes of. the student and any.of his groups. This wasrtruewemenrin the case of fathers where.a high positive relationship was found between student and father Language Liberalism Indices,sv The.explanations for.this are several. .Eirst, the groups.maymreally not be.reference groups on language issues for the student. Some.other group, such as college.peers,. may be the reference group. Second. theqmeasures were gross. Either.or both of the importance and language measures may shave beenminappropriate. Third, the importance measures.wares skewed. This.restricted their range and made interpretation. .of the statistical computation difficult. Fourth, socially. acceptable answers may have been given to the language -194 questions or importance ratings. .This.is especiallyilikely in the case of.importance ratings where almost everyone con- sidered the family quite important and the social class and subcaste as more or less unimportant. The relative importance ratings reflected the same phenomenon: .family ranked very high, region and social class were intermediate and subcaste very low. 195 HYPOTHESIS IV. ILANGUAGE SPECIKLIZNTION IS RELATED TO THE PERCEIVED ATTITUDES OF REFERENCE GROUPS AND TO FEELINGS ABOUT THE REFERENCE GROUPS Six language specialization measures were used to .test this hypothesis: a) number of languages usedwith parti- cular people, b) number of people with whom particularlangu' ages were used, c) number of languagesrused.in.particular places, d) number of places in which particular languages were used, e) number of languages used on particular occasions, and f) number of occasions on which particular languages were used. Hypothesis IV a) Lgnggagg specialization is relatedrto thg per- ggivgd diversity of lgnggagg preference gmgng family, subcaste, social class and region, The language preferences of the father, mother, subcaste, social class and region were listed for each student and a measure of diversity of language preference made. The diversity of.language preference measure was the total number of languages an individual.would.have to know in order to converse with members of his family, subcaste, social class and region in at least one of the languages they preferred. Thus if three of an individual's groups preferred.fiindi and the fourth group preferred Gujarati,.it was considered that the individual would have to know two languages to please all groups. If.three of the groups preferred Hindi.and Gujarati while the fourth preferred Gujarati, then the.individual would need to know'only one language to please.all groups. ‘196 The diversity of language preference measures were compared to the Language Specialization Indices by means of the chi square statistic. Table 95 shows the result. A significant chi square between the diversity measure and number of peeple with whom.particular languages were used was found. The relationship was positive. _The greater the number of people with whom the individual used particular languages, the greater the number of different languages he perceived his groups as preferring. The sample size was 90. TABLE 95 Relationships between Language Specialization and Divere sity of Language Preference Language Specialization Diversity of Language Measures , P M Languages 2 People X * Languages Places Languages Occasions tSignificant at the r05.level Hypothesis IV b)L Language spe giglizgtign is related to thg per . ggivgd hgtgzgggggi ty of language liberalism l. - ub .ste ”social.class and“ ion. The.second part of Hypothesis IV was tested by dividing theerespondents into two.groups: “almindividuals who,sawutheirwreference,groups.as.homogeneous,mi-e.,rall.thet _ reference groups had liberal language Indices or.all_had .197 non-liberal language Indices and b) individuals who saw.some of their groups as liberal and some as non-liberal. Whether or not a group had a liberal language Index was determined in the following manner. The distributions of‘Language Liberalism Indices for the fathers, mothers, subcastes, social classes and regions were each dichotomized as close to their respective medians as feasible. Higher indices (indiceswabove the.median) indicated liberal attitudes: lower indices, non-liberal attitudes. The cut-off scores for liberal and non-liberal indices are given in Table 96. TABLE 9 6. Cut-off Points for Dividing Language Liberalism Indices of Reference Groups into Liberal and Non-Liberal Categories W W *W ' m E Sense 9?. Father 20-29 11 9-19 13 Mother 19-30 5 15—18 6 Subcaste 16—30 9 .llelS .10 Social Class .19-23 .8 9—18 8 Region 19-25 10 14—18 1.2 Meanflspecialization.measures.werewcalculated,fior the.two groups of respondents whoseareference groups were whomogeneous : those whose reference groups were.hetero- geneous. The means were compared by t.test. Four of the .six comparisons yielded significant.differences. Table 97 shows these comparisons. 198 It is apparent that all kinds.of specialization“ peeple, places and occasions,were related to how heterogene- one an individual's groups were seen as being. Specialization by place was especially related to heterogeneity of language attitude. ‘ Far fewer individuals saw their groups as hetero- geneous than as homogeneous. On all of the specialization measures, respondents with heterogeneous groups had higher means, i.e., Indians who sawtheir groups as heterogeneous said they practiced more specialization of language usage than did Indians who saw their groups as homogeneous. Although the hypothesis made no prediction regarding differences in language specialization between those whose language groups were entirely liberal and those whose groups were entirely non-liberal it is interesting to compare the means for these gnoups with.each.other and withthe mean for ‘ heterogeneous-groups. Table 98 shows these means. The table indicates.that.in general. the least langu- age specialization was practiced byerespondents whosemrefer- ence.groups all had non-liberal,indices- Higher means were found where groups all had liberal indices and the highest means were found where the groups were heterogeneous- Four out of six times individuals whosaid their groups .all had non-liberal.attitudes.claimed to practice the least special- ization of language usage. Individuals whonfelt their groups all had liberal language attitudes professed more specialization of language.usage.and those who perceived 199 a mood HH.m mm 0N.H HmoN mo. 0 mm. HH.m mm 50. Onom Ho. Ho. no. u nouaocn emesmcsa Heuonfinlcoc e>sn meow m mo.H mh.m m mn.a -.v «H oo.a mo.m ma ¢N.H «w.m one amuenaa «>0: 080m .mosouo ouceueuem usoecemeuoeem neuaocH eoesmcea . Heueanncec om mm.a mo.m om ma.a mm.~ mm om.a he.~ mm mm. oo.~ e>un Has no nooaocH omenmcen damned 25: Had .emsouo eoceuouem . msoecemosem m die s. m .olqm 3 -..m .old g a dim 3 133me Ilmuuflmll damaged. Jamaal lemonade—J succueuem neusesez.ceaueuaasaoomm omesmssa one: hm Manda ~ ~v- “NW“ , ... em...“ _ ~MA.‘~" . . ... nun-.....w _ goo. . 200 HH.m Adom mmom vmom coon 0N.N .Idnmaflll. Ilddumflll oceaesuuo nemnsmcen .UOUHUGH emesmceq Heuenfinlcoc mn.m mm.v mo.m Nw.m o>e£ case one asuonaq e>sn 036m .omoouw euceueuem useecemoueuom mooaoca emeomcsq Heuenan oo.~ «m .m oo.m mm .m 28s 34 .335 oocoueuom nooocemoeom . oeuwocu emssmcsa asuonaa oo.m oo.m om.~ mm.~ Ices o>e£ Had .eosouo euceueuem nooecemoaom landfill .Ildflflnfll. Iduflml .JIIAuflafil. neusam nemesmcsa. oaooem nommsmcen umnauwwuuauuuuflm mandamus madmmuununuuum umuzmauq . g museum oucoueuom assocemeueuem use Heuenaqr.asuenuqlcez Mom eeuoemez.coaueuaaouoemm emesmceq.ceez mm mdmdfi .‘-__"n. .ra-w. —‘ _ ,- .fp-~~r" 201 some, of .theirgroups to be liberaland some to be non-liberal said they did the most specializing of all. A second test of the hypothesis was made. For each respondent. a range of language liberalism was calculated by subtracting the lowest Language Liberalism Index from the highest Language Liberalism Index found among his groups. The ranges were then correlated with the Language Special- ization measures. The correlations were based on a sample of 20. Table 99 shows the only significant.correlation to be between the range of Language Liberalism Indices and number of languages used on particular occasions. 'The greater the range of language liberalism among an individual's groups, the greater the number of different languages he used lonparticular oceasions. TABLE .99 Correlations between Language Specialization and Range of Language Liberalism Among Reference Groups W Language Specialization Range of Language M33§U£fl§ ‘ Liberalism Indiges Languages Reople Languages Elaces Languages Occasions r = .46* {Significant.atrtbeuiosnlevel. 202 Hypothesis IV c) L ia izatio is r at d ' e c ‘l s c ste soc cl ss, 3mm This part of the hypothesis proved difficult _to test because of the.skewness_of.the importance and the relative importance ratings given the family, subcaste. social class, and.region. Only three out at.the possible 48 relationships were significant.” The importance ratingswfor_family were.very. high.. Table loo.shows that no relationships were found be- tweenfilanguage specialization measures and themimportance ratings.ofcfamily. No relationships.were found between.langu- age specialization and subcaste importance. either. Two significant correlations appeared between social class and language specialization. -The correlation of -29* (based on a sample of 47) indicates that the higher the relative importance of an individualts.social class. the more language specialization he practiced in.particular places. The.negatiye coefficient.of ~r36* Llample.size was 401Mindicateswthat«thembigher_theurelativemimportanceof one's social class, the fewer the languages used for parti-. cular occasions. ‘ I Jansignificant-relationship.also“appearedwbetwesn.the .importancezratings.given.the,regionwandramlanguageespecial-» ization measure- In general, ..region ,.receixz.ed arather_.low' importancewratings—egenerallyuin“the”onemtocthree[Laamewhat important -to unimportant) Jugs. unethofithe4peoplernamed two or three peeple with whom particular languages were used. -2Q3 Thus,.there was a heavy clustering.at the low end of both scales-emany low'importance ratings and few people. ‘As might be expected. the significant.chi.square was for the number of different peeple named. The sample size for this ,chi.square was.87. Visual.examination of the chi square: table indicated the relationship to be negative: the fewer the people with whom particular languages were used, the higher theimportance rating given the region. DISCUSSION: HYPOTHESES IVa, IVb. IVc This hypothesis which had predicted relationships be- tween the perceived language attitudes of reference groups and the specialization of language usage practiced by students received much support. I an y onscrelationsh p between.the specialization. measures and diversity or.language“preferencecmeasures was found. It occurred for the.9people“.specializatien measure. Diversity.ofmlanguage preference was_acmeasure of what.the . individual thought that.hit groups liked. The {people” language specialization msasurewasmawstatement of what the .. individual mean. didwith differentpeople. The positive .relationship between the.twn.measursscindicates that the respondent .said .11.. did with people what .he thoughthiscgroups The.relationship.found.between“diversity.ofngroup ,preference-and the fipeople" specialization measure.wes.not .repeated for.other specialization measures. Thewfact.that 204 .ao>od mo. ops an unuuauacmam. amm.luu nooaenouo nemesvoen sm~.nu , eeomam oomnomcna aux oamoem memosmceq «uuuuuuusq mileage. umaeuuuqam.-uuuuqq_nuuaumuumam.n:1qu e mannequuaaminliquqq. unnunnmammmueui e>wueaem auueoaH_oedmb umaem luuomam e>ausaem quomfia -e>«AWaem_luuemEH coauouaamaueom one enmao Heaven on monfim .haflfimm OmdflmnOA . _ . seamem one unsao Heaven .eueeunom .aaaaem we coomuuomaH one coaueudanauemm swoomooa oee3uen anacondauoaem OOH NAQGB 205 the significant.relationship occurred only with.what appears to be the most appropriate of the specialization measures (“peeple". as compared with “places“ or "occasions“, suggests the division of specialization along these lines to be a.mean- ingful one. All three types of specialization measures (people, places and occasions) related to heterogeneity of Language Liberalism Indices among reference groups. The most special- ization was done by those whose reference groups were hetero- geneous with regard to liberalness of language attitude. It may be that theseoindividuals pleased their conservative groups and then did what pleased themselves elsewhere. The BBCDDdJBOBt specialization. was practiced by respondents who Aperceived all of their groups to be liberal. A possible interpretation.here is.that if an individual's groups were all very.liberal. than he was free to use whatever languageLs) he chose.all the time. The.least specialization was practiced by those whose groups were.seen as beingmentirely conservative. This result.is understandable if one remembers that the refer- ence groups were both conservative and homogeneous. In such a situation the language behavior appropriate for one was probably apprOpriate for all. It wasmfound that the greater.thewrelativeiimportance of social class,.the greater theinumber.of places.in which particular languages were -used .and the.fewer .the._number of languages used.for particular occasions. Rresumably._ifhsocial class were important to an,..individual in India. ..he would ”be .206 careful to use the proper languages in the preper places - with inferiors, with equals and with those who rank above him. For such an individual it would not be the occasion or ceremony which would determine the language to be used, but the place instead. The greater the importance of region the fewer the peeple with whom particular languages were used. This find- ing makes a great deal of intuitive sense. If the.region is important to the respondent. then he probably uses the ' regional language at every Opportunity. This in.turn means, that there would be very few peeple for whom he would employ other languages. 207 HYPOTHESIS V. INTERACTION IS RELATED TO NATIVE TONGUE, PERCEIVEDILANGUAGE LIBERALISM-OF REFERENCE GROUPS AND‘TO FEELINGS ABOUT THE REFERENCE GROUPS ‘ Hypothesis V a I i s r d to native ton u t in the Indian.ggmmugity on the.Michiggp Statg ggmpgs Two measures of interaction were used to examine this hypothesis. The first was,a measure of frequency of contact. The respondents at Michigan State University were presented with a list of Indianson campus and asked to indi— cate who they saw: a).occasionally. b) fairly often. and c) very often. The preportion of individuals within their own native tongue group they saw either occasionally, fairly often. or very often was calculated for each respondent. (Native tongue was determined from the item on the question- naire asking what language the.respondent considered to be his native tongue). Theproportions were.averaged for eadh native tongue group. i.e., a mean proportion of within-group contact was calculated for each.language;group. There.were five languages listed as the native tongue by.five or more Indians! Hindi. Gujarati, Punjabi, Tamil, ..and Telugu. Table 101 shows.the mean proportion of contact.in.adl categories within.eadh-of these native tongue groups. The same procedure.was followediinmthe calculation of two.other.mean proportions of contact. The proportion of individualsAwithin{theinative”tongueugroup.seen.fairly.often or very often was calculated as was.the proportion of.indi- viduals within the native tongue group seen very.often. Table 101 also shows these mean preportions. 208 The table shows that except for Punjabi speakers,the speakers of these languages who had any interaction at all in the Indian community did so with more than half of their language group. The same is true when it was the 9fairly often" and "very often“ interaction category that was being considered or when it was the "veryoften" category. The exception was the Hindi group which was quite a bit larger than the others. Table 101 also-shows the two South Indian languages. Tamil and Telugu. to have the highest within-group interaction rates. This substantiates informal reports from the south Indians that they "stick together” because they feel North Indian students didn't really like them. TABLE 101 Mean Preportions of Contact in Native Tongue Groups c ' N 'r G Native Number in Mean Rropor-r Mean Eropor- .Mean Progor- Tongue Native tion of Con- tion of Con- tion of on- Group Tongue tact: tact! Fairly tact! Very Group Occasional, Often and/ or Often Fairly Often, ‘Very Often and/ or Very __9f..ten___ _ Hindi 14 .72 .44 .21 Gujarati 7 .82 .60 .57 Punjabi 5 . 31 . 20 .06 Telugu 5 .9 s .95 .50 The second measure .of interaction ..used was a measure of friendship choices. The students were asked to indicate 209 their casual and-close friends on the roster of Indian stu- dents. The. proportion of casual inand/oruclose. friends within the native tongue group was.calculated for each respondent. ..A.mean proportion of casual and/or close friends was calcu- lated for each of the five language groups: .Hindi. Gujarati. sPunjabi. Tamil. and Telugu. The same procedure was followed ..to calculate a mean proportion of close friends for each of the five language groups. Table 102 shows these mean propor- tions. The friendship results parallel the interaction ,results. Except for the Puniabis. those Indians wheelisted any friends in the Indian community did so.for.more than half .of their own.language group. .Again it was the Tamil and Telugu speakers who had the highest proportions for the ”casual and close friend" category. For the”close friend“ .category.the picture changed somewhat. .It was the Tamil and .Gujarati speakers who had the highest withinqgroup.ratio. No statistical test was made for the hypothesis that cinteraction_generally follows native tongue within the Indian M-community on the campus. .It appears. however..from Tables i101 and l02.that there.is support for the hypothesis. Hypothesis.V b).I act [related e~h era 1 a 'a er s fa l * su cast , £99131 glasg.gpd rggiop. ' .The measure of interaction used to.test-this hype- . thesis was.constructed to.indicate hownmuch.ofneach.studsnt‘s l discussion about.Nehruls death took.place in the Indian com- “munity. .The measure used was the Nehru.Discussion Index described in Chapter V. 210 Individuals were divided into.categories according to whether their reference groups had: a) all noneliberal \Language Indices. b) all Liberal Language Indices or c) some TABLE 10 2 Mean Preportions of Friendships in Native Tongue Groups m N T G Native Number in Mean Preportion Mean Proportion Tongue Native Tongue of Close and/or of close Friends ..Em. Groun W: i a Hindi 14 .53 .22 Gujarati 7 .9 3 . 57 Pun jahi 5 . 32 .10 Tamil 6 .95 .84 Telugu S .l.00 .31 Liberal and-aomeinon—Liheral Indices. A mean Nehru Discussion Index was-calculated foreachof the .threecategories. Table loamshowanthese means. .Differences between.pairs of.means were tested for.significance (t test) and found not to be significant. Hypothesis V c) I i s r t to e im c o The importance ratings anuindividualwgave his.family, .subcaste. social class and region were.summed and a mean .calculated forethe individual. This mean was an indication .offihowhimportant the-student considered his groups to be.to him. 211 TABLE 103 Relationships between Heterogeneity of LanguagelLiberals and Discussion about Nehru' s Death WW Nehru Disgpssion Igdgx Woe dean ......SD N. Homogeneous Reference Groups: All have non-Liberal .45 .33 13 Language Indices Homogeneous Reference Groups: All have Liberal .45 .37 9 Language Indices Heterogeneous Reference Groups! some have Liberal and some .63 .43 7 have nonvLiberal Language Indices To test Hypothesis Vc. the mean importance ratings were correlated with.the measures of discussion concerning Nehru’s death. The correlation was.based on.a sample of 42. Table 104 gives the result. . The significant correlation indicates that the.more . important the.individual felt his reference groups to be to him. the.less his.discussion within the Indian community .relative to his.discussion on the American.community about the death of Nehru. TABLE 10 4 The Relationship between Discussion about Nehru Death mportance of Reference Groups and Nehru Discussion Index er a -.44** **Significant at the .01 level. .212 DISCUSSION: HYPOTHESES Vao Vb” Vc There was some evidence (not tested for statistical significance) that native tongue was related to friendship and contact patterns within the Indian community. This may have been due to the.influence of reference groups and/or it may have been a psychological factor--a.relaxation from the strain of constantly speaking in a foreign tongue. The.heterogeneity of LanguagelLiberalism Indices among reference groups was not related to the patterns of discussion concerning Nehru°sdeath° Those whose reference groups were perceived to be homogeneous with regard to langtage attitudes talked to the same proportions of Indians and Americans as did those who sathhe r reference groups as .heterogeneous. It may be.that the reference groups in India were too remote to affect interaction patterns in the U.S. It may be, too, that the question of interaction with .Indians vs. interaction withAmerieans was.outsideithe usual _ concerns of.the reference groups. That is, the reference groups may haveqhad much to say about interaction with Indians, but little aboutrinteraction with nonelndians.he- . cause the issue arose infrequently in.lndia. , Interaction. washrelated .to .the impertanceof .the . reference groups. The more important theiindividual felt.his . reference groups,to he to him,,the more discussion about knehruis.death he had within.the American community relatiVe to that withinithe.1ndian.community. It isudifficult.to 213 interpret this finding. It may be that the available American and Indian audiences were of different kinds and that the respondent was searching in.the American community for sub- stitutes for his reference.groups.in India. .The largest number of.llndians present were students .and.it is. a fair assumption that for most respondents the Indian audience was a student audience. The American.audience.could have been any audience--perhaps the same kind of audience as his reference groups in India—-eog., families, professors, etc. The data do not include any.information about the American audience. 214 HYPOTHESIS VI. LANGUAGE LIBERALISM IS RELATED TO AETITUDES TOWRRD CHANGE AND POLITICAL CONTROL Hypothesis VI a) Language liberalism is related to change stitntatien. The hypothesis was tested by relating the respondent's Orientation Toward Change Index. Confidence,.Motivation and Self Percept scales to his Language Liberalism Index. chi squares and product moment correlations were calculated. Table 105 displays the findings. The only personality scale relating to Language Liberalism was Self Percept. The relationship was negative. Individuals who saw themselves as adventuresome and liking to try new things professed the most conservative language attitudes. ,Conversely, those students who were the most conservative about trying new things were the most liberal when it came to the question of language attitudes. The sample size for the correlation was 69. TABLE 105 ‘_R=_: sot-o .; 9’t see Ioaoo-t-:- L o; - rm. "’7 no.0: O _-.; *9 Change 9‘1““an Laggmg gm: align hag“ ' Confidence Motivation Self Percept .r = -.29* Orientation Toward Change Index fl fir *Significantiat‘the”,05fllevel. 215 Hypothesis VI b) Langgage liberalism is related to approval of .Wiai The students" Language Liberalism Indices were com- pared to their Indices of Social Change Approval. Language Liberalism was also compared to the items comprising the Indices of Social Change Approval! Approval.of Social Change Speed, Approval of Social Change Direction and Approval of Social Change Soaps. The comparisons were made by means of chi squares and correlation coefficients. These findings are also displayed in Table 106. The positive correlation coefficient between the approval of the direction of social change and the language liberalism of the students indicates that the more the indi- viduals were inclined to approve of the direction of India's social.change, the more liberal were their language attitudes. The correlation was based on a sample of 79. The significant Chi.square betweenthe approval of the. scope of India°s social .,.change and..languag.e..liberalism appears to reflect a curvilinear relationship. Indian students who expressed liberal attitudes about language tended to feel one of two ways about the.scope of India's socialichangsp they either approved very much or.they.dieapproved very much. Students with conservative language attitudes expressed more moderate approval .of .the scope of. changes. The mp1s size for the chi square.statistic was 76. .llliv Ill-ll} I‘llllellllll ' {it} 216 TABLE 106 Relationships between Language Liberalism and Approval of Social Change in India W W W Approval of Social Change Speed ApBroval of social Change irection r = .24* Approval of Social Change Scope X2* Index of Social Change Approval *Significant at the oOSlevel Hypothesis VI ‘ c? ngggagg liberalism is :glatgg to attitgggs concerning ghggg ggvernmggt pgwer shogld rggider Four sets of measures involving.attitudes concerning where government power should reside were used to test this hypothesis. These sets of measures are described in Chapter v. Briefly they are: i) GoVernment Power measures: ratings of how'much power the Central, state and local governments each should have. ii) Difference-Power measures: differences between the power ratings for each pair of govermment levels. ’ iii) Outcome measures: ratings of,howumuch, if at all, the higher government levels should prevail when two government levels cenflict. iv) Power Index! the sum of the three Outcome measures calculated in a manner such that the higher the Power Index, the stronger the~higher government levels should be in comparison to the ,lower levels.- LanguagelLiberalism scores were examined.in relationship to the.four sets of power measures. CorrelatiOn 217 coefficients and chi squares were used to test the' hypothesis. The results are found in Table 107... 1) Government Power Measures: Significant and positive corre- lations were found between Language Liberalism and two of the three Government Power measures. Individuals with high Language Liberalism Indices accorded the state and local governments a high degree of power. The correlations were based on samples of 78 and 77 respectively. ii) Difference-Power Measures: A significant and negative correlation was found between the Language Liberalism Index and the difference in ..power between the Central and state governments. This correlation indicates that the most liberal language attitudes belonged to individuals who accorded the Central government less power than the state government or who accorded the two government levels the same power. Conservative language attitudes were found for those . who granted the Central government. more power than the state government. The correlation wasbased on a sample of 78 power ratings. iii) Outcomemeasures: The significant -..chi square between , Language Liberalism and outcome .in . conflicts.between.-Central .. and state. governments reflects .a curvilinear relationship. . . Themost .liberalJanguage. attitudes were -held,.by. those indi- viduals who felt that..the. higher level . of government. should = ..usually dominate the.lower- S.tudents.with-conservative . language attitudes. tookextremes .on the government ..oonflict question. They.chose-.either to minimize or .maximize...the 218 dominance of the Central government .over the state government by indicating that the Central government should always pre- vail or it should prevail one half or less of the time. The sample size for the chi square statistic was 77. iv) Power Index: Neither the chi square nor the correlation , coefficient .calculated between.the Power-Index and.,the,.Langu- age Liberalism Index was significant. TABLE 10 '7 Relationships between Language Liberalism and Attitudes Concerning Where Government Power Should Reside ,r___,- .___.__. ..— « .....— Attitudee Concerning Govern— .Eaassass_Libsralism_lndex_ .... ment 1’er .sts£_meass£ss Central Government Power ‘. ' State Government Power r = .24* Local Government Power r = .24* D - o . .. L's Central minus State Government 1' =-. 23* State minus Local Government Central minus Local Government Qntssmsnmeassgss Central vs. State Government Xz" State vs. Local Government Central . vs. Local Government Power Index fl 7 *..5ignificant-.wat the. .115- level DISC08810NI srpomsssss VIa, VIb. VIc Hypothesis V1 .pnadigted relationships between language 7 attitidesand attitude--ttoward change..-and-npolitical_.control. .219 Most of the findings were in the direction opposite to that expected. The individual who pictured himself as wanting to explore and liking to try new things had non-liberal or con- servative language attitudes. This was contrary to aspects- tion and is difficult to explain. Indians who hadalfberal language.attitudes approved. of Indials social change. This finding was consistent with the prediction. Presumably. liberalness in the realm of language was associated with the.desire to.see.Indiammoderh- .ized. The validity of this interpretation depends on the interpretation of “direction of.aocial.change“. The respond- ‘ents may or may not have viewed Indiaispsocial.change.as directed toward modernization. In congunction.witb this was a second finding.ooncern- ,ing.social.change.. Respondents with liberalilanguage atti- tudes took one of two.extreme positions. They either approved very much or they disapproved very much of the scope of India's social change.. Interpretation of this finding is difficult. More information about how social change was viewed is necessary. It may be that respondents with libera1.language attitudes would haveeliked to see large scale social changes..but.differed among themselves as to the scale the changes should be at.that point in Indials progress. That is._some.may have.felt change--Ihould start in all areas “at-onceowhile.othershfelt.thatathe”scopeflshouldibeasmaller at the- start and “graduallyincrease. Bytheasametoken. ,220 respondents with liberal language attitudes might all have agreed that the scopeof India's social change should be small (er any other degree of scope) yet differed among themselves as to how'large they perceived the scope of change actually to be. , Indians with liberal language attitudes were found to accord high power to the state and to the local governments. They also minimized power differences-between Central and state governments. The findings seem to reflect one of two things. Either the respondents with liberal language atti- tudes simply were liberal and wanted to give every government level much power or they were identified much more at the regional and somewhat more at the local level than were their fellow countrymen who possessed more conservative langu- age attitudes. 221 HYPOTHESIS VII. LANGUAGE SPECIALIZRTION IS RELATED TO ATTITUDES TOWIRD CHANGE AND POLITICIL CONTROL Hypothesis VII a)L l to s latd c'e merisntatien. To test this hypothesis. the six language specializa- tion measures concerned with language usage in India were each related to the Orientation Toward Change Index and its scales. The statistical tests used were chi squares and correlation coefficients. Table 108 presents the.findings. The only significant comparison out of the 24 made was be- tween Self-Percept and number of places in which an individual said he used certain languages. The Self Percept data were fairly normally distributed. The number of placesmnamed was skewed_with most peeple naming two or three. It is not clear what pattern.exists between self Percept and the . number of places although it.may well be.a positive one with Self Barcept increasing.as the number of places increases. The chi square was based on 70 individuals. TABLE 108 Language - ~ Specialization { 4 Self . ‘ Orientation ‘Measures Confidence Motivation ~Rercept. Toward Change . .. . ' _ , . Languages Panel-a Languages . Elaces' 82* ' Occasions 222 Hypothesis VII b) ngggggg-gpggialization is gelated to approval ”Language specialization measures were examined in relation to the Indices of Social.Qhange Approval_and their components. The statistics used were correlation coefficients and ohi squares. The 24 comparisons yielded four significant findings. Table 109 presents the findings. ~A significant Chi square was found between the Index .of Social Change Approval.and the number of different people with whom an individual used.perticular languages. The pattern of responses was not clear, but appeared to be curvi- linear. Students who expressed little or great approval of social change also indicated that there were fewhpeOple with whom they used particular languages. _Conversely, individuals expressing moderate.approval of.the social change in.India were the individuals who said that they used particular languages with a large number of peeple. The chi square.was based on.data from 76 respondents. The number of languages used.in particular places related to all three components of the Index.of-$ecial.Qbange Approval. The.relationdhip with Approval of.aocial.Qhange fipeed.tended.to be U shaped and themrelatienships with direction and scope were negative. In the first instance. feelings. that India's changes werencnurring .either too ...slofly or too.quickly.werewassociatedsaithMgreater-languagewspecial- ization whilermoderete approval of social changewspeed was . associated with lesser Language specialization. .Appnoval,ef 22.3 the direction of India's changes and approval of the scope of India's changes were paired with a lesser amount of langu- age specialization. The Indian student who said that he used a large number of languages in particular places tended to feel that India's social changes were occurring either too slowly or too quickly.that.about half the changes were in the right direction and that these changes were somewhat too small in scope. The chi square calculated between Approval of Social Change Speed.andlanguage specialization by place was based on 77 respondents. The correlation coefficients for direction and scope were based on samples of 79 and 78 respect- ively. TABLE 109 Relationships between Len age specialization and Approval of social ange in India Language Ap roval of Apgroval of Ap roval of *Tndex of special- ocial ocial ocial Social ization Change Speed change Change Change W ,l 21.95229... .2222;— Eternal Languages People x2* Languages 2 Elaces x * r = -.26* .r = -.28* Languages Qccasions *Significant-at-the n05.1eve1. 224 Hypothesis VII c) Langnagg specialization is related to attitudes concerning finggg government power should reside. Hypothesis VIlc was tested by examining the relation- ships between languge specialization measures and i) Govern- ment Power:measures, ii) Difference-Power measures. iii) out- come measures and iv) the Power Index. Correlations and chi squares were computed between each of the language special- ization measures_and each of the power.measures. The signi- ficant results are diSplayed in Table 110. 1) Government Power measures: Two significant relationships were found between the power accorded the Central government and the language specialization measures concerning languages used in particular places. Both the correlation coefficient computed for the.number of languages used and the coefficient computed for-the number of different places in which languages were used were significant. The correlations were both computed on data for 80 respondents. The correlations were significant at the .05 level- Table 110 shows these relationships. In.bothminstances,Mthshcorrelations.aremnegative. indicating greater specialization.(i.e. more languages named .and more.places.cited) was associated.with.a desire.to.accord the Central government.less power. No relationships appearedcbetween the language -specialization measures-and.the_pewer-accorded the state _goyernment. 225 For the local government, two statistically signi- ficant relationships emerged from the six comparisons. Chi squares, calculated on samples of 80 and significant at the .001 level, were found between the power granted local govern- ments and the number of languages used a) with particular peOple and b) in particular places. The exact pattern of relationships was not clear from the chi square tables. ii) Difference-Power measurest Language specializa- tion measures did not relate to any of the difference-Power measures. iii) Outcome.measuresz No significant relationships were found between.the language specialization measures and‘ the questionnaire items asking what the outcome should be when pairs of government levels were in disagreement. iv) Power Index! A positive and significant correla- tion was found between the number of languages used with particular peeple and the.degree to which oonflicts.involving government levels should be.resolved.in favor of the higher government levels. The correlation was based on data from 79 respondents. It indicated that respondents who-tended to .resolve government.conflictsnin-favor of the higher govern- ment levels-also tended.to—use.a relatively.large.number of languages when conversing with.particular people. DISCUSSION: HYPOTHESES VIIa, VIIb, VIIc Hypothesis VII predicted. .relationships .between...langu age specializationvand.attitudes.towardrchange.andhpolitical control. Support for this hypothesis.was.found. Muchwlanguage ,226 Specialization was practiced by students who saw themselves as adventuresome and interested»in trying new things. There was nothing in the data to explain why this might be. A possible explanation is that an individual with such a personality might get himself into a.variety of groups and social situations and such.groups.and situations might require greater specialization of language usage. Conversely, stu- dents who didn't like new eXperiences might.have had more restricted ranges of social settings, i.e., homogeneous social settings which required the use of very few different languages. Language specialization related to approval of India's social change in a rather bewildering manner. Great .Specialization was associated with feelings that social change was occurring too slowly. Little specialization was associ- ated with feelings that social change was in the right direction, but that it was too small in scope. Interpretation of such.findings is difficult. It was asswmed that the social change occurring in India was primarily Western and flmodern“ intnatnre. Indian students may not.have viewed their social change in this light. They may have.thought of it as occuring.in some.other.manner. Un- fortunately, no.questions were asked.to determine justihew . the changes were.viewed. If.the students.didiindeed view the . change as Westernization. those amok felt .it was not occurring rapidly enough may have been sitting in the middle of the change, i.e.. they may have lived in very heterogeneous worlds 227 NGGHMMMMO mflmlmMMMl tMN 0|".H g sssNx .mNO' " “ nfimwmmcnfl .Ho>oa Hoo. as» no ucnoaussmam.cc .H¢>¢H mo. an» an uauouwaamam . L, «mm. «H NewcH.uesem ucofisuo>ow Hausa .m> Huuucou ucoacwe>sa aeueq .n> ousum uceasuobou ensue .m> Heuuseu mludflnflfidflfiqufim .u.>ew Hausa asses Heuucoo .u.>ew Hausa asses eumum .u.>ew ensue asses Heuusoo , m n essNN H030“ HGOEGHO>OU HGUGA H030m HCOEflBP>OO Cumum Hm30& unmfidhfibew HGHHQGU mmuumumawuuuam Idmmfll .Ifldflflfidfll , aided-dad; a; g enamem vasonm Hosea OHM mdmda adoauaaem Inerzimsucueuceu sensuauea use ceauuuaaeaoeom emmsmcmanceosuen unanncoausmem ..,._ -‘Wfq'r-wyw‘ ----;--r—‘ "; “—‘ikamrn * ‘ -_ 3 -.;.'_~————~..._‘L~-.o‘- '» ... .--~o~ , ' -“-‘.J-V’ ~_. ,.o._.g‘.~.~,_ . « 228 which necessitated the use of many languages. Further, those who needed to practice little specialization might have been living to a large extent in the traditional India or to a large extent in the “Western“ India. _The result of eitheriof these alternatives could have been social situations which _were rather homogeneous in terms of their demands for parti- cular languages. Another interpretation of the findings about language specialization.and social change.approval concerns how itired the respondents were of the social rules regarding their language use. If the respondents were tired of follow» ing a multitude of such social rules, then the feelings of ”the great ”languagespecializers".that social changes were too slow.in coming.are quite understandable. Inilike manner, .feelings that social changes were in the.right direction but _too small in scope on the.part of those practicingilittle language specialization are .also .understandabl e. Language.specialization related to where the student Ifelt.government power should reside. Those who generally favored the.balance of power.at higher government levels practiced much language.specialization. Little-specializa- .wtion. however. was practiced by.those according the Central .government much power.in.general. The.findingsnappear.to be contradictory.n this is not completely true because.the .-measuresiused to test the hypotheses differed. the first Igovernmentmpower.measuremdiscussed washbasedwnponirelative power..i.e., Central vs-.state, state vs.ilocal..and.central vs. local govermments. The second power measure was not a ‘229 relative one. It was simply airatingiof_howumuch power the respondent felt the Central government should have. The specialization measures differed too. It was the number of languages used and the number of places in which they were used that correlated with the Central government power rat- ings and the number of languages used with particular people which related to the Power Index. To summarize. the. somewhat .confusing...findings: A high degree of language specialization was practiced by those who generally accorded the Central government a low.degree of power but who favored the upper government levels when there was conflict between levels. 230 HYPOTHESIS VIII. INTERACTION IS RELATED TO ATTITUDES TOWRRD CHANGE AND POLITICAL CONTROL Hypothesis VIII a) Inte tio s elated to chan e rientation. The Orientation Toward Change Index and its scales were correlated with the Nehru Discussion Index. Chi squares were also calculated between the personality measures and the Nehru Discussion Index. None of the.relationships was .significant. The personality measures were not.related to the proportion of discussion occurring within the Indian com- .munity.relative to that occurring within the American community. Hypothesis VIII b) Interactiog is related to approval of social Chi squares and correlation coefficients were.calcu- lated between the Nehru Discussion Index and i) the Index of Social Change Approval and ii) the items comprising the index of Social Change Approval. Table 111 presents the results, IA positive correlation, significant at the .05 level, was found between the Nehru Discussion Index and approval of the direction of IQEI;VE' social change. _This means that Indian students who approved of the direction of India's social change conversed mainly with other Indian students about Nehruls death.-.The sample size was 57. Hypothesis VIII c) I t a t s ated to attitudes c ncern n where government power should reside. To test this hypothesis the Nehru Discussion Index was paired with each of the 10 power measures. Both chi squares and correlations were. calculated betweenetheNehru Discussion Index and the power measures. Table 112 shows the results. 231 TABLE 111 Relationships between Discussion about Nehru's Death and Approval of Social Change in India W W Em: , Discussion Index Approval of Social Change Speed Approval of social Change Direction r = .28* Approval of Social Change scope Index of Social Change Approval *Significant at the .05 level. i) Government Power measures: A significant chi square was found between the Nehru Discussion Index and the power accorded the local government. The.relationship appeared to be curvilinear with the extremes.of almost no interaction with Indians and almost all interaction with Indians associated with the least power for the local govern- ment. The middle range, where Indian-non-Indian interaction was about evenly balanced, was associated with the greatest amount of power for local government. The chi square was based on a sample of 56. ii) Difference-Power measures: No significant relationships were found between Difiference-Powar measures and the NehruDiscussion Index. iii) Outcome measures: When the.power-measure was one of who should prevail in a situation of clash between “.government levels, significance was found for the outcome .concerning.Qentral.and state governments. Once again the relationship was curvilinear with the greatest and the least 232 proportions of interaction with Indians matched with the least power difference between the Central and state govern- ment. The middle-of-the—roaders whose interaction was about half Indian and half non-Indian gave the Central government level a greater edge over the state level. This chi square was calculated on 55 individuals. iv) Power Index! ‘A.significant chi square was found between the Power Index and the Nehru Discussion Index. The chi square was based on 45 individuals. Visualoinspection of the chi square table showed the relationship to be curvi- linear. Very great and very little discussion in the Indian community were associated with a low Power Index. A moderate amount of discussion of Nehru's death within the Indian com- munity relative to the amount of discussion within the American community was associated with a higher Power Index. TABLE 1 1 2 Relationships between Discussion about Nehru's Death and Attitudes Concerning Where Government Power should Reside Attitudes Concerning Government Power 1N3h29_21§cu§§ign_gngex Power measures Central Government Power State Government Power 2 Local Government Power X ** W Central minus State Government State minus Local Government Central minus Local Government Outcome measures Central vs. State Government X2* State vs. Local Government Central vs. Local Government Power Index x2* *Significant at the .05 level. **_QJ~-\4“A-n‘n go‘- ‘DLA n1 1 Aer1 233 DISCUSSION: HYPOTHESES VIIIa,VIIIb, VIIIC Hypothesis VIII predicted a relationship between interaction and attitudes toward change and political control. No relationship was found between interaction and change orientation. It was expected that reapondents with a posi- tive change orientation would seek.out Americans. The leak of support for this hypothesis may be a function of the.inter- action measure used to test the hypothesis. The interaction measure employed was not a general one. It was a very specific one about a very Indian event~—the death of Nehru. Relationships were found between interaction and approval of India°s social change. .Indian.students who ,approved of the direction of India°s social change discussed Nehru°s death with Indians more than with Americans. ,Inter- pretation of such a finding requiresmore information about the discussion than was available in this study. Possible reasons for such a finding are that the ”approvers" of India's social change credited Nehru with the change and as such mathave wished to mourn India's loss tagether. ,Eurther, they may have been.verynconcerned about India's future-~who 'would direct India now and would these changes they liked be continued? The respondents at the other end.of the continuum ..were,not-strongHdisapprovers—-they”thoughtcthat«about.half.the changes were in the rightwdirection and.half in the wrong -direction...Perhaps.for them, the worry that the changes might not.oontinue or continue in the.same manner was not so great. Other kinds of changes might be just as welcome. 234 Interaction also related significantly to attitudes concerning where government power should reside. Indians who discussed Nehru's death in about the same prOpertion in the Indian and American communities gave the local government more power than did their compatriots, but tended to favor the Central government over the state government when the two conflicted.. Indians whoseninteraction was primarily within the Indian. communityor primarilywwithin ..the Anterican com- munity accorded the least power to themlocal government and ”desiredithe least power difference between.the Central and local government in situations.of cenflict. If discussion of Nehru"s death can be taken as.an _indication of cultural.identification,.then it was the Indians who identified.with both the American and Indian cultures who wished local governments to.havena fair degree of.power and who wished also to see the Central government dominate the state governments. Their compatriots.who«presumably identi- fied primarily with either the American.or Indian.culture generally.gave the local government little power and gave thev Central and state governments the same power. They.seemed to be interested in less power at both the local and Central levels. This may meanethat their identification in India was mainly at the regional level. 235 HYPOTHESIS IX. ATTITUDES TOW3RD CHANGE AND POLITICAL CONTROL ARE RELATED TO SOCIO-CULTURAL FACTORS Hypothesis IX a) C an r ‘n o s related to rural-urban insides“.— Contingency tables were.made and a chi square and correlation coefficient calculated between the rural-urban measure and the Orientation.Toward Change Index. The same procedure was followed between the ruralaurban measure and each of the three major.components of the Orientation Toward Change Index. The sample sizes for analyses of urban resi- dence with the Orientation Toward Change Index and its compo- nents ranged between 57 and 84. None of the statistics was significant. Hypothesis IX b) Change orientation is related to amount.gf 55:231. Correlation coefficients and chi squares were.calcu- lated between the Orientation Toward Change Index and seven travel measures. The seven travel measures were those used to testhypothesesI and II. Three measures concerned travel inside India,.two concerned.travel outside India and two.were measures.of vicarious.culture contact. .Chi squares and correlations.were also computed between the scales on.the Orientation Toward.Change Index-andlthettravel.measures: Table 113 showe.themeignificant.relationships found. Sampleesizes.were.57-erc58-for.etatistics.invnlving the Orientation Toward Change Index. 66 or 67 for Confidence, 84 or 85er Motivationaandhalor 82 for Self Rercept Baal-8,. ,236 The general measure regarding the amount of travel in India correlated significantly with the Orientation Toward Change Index. The higher the respondent's Orientation Toward Change Index, the more general travelmhe had done in India. No other measures of travel inside India related to the Qrientation To- ward Change Index or its scales. One of the measures of travel outside India related to change orientation measures. The number of trips.made out- side India correlated in.a significant and.positive manner with the Orientation Toward Change Index and with the Self Percept scale. AThe higher the individual's Orientation To— ward Change Index.and the higher his Self.Percept.score. the greater the.number of trips he had made outside India. One of the measures of vicarious culture contact.a1— so.related to the Orientation.Toward Change Index and the Self Percept scale. The relationships were positive. The more reading aboutwother cultures an.individual.had done. the higher his Orientation Toward Change Index and the higher his Self Rercept score. Hypothesis'IX c) C An.index.of social class was.created by summing z scores on seven variables: .perrcapita family.income..number of books in family library, number of.aervantsuin family, .father's occupation. father's educational-level. mother's _educational level. and proportion of college—age brothers and sisters.attending college. ‘A fuller description.of.this index appears in Chapter V. .Ho>oa mo. ecu an unmoauasmem. “J, I cmm.uu you." w eom.nu xeodH emcsno n ousaea coausunewuo a W .mm.uu .m~.nu udeuuon «How i sowus>auez.w sudevamsou w gag; gldAGlelJaquluJallddIu-fldwl gag. saga you» nesuo.use34 neauucsoo enaeuso ca He>sua oe>aa nemsaaa>_ods ; Icoo maafism mcaosom eosz_nd«ue asueooo «ensue ed3ee .eewuuo M .30 so. , >43 Iago a W. E He>swa one sequenceauo emdsco ceezuon.ndwnnsoHumaem maa mqmdn. 238 Chi squares and correlation coefficients were computed between the change orientation measures and the Social Class Index. The Social class Index did not show any relationship to the Orientation Toward Change Index or to its components: Motivation. Confidence or Self-Percept. Although the-total measure of.aocialmclass did not relate to.change orientation.msome of.its components did. Table 114 shows these relationships. The.relationship.betweenSelf.PerceptWandrnumber.of servants appears to be.awpositive one with fewer servants associated .withlessinclination .to .seeoneself as liking to try.new things. A significant.positive correlation was found between mother!s educationiand the respondent's Selszercept score. The correlation indicates that the greater themeducation.of the mother, .the-more...the.son...saw..-himself.asinterested in experimenting with new things.. Sample size for.the corre- lation was 79.. Althou h thensamp e was not.small. there was a concentration of mothers in the low education categories. Beyond.therevelpomeoderate.education.the.data.were.not .sufficient,to.test-theirelationship-betweenWmotherls-educa- tion and other.variables. Hypothesis IX d) C t s ed to l n . .. It A mean Orientation Toward Change Index waswcalculated for students who were.thefirst...in .their. families .to study abroad end.a mean was calculatedwfor those who were not the 239 TABLE 1 14 Relationships between Change Orientation and Social Class W Shanda—033m Social Class Confi- Motiva- ' self Orientation dence tion , Percept Toward Change _lndss___ Per Capita Income Books in Family Library 2 »-Number of Servants X * Father's Education Mother's Education. r=.23* Fatherls Occu tion‘ Brothers and isters in.College Social Class Index *Significant at the .05 level first-in-their families to study abroad. The means were compared by t test. The same.procedure was repeated for the scales forming the Orientation Toward Change Index. Mean Confidence, Motivation. and $elf.Percept.scores were calcu- lated for both groups.of students and compared by t test. eflo.significant.differences were.found. Whether or not an individual was thveir t in his family to study .abroad bore no relationship to his.Orientation Toward Change Index. Confidencerscore,‘Motivation.score or SelfPercept score. Hypothesis . IX e) ' t N -8 The Orientation Toward Change Indices were examined separately for North Indians and for South Indians. A.t test was calculated between the mean Orientation Toward Change Indices 240 rfor the two groups of students. The t test was not significant. “—..; The same procedure wastollowed with the Confidence, Motivation, and Self.Percept scales. North Indians-and South.1ndians differed significantly on only one scale, Motivation. The t. significant at the .05 level..indicated that lower motivation was associated with residence in the South and highem motivation with.residence.in.the.North. Hypothesis IX f) A v f a h n n I d s el te 'eur. . de c . _, ..Correlationsand -..chi. squares- were- calculated- between rural-urban residence and the Index of Social Change Approval. It.will he recalled-from Chapter V that this Index is a.cemposite ofWresponses torthree items concerning uapproval”ofithewspeed.wdirection”andwscOpe.ofrIndiaisWsocial _change. Correlations and.chi squares were also calculated .WbetweenWruralsurbanmresidencewandmeach.of.thewthreeritems comprising the index. iTheicalculations were basedon samples of 88 and 85. Nousignificant“relationship.wasmfoundmbetween.rural- urban.residence and the Index of Sociaerhange Approval. ‘One of the.three ommponents.of the index related torurban .Mresidence..“Amnegativercorrelation-wasrfoundmhetween-approval ..of the speed at which.change.was progressing and rural ..residence.. Individuals w o felt that thatchanges.were be- .. . ing made too ..quicflcly.-.or.-at..ahout.-the...right....speeimutnoften .came.from.smallmtowns.MWTahlemlls-showsmthisirelationship. . II I VII- I 4 ‘V ll-l l 241 TABLE 11 5 Relationships between Approval of Social Change in India and Rural-Urban Residence W I ADDIQXI] 9f SQQIQJ @3392 . Approval Approval Approval Index of Social of Social of social of Change Change Change social Speed Direction Scope Change Amman. Rural-Urban Residence r=-.25* #7 7 *Significant at the .05 level Hypothesis IX 9) A i I W- The travel measures wereirelated.to.theisocial change measures by means of correlationmcoefficients and chi squares. The-statistics-were-hasedflon.samples-ranginguinmsizemfrom 88 to 92...Table lls.shows the significant results. ..The number of states lived in by the student related ”negatively to his approval of.therdirection.of social change -in India. Indian sojourners who had lived inrnumerous.states tended to feel changes were in_the wrong direction. .The same relationship.appeared when.individMals.were ..asked how much travelling they had done in India. .Those who ..said“they-had”donemthewmost-travellingitendedutomfeel.less .appnoval for therdirection of India's social.changes than ,thoseewho.had.travelled.less. .242 Negative relationships were found between the number of trips an individual had made outside India and-his Index of Social Change Approval and between the number of trips outside India and Approval of Social Change Speed. Indian students who had made.the fewest trips outside India.had high Indices of Social Change Approval and.felt that India's social change was occurring too rapidly. The Index.of Social Change.Approval did.not relate .-to eitherwofmthemvicarious-culture.contactheasures..but.the 'Approval.of Social Change.Speedudid.. Indians who had.done much reading about other cultures felt social changes were occurring too slowly. Hypothesis IX MW tg-ggg;a;.gigas. Chi squares and-oorrelation coefficients were calcu- .lated.between the Index of Social Change Approval and the socialmclass indices. -No significant“relationships were found: approval of.Indials social.change did not relate to social-class. Hypothesist . 1) Approval of sggial Chang; in India is gglgtgg tgrfamily.traditiog-gf-foreign.stggx. A~mean.Index.of Social ChangeuApproval was calculated for Indian students whoiwere the firstWinmtheir families to .-study.abroad-andrawmean“Indexgwasmcalculatedmfornthewstudents .-who wmre not thveir t inrtheirwfamiliesetoretudyWahroad. . The means . -.were compared by .. t ..test -and.the..-difference.,between -them found not to be significant. .Thetsame.pnocedure1was .Hu>oa do. on» as unmouuaomam .. HO>OH mo. 2.? no unmodudnmfim a anboum@<,omcsnu Houuem no ersH tflm .IHH emoum m. emcmno Heaven mo.au>ou ea “4 c x ..m~.uuu coauuouflo .- - N «9330 Heaven mo Hosea—mad «Hm.lun. aom.|nu oeemm omsmno deacon no Hn>euddd Jqflaldlo H 30 '64.“qu Ices suds henna eoaeuso saccH c« c« ca oe>fiq nouusoo usond oeuaea> eon: He>uu9 oeban nemmHHa> 0cm haasmm msaosem mefiuussou edema Hmueceo aeumum «S369 .ueauau IIIIIMIIQIIIIJ .muqmaamlquqmaqmlmquauuuaw_ an H mamam:qmauq«naqnuum. amounmlauaummluunamuamqqfl. illlllf n'l‘ll..l.llll'v'llIl."li soaeuso ao>nua H¢>aua use oHosH ca emsono Hmauom no au>oumdd seesnen emdnncoaunaem 0.: mAm49 .n ".‘ W‘w-‘x --—¥—.'.4tl—«.—.b;-‘_-lb-tt ‘flI—V-"* WW - ,. _----— ’9‘ ~ ‘_--'»u‘ 244 followed with each of the components of the Index of Social Change Approval. No differences were significant. The respondent's approval of India's social change did not re- late to whether or not he was the first in his family to study abroad. Hypothesis IX 1) Appggvgl 0: social ghangg in India is related 119.—132W A mean Index of Social Change Approval was calculated for North Indians as were means.for each component of the Index. They were compared by t test to corresponding means for South Indians. The t tests were not significant. The approval of Indials”socialmchangemexpressed.by north .WIndians did not differ from that expressed by South Indians. Hypothesis IX k) A t u s. ncer where ove— ent s d r re at d‘to ru 1- rb r si- dence. .The four sets of.measures.about power were related to “rural-urban residence by.means of chiwsquares_and correlation .coefficients. Theonly-significant“relationshipifnundebetweenuurban .residence.and.anyiof.the.measuressconcerningwpowermand .. government levels. was "with. the . power A the Central Comment .should have. Asmmight he expected,.individualsmfromimore "urban“locales favored a.stronger Centralwgovernmenttthan did individuals--from,.mral...areas. . The..-correlation..was. .-31 .... (significant ..at the. ins -.level) v. and. -....waabased on a.-sample .of 91‘. Hypothesis Ix 1) A Correlation coefficients and chi square values were calculated between travel and the power.accorded government levels measures. Table ll7.shows the significant.statistics. The statistics.were based.on samples ranging in size between ,86 and 92. ' 1) Government Power measures: .The relationshipbe- -wtweenmnumberhof“statesilived.inmandnpowerwaccorded.the i.Central.governmentuwas“highlywsignificant.andtappeared to be ‘curvilinear. -Themgreatest andithe least power.was accorded the«Centraligovernmentmby.individuals having lived in more rstatesmand”moderatempower.wasiaccordedmthe‘Centralrgevern- .ment by.individuals.havingulived in fewer states. . . The.v-amount.l..of_...contact themrespondentts family had .with.non-Indians”wastrelatedmto.thewpewer-themindividual “accorded.the,state government. -Thewchi square wasmcalcnlated on a sample of 92. The..chi..aquare.table indicated a nega- tive relationship between the two variables. The greater the power accorded the state, the less the family contact with non-Indians. A”significant“negativeccorrelation“manifoundmbetween .ipowermaccorded.themlecal.government.and the.ameunt of read- “ing the individual.diduabout-foreigntcountries. Theugreater wthe power accorded the local government, the less the read- .ing done about foreign countries. .246 ii) Difference-Power measures: The difference in power accorded state and local governments and the differ- ence in power between the Central and local governments was related to the number of states in which the individual had lived. Individuals who accorded the state government con- siderably more power than the local governments had lived in relativelymore states. Individuals who accorded the state government. verylittleipowenmorethan,..the local governmenthadalsolivedin.more..states... Individuals who accorded . the .st ate government a moderate -amount . of power more than the local government hadlived. in fewer states. Indians living in either very few or very many states accorded -the Central ..and .locaL governments a smaller . differ- ence in.power than .didIndians living .in -an .intemediate number of states. . .iii),..0.utcomel.measuress Onesignificant relationship was found. between. the Outcome. measures «and the travel measures. The.relationship.was..between the state vs. local government measure and. the number .of stripsmade outside India. The greater - the . number. .of. times an, individual ‘had . left . India, -.the ,.less...often.-.the..outcome favor.ed...the state. This. correlation..-coefficient.Awas..computed .ondata for 86 individuals]. . iv} Power Index: ”No significant relationships ..be- tween the. Power Index .and..the...travel _ measures were found. .H¢>¢H sec. on» no uauoauacmam..e age.” .3. 2.3 as useueuusmam 2., Ho>oa no. enu.uu unuoausamam . XOflcH H.30m uceEcHe>eo Hausa .I> Heuuceu cemm.luu usescue>eo Hausa .m> ensue usessueseo eueum .e> aeuuceu e as u o cchx. usessue>eo Heooq asses Heuuceo «canx usescfie>oo Hausa asses eueum 7 ; . . fluudluqaluflaulufluadudldl Mm m m u - cum.luu Hosea usescue>oo among a x Hosea usescue>oo eueum N ....Nun Hgafl “GOSHCKVQU HGHUCOU MflMdmuflaluflamM. flflHdeflU . MdfifiH eceaocHlsec sesuo soaeuso eaosH ca ca cu oe>aq page you» usage oouauH> coo: au>oue oa>aq mommnafl> can 1:00 wageem ocaosem seauucsou moans asheseo -neusum us309.me«uuo llllHflHdMlmaqfiaHflNam muuuauo quamflo «saunas» lllmnuqllls H i: ..r 3 made—cocoa occupants eoueuso Hebsua é He>sua use eoueem oasosm nosed usescue>eo chest mcacueucou eeosufiuud coesuen emanecoausaem 5 Hz... "H28 -fiA .‘fl-,-. ‘HC‘ —--'--.. _flo-q‘ __._.- 248 Hypothesis IX m Attitgdgs gogcerning where government power should reside are rgiated to social class. The Social Class Index and each of its seven compon- ents.were related to each of the 10 power measures by means of correlation coefficients.and chi squares. The signifi- cant relationships are shown in Table 118. 1) Government Power.measuress IA.significant chi square wasfound between fatheris education and power “accorded.the-Central,government...The_chi ..... square was calcu- lated.onhaisamplemsizeuofmaz..iThe“relationshipflappears.to be curvilinear with both greater.and lesser power accorded the Central.government by individuals whose fathers had less education. sabderate.power.was.accorded the Central govern- ment by individuals whose-fathers had more education. .A.eignificantwnegativeicorrelation.waswfound.be- .tween.per capita.income-measures and power accorded the local government. The higher-theiindividual's.per.capita income, the less the power he accorded the local government. The sample size.for this relationship was 74. .A significant negative correlation was also found between the number of.family servants and.power accorded the local government. The correlation indicated that the more servantsuawstudent!s family.had,.the less power he.was.will- ing for theilocal,government to have. ‘The correlation was calculated for 92 Indians. ii) Difference—Power measures! The per capita.incame _measure.relatedwsignificantly.tomthenstate4minusllocal 249 govermment measure. Visual.inspection of the chi square table showed the relationship to be curvilinear. High and low per capita income were associated with small power differ- ences between state and local governments. Calculations were based on a sample of 74. The final significant relationship was a positive correlation between the number of.family servants and the difference in power between Central and local governments. H.The larger the number of servants his family had, the Imgreateruthelpower,thewindividual”accordeduthe.Central.govern- ment relative to that accorded the local government. The sample size for this statistic was 90. iii) Outcome measuresx. No significant relationships . were.found.between the Outcome.measures and any of the ”socialtclassnmeasures. iv) Power Index: No significant.relationships.were ..found between the.fiower Index and any of the social class . measures. Hy path as l 8 . IX n) Attitudgs concerning where government pgwe; d r s d e at to famil trad tion . W .For.eachmof-tbe.ten,powerflmeasuresnseparate;means .wweremcalculatedlforethemstudents.whomweremthemfirstmin.their tfiamiliesttemstudyrebroadmandmfior,thosehwhetwerewfirom wfamilies.wherewothersihad”studiedmabroad. .The differences _mbetween-the“pairs“ef“meansmfior-eachtof"the.tenapowerumeasures ”mwere tested for significance by,a t test. .The.tests.were .tnot.significant. Whether ornnot the individual was.£rom a 250 family where others had studied abroad did not relate to where he felt government power should reside. Hypothesis IX 0) Attitgdgs anerning whgre govermw WWW W Means were..-calculated forthe North Indians and for the south Indians for .each.of_.the ten power measures. The means for the.North and...for the South Indians for each power measure ..were- compared. by t test. The differencesbe- tween the means .were. not significant. DISCUSSION: HYPO‘I'HESES 12:3, 12:1,, Ixc, de, Ix., Ixf, ng, Ixh, 1x1, 1):), xxx, 1x1, Ixm, Ixn, 1x0, Rural-urban _. residence was. not ..a..correlate of change .proneness,. .but-..did showsomerelationshipmto approval of . India! s... socialwchange endeattitudeswhconcerning where. govern- ment power...shoul.d reside". Relationships “between. mral-uurban. residence .-and ....changeproneness had ..b.een‘..-exp.ected. . The. failure to find - relationships. may “beta function ..of the heavy. concentration .. of the.samplein-the...larger....urban areas. The,..lad£ of .relation- ships for_.the.sample, too, may -refl.ect..an absencehof any . . correlation between-thewpersonality. -and residence. ..uariahles. ‘ As expected. -Indian.-students -from. the.norerural areas . weresnore -lik.ely...to feel ..that .India'.s..-social ...changes .-.were“occurring“,tootrapidly, .or....at..about the. tightened. To . .these-individuals., ...the ..changes. must .have...appeared more ....drastic than they.-did-.to .their..more..urban-.counterparts. .ao>oa so. on» as unuu«u«cm«m.. 40>: mo. on» an ~533an . 251 KIUGH INMHU HGflUOM «mm.uu eeOMoluH aux .3 mneumam see» one» code nuse>uem use Imosooo Imusom leuspm mo muonuoum m.um£umm m.ue£uoz w.ue£umm Honesz dunno nunuum eeeao mueunaq hawemm ca exoom ll ROUGH HQSOG usessue>eo Haven .e> Heuuceo ucossue>oo Hausa .e> eueum uceesue>oo eueum .o> Heuuseo an” 0 O ”CUE ICH0>OU 109g QSGdE HUHUCCU 'CN” OUC.E IfiGDGU H000..." flauCH—fl .000m A 093.5 lonesoo euatm msoas Hmuuceu emm.Luu Hosea usescHe>eo Hausa Hosea usessue>eo oueum Hosea usessue>oo Heuuseo a. O m .EOUGH OUHQMU Mom HflfidMIMflIEQHflNdm mafiaufludamIMIGQMHHHfl Heaoom one opener paoocm meson usessue>eo euesz mcasueosoo neosuausd ceoauen amazecOAHeHem mad mnmda 252 Also as predicted, urban dwellers accorded more power to the Central government than did their less urban counterparts. A possible interpretation is that the focus of rural folk was more often on local and/or regional affairs while that of the urbanite was broader. To the dweller in large cities, the need for a strong Central government might have seemed more obvious than to less .metropolitan dwellers for whom the Central government.may have seemed rather remote. Travel related to the personality andmattitude measures. Greater travel inside India, a greater number of .trips made outside India,.and greater reading.done about other cultures were associated with a highequrientation Toward Change Index for the individual. Intlike manner, the .greater the number of trips he had made outside India.and .the.greater the reading he had done about other cultures, “themmore likely the individual was to see.himself-as a per- son who liked to try new'things. These.results suggest rather clearly that the broader the range ofuculture contact, the, more change prone the individual felt himself to be. ..This supported the hypothesis which.was based on.the theory "that exposure to many ways of doing things and.to.many~ideas fibroadened the individual's perspective and his interest in .trying new things. Travel also related to approval of India’s social . change. ReSpondents who had lived in many states or -travelled a great deal in India seemed to feel that changes 253 were in the wrong direction. Again explanation of these relationships is hampered by lack of information on how the reSpondents viewed the changes. Indians who had made many trips outside India seemed to feel that India's change was too rapid and these Indians had low Indices of Social Change Approval. In contrast,.respondents who had done much read- ing about other cultures felt.that social change was too slow. The difference between these two sets of.findings .seems to be one of theory and practice. Reading about other cultures apparently.imbuedlthe student with the desire for faster change, but travelling inside and/or outside India .resulted.in dissatisfaction with the direction of change and in feelings that change was too rapid. .Why.travel resulted -in these feelings was not apparent from the data collected in this study. Travel related to attitudes concerning where govern- ’ment power should reside as well.as to change proneness and approval of India's social.change. Respondents who had lived in many states accorded either very.much or very little power to the Central government, accorded.the state govern- ment.either much more or.very little power relative to the local government, and accorded the Central and local govern- ments nearly the.same.amount of power. (The last relation- ship was curvilinearnwith respondents who lived in few states also according Central.and local governments nearly the same amount of power). The pattern of relationships seems to be fairly consistent. Individuals who.have lived in relatively 2 54 ~are states either favored government power at the highest levels or favored it at the lowest levels. The inter- mediate position belonged to the residents of the fewest number of states. Residence in many states may have lead students to feel that India's subcultures were so diverse that India must have strong Central government in order to stay unified and progress. In contrast, it may also have resulted in feelings that no Central government could understand and administer to the multitude of subcultures and, therefore, these groups should.be.left alone to manage their own affairs. Travel outside India was associated with a tendency to less often favor the state in stateelocal government --disputes. .The implications of this finding are not clear. Less power.generally was accorded the state by respondents with the most non-Indian contact in India and less power generally wasmaccorded the.local.government by respondents who read the most about other cultures. One interpretation .of this is that vicarious culture contact resulted in a ,broader perspective which in turn was associated with less concern for state and local affairs. Social class was related to some of the attitudes toward change.and political control measures. The number of ,family servants and the amount of.education of the mother ,were,associated-with Self Percept Scores. The greater the ,-number-of servants and the higher the mother's education, the more inclined the respondent was to perceive himself 255 as liking to try out new ideas. This result tends to support the observation in the literature that the upper classes actually do the changing. The results of this hypothesis indicate that it was the upper classes who more often saw themselves as liking to change. There were no relationships between approval of India.'ssocial change and social class measures. If class differences in change proneness exist, they did not show up here in feelings about India's current change. Explana- tions for the absence.of c1ass.differences start with the possibility.that no such differences in fact, exist. Another possibility, however, is that.individuals in the sample viewed the changes in different lights and this in turn was linked to their approval and disapproval of the speed, direction,.and scope of the changes. Social class differences were.feund for attitudes ,concerning where government power should reside. Respond- ,“entsuwhose fathers had the.highest.education wished to ”confer.moderate.powermon the Central government. The'ex- tremes (greatest and.least) of power for the Central - goernment.were.the desires ofirespondents whose fathers had . low education.. High family income was associated with less "‘power for the local government and little difference in . power between state.and local governments. A larger staff of family servants was associated with less power for local governments and.large differences in power between the Central and local governments. The picture emerging from 256 these findings is that respondents of higher social class wished to accord the Central government a moderate amount of power and wished the state and local governments to have considerably less power. Respondents from lower social classes desired relatively strong local governments and state governments. Respondents of lower social class were Split on the power issue for the Central government: most wanted it to be relatively weak: a few wanted it to be ex- tremely powerful . The interpretation of the relationships found no doubt rests on a number of social and cultural factors not examined in this research. This research suggests that one such factor, however, may have been the realization of the need for Indian unity and with it a feeling that only a strong Central government could achieve this. It may be too, that North-South residence was a factor in conjunction with social class which affected feelings about government power. With the Central government located in the North of India and with the existence of North-South friction in India, geographic residence and social.class may have Operated together with regard to feelings about government power attitudes. Whether or not the individual was the first from his family to seek education abroad more no relationship to any of the.attitude toward change.and political control measures. The hypothesis . made because of observations in the literature, reCeived no support from.the data. 257 The only difference found between North Indian residents and South Indian residents was in terms of motivation. North Indians said they had more motivation than South Indians. The motivation finding tends to substantiate the informal picture of the South Indian who doesn't feel he will be given a fair chance. It had been expected that south Indians would want to accord the Central government less power than would North Indians. Thiswas expected for two reasons: general resent- ment if the North by the south and the location of the Central government in the North of India. The failure to find North-South.differences on the question of power for the Central government may have been a.realization on the part of a majority of the students from the South as well as from the North that a strong.national government is neces- sary to unify India. In addition, there may have been a number of North Indians who were "states' righters“ who would not have wanted to see a strong Central government. 258 HYPOTHESIS X. ATTITUDES TOWARD CHANGE AND POLITICAL CONTROL ARE RELATED TO THE PERCEIVED LANGUAGE LIBERALISM OF REFERENCE GROUPS AND TO FEEL- INGS ABOUT'THEVREFERENCE GROUPS. Hypothesis X a) Change grie fin gti 93 is rglgted to the language subc ste soc class Wf The Orientation Toward change Index was correlated with the Language Liberalism Indices of each of the indivi- dual's groups. Confidence, Motivation, and self Percept scores were also.correlated with.the language indices. Table ll9.shows the significant relationships. The statistics were based on rather small samples. A.Language Liberalism Index was calculated only when reSponses for all six of its components were present. In like manner, Confidence, Motivation, and Self Percept scores were com- puted only when all items comprising the scale were answered. To have an Orientation Toward Change Index the student must have reSponded to every item in the Confidence, Motivation, and Self Percept scales. A missing reaponse to any of the items in one of the indices or.scales discussed above ex- cludedwthe individual from consideration in any of the analyses involving that index or scale. The result of this procedure was a severe reduction inmsample size for many of the statistical comparisons. Sample sizes for statistics involving the Orientation Toward Change Index ranged from 8.to,18: those for Confidence from 8 to 19: Motivation, ll to 23 and Self Percept, 11 to 22. 259 The Orientation Toward Change Index of the respondent related to the language liberalism of only one of his refer- ence groups, his father'. The negative correlation between the measures indicates that students with a positive orientation toward change considered their fathers to have _conservative language attitudes. The correlation was based on a sample of 18. A.significant chi square was also found between Self Percept and the perceived liberalness of the fatherb language attitudes. Here also the relationship tended to be a nega- tive one. Students who saw conservative or provincial attitudes on the part of their fathers saw themselves as peOple who liked to try new things. Confidence bore no relationship to the.language attitude of the individual's ‘membership groups nor did the motivation measure. TABLE 119 Relationships between Language Liberalism of Reference Groups and Change Orientation W M W191: 2125123922141: 1'. a..1is_m_ _In. d. _ices. Father Mother subcaste Social Region Class Confidence Motivation 2 Self Percept X * Orientation Toward Change Index r=-.Sl* * Significant at the .05 level. 260 Hypothesis X b ) Engage orientation is related to the hetegg- t f a a i ralism on f , subcastg, sogial glass agd regionI Individuals were divided into categories according to whether their reference groups had: a) all non-Liberal Language Indices, b) all Liberal Language Indices or c) some Liberal and some non-Liberal Indices. A mean Orientation Toward Change Index was calculated for respondents in each category. Mean Confidence, Motivation,.and Self Percept scores were also calculated. Table 120 shows themeans. A significant difference (t test significant at the .05 level) was found between the mean Orientation Toward Change Index for respondents whose groups were all perceived to be liberal and respondents whose groups were.heterogeneous. Respondents with heterogeneous reference groups had higher ‘ Orientation Toward Change Indices. Hypothesis X I ‘33g_1ihg;31 f 1 sub st so i Wingless. The Orientation Toward Change Indices.of the respond- ents were divided at the median. Scores in the “high".group ranged from 156 to.lB7. "Lowu group scores ranged from 123 to 155. There were 29 scores.in each group. ' The mean.importance attached to groups with liberal language attitudes by Indians with high Change Orientation Indices was calculated as was the mean importance attached to .groups with liberal language attitudes by Indians with low Change Orientation Indices. The means were compared by t test. 261 m ~a.aa ss.osa as ma.m mm as om.m aa.oa ma mm.ea oo.~ms ma we.m am we ob.~ oH.H~ on ms.ma ms.vms ma mv.v am as mo.~ om.om m 4mqm amen a .m4m. mama -nmomH «mango .ummoumm 30in UHmEOB COHumucmfiuo Amaummcimfiullnmaunlo m 0 DNA mdmdB wooeosH omnsmsmq m mm.ma mo.Hm amassed o>mn osom “museum coco , uuomom msoocomoumuom novaosH omszmcmq we mm.ma om.om amuonuq e>o£ Had .mdsouo ‘ ousououem msoesemOEom weoHosH emssmcmq ma ev.ma Ho.mm Houonaqnsos sew: Had «mQSOHG oucouomom wsoocomoeom mmmowmlMummHuWMm i museum msoocomoumuon com Houoan .Hmuonaqusoz now mousmmez cofiumusofiuo omcmno does. 262 The difference was not significant. The same procedure was followed for the importance ratings of groups with non- liberal language attitudes. This difference was also not significant. Table 121 shows all four mean importance ratings. For individuals with a high Orientation Toward Change Index a t test was performed between the mean importance they attached to those of.-their reference groups with Liberal Language Indices and the mean importance they attached to those of their groups with non-Liberal Language Indices. The t was not significant. The same procedure was followed.fer?individuais with a low'orientation Toward Change. This t test was not sig- “w‘i- nificant. Individuals with a high Orient atien Tewerd Change Index did not consider their groups whose language attitudes were liberal as more important than their groups whose language atti- tudes were nenIliberal. The same was true for individuals whose change orientation was low. TABLE 121 Importance Attached to Liberal and Non-Liberal Reference Groups for Respondents with High and Loerrientation To- ward Change Indices Orientation Toward Importance of groups Importance of groups Change Index with Liberal Langu— with Non-Liberal W LWa I age; 2.2. .13. in.a: 52.2. a High ,2-71 1.60 24 2.63 1.53 24 Low 3.50 1.06 8 3.50 1.64 6 263 Hypothesis X d)A Approval gf 39; lg; changg in Igdia is gglgt gd m2_t he langua gg attitude; of family, subcaste, W. The indices of perceived language liberalism for the reference groups were correlated with the Index of Social Change Approval and the items composing the Index.Contingency tables were prepared and chi squares calculated. Table 122 shows the only significant relationship found. The Index of social Change Approval did not relate to any of the Language Liberalism Measures. One of the three components of the Index of Social Change Approval did relate, however. Approval of the direction of India's social changes related positively to the father's Language Liberal- ism Index. The greater the individual's approval of the direction of India's social change, the.more liberal he con- sidered his father's language attitudes to be. The correla- tion was based on data from 24 respondents. TmBLE 122 Relationships between the Language Liberalism of Reference Groups and Approval of social Change in India _pnrgA xal_9__f 5.922;...th a ge WW. Father Mother Sub- Social Region castg Class Approval of Social Change Speed Approval of Social Change Direction r=.4l* Approval of Social Change Scope Index of Social Change Approval *Significant at the .05 level. 264 Hypothesis X e) A-- ~v- - soc - g ; , . : r,< r t9 the hetgrogeneity of l nguage liberalism gmggg fgmily, subcastg, social class agd region. To test this hypothesis individuals were divided \e into three groups: those whose membership groups were all ' perceived as having Liberal Language Indices, those whose membership groups were considered heterogeneous and those whose membership groups were felt to all have Non-Liberal Language Indices. Chi squares were calculated between this measure of heterogeneity of language liberalism and 1) Index of Social Change Approval, ii) Approval of the Speed of Social Change: iii) Approval of the Direction of social Change and iv) Approval of the Scepe of Social Change. None of the chi squares was significant. The perceived heterogeneity among reference groups regarding language attitudes bore no relationship to the approval felt for the current social change in India. Hypothesis X f) 'r v f s a I d 3 t9 thg impgrtance gnd language liberaligm of family, subcaste, social class and rggiogz The Indices of social Change Approval were.divided into three groups: ”high" (scores above 10), “medium” (scores of 10) and “low” (scores below 10). For individuals in the "high“ group, two means were calculated: the mean importance they attached to those reference groups whose language attitudes they saw as liberal, and the mean importance they attached to those 265 reference groups whose language attitudes they saw as non- liberal. Table l23 presents the means, A t test was calcu- lated to test for significance between the two means. The t was not significant. The same procedure was followed fer individuals in the “medium“ and "low“ groups. The t tests calculated for the "medium” and "low" groups also were not ’significant. Individuals with varying degrees of approval for the current social change in India did not.attach a different degree of importance to those reference groups whose attis tudes they felt were liberal than they did to those groups whose attitudes they felt were not liberal. TABLE 1 23 Importance Attached to Liberal and Non-Liberal Reference Groups for Respondents with High, Medium and Low Indices of Social Change Approval Index of Importance of Groups Importance of Groups Social Change with Liberal Lan- with Non-Liberal Apprgyal guagg Indicgs La gnggg Infiiggg than 1542.... E m 2.2.. .13. High 3.15 1.5 20 3.20 1.5 17 Medium 2.80 1.6 6 3.20 1.6 13 Low 2.60 1.7 10 2.30 1.5 15 Hypothesis x g) Attitudgs concerning where governmegg pgwer s o d r side ar ated to t e n us be alis f fa sub ast ‘ so ial c ass W. Correlations and chi squares were calculated between the power measures and the Language Liberalism Indices for- -266 the father, mother, subcaste, social class and region. Four relationships were significant. Table-124.shows these relationships. 1) Government Power measures: No significant relation- ships were found between the Government Power measures and the Language Liberalism Indices. vii) Difference-Power measures: A significant nega- tive correlation was found between the Language Liberalism Index for the father and the difference in power the respond- ent gave the Central and state governments. Sens whose fathers had liberal language attitudes granted the Central and state governments similar degrees of power. Sons who saw their father's attitudes as conservative wished to give the Central government.more power than the state government. The sample size for the.correlation was 22. A significant relationship was.also found for the liberalness of Mothers' attitudes and the Difference-Power measure calculated between the Central.and state govern- ments. The pattern of relationships was not clear in the chi square table, however. The sample.size was too small (only 11 individuals) to determine anything conclusive about the relationship. iii) Outcome measures: A significant chi square was found between the Language Liberalism Index of the father and the outcome in Central and state government conflict. The sample contained only 21 people. It wasrnot clear from the Chi square table what form the relationship between the two variables took. 267 When disagreement exists between state and local government, the relationship between the desired outcome of this disagreement and the perceived liberalism of the father was clear. The student who felt both levels should prevail about half the time saw his father as having con- servative language attitudes. The son who felt the state should usually or always prevail considered his father to have liberal attitudes about language. The correlation was calculated on a sample of 21. 1v) Power Index: No significant relationships were found between the Power Index and the Language Liberal- ism Indices of the reference groups. Hypothesis x‘ h) Attitudes concerning where govgrnmegt power ghggld reside are related to the hetero- e t f an ua e liberalism il , subcastg, social glass and rggion. There were two testsof this hypothesis. Ear the first.test, heterogeneity among the LanguagelLiberalism Indices of reference groups was.cerrelated with the differ- ence in poweraccorded the Central and state governments. The correlation was not significant. .A correlation was also computed between the Outcome measure for conflict between the Central and state,govern- ment and the heterogeneity of language liberalism measure for the second test of the hypothesis. VThe correlation was not significant. ao>ua do. on» an ucuowuwamwmr. Ho>ca no. on» an ucuuwuwnmwu . XOOQH Hgflm esessue>eo Haven .e> Hewuseo ah¢.uu asessue>eo Hosea .e> iusum aux usessneoeo eueum .e> Houuseo muummqqawusmmmam. 268 usessueboo Haven essda Heuuseu uneasue>eo finned ensue ensue ecmx cmv.uuu aseasue>eo eusum eases «unseen Meson usessue>ew Hausa Hesom useesue>eu eusum nosed uneasue>eo asuuseu e an .m museum eoseuemem we Eeuasuenaq emosm Issqnesu can ewe-em oasonm meson usessue>ow euerk measuoosoo newspauec seesuen amazeseaumaem vN H magma ..- '_.-..._-—w~ »-- 1 269 DISCUSSION: HYPOTHESES xa, xb, xc, xd, Xe, Xf, xg, Xh, X1, xj' xkl x1! “I xnl x0. The first parts of Hypothesis X predicted.relation- ships between change proneness and the language attitudes of reference groups. It was found that the individual with a high degree of change proneness saw his father as having conservative (non—liberal) language attitudes. this was a finding opposite to that expected. It had been.expected that individuals with positive change orientations would have reference groups with liberal language attitudes. The findings suggest that the change prone person may have con- sidered himself to be breaking tradition. Instead of a liberal background fostering an interest in new things, it appears to have been a.traditional.conservative background which imparted to the individual a desire for change. It should be remembered that the findings.were all based upon the respondent's perceptions of himself and of his reference groups. Whether or not the significant statistical relation- ships found.reflect the actual state of affairs is not known. A further.finding was that respondents who saw their reference groups as heterogeneous, i.e.,.some.reference groups possessing liberal and some possessing noneliberal language attitudes, were more change prene.than were.respend- ents who considered all.of their reference groups to be liberal. Individuals with homogeneous non-liberal.reference groups occupied an intermediate position in terms of change proneness. This finding mirrored the previous.ene. .Lgain 270 it was.the liberalness of reference groups which was assoCi- ated with a low change orientation. One interpretation of this result is that having a wide range of opinions on language attitudes available means that no matter what the reopondent does, he deviates from one group. Deviation is easy, the pressure of all groups is not strong--the individual is freer to make his own choice. Another possibilitylis thatibavinglreference groups with divergent opinions is like having.a buffet-~an.intro- duction to a wide variety of positions which might not have even been considered had all the reference groups been more or less homogeneous. ‘ Change orientation did not relate. to the..importan‘ce and liberalness of reference groups. .Individuals with a high degree of change proneness did not consider their groups whose language attitudes were liberal as more import- ant than their noneliberal.groups. Importancenandeliberal- .ness were not related for respondents with low change proneness. This finding.is.notmsurprising for two reasons. First, it was only the father's language attitudes which wereassociated with the change proneness.of thelrespondent (attitudes of ”themmother, subcaste, social. class andregion did.not.relate). Theiimpertance ratings given all the respondent'sireference“groupswereuaveragedntomtest,this hypothesis.whichuprobablyrobscured any relationships which might.have been.present.fer any particular.reference.group. Second is.the.problem of thevdistribution at.themhmportance 271 ratings. As mentioned earlier, the importance ratings for some reference groups were rather skewed. Approval of the direction of India's social change was paired withliberal language attitudes on the part of the father. This was a...relationship -.in-the direction ex- pected. It supports the expectation.that.liberal language attitudes on thepart of.-reference groups will be reflected in liberal attitudes on the part ofthe respondent.v It.is noteworthy that the significant .relationshipoccurred for the approval of social change .W measure rather than for the approval of 22.119. or approval ofm measures. The heterogeneity perceived amongreference groups in terms of language attitudes didnot .relate. to approval of India's social change. Approval of social.change in India measures were measures of attitudes. toward certain phenomena. The heterogeneity of .. opinionseof reference. groupson another issue such as language would “not, necessarily .relateto the respondent's ..feelings aboutisocialchange. This is parti- cularlylikely whenit is .renembered that -it was with only one of the reference groups (father) that .the respondent's opinions on language agreed. The heterogeneity of..reference group languagewliberalism was a-mcerrelatetof change. prone- .ness. Changeproneness, however, was a personality vari- able-~a way of looking at situations. ”insgeneral, rather than an. attitude toward specific phenomena such as..attitudes to- ward India's social.change. 272 Approval of social change in India did not relate to the importance and liberalness of reference group language attitudes. There are two reasons which might account for this finding. Social change approval did not relate to the liberalism of reference groups at all, so a.consideration of the importance of the groups didn't sharpen any existing relationships. As mentioned several times earlier, some of the importance ratings were quite skewed,.thereby.restrict- ing statistical manipulation. A few relationships were found between the language attitudes of reference groups and the.individual's attitudes concerning where government power should reside. Individuals “who sawrtheir fathers as having liberal language attitudes granted the Central and state governments similar degrees of power. but felt that the.state should.usually or.always pre- .vail when disagreement occurred between state and local governments. It would appear.tbat thisnrelationship, like the one between the.individual's language.attitudes.and attitudes concerning where government.power should reside, reflect a general liberalism of-attitude about government power. A fair degree of government power was granted.to.all levels of government by respondents whose fathers had liberal .language attitudes. For those respondents the upper level of government retainesthe edge over the lewer.in cases ef dispute between them. Attitudes concerning where government-power should . reside were not related te.the heterogeneity of reference 273 group language attitudes. Again there is evidence that a variety of reference group attitudes on one issue (language) did not relate to the respondent's attitudes on another (delegation of government power). Again the attitudes of the respondent concerning language agreed with only one of his reference groups and so the positions of the other reference groups might or might not have had any.effect on the individual. CHAPTER VII SUMMARY In the preceding chapter, the findings from each part of every hypothesis were presented. In the discussion at the and of every hypothesis, an attempt was made to knit to- gether the findings from the whole hypothesis. In this chapter, there is a further attempt to summarize by taking an overviewrof several hypotheses at once. Summaries for the.Langnage_and Qulture.hypotheses, for the Language and Personality hypotheses, and for the Culture and Personality hypotheses are presented. The sum- maries are followed by brief comments.on the common threads running through results of these hypotheses. Following this is an overview of each of the five language variables and the three attitude toward change and political control variables in turn. The overview is pre— sented in.the form of characterizations of the respondents in terms of the correlates for each variable. We TheJLanguage,and Culture hypotheses examined corre- lates of the-languagergttitggg.uariablesntprestige..affec- .tion and liberalism) and of the.languagernagg. variables (specialization and interaction.) 274 275 Socio-cultural“correlates (ruralaurban residence, travel, social class, family tradition of foreign study and North-South residence) were examined for the language attitggg variables. Aspects of perceived reference group language attitudes and the respondent‘s feelings for these groups were examined for one language attitude variable (liberalism) and for the language‘ggggg variables. Summaries of the findings from the Language and Culture hypotheses are presented below. In Hypothesis one it was found that travel, social class, North-South residence, and to author extent, family tradition of foreign study‘were-correlates of the prestige ratings given languages. In addi- tion, rural-urban residence, travel, social class, and to a minor extent, North-South residence and family tradition of foreign study were correlates of the affection felt for languages. In.Hypothesis two, social class emerged as the only socio-cultural correlate of language liberalism. Rural-urban residence, travel, North-south residence and family tradition of foreign study and not re— late to the language liberalism of the respondent. In Hypothesis three, the respondent‘s language liberalism was found to correspond with that of his father, on the whole. Specific parts of his Index correlated with those of each of his reference groups. Particularly noteworthy were the acceptance of Hindi for India's official language and the prestige attached to bilingualism issues on which the position of each reference group corresponded to that of the respondent. The importance of refer. ence groups did not relate to the degree of the respondent's deviation from the language liberalism of the reference groups. In.Hypothesis four, diversity of language preference among reference groups was found to be associated with the language specialization.the respondent practiced with different people. Heterogeneity among the LanguagelLiberalism Indices of the reference groups related to Language Specialization of all kinds. The importance attached to the region and social class were related to specialization. 276 In Hypothesis five, native tongue appeared to be re- lated to friendship and contact patterns within the Indian community. The heterogeneity of Language Liberalism Indices among reference groups did not relate to the patterns of discussion about Nehru's death. The importance with which reference groups were regarded was related to the discussion patterns. When his reference groups in India were deemed important, the respondent's discussion about Nehru's death occurred mainly with Americans. Socio-cultural correlates were found for all the language attitpgg.variables (prestige, affection and liberal— ism) hath travel and.social.class emerging as the primary correlates. Language liberalism also had another set of correlates. The overall perceived language liberalism of fathers correlated with the respondentls language liberalism. On specific language attitudes, all of the respondent's reference groups were often in agreement with the respondent. The importance of reference groups was not a factor in the correspondence between the language liberalism of the respond- ent and his reference groups. .Reference.group attitudes were associated strongly with.both.language‘g§331 variables (specialization and inter- action). Diversity.of language.preference.and heterogeneity of Language Liberalism Indices among reference groups related to specialization. Native tongue.related (non—statistically) to interaction. The“importancemofireference groups was a factor.in both specialization andhinteraction. §nmmn£x. It.is apparent that socio-cultural factors were important in language attitudes. It is further apparent that perceived reference.group attitudes related to both 277 language attitudes.and languagewusage and that.tbe import- ance of reference groups were.associated with language usage. Languagg and Pgrggnglity The Language and Personality hypotheses examined the attitude toward change and political control correlates of one language attitudg variable, (liberalism) and both lan- guage pgagg variables (specialization and interaction). The attitude toward change and political control variables examined were change orientation, approval of social change in India and attitudes concerning where government power should.reside. . Summaries of the.findingsifrom the Language and Personality hypotheses are presented below. In Hypothesis six, Language Liberalism was found to relate to personality and attitudes concern~ ing social phenomena in India. Liberal language attitudes were associated with liking to try new things, high approval of the direction of India!s ‘social change and either very high or very low ' approval of the scape of social change in India. Liberal language attitudes were also associated with according a high degree of power to both state and local governments and to minimizing differences in power between Central and state governments. In Hypothesis seven, language specialization was found to relate to personality and attitudes to- ward social phenomena in India. Considerable language specialization in particular places was associated with liking to try new things. It was also associated with feelings that social change .in India was occuring either too.rapidly or too slowly, that social change was in the wrong direc- tion, and too.large in scope. Greater specializa- tion was associated with according the Central government relatively less power but favoring the higher government level when conflict occurred between government levels. 278 The results of Hypothesis eight showed interaction not to be related to the personality measures, but to relate to the attitudes concerning social- phenomena in India. Discussion of Nehru's death occuring primarily within the Indian community was associated with approval of the direction of India's social change. Discussion occuring either primarily within the Indian community or primarily within the American community was associated with a desire to grant the local government little power, grant the Central and state governments about the same power and grant the higher government level the same or less power than the lower government level in cases of dispute between them. Liberalism and specialization related to change orientationze interaction did not. The relationship between .liberalism and change orientation was negative. The relation— ship between specialization and change orientation was positive. Liberalism, specialization and interaction all .related to approval.of India's social change. In particular, .approval of the direction of India's social change was a correlate of.all.three language variables. Government power measures related to liberalism, ..specialization and interaction. (Sometimes the relationships were with the simple Eower measures, sometimes with the Difference-Power measures, sometimes-with the Outcome measures and sometimes with.the Power Index. §nmmarx. The attitude toward.change and political control variable was a correlate of the language attitggg .(liberalism) measure. All of the attitude toward change and .npoliticalncontrol measures.werewcorrelates of.one or both _ language ‘ggggg_(specializationand interaction).measures. 279 W._, The Culture and Rersonality hypotheses examined relationships between the attitude toward change and politi- cal control measures and the socio-cultural factors and between the attitude toward.change and political control ”measures and reference group attitudes. The attitudes to- ward change and political control were change orientation, approval of social change and attitudes concerning where government power should reside. The socioecultural factors were rural-urban..,residence, travel. social class, family tradition of foreign study and NorthQSouth.residence. Reference.groupattitudes.were language liberalism attitudes. Summaries of the findings are presented below. Hypothesis nine found travel and social class to be correlates of change orientation. Urban residence and travel were correlates of the approval of social change goggg. Travel was a correlate of approval of social change direction. There were no socio-cultural correlates of approval of social change sggp . Urban residence. travel and social class were correlates of the power granted the Central government. Travel was the only socio-cultural correlate of power granted states. Travel and ’ socialwclass were the correlates of local govern- ment power. No socio-cultural correlates were found for the difference in power between Central and state governments. Iravel.and social class correlates were.found_for.the.difference in power granted ’ the state and.local governments and for the differ- ence in power granted the Central and local governments. No socio-cultural correlates were found for the outcome in disputes between the Centralwand state governments or between the Central and.local governments. Travel was a correlate for the 280 outcome in state and local government disputes. No socio-cultural correlates were found for the general tendency to favor the higher or lower government level in cases of dispute between government levels. Hypothesis ten found a positive change orienta- tion on the part of the respondent to be associ- ated with conservative language attitudes on the part of fathers and with reference groups which were heterogeneous in the liberalism of their language attitudes. The importance of reference groups with liberal and with non-liberal attitudes was not associated with change orientation. Approval of the.direction of India's social change was associated with the perception of liberal ,language attitudes on the part of the father. social of the Neither the heterogeneity of the liberalness of reference group language attitudes nor the import- ance of the reference groups to the respondent was related to approval of India's social change. The difference in power between the Central and state governments and the outcome when state and local governments conflicted were also associated with the perceived liberalness of the fathers attitudes. Heterogeneity of language liberalism among reference groups did not relate to attitudes concerning where government power should reside. The correlates of change orientation were travel, class, fathers.language liberalism and heterogeneity language liberalism of reference groups. Urban residence, travel and father"s language liberal- ! ism were.the correlates of approval of Indidk.social Change, Urban residence,-travel, social class and father‘s language liberalism were the correlates of attitudes concern- ing where government power should reside. §nmma£xt Socioqcultural correlates were found for all the attitude toward change and political control measures. . Travel emerged as the most important socio-cultural.factor. 281 A further.kind of summary is offered here.in the form of characterizations of individuals reSponding at the ex- treme of each language and attitude toward change and politi- cal control variable. L ua P st e The individual who accorded Hindi and/or Hindustani high prestige: -had lived in few cities, towns or villages in India. -had a high.pr0portion of brothers and sisters in college. -was not the first in his family to study abroad. -lived in the North of India. The individual who accorded English high prestige: -had read a great deal about other cultures. -had a high Social Class Index The individual who accorded Urdu high prestige: -had lived in many states or in few states in India. -had read a great deal about other cultures. The individual who accorded Bengali high prestige: -had read a great deal about other cultures. -came from a.family with a.large library. There were other relationships between language prestige and socio-cultural factorso Some relationships were found for nearly every language in the analysis. 282 Travel and social class emerged as the important socio-cultural correlates of prestige. Travel and social class did not relate in the same manner to the prestige attributed to all languages, e.g.. increased travel was not associated with increased appreciation of the merits of all languages--at least not if prestige ratings are any evidence of such appreciation. Travel might relate to prestige ratings differentially for two reasons. Eiret,.travel gives the rater a new perSpective. Travel gives.himiamchanCe ,to stand.back.and look the situation over. It also.gives him personal contact with.more.languages which widens his base for making comparisons. ,Second, it brings him into .contact.with a new set of stereotypes regarding the prestige of various languages. These stereotypes may reinforce those he has.already acquired or they may.conflict. Rreetige appears to.be a stereotype.to alconsiderable extent. Social class.di£ferencee in prestige attached to various languages increase.support for.this interpretation. Weiss -The.individual who expressed high affection.for Hindi: ~had low Social Class Index -came from a family with a relatively low per capita income. scame from a family.wheretthemother had less education -came from a family where the father had a relatively .low'occupation. ' -lived in the North of India 283 The individual who expressed high affection for English: -had lived in several states in India -lived in the South of India The individual who eXpressed high affection for Urdu: -had lived inllarge.urbaniareas.most of his life -had made trips outside India -had visited other countries The individual who expressed high affection for Bengali: -came from a family with a relatively low per capita income Many other relationships were found between language affection and socioecultural factors. Eor example, indivi- duals expressing strong.affection for Gujarati came from the more rural areas in India and came from families where the father had a relatively low occupation. The affection ex- pressed for a language was no doubt based on.a.complex set of factors. In studying the findings, it appeared that the amount of usage of a language might well be a factor in the affection felt.for the language. W The individual who had liberal language attitudes: -came from a family where the mother had less education. -came from.a family with a smaller library -perceived his father to have liberal language attitudes. -perceived much correspondence between his own specific language attitudes and those of each of his reference groups. 284 -perceived himself as an individual who was not adventuresome and who did not like to try new things. .-approved of the direction of Indies social-change. -either approved very much or was at the other ex- treme and disapproved very much of the scepe of India“s social change. -accorded a high degree of power to the state govern- ments. ‘ -accorded a high degree of power to the local govern- ments. -minimized differences in power between the Central and state governments. Language liberalism.proved to be a concept related .to a wide variety of variables. It had socioecultural, reference group,.personality.and specific attitude correlates. The relationships between the respondenth language liberalism, his Self Percept and the perceived.language liberalism of his father.were not.always in the direction expected. Language Spgcialiggtion S. t b P o: e ,The.individual who used a large number of languages with particular people or who listed many people with whom.particular.languages were used: -considered his reference groups to be father dis- parate in the number of different (non-overlapping) languages preferred among them. +saw his reference groups as having great hetero- geneity in.terms.of.liberalness of language attitudes. -attached.little,importance to his.region. -considered.social change in India in somewhat negative terms: disapproval of the direction of social change, approval of the speed and scope of social change. 285 -wished to see the higher government levels dominate the lower when conflict occurred. Wee. The individual who used a large number of languages in particular places or who listed many places where particular languages were used: -saw his reference groups as having great heterogeneity in terms of liberalness of language. -attached great relative importance to his social class. -saw himself as a person who liked to try new things. —considered social change in India to occur either much too slowly or much too quickly. -oonsidered social change in India to be in the wrong direction. —felt social change.in India to be too large in scope. -acoorded a low degree of power to the Central government. 3 . z . b 0c n. The individual who used a .large number of languages on particular occasions or who listed many occasions for which he used particular languages: -saw'his reference groups as having great heterogeneity in terms of liberalness of language attitudes. -saw'his reference groups as having a great range in .terms of.liberalness of language attitudes. -attached little relative importance to his social class. Language specialization also turnediout to be.a.use- . ful concept.. Reference.group language.attitudes were.parti- “cularly.important correlates although some relationships to .personality and specific.attitudes were found. 286 Interaction The individual who interacted within the Indian community concerning Nehrus death far more than he interacted within the American community: -attached little importance to his reference groups. —approved of the direction of Indies social change. -accorded the local government little power (as did his opposite who interacted mainly in the American community). -accorded the Central government and the regional governments about the same power in situations of clash(as did his opposite who interacted mainly in the American community). -accorded the higher government levels less power than the lower government levels in cases of con- flict (as did his opposite who interacted mainly in the American community). Of all the language concepts, interaction was the least useful. It related only marginally to reference group language attitudes. not at all te.the personality . measures and to a fair degree to attitudes.toward Indies ‘social change and political control. Qhange Origatgtign We ‘_' - no relationships - Minam- -ino relationships - 83;: ngcdpt. The individual who perceived himself as adventuresome and liking to try new things: .287 had conservative language attitudes. practiced much language specialization. had made trips outside India. had read a great deal about other cultures. came from a family with a relatively large number of servants. came from a family where the mother had more education. perceived his father to have conservative language attitudes. O t T C Inde The individual with a high degree of change proneness: had done some travelling inside India. had made.trips outside India. ' had read a great deal about other cultures. perceived his father to have conservative language attitUd“. ' perceived his reference groups to be heterogeneous in terms of language attitudes. The findings.show change orientation to be a fruitful concept. .Change orientation related to the.respondents language attitudes..to.his language usage, to socio-cultural factors and to the p.erceived-,language.attit udes of the _reference gnoup. WW . v . f S S eed. Thenindividual who.feit the speed of Indiis social change to be much too slow: -practiced great language specialization (as did .Hindividuals.whocfelt.social change to be too.fast). 288 -had lived in large urban areas most of.his life. -had made.trips outside India. -had read a great deal about other cultures. W Th- individual We approved highly of the direction of Indidk social changes -had liberal language attitudes. -practiced little language specialization -discussed Nehru‘s death with.1ndians more than with Americans. -had lived in few states in India. -had travelled little in India. -perceived his father to have liberal language attitudes. W. Th- individual who approvod highly of the scope of India‘s.social changes -had liberal language attitudes (as did Indians who disapproved of the.scope of India‘s social.change). -practiced moderate or little language specialisation. W. The individual wholXPIB33fid high approval.of Indies social changes -practiced little language specialization (as did individuals who exprassed little approval of India s .social.change). Approval of.aocial.change was a.fairly useful concept. .CorrelatesMofisocialichangeiapproval were.primarilywsocio- “,culturalavariables.and to a-miner extent,.language.attitudes. .language.usage.andwreference.gnoup.language.attitudes. 289 A s C c in Gov me t Pow r Qantrgl Ggggzgment Rowgg. Theiindividual who accorded the f Central government high power: -practiced little language specialization. -had lived in large urban areas most of his life. -had lived in many states (as did respondents accord- ing the Central government little power). -came from a family where the father had less educa- tion (as did those wishing to accord little power to the.Central governments.) ‘§:at§_§gg§£nmgnt_gggg;‘ The individual who accorded the state government high power: -had liberal language attitudes -had little nonelndian.contact in India. Lg:al_§g1§£nmgn§_ggwgz‘ The individual who accorded the local government high power: -had liberal language attitudes. -had read little about other cultures. -came from a family with a relatively low'per capita income. -came from a family with relatively few servants. -discussed Nehru's death in about the same proportions in the Indian and American communities. vidual.according the Central.government.much more power than the state government33. -had-conservativeilanguage”attitudes. -perceived his father to have conservative language attitudes. .D ..o er: 8 . n .Gover nt. The indivi- ..dual according the state governments.much more power than the local governments: ”290 -came from a family with a relatively low per capita income. -had lived in several states in India (as did respond- ents who accorded the two government levels about the same power). -P : C t .n s Loc ve e t. The indi- vidual according the Central government much more power than the local governments: -had lived in several states (as did respondents who accorded the two government levels about the same power). -came from a family with a relatively large number of servants. gnpggmg: antrgl vsI Sgate Govgrgment. The individual who ,strongly favored the Central government in.disputes between the Central and state governments: -had conservative language attitudes (as did respond- ,dents.who.favored the state government). -discussed Nehru's death in about the same proportion in Indian and American communities. Outggge: Spaggv vs. ngal vaWernmen. The.individual who .-strongly favoredthe a.state...government..in- disputes between the stateiandilocal governments: -had.not travelled outside India. -perceived his father to have liberal language attitudes. : C . _L c -,nowrelationships _ggug;_;nggxinhemindividual,whomfavoredmthemhigher.of.two -government.lsvels.when there.wasconflict between two of . them: - practiced.much language specialization. 291 -discussed Nehrdb death in about the same proportion in Indian and American communities. Although there was a wide variety of correlates for the attitudes concerning where government power should reside, the patterns of relationship were not very clear. Significant relationships were found among all three sets of variables: language, culture, and personality. In many cases, only a few of the many possible relationships tested were significant. Usually, the patterns of significance and non-significance were consistent. In the case of Language Liberalism, it was the father with whom the respondents attitudes usually corresponded. It was also the perceived.language liberalism of the father that related to the change orientation, approval of social change in India, and attitudes concerning where government power should reside. In the case of the Qrientatien Toward Change Index, the pattern was also consistent. The relationships occurred with one component, the Self Percept Scale, or with the Index itself. No relationships occurred with.the Confidence or Motivation scales. Another case in point.concerns the travel measures. The travel measures generally.showed a.number of relation- Ships to other variables. One of the vicarious travel ,measures,.family.contact with non-Indians, never relatedsto anything elser however. .Thewsocial”classmmeasures“provide.stilluanotherwillus- ,.tration. The Social Class Index rarely related to.other .i92 variables. The components of the Index were another matter. They all related occasionally to other variables. The relationships were sporadic for every component. This brings us to the point of suggestions for future research. One of the most apparent of the suggestions is for refinement of the measures used in this research. The Social Class Index, for example, was.not a very satisfactory one. The»9rientation Toward Change Index needs further work. The Orientation Toward Change Index followed the rules for multi—dimensional indices--its components had very low Aintercorrelations.and correlated well with the total Index. Unfortunately, only one part of the Index.showed any relationship to other variables.. Whether the theory involved in deciding.upon the components needs revision or whether ruthefor each lan- 2 a Little affection 1 = No affection (a) Which of these.1anguages did you learn from your (b) Why did you learn each of the other languages? 303 Can you read any of the following: (Check appropriate answer for_each.languags). No Yes Number of Years of Study Sanskrit: Pali: Prakrit: It is recognized that people use different langua es in different ways. In the following questions (4—16 we are interested in the different languages you use.in communicat- ing with various kinds of people and in various kinds of situations. 4. .5. 11. In which language do you write home? i -- When you were home in India, which language did you speak.with3 (a) your family? (b) home town friends? (c) college friends?. In which.language do you dream? ._ fi _fi Have you ever heard yourself speaking in.dremms? N0 Yes (in.what language?) Which.language do you consider to be your native tongue? In which language are.you.most.comfortab1e? With what particular people in India do you use what languages? (a) Please explain __ (b) I use only one language in India. (specify) In what particular places in India do you use what slanguages? (a) Please explain 12. 13. 14. 15. 16. 17. 18. 19,. 20. 304 (b) I use only one language in India. (specify) On what particular occasions in India do you use what languages? (a) Please eXplain (b) I use only one language in India (Specify) Which language do you think will be most useful to you in your‘profession in India? Whidh language will be the most useful socially in non- professional.situations in India? fl (a) What language(s) do you prefer? (b) (in addition, are there other languages that you like? (specify) (c) What languages are not appealing? (a) What languages do you use unwillingly? (b) When and.why? fi—fi ’— Are there other languages you wish you knew'or you would like to learn? (a) Please specify: (b) Why? . ¢_ Do you feel that there is any value to studying ether languages? ‘Yes Why? No. Why? w—Y At what age did you first start learning English? ( circle one). . l 2 3 4 5 6 7 8 9 10 ll 12 l3 14 other. Howumany years of janmgl English classroom instruction have you.had? 'Were your English instructors: Indian 305 British Other European (specify) American Other (specify) 22. How similar do you feel the spoken English you learned is to the English used in America? Identical Very similar Somewhat similar Somewhat dissimilar Very'dissimilar Other (specify) 23. How adequate do you feel your English instructienfwaa? Completely adequate Fairly“adequate Somewhat inadequate Fairly inadequate Completely.inadequate 24—25. In India, it is necessary for many people to use more than one language. In other countries this is not the case: only one language is necessary. How do you feel about the fact that it is necessary for people in India to use more than one language? 24. For the nation in general: It is'a very good thing It is‘a good thing It is neither a good nor a bad thing It is a bad thing It is a very bad thing 25. For me as an individual: It is a very good thing It is a good thing It is neither a good nor a bad thing It is a bad thing It is a very bad thing 26. 27. 28. 29. 306 How'much prestige do you attach to being bilingual? Very'much Much Some ___;L1ttle ____None When people know and use several languages, there is a possibility that they will use words and phrases from one language when using another language. How'much do you approve of using words and phrases from one lan- guage when communicating in another language? ____Approve highly _Approve somewhat ____Indifferent or neutral ____Disapprove somewhat Disapprove highly Other (specify) How'much do you think that you do use words and phrases from one language when communicating in another language? Very much Much ____Some ___;Little ____fione How concerned are you personally about the language prob- lem in India at the national level? ____Extremely concerned ___;Very concerned ____Somewhat concerned ____Not very concerned ____Unconcerned. 30..How'willing are you to accept Hindi as the official lan- guage of India? ____Completely willing ____Somewhat willing ____Noutral or indifferent ____Somewhat unwilling! ____Completely unwilling ____pther (specify) 307 31. (a) In your opinion, what language should be India‘s official language? (b) Do you have a second Choice? (if yes, specify) (c) Please explain your answers to (a) and (b) above 32. How strongly do you feel about your answer to question 31 (a) above? (i.e., how strongly do you feel that the language you named should be the official language?) As strongly as possible Very strongly Rather strongly Somewhat strongly Mildly 33. (a) How much do you approve of the practice of drawing state lines according to language boundaries? l ____Approve highly ____Approve somewhat ____Indifferent or neutral ____Disapprove somewhat ____Disapprove highly (b) Why? 34. Please rate...“ the languagebelow onhow much prestige you accord them. Add any others that you know which do not appear on this list. 5 = High prestige .2 a Fairly low prestige 4 = Fair1y_high.prestige ’ ‘ 3 a Medium prestige l = Low prestige ____Assamese. Oriya Bengali Punjabi Gujarati Tamil Hindi Telugu 308 Hindustani ____Urdu Kannada ____English Kashmiri ____ Malayalam ____ ____Marathi 35. In your opinion, is there anything that any of the lan- guages is noted for, i.e., any particular characteristics that you associate with a language? Please specify. Language Characteristic 36. (a) Please list below the languages, dialects, sub- languages and vernaculars that your fathg; knows. (b) Indicate the extent of your fathers knowledge by putting an “X ” in the appropriate columns. E = Excellent F = Fair G = Good P = Poor-, UNDERSTANDING READING WRITING SPEAKING SPEECH LANGUAGE E G F P E‘g F P B G F P E G F P -1 --—- e-—-.‘ ——-— nt--u. P--- 1--—- ----1--—- r.--- 1-—-- t'lf‘trhv'rr 308a 37. (a) Please list below the languages, dialects, sub- languages and vernaculars that your moths; knows. (b) Indicate the extent of your mothers knowledge by putting an "X“ in the appropriate columns. 3 = Excellent F = Fair G = Good P = Poor. UNDERSTANDING READING WRITING SPEAKING SPEECH ‘ W.— EEEE ESE}: .5522! p922. 1“ JF---1----T--- db--— 2. 1..--.. +--- ..-..- ...... 3". h--- ---d _--1 ---‘J L --» -- —-- 1---- )L +— -( -+ Q. u—-—+r run-q I———ed nun—q 2* Jun--J» ----i —-—.—J (—--J 38. (a) (b) (c) (d) (e) (f) 39. (a) (b) (c) (d) 309 What language(s) is (are) preferred by your father? In addition, are there other languages that your father likes? (specify) __ __ What languages are not appealing to your father? What language(s) is (are) preferred by your mother? In addition, are there other languages that your mother likes? (specify) What languages are not appealing to your mother? Are there any languages that your father uses un- willingly? When and why? Are there any languages that your mother uses un- willingly? When and why? 40. How'much do your parents approve of using words and phrases from one language when communicating in another language? Father Mother Approve highly Approve somewhat Indifferent or neutral Disapprove somewhat ______ Disapprove highly Other (specify) 41. How’much do you think that they use words and phrases from one language when communicating in anOther? Father Mother Very much Much 42. 43. 44. 310 Some * Little NODO How much prestige do they attach to being bilingual? Father Mother Very much Much Some Little None How concerned are your parents about the language prob- lem in India at the national level? Father Mother Extremely concerned Very concerned Somewhat con cern ed Not very concerned Unconcerned How willing are your parents to accept Hindi as the official language of India? Father Mother Completely willing Somewhat willing H H!!! Neutral or indifferent Somewhat unwilling Completely unwilling Other (specify) 311 45. What language do your parents feel should be the official language of India? 46-47. How do you think that your parents feel about the fact that it is necessary for people in India to use more than one language? 46. For the nation in general they feel: Father Mother It is a very good thing ______ ______ It is a good thing ______ _____‘ It is neither a good nor a bad thing _______ __ It is a bad thing It is a very bad thing 47. For them as individuals they feel: Father Mother It is a very good thing It is a good thing It is neither a good nor a bad thing H! I It is a bad thing It is a very bad thing Every society is composed of systems of social units. In the following questions we are interested in how some of Indidh social units feel about language. The first social unit is subcaste (jati). ' 48. (a) Which language(s) is (are) preferred by the members of your subcaste (Jati)? v In addition, are there other languages that they like? (specify) .— (b (c) Which languages are not appealing to them? 49. (a) Are there any languages that the.members of your sub- caste use unwillingly? (b) When and why? __ 50. 51. 52. 53. 312 How'much do the members of your subcaste approve of using words and phrases from one language when communicat- ing in another language? ____Approve highly ____Approve somewhat ____Indifferent or neutral ____Disapprove somewhat ____Disapprove highly Other (specify) How'much do you think that they do use words and phrases from one language when communicating in another? Very much Mudh Some Little None How'much prestige do the members of your subcaste attach to being bilingual? Very much Much Some Little None How concerned are the members of your subcaste about the language problem in India at the national level? Extremely concerned Very concerned Somewhat concerned Not very concerned Unconcerned 54. 55. 313 How willing are the members of your subcaste to accept Hindi as the official language of India? Completely willing Somewhat willing Neutral or indifferent Somewhat unwilling Completely unwilling Other (specify) -— What language do the members of your subcaste feel should be the official language of India? __ 56-57. How do the members of your subcaste (jati) feel about 56. 57. 58. the fact that it is necessary for peeple in India to use more than one language? For the nation in general they feel: ____ It is a very good thing ____ It is a good thing ____ It is neither a good nor a bad thing ____ It is a bad thing It is a very bad thing For them as individuals they feel: ____ It is a very good thing It is a good thing ____ It is neither a good nor a bad thing ____ It is a bad thing It is a very bad thing In India, as in all countries, there are several social classes; In your judgment what are the major social classes in India? 59. 60. 61. 62. 63. 64. 314 Now, how would you place yourself in this scheme? What is your Q23. social class as an individual? (a) (b) (c) (a) (b) How Which language(s) is (are) preferred by the members of your social class? In addition, are there other languages that they like? (specify) Which languages are not appealing to them? Are there any languages that the members of your social class use unwillingly? ._ When and why? _ much do the members of your social class approve of using words and phrases from one language when communicat- ing How in another? Approve highly Approve somewhat Indifferent or neutral Disapprove somewhat Disapprove highly Other (specify) much do you think that they do use words and phrases from one language when communicating in another? Very much Much Some Little None How much prestige does your social class attach to being bilingual? Very much Much Some Little None 315 65. How concerned are the members of your social class about the language problem in India at the national level? 66. 67. Extremely concerned Very concerned Somewhat concerned Not very concerned Unconcerned How willing are the members of your social class to accept Hindi as the official language of India? Completely willing Somewhat willing Neutral or indifferent Somewhat unwilling Completely unwilling Other (specify) L T ' What language (do the members of your social class feel should be the offiCial language of India? 68-69. How do the members of your social class feel about the fact that it is necessary for peeple in India to use more than one language? 68. For the nation in general they feel: It is It is It is It is It is a very good thing a good thing neither a good nor a bad thing a bad thing a very bad thing 69. 70. 71. 72. 73. 316 For them as individuals they feel: ____ It is a very good thing ____ It is a good thing . ____ It is neither a good nor a bad thing It is a bad thing It is a very bad thing (a) Which language(s) is (are) preferred by the peOple in the region where you live? (b) In addition, are there other languages that they like? (specify) ~ (c) Which languages are not appealing to them? _ (a) Are there any languages that the members of your region use unwillingly? (b) When and why? How'much do the people in the region where you live approve of using words and phrases from one language when communicat- ing in another? Approve highly Approve somewhat Indifferent or neutral Disapprove somewhat Disapprove highly How much do you think that they do use words and phrases one language when communicating in another? Very much Much Some Little 0 3 None 317 74. How'much prestige do the people of the region where you 75. 76. live attach to being bilingual? Very much Much Some Little None How concerned are the peOple in your region about the language problem in India at the national level? Extremely concerned Very concerned Somewhat concerned Not very concerned Unconcerned How'willing are the people in the region where you live to accept Hindi as the official language of India? 77. What feel 78-79. Completely willing Somewhat willing Neutral or indifferent Somewhat unwilling Completely unwilling Other (Specify) t_ language do the people in the region where you live should be the official language of India? How do the peoplelin the_region where you live feel about the fact that it is necessary for people in India to use more than one language? 78. For the nation in general: It is a very good thing It is a good thing 79. 318 It is neither a good nor a bad thing For 80-82. 80. 81. 82. The It is a bad thing It is a very bad thing them as individuals 8 It is a very good thing It is a good thing It is neither a good nor a bad thing It is a bad thing It is a very bad thing India is changing in many ways. How do you personal- ly feel about the changes that are being made? changes are being made: Much too quickly Somewhat too quickly About right Somewhat too slowly Much too slowly Among the changes being made: The All are in the right direction Most are in the right direction About half are in the right direction and half in the wrong direction Most are in the wrong direction All are in the wrong direction changes being made are: ____,Much too large in scope (size) Somewhat too large in scape (size) About the right scope (size) 319 Somewhat too small in sc0pe (size) Much too small in sc0pe (size) 83. How much power do you feel each of the following levels of government in India should have? (Check one alter- native for each level of government) Central State Local Gov't Gov’t Gov't Absolute power Great power Some power Little power l-Hl No power 84. When two of the governments are in disagreement, what do you feel the outcome should be? (Check one alternative for each column) a. (A) Central (A) State (A) Central vs. vs. vs. (B) State B cal (B) Local ' A should always pre— vail over B A should usually prevail over B A should prevail about half the time over B A should rarely prevail over B A should never pre- vail over B 85. (a) Below are some major groups in India. Please indi- cate the importance of each group to you by placing an “X“in the appropriate column. 5 3 Extremely important 2 = Not very important 4 - Very important 1 = Unimportant 3 Somewhat important 5 4 3 2 1 mn- _ “*__* We... ___.._..__...._ Family, J ___ ___.___.__. ___ 320 Sybcaste (jati) 5 .....Eesm __.._______.___ WW): _____._____.__. 86. 87. 88. (b) (a) Indicate the relative importance of each group to you by ranking them, using a S for the most import- ant, a 4 for the next most important, and so on. Place these ranks on the line at the left of each group. As peeple grow older sometimes their level of drive and motivation changes. As for yourself, would you say that your drive and motivation is: Much higher now'than it used to be Somewhat higher now than it used to be About the same now as it used to be Somewhat lower now than it used to be Much lower now than it used to be (b) Please explain your answer. Some people are said to be strongly motivated and to have a lot of ambition and drive. Others are said to be less motivated and to have less ambition and drive. How mudh motivation would you say that you have? Very much Much Some Little None (a) As compared with the rest of your family would you say that you are more or less ambitious than they are? Much more ambitious Somewhat more ambitious About the same in ambition Somewhat less ambitious Much less ambitious 321 (b) Please explain your answer. 89. As far as you are concerned, how important would it be to have more drive and ambition than you now'have? Extremely important Very important Somewhat important Not very important Unimportant 90. Some people are always trying to get ahead and improve themselves and others seem to be satisfied with their present condition. As far as you yourself are concerned would you say that you are: Much more ambitous than most peOple Somewhat more ambitious than most people About the same in ambition as most people Somewhat less ambitious than most people Much less ambitious than most peeple 91-108. Please answer the following questions by placing a check (v?) on the continuum at whatever point you think best describes your feelings. fismnlst How'much do you enjoy lectures? ‘/, very much not at all 91. Do you enjoy starting conversations with strangers? very much not at all 92. In your social group at home, how embarrassed does it make you to express a deviant opinion? very little very much 93. Do you like to take changes? very much very littlefl 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 322 Do you seem to get flustered when the attention of a group turns to you? always never Do you like to stand out in a crowd? not at all very much Do you like to act on hunches? very much not at all— Are you often praised by other peOple? never > very often Do you usually mull over decisions you‘ve made and worry about whether or not you‘ve made the right decision? always never Do you complain to the person.in.charge when food, service or merchandise is not as good as it should be? never always Do you enjoy meeting new people? very much very little In your present business or school group, how'much do you think others will look down on you if you express a deviant opinion on an important issue? very much not at all Do you feel self—conscious very often? Very often not very often Have you found that other people often praise you for your ingenuity? very often never 323 104. Do you find it very disturbing to make errors in public? very much not at all 105. Howumuch confidence do you have that you will succeed in life? complete confidence no confidence 106. Would you describe yourself as the shy type? very much not at all 107. How'much do you enjoy giving speeches or talks before groups of people? very much not at all 108. Do you ever wish you had more self confidence? rarely often 109-124. Please indicate how much you agree or disagree with each of the following statements according to the Agreement Code. Check (J') the column which corres- ponds to the appropriate Agreement Code. AGREEMENT CODE 1. Disagree very much . Disagree somewhat , Neutral . Agree somewhat . Agree very much .l.._2_._§_,_£_._§_ 109. I usually know where to go to find out about important things. U‘Ith llO. Unless there is a good reason for changing, I think we should continue to do things the way they are being done now. __ __ __ __ __ lll. Youth is usually too impatient and too much in a hurry. ___ ___.___ ___ ___ 112. When new ideas are going around, I am usually among the first to accept them. _______ ___ ___ ___ 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 324 I watch my expenses very care- fully and know' where my money goes. It is not particularly import- ant to me to know all the answers to things people are talking about. A penny saved is a penny earned. Among my friends, people usually take me to be younger than I am. I an a little bit suspicious about peOple who always want to have the latest in everything. Among my friends, I am con- sidered to be a free spender and liberal in the use of money. The enthusiasms of younger peeple are hard for me to understand. Others seem to be better than I am at knowing about everyday affairs. In general, modern ways of living are much superior to older ways. My interests seem to be more with those people younger than I am than with older people. I seem to understand about the way in which the world is developing better than many peeple. Money is made to be spent. General Demographic Information Age Sex: (circle) M F Religious affiliation: .1. .2. 325 128. Caste: 129. subcaste: (jati): 130. For the purpose of our survey we need a rough indication of the income of your famil . Please indicate in which of these ranges your familys income falls. 131. 132. Yearly Income in Rupees 1200 and under Rs. 1201 to 2400 Rs. 2401 to 3000 Rs. 3001 to 4000 Rs. 4001 to 6000 Rs. 6001 to 10000 Rs. 10001 to 16000 Rs. 16001 and up Rs. In many families, the family income is used for the support of a number of people. How'many adults and children use the income that you have indicated as your family income in the question above? (a) Place an “X"by the category which best describes your father's education. (b) Place an "X"by the category which best describes your mother's education. Father Mother Below high school (8 years or less) ______ .Matriculation ( 10 years) ______ Intermediate (11 to 12 years) Bachelor's degree or equivalent Master’s degree or equivalent Doctorate 326 GENERAL INFORMATION (Please Print) NAME 1. How’many months have you been in the United States? 2. For which degree, if any, are you now working? ____ 3.3. or 3.5. M.A. or M.S, Ph. D. Other (eXplain) None (explain) 3. What is your major? 4. Where did you attend intermediate school (high school)? Town or City Country 5. What were your major subjects in intermediate school (high school)? 6. What schools above intermediate level did you attend before coming to Mishigan State University? Name of School Place Major Subject Degree . Hllgé‘ 7. Please list any academic honors you have received. 8.‘Were you working, i.e., did you have a job, before you came to the U.S.? Yes (specify) No 9. 10. 11. 12. 13. 14. 15. 16. 327 Do you have a definite job to go baCK to in India? Yes (specify) No Where have you lived most of your life? (check one) in a large city (300,000 or more population) in a large town (100,000 to 300,000 population) in a medium town (20,000 to 100,000 population) in a small.town (1,000 to 20,000 pepulation) in a village (l,000 or under pepulation) Place an "X" on the map (opposite page) where your home is. ' State: City or Town: In how many cities, towns or villa as have you lived for over six months in your country? Circle one) 1 2 3 4 5 6 7 8 9 10 or more Place numbers on the map to indicate each place you have lived, starting with a "l“for the first place a "z“for the second place and so on. - How'much traveling to different places have you done in India? ____ Very much Much some Very little None How'many times have you been out of your country? (circle) 1 2 3 4 5 6 7 8 9 10 or more (a) In how many countries have you stayed for longer than one month? (Do not count India or the United States) 17. 18. 19. 20. 21. 328 (b) Names of countries: (c) Total time outside of India in years: How much reading about foreign countries have you done? ____ Very much ____.Much ____ Some ____ Very little None How'much contact with non-Indians would you say that your family has? Very much Much Some Very little None Are you the first in your family to seek education out- side India? - ‘ Yes ____ No (please explain) Hew'many brothers and sisters do you have? Brothers (specify how’many are over 18 years ofage) Sisters (specify how'many are over 18 years of age) Howwmany of your brothers and sisters are in college or have completed college? Brothers Sisters 22. 23. 24. 25. 26. 329 How’many of your brothers or sisters are studying or have studied abroad? Brothers (place of study): Sisters (place of study): _ I Did any of your parents, grandparents or uncles and aunts study abroad? Father: No Yes (Where?) Mother: No ____Yes (Where?) Grandfathers: No Yes (How'many and where?1___ Grandmothers: No ____Yes (How'many and where?)___ Uncles: No ____Yes (How'many and where?)___ Aunts: No ___;Yes (How'many and where?)___ About how many books, excluding school and college text books and help books, would you estimate are in your family's home library? HoWMmany household servants does your family in India have? Which of the following best describes the occupation of your father? (Check one) If father is deceased, check what his occupation was. Professional (lawyer, medical doctor, university professor, etc.) Owner or manager of a large business Government official (specify) Owner or manager of a large rural estate Owner or manager of a small business Office worker (other than above) or clerk in govern- gggice worker (other than above) or clerk in business Skilled craftsman (carpenter, tailor, etc.) or foreman Owner of a small or average farm Unskilled urban worker Agricultural worker not owning land 330 Agricultural worker renting land or sharecropping Semi-skilled urban worker (truck drivers, waiters, machine Operators, etc.) ’ Other (specify) On Wednesday, May 27, Jaweharlal Nehru died. We are interested in reconstructing the patterns of communication about this event. Please think back very carefully as you answer the following questions. 27. 28. 29. 30. 31. 32. 33. At approximately what time (day and hour) did you first hear about it? How did you first hear about it? __ Radio Television Newspaper Telephone (from whom?) Word of mouth (from whom?) Other (Specify) With how many Indian.friends did you discuss the event? Specify names: With how many American friends did you discuss the event? Did you attend the commemoration service? Yes NO Did you discuss the event when you wrote home? Yes No How important to India do you feel Nehru was? Extremely import ant Very import ant 34. 35. 36. 37. 38. 331 Somewhat important Not very important Unimportant In your opinion what was Nebru's single most.important contribution to India? How well did you like Nehru a; 3 political figure? Very well Well Somewhat Not well Not very well How well did your parents like Nehru as a political figure? Very well ____ Well ____ Somewhat Not well Not very well Three men whose names were mentioned as possible candidates for prime minister were Shastri, Nanda and Desai. In your judgment what does each of these three men represent in Indian political life? Shastri: . ‘_ .__ Nanda: Desai: How’well do you think each of these three men would serve India as prime minister? Shastri Nanda Desai would do an outstanding job would do a good job would do a mediocre job 332 would not do a good job would do a very poor job 39. In your judgment, is there someone other than Shastri, Nanda and Desai who should have been considered as a candidate? NO Yes (Specify) APPENDIX B Age Number in Percent of Ygags ngg13 ng212 22-24 25 26.0 25—27 28 29.2 28-30 16 16.7 34-36 7 7.3 40-42 1 1.0 43-45 1 1.0 No Response 2 2.1 Total 96 99.9 Mean Age 27.5 Std. Dev. 5.6 Median Age 26.4 333 APPENDIX C ACADEMIC OBJECTIVES Number in Sample Bachelor of Arts or 16 Bachelor of Science Master of Arts or 26 Master of Science Doctor of Philosophy 50 Other: Diploma in school administration 1 General Experience in field 1 Post doctoral research 1 Research Associateship 1 Total 96 334 Percent of mg... 16.7 APPENDIX.D ACADEMIC SPECI Engineering Marketing and Transportation Industrial Administration Agriculture Horticulture Fisheries and Wildlife Dairying Food Science Biochemistry Biophysics Biology Chemistry Pharmacology Veterinary Medicine Geology Physical Science Mathematics and Statistics AnthrOpology Communications Economics Education Psychology Social Science Sociology English History No Response Total* ALTIES JQ‘P , )i ' 1 fly . Number in Percent of nggls Sample uh» U :5 h e ee e 00 (D e OOHO O OwONPPH O I O O O O O O 0 L00 col—'0 \l O 00 NI—‘P‘H H HP ounwwww map PumNNw HHNH HH H HP pNNNNHH mww HumNNw HHMH PH KO ‘1 H O O *One person listed two majors. 335 APPENDIX E NUMBERKOF MONTHS SPENT IN THE U.S. Number in Percent of Months Sign}: Sample 1 - 6 9 9.4 7 -12 23 24.0 l9-24 15 15.6 37-42 10 10.4 43-48 8 8.3 over 48 6 6.3 Total 96 100.0 Mean Months in U.S. 24.7 Std. Dev. 16.3 336 APPENDIX F RELIGIOUS AFFILIATION Hindu Islam Jain Sikh Zoroastrian Christian Unspecified Catholic Methodist Presbyterian Seventh Day Adventist Atheist and no religion No Response Total 337 J I w .- ...~ “ 7.- fie % Number in _éamnlL. 5 whflwo m m HHHwb 96 APPENDIX G Two items on.the questionnaire were concerned wdth the employment of the respondents before and aftertheir sojourn to the U.S. The items and the responses to them appear below. 17. Were you working, i.e., did you have job, before you came to the U.S.? Yes No 18. Do you have a definite job to go back to in India? Yes No EMPLOYMENT BEFORE AND AFTER SOJOUBN _— .. , . . Had Job before U.S. Had Job to Return to sojourn in India Number in Percent Number in Percent Sample of Sample Sample of 8:321: Yes 66 68.8 30 31.3 No 30 31.3 66 68.8 Total 96 100.1 96 100.1 338 APPENDIX H Item 16c on the General Information part of the questionnaire resulted in considerable confusion. Many reSpondents appear to have included the number of years they had lived in the U.S. in their response to the item. (See page 55). Item 16 in its entirety is given below. Also given below are.the responses to part c of item 16. Data from 16c were not included in further analyses. 16. (a) In how many countries other than India and the U.S. have you stayed for longer than one month? (b) Names of countries: (c) Total time outside of India in years: NUMBER on" was SPENT ’ oumsmfmom Number of Number in Percent of Yggga Sample Sample 1 9 9.4 2 11 11.5 3 15 15.6 4 4 4.2 5 1 1.0 6 0 0.0 7 2 2.1 8 0 0.0 9 l 1.0 10 0 0.0 11 l 1.0 No Response 52 54.2 Total 96 ,100.0 339 APPENDIX I Distributions of Responses to Items Comprising the Language Liberalism Indices (Discussion of the Indices appears in Chapter V, pages 104 to 119 Number of Social ganggaqgg. £31; Father ,Mother Sgbcggtg Class Regign 0 36 12 11 10 8 8 l 15 8 6 ll 7 8 2 7 3 5 4 2 6 3 7 2 1 3 2 4 2 l l 5 1 l 1 Miscellaneous 6 4 9 7 South Indian Languages 8 6 2 5 5 9 Many Languages 3 1 1 l l 1 All but. . . languages 4 3 4 12 8 13 340 341 LANGUAGES USED~UNIELLINGLY ~~.... - k Number of Social Languages ,§2;§_ Fathg; Mgther Subcgste Class Regign 0 52 55 53, 15 17 ll 1 19 13 16 11 8 14 2 10 2 l 4 5 7 3 1 3 l 4 2 1 Miscellaneous 5 4 1 South Indian Languages 1 1 Many Languages 1 2 1 All but ... Languages 5 1 l 2 4 l The intermixing of languages was greeted with in- difference or slight approval by the students and they saw their groups as reacting the same way. In general the reSpondent saw'himself as approving the most and his mother, the least, although the mean approval ratings among the groups did not differ greatly. Variability of ratings within groups was not high. According to the Indian students the most inter- mixing of languages was carried on by social classes. The least was done by mothers and fathers. 342 me. No. om. mo. mH.H mm.a .ouo .ouo m~.m mm.m o~.m mo.~ mo.m om.m Hooouoee one: o.ooH o.ooH o.ooH o.ooa o.ooa o.ooa Hosea o.oa m.ma H.om o.od o.ma H.~ cocoon-m oz 0.0 o.H H.m H.m H.m H.m servo H.m ~.m m.v m.aa m.n masons ooouooouao a m.> N.¢ m.m m.m 5.0H possoeon e>oumomofln m a.sv a.mm m.mm m.am «.om Houston no nauseousooH m m.mm «.om m.ma a.mm H.b~ nonsoeoe e>oumm< o o.o ~.m m.o m.» o.oa.z seamen «tosses m mmflmflm memao Heaven emmummsmauuuwnxoz :kmmngn-:MHmw dmwmmmmfllwmlddmmflm A<>Oflmm4 OZHXHszBZH 3 m“ mm. oo.a oa.a ~m.a oo.a ..>oo .oum oa.m mm.m em.~ om.~ oa.m moaneououoa one: o.ooH o.ooH H.ooa H.ooH o.ooa sou-a m.o a.ms o.oa m.HH ~.o cocoon-m oz 0.0 0.0 o.o o.o a.” mango a.m H.m o.m~ o.H~ H.m «cos a e.oa o.oa a.mm o.oa o.am canvas a m.¢o o.om a.sm o.o~ o.oo seam m m.a~ o.a~ m.o o.oa o.oa sous o ~.m m.ms m.o m.o m.ma nose suo> m nldlen» .m . - lllum Hem u. wanna ejall: «x gimme. IH OZHXHEMBZH i984 344 The moderate prestige associated with bilingualism on the part of the students was echoed in the.prestige ratings they claimed their groups attached to bilingualism. The tables show the prestige ratings. The mean prestige ratings of all the groups clustered around 3.00 indicating that on :the average respondents felt all groups to consider bilingual- ism to have ' 'some" prestige. There was considerable spread of cpinion in all groups, however. A comparison of the individual and his groups on willingness to accept Hindi as the official language showed ‘even the lowest group to be part way between indifference and some willingness. The most willing to accept Hindi were the respondents themselves. The students saw their region and social class as almost as willing as they were. The least in perceived willingness were the subcastes, fathers and mothers. Over 20 per cent of each group, however, was seen as at least. somewhat unwilling to accept Hindi and at least half of these were viewed as completely unwilling to accept it. The vari~ ability within the groups was large. 345 oo.a Ha.a mm.~ mm.a mm.a .>eo .ouw mo.~ HH.m mo.~ vs.m aa.m omauooum ones o.ooa o.ooa o.oo H.ooa a.ooa Hosea m.~a o.ma o.o~ o.oa m.o~ ooooomom oz o.o o.o o.H o.o o.o nosuo m.o ~.m H.m o.va m.~a cooz a m.oa m.ma e.o~ o.a~ s.oa «Hausa m 06¢ v.mm Nam ozvu mod seem m m.ma m.HH m.HH v.o p.oa suns. v m.p a.ms m.s v.oa o.ma rooz.suo> m Isolaqeml .n__ I; .3 .... om “lumdwlmn m: sled. la»: Spunzmmmwmm .. ...J—dlw: u m: on mfllumfllllullaum m e O ( . ' [I . 1 n I! 0.1! I. e v I on: C O l SMHAoa Hoe. on» an uaaoauecmuw..u “ --- H¢>oa no. an» an unnuamaemam . p Ammv mo. Ahmv ha. Ahmv ma. “on” ma. “va AN. nu: Ammo so. Ammo any Ammv.H¢. Ammo ea. It! Ammveeibvo Admv Nm. AHMV mm. III AHmV fin. Ammv NM. III «may fig. lumdmmflmw mdflmmamwmmm Ilflmmmmadfll NdmdddddmdmimaAAMNQQQ_ Ema donned moaxueueucH coon - uoz- Iamomcfiaam uemmsmomq eommomcmq ImaQ xeo H Em amuonaq e no can . m.uceocomuem ea mseuH mcoec ecoquoaeuuouueucu xeooH Euadnuenaqnemeomcmn M aocam no euoeuoeuud.i emaueeum Bewamswcfiaammm mcustuewou Hanu04.w Hm>ouoo< mcuxaeueucH u hamcaaaaaco dean memmomceq H mcaaeemmd uez memosmomq W .oaa omWémoa-eOmmm no pose“ no ecofiumaounouueucd en» mo code-nu .ooa_o:mmoa momma co oooom sum ocean sum nooauoaenuouneuca any nvqga co eeeua e39 b NHQZMQAQ ao>ua Hoe. as» an uaoueuacmamcmm Ho>oa Ho. on» an ucmuauacmew .m Hosea mo. on» an ucmuauacmam . Avmvemv. AVNvefifi. Agmvem¢o “MN” 00. Awdveam. ' xoocH Euaaeuenaq emmamceq Agaveehmo III Amwv mo. Amhv No. Amov ho. .Aawvom.I «wavav.I aocam we encouneuufi III Avsu .om. .Hev ma. ”may -.I wvHme.I codename ameawsmcaaam 9 III Ahhveeemh. “may mN.I wwav wwo mdaxdsfioucH Hmsu04 M II. Amav mo.u Away mm. amsouoo4.mc«xusuouam III on oo. mamcaaaazco neuo memoomoea III madameQQG uoz mommawcem Immmdmw. ImmmmmumMI.mmwmmamumom.mmwmqmmmIII_N4mm444 mom mmaamummfl we soon Emu Hoouud moaxqeueuoH ween , uoz Iumouoc Iaoomcaadm nemmamoaq eemmomcmq smaamuenaq emmomcmq u.ue£umm ca neeuH moss“ mooaueaeuuooueucH Addy emm. 350 «Gena mo 00cm lumeuod Add. mm. “aav.mm.I “Hay HH.I Ammo Ho.I Ammo Ho. Ammo ca. III .wsv om. A~sv no. III «who...¢¢. flmflmmmwMI mmmmmauowmw Ilflmwmumme Ema Hmouo4 wcaxHEHeucH Ismsmcaaam WHHM mm. ..vavmo.I onV¢M.I $9.2. «Fae oo. Ho>oa Hoe. an» on acmuauucmfim..w Ho>ma mo. on» an panofiuecmam . Amy no. xoUcH Euaaeuenaqflemeomcea wvamo.I «noun mo euceuQeUUA ..«an~.I omens-um sueamsmcaaem «mav.oo. mcaxaeuouaH «cause “may mm. ao>ouo94 mcaxdfiueucm Amy 00. mamcaaaascn Uses oemmamcea mcaaeemad uez eewmsmoen fluid. duos 3 GOOD - uoz . mommamoma mommamcmq anamqu A omeomcoq o.ue£uez o« e ” xovcH EeuH wooew oceaueaouuoououoH N‘— 7 A “Pic—Iv -.. .-- Ho>¢a Hoo. any on ucmufiuacmam..m Hosea no. as» an pancauacmum . xeooH Emaaeuenaq emeomcen «poem me eoceumeuud omauooum aoaasamoaaam maaxeauouaa finance ae>onm04 moax«ahevcH mamcaaadsco coon nemmomcem mafiaoemod uoz enmeomcen .mHV hm. Amav mm. “may ma. Amav emv. “gay cmo. Amy emu. III Away mo.I «movmo.I Amov no.I .AmflvmmoI “mavemm.I III Ahov chm. Ammo emm. .Aoflv mo. “may oa.I III “moveseao. “may mm.I mmav ca. 1 III Amav 0H.I “mav.¢o. E. III a: .E. 3 g 3 l3 3% 3.13% we once Ema amou04 mcaxaaueucH can: - uez - Iuaeuu< Iamomcuaam nemesmcoaveemmsmcmq I31 Emwanuenanflemoomcmq e.eueeunom ca useuH wooed nooaueaenuouuoucH ..§ - .. ( ' v"*—*""" ‘e-o-<..v - H¢>oa goo. as» so ucmuauaaoam.mm Ho>oa so. on» an noncomacmem cm Ha>oa mo. on» no ucmuwuaamgm . Away mac 352 «coax we even Iumouud lode mm. Asa. .mm. “may .00. Ame no. .Ao. “use so. impo NH. Aesv ma. .mav nH.I mode III Aamveeedmo atheemm. “may mmo III Ashe...mm. ..eoo NH. AHHV hN.I mmoI xoocH Emuamuenaq emesoceq mmoI «poem mo euceuoeoo4 «Ody mm. omaueeum Euadwomcuaam woao oH.I mcfixasuoucH amoeba Aoav mm. He>ouma4vmc«xuaueucH “no Hg. mamcfiaafi3cs neon memmomcen III mcuamommd uoz memmomceq lawmaummum mdamflammumm Ilflmwwmmmfla Namdflwdflmmm mdeMMNMfl uoz Iamomcdafim memmomcmq uommomcwq Ema Husuud moaxaeuoucH coma xuu H Euaaouonaa e on cmq.mmmao do uom H memuH 08¢ mo aumHouuoououcH .‘V V. ,- ...-«paw. .» Hosea Hoe. an» an unnuaudamumIcm Hosea mo. as» on unnuauacmam . Ammvmo. 353 «Och we even Iuaouo4 ANNV NN. ANNVemwo ANNvemv. «ONV fin. Aomv NOII Ammv Odo! “any 00. AHNV mHo III Ava eMNo Amhv oo. .aowv vac III Amso ...me. w-~ HooI III .oav mo.I Away so.I xooaa amaamuonao newsman; «hay «H.I «ocum.uo GUGMHQOOQA «may as. moduuoum aueawamcqaam mhd~ mg. mcdxafiueucH Husuvfl sod. om. He>ofimmd mcwxdaueucm Amy mNII hachHHa3oD uflen newcomceq III mdflddomnd uoz «Confimcod ImlIIIIw 33am mIcIIIIIIIHxEuuucH. Elwmummlla j; 5: o 3% Ema . HmouU¢ mcwwaHoucH can: uoz Iamzmcaaum oomesmceq womesmoon xe H Emaaeuon A o e and once m c« mEouH a .fi m oHumaouHooueucH APPENDIX K CONFIDENCE SCALE (Discussion of Confidence Scale appears on pages 124-126). 91-108. Please answer the following questions by placing a check (II) on the continuum at whatever point you think best describes your feelings. gaggle: How~much do you enjoy lectures? J very much not at all 91. Do you enjoy starting conversations with strangers? 92. 93. 94. 95. 96. very much not at all In your social group at.home, how embarrassed does it make you to express a deviant opinion? very little very much Do you like to take.chances? very much very little Do you seem to get flustered when the attention of a group turns to you? always never Do you like to stand out in a crowd? not at all very much Do you like to act on hunches? very much not at all 354 97. 98° 99. 100. 101. 102. 103. 104. 105. 106. 355 Are you often praised by other people? never very often Do you usually mull over decisions yoJVe made and worry about whether or not you’ ve made the right decision? always never Do you complain to the person in charge when food, service or merchandise is not as good as it should be? never always Do you enjoy meeting new people? very much very little In your present business or school group,.how'much do you think others will look down on you if you express a deviant opinion on an important issue? very much not at all Do you.feel selfpconasious very often? Very often not very often Have you found that other people often praise you for your ingenuity? very often never Do you find it very disturbing to make errors in public? very much not at all How’much oonfidence.do you have that you will succeed in life? complete confidence no confidence Would you describe yourself as the shy type? very much not at all 356 107. How much do you enjoy giving speeches or talks before groups of people? very much not at all 108. Do you ever wish you had more self confidence? rarely often APPENDIX L KOSSOFF SELF PERCEPT SCALE (Discussion of Self Percept scale appears on pages 126-128) 109-124. Please indicate how mudh you agree or disagree 109. 110. 111. 112. 113. 114. 115. 116. with each of the following statements according to the Agreement Code. Check (\I) the column which corresponds to the appropriate Agreement Code. AGR§§QENT coo; . Disagree very much . Disagree somewhat . Neutral . Agree somewhat . Agree very much _l__2__.3__4.._.§_ Uheuuknd I usually know where to go to find out about important things. _ Unless there is a good reason for changing, I think we should continue to do things the way they are being done now. m _ ___,___ Youth is usually too impatient and too much in a hurry. When new ideas are going around, I am usually among the first to accept them. I watch my expenses very care- fully and know where my money QOCSe ' ___-___— It is not particularly import- ant to me to know all the answers to things people are talking about. A penny saved is a penny earned. Among my friends, people usually take me to be younger than I am. 357 117. 118. 119. 120. 121. 122. 123. 124. 358 I am a little bitisuspicious about people who always want to have the latest in everything. Among my friends, I am con- sidered to be a free spender and liberal in the use of money.___ The enthusiasms of younger people are hard for me to understand. Others seem to be better than I am at knowing about everyday affairs. In general, modern ways of living are much-superior to older ways. My interests seem to be more with those people yOunger than I am than with older people . I seem to understand about the way in which the world is developing better than many people. ' Money is made to be spent. _l._2.._§_ APPENDIX M DISTRIBUTIONS OF RESPONSES TO ITEMS COMPRISING MOTIVATION SCALE (Discussion of Motivation scale appears on pagesylZB-132)N W. The latent percent- age of Indians rated their motivation.and drive as.much higher than it had been in the past. Nearly one-half the sample checked the highest category. "Mush higher now.than it used to be.“ The question and the responses to it appear below. Explanation for the response.categories checked were not always informative. ,Many respondents, however, credited travel, increased education and/or contact with many more people as primarily responsible for the increaseein motivation. 86. (a) As people groonlder sometimes their level of drive and motivation changes. As-for your- self, would you say that your drive and meti- vation isf Much higher nowIthan it used to be Somewhat higher now than it used to be About the same now as it used to be Somewhat lower now than it.usad to be .Much.lower.now than it used to be (b) Please explain your answer. 3 59 Number of Percent of Motivation now is: ‘§g£BQB£££' _éemsle___ 5 Much higher now than it used to be 47 49.0 4 Somewhat higher now than it used to be 32 33.3 3 About the same now as it used to be 9 9.4 2 Somewhat lower now than it used to be ‘ 2 2.1 1 Much lower now than it used to be 1 1.0 No Response 5 ‘ 5.2 Tetal. 96 100.0 Mean Change in Motivation 4.34 Std. Dev. .83 flgsigg§i2n_£gxgl. The second question asked concerning motivation.appears below. Following the question is the dis- tribution of responses to it. Estimates of motivation were I u very high. oh. half the individuals felt they had "much and another third felt they had "very muchumotivation in general. 87. Some people are said to be strongly motivated and to have a lot of ambition and drive. Others are said to be less motivated and to have less ambition and drive. How much motivation would you say that you have? Very much Mudh Some Little None 361 KMQQI¥5E19§§§I¥iyf’ I * Number of Percent of amount of Motivation Responses Sample 5 Very muCh 32 33.3 4-Much 49 51.0 3 Some 13 13.5 2 Little 1 1.0 1 None 0 0.0 No Response 1 1.0 T'P‘l’ 96 99.8 Mean Motivation 4.18 Std. Dev. .70 ambition Relative to that of Family. The third motiva- tion item asked the respondent how he felt his ambition com- pared to that of his family. The item appears.below'followed by a distribution of responses to the item. When the students compared themselves to their.families, about.58 percent rated themselves as having more.ambition and a third saw little difference between themselves and their familieS. Explanations offered for responses to the family and self comparison indicate that those who saw’no.difference between themselves and their families considered their families as very ambitious. Those who saw themselves as having greater ambition than their families saw travel as the key. Some used travel as the “proof', saying they were the only ones in their family to travel abroad for education and this was evidence of their higher motivation. Others saw trave1.as the motivation. In traveling they saw'other cultures, other 362 ways of doing things and this.made them want to do.something great for their country or achieve some personal goal. 88. (a) As compared with the rest of your family would you say that you are more or less ambitious than they are? Much more ambitious Somewhat more ambitious About the same in ambition llll Somewhat.less ambitious Much less ambitious (b) Rlease explain your answer ambition Relative to that of Family_. Number of Percent of C r d F 1 34mm _Jmnls—S 5 Much_more 22 22.9 4 Somewhat more 34 35.4 3 About the same 33 34.4 2 Somewhat less 4 4.2 1 Much less 0 0.0 No Response 3 3.1 Tgtal 96 , 100.0 Mean Ambition 3.80 Std. Dev. .86 I . f Increas D v tio . Agreement on.motivatien.was not nearly so high when the question was one of how desirable increased drive and ambition was. ‘Most respondents.felt it to be of some importance, but they did not agree on Just how much. The responses average about half-way between somewhat important and very important. 363 The question concerning the importance of increasing motiva- tion and the reSponses to it appear below. 89. As far as you are concerned, how important would it be to have more drive and ambition than you now have? Extremely important Very import ant Somewhat important Not very important I Unimportant c f Incre s n Dr v nd ti n Number of Percent of ImmEWf I as n Dr V W8 _9.IB2L_—3 ° 5 Extremely.important 23 .24.0 4 Very important 33 34.4 3 Somewhat important 21 21.9 2 Not very important 12 12.5 1 Unimportant 4 4.2 No Reaponse 3 3.1 Total 96 100.1, Mean Importance 3.63 Std. Dev. 1.12 ambition Relativg to That of Others. The last item in the.Motivation scale asked the individual how he felt his motivation compared to that of others. The item and the dis- tribution of responses to it appear below. Only one person in the entire sample considered himself to have less motivation than most people. Over 75 percent considered themselves to have more ambition than most people. 364 A comparison of the individual's ambition relative to that of his family with his ambition relative to that of others suggests a tendency for the respondents to view their families as more ambitious than most people. 90. Seme people are always trying to get ahead and improve themselves and others seem to be satis- fied with their present condition. As far as you yourself are concerned would you say that you are: Much more ambitious than most people Somewhat more ambitious than most people About the.smme in ambition as most people Somewhat less ambitious than most people Much less ambitious than most people ambition Relative to that of Othgrg fl Number of Percent of t C d s Rgsponsgs éfiggp;g___ 5 Much more ambitious than most people ' 28 29.2 4 Somewhat more ambitious than most people ' 45 46.9 3 About the same in ambition as most people 20 20.8 2 Somewhat less.ambitious than most people 0 0.0 l Mush less ambitious than most peeple l 1.0 No Response 2 2.1 Tetal 96 100.0 Mean Mbition 4.05 Std. Dev. .78 APPENDIX N IMPORTANCE AND RELATIVE IMPORTANCE RATINGS (Importance and.relative importance ratings are discussed on page :90) Importance and relative importance ratings were obtained by means of the following questions V 85. (a) Below are some major groups in India. Please indicate the importance of .333 group to you “.x by placing an in the appropriate column. 5- Extremely important 2 = Not very important 4 = Very important 1 a Unimportant 3 2 Somewhat important _§_._i_._3_._2_ .1. ....Besisn_ ‘ .___.___.._..___.___ ___énsial.21ass.____. ___ ___ ___.__. ___ iffzsmilge .__.__..__..___._._ W ___—___... ._;£snntr! .__.__. ___.__;..__ JLLQShsIAnssitvhw ... ___.__. ... ... "'"(b’ Indicate the.relative importance of each group to you by ranking.them. using a 5 for the most important. a 4 for the next most important. and so on. Place these ranks on the line at the left of each group. 365 366 Hm.a am.v A~.mv m Am.n V s mo.” mo.a Am.mv m Am.~ov om -.H mote Amqev e .m.e v a 5H.H o¢.m_ “a.mv m .o.omv mm wmé S.m 3.3 v 8.02 mm .o m ewes. moummum._m mmumuum. z flwmmmmomam “mammoem ez_mmflmmmaaddm H emoouo Hofiuom Hons: us eocouuomBH Am. m V m in. adv as Am. adv Ha .v.¢mv mm .m.mmv mu wdomnom m mmumuaqam worseeom m Amomdv .NH Am.o v m Aoovmv MN Amom v m Avom V m mauuumm._m mdfldMfldaM >H0> w |I||I[ } Ah.oov om sausage AH.NNV N Assent ouuuunsm Ao.mvv he . manage Am.o v 0 sense seduce .v.oav as cosmos .maquuem .m mmummmuam haoeouuxm m . lillll,‘ 367 mv.H mmom mm. Hm.H hN.H wh.m ¢H.H boom mood mm.~ in.oeV av Am.¢vV mv .m.va we “a.meV «a .e.meV He .86 a. sex. sol-0.3m m eucsuueofin eeoomoem acouuomaH e>«uoaem OZ .m.~ V s Am.o V o «o.H V a .v.m V m Am.mmV em .¢.m V m A~.v V o .H.~ V m Am.m V m A~.¢ V v .v.m V m .o.mHV ma .¢.oaV as .o.¢HV vs .e.n~V so av.oaV as .v.m V a as. oaV we .s.5HV ha .m.HHV as mmummum mm .mmemmum mm mmemuem mm matmmew am mommmwmom mommmummm. mmummmmsml. one: one: uses. Domed Bunsen awash ocooem H . u m to sodium Heaven some: we eossuuonEH sown-o3 - . .m.mmV «cod V H Ae.oHV we .H.m V m .H.~ V N ~n.~4uonVsuuasou eueoonsw.. saaeum oIflHU afldoem p cadmom.e