|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 SPANISH IN CONTACT WITH FRENCH AND ENGLISH IN MONTREAL: LEXICAL BORROWINGS, SEMANTIC EXTENSIONS, LOAN TRANSLATIONS AND MORPHOSYNTACTIC VARIATION By MARIA IDA FIONDA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSI TY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2008
2 2008 Maria Ida Fionda
3 To my parents: Franco Fionda and Cesira Bruno Fionda
4 ACKNOWLEDGMENTS This project was made possi ble thanks to the continued assistance and support of several people. I thank my supervisory committee chair, Dr. Gillian Lord, for her insightful commentary on the development of this thesis and most of all, for her excellent advice on some of the many di fficult decisions involved in the execution and writing of this project. I also thank my cochair, Dr. Hlne Blondeau, whose invaluable expertise in the field, continual encouragement and positive feedback on the topic drove the project from an idea to a r eality. I thank my mother, Cesira Bruno Fionda; and my father, Franco Fionda for their endless caring and encouragement. Without my father's weekly trips to stores and supermarkets to collect the periodicals needed, many times in the subzero freezing tempe ratures of Montreal winter, this project would never have been possible. His role in this investigation is priceless. My mother I thank for teaching me to never give up through the many challenges that arose during this project. I thank my partner, Vinodh Venkatesh, for his daily support and encouragement and the many times he said, "You can do it." Finally, I thank Luigi Valente, station manager and technical director of CFMB radio, for compiling a year's worth of daily radio shows, without which the oral component of this project would have been impossible.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ........... 4 LIST OF TABLES ................................ ................................ ................................ ...................... 7 LIST OF FIGURES ................................ ................................ ................................ .................... 8 ABSTRACT ................................ ................................ ................................ ............................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ............. 10 1.1 Introduction ................................ ................................ ................................ ............... 10 1.2 Hispanic Presence in Canada ................................ ................................ ..................... 14 1.2.1 Immigration to Ca nada ................................ ................................ .................. 14 1.2.2 Hispanic Immigration to the Province of Quebec ................................ ........... 21 1.2.3 Quebec Language Laws ................................ ................................ ................. 23 1.3 The Hispanic Population in Montreal ................................ ................................ ........ 25 2 THEORETICAL FRAMEWORK and review of the literature ................................ ........... 35 2.1 Linguistic Outcomes of Language Contact ................................ ................................ 35 2.2 Research ................................ ................................ ................................ .................... 46 2.2.1 Research on Spanish in Quebec ................................ ................................ ..... 46 2.2.2 Research on Trilingualism and Multilingualism ................................ ............. 47 2.3 Concepts and Definitions ................................ ................................ ........................... 49 3 METHODOLOGY ................................ ................................ ................................ ............ 51 3.1 Research Ob jectives ................................ ................................ ................................ .. 51 3.2 Sources ................................ ................................ ................................ ..................... 52 3.3 Data Selection ................................ ................................ ................................ ........... 53 3.3.1 Qualitative Analysis ................................ ................................ ...................... 53 3.3.2 Quantitative Data ................................ ................................ ........................... 55 3.4 Data Organization and Analysis ................................ ................................ ................ 56 3.4.1 Cross References ................................ ................................ ........................... 56 3.4.2 Spelling Variations ................................ ................................ ........................ 56 3.4.3 Categorization ................................ ................................ ............................... 57 3.5 Hypotheses ................................ ................................ ................................ ................ 59 4 RESULTS, ANALYSIS AND DISCUSSION ................................ ................................ .... 64 4.1 Collective Findings ................................ ................................ ................................ ... 64 4.2 Established Lexical Borrowings ................................ ................................ ................ 66
6 4.3 English origin Non established Items ................................ ................................ ........ 70 4.4 French origin Non established Items ................................ ................................ ......... 72 4.5 Semantic Extensions and Loan Translations ................................ .............................. 75 4.6 Morphosyntactic Innov ations ................................ ................................ .................... 82 4.7 Additional Changes/Processes ................................ ................................ ................... 83 4.8 Quantitative Analysis ................................ ................................ ................................ 94 5 CONCLUDING REMARKS ................................ ................................ ........................... 105 5.1 Summary of Findings ................................ ................................ .............................. 105 5.2 Limitations of the Study and Directions for Future Studies ................................ ...... 108 5.2.1 Limitations ................................ ................................ ................................ .. 108 5.2.2 Directions for Future Studies ................................ ................................ ....... 109 5 .3 Conclusions of the Study ................................ ................................ ......................... 111 APPENDIX: SAMPLE DATA NOT SELECTED FOR ANALYSIS ................................ ...... 114 LIST OF REFERENCES ................................ ................................ ................................ ........ 116 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ... 119
7 LIST OF TABLES Table page 1 1 Population in Canada: Spanish mother tongue ................................ ............................... 31 3 1 Periodical dates ................................ ................................ ................................ ............. 61 3 2 Radio recording dates ................................ ................................ ................................ .... 61 3 3 Competing variants ................................ ................................ ................................ ....... 62 3 4 Categories ................................ ................................ ................................ ..................... 63 4 1 Total data ................................ ................................ ................................ ...................... 95 4 2 Assignment of ranks ................................ ................................ ................................ ...... 96 4 4 English origin non established items ................................ ................................ ............. 98 4 5 French origin non established items ................................ ................................ .............. 99 4 6 Semantic extensions and loan translations ................................ ................................ ... 101 4 7 Morphosyntactic innovations ................................ ................................ ....................... 101 4 8 Additional changes/processes ................................ ................................ ...................... 102 4 9 Quantitative data ................................ ................................ ................................ ......... 104
8 LIST OF FIGURES Figure page 1 1 Mexican restaurant sign ................................ ................................ ................................ 28 1 2 Lawyers office s ign ................................ ................................ ................................ ...... 29 1 3 Alternative medicine store sign ................................ ................................ ...................... 29 1 4 Map of Canada ................................ ................................ ................................ .............. 30 1 5 Map of Qubec ................................ ................................ ................................ .............. 31 1 6 Map of Montreal ................................ ................................ ................................ ........... 32 1 7 Distribution of Latin American population, 2001 ................................ ........................... 33 1 8 Native language distribution in Montreal ................................ ................................ ....... 34 3 1 Analyzed bilingual advertisement 1 ................................ ................................ ............... 61 3 2 Analyzed bilingual a dvertisement 2 ................................ ................................ ............... 62 3 3 Data in an article ................................ ................................ ................................ ........... 62 A 1 Monolingual French advertisement not selected ................................ .......................... 114 A 2 Bilingual French/Spanish advertisement not selected ................................ ................... 114 A 3 Sample indiscernible origin ................................ ................................ ......................... 115
9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts SPANISH IN CONTACT WITH FRENCH AND ENGLISH IN MONTR EAL: LEXICAL BORROWINGS, SEMANTIC EXTENSIONS, LOAN TRANSLATIONS AND MORPHOSYNTACTIC VARIATION By Maria Ida Fionda December 2008 Chair: Gillian Lord Ward Cochair: Hlne Blondeau Major: Spanish While Spanish language contact phenomena have been the focal point of many linguistic studies in different parts of the world, virtually no studies have addressed this topic in Canada, where prominent Spanish speaking communities reside. Furthermore, few Spanish linguistic studies address contact between languages in a setting where three codes are present. This study breaks new ground in the field of Spanish language contact studies by investigating the Spanish language in Montreal, Canada through Spanish language periodicals and radio shows in the Montreal Hispani c community. It looks for linguistic evidence of Montreal Spanish in contact with French and English, provides an inventory, proposes categorizations and analyzes the distribution patterns of the data. Furthermore, it investigates which of the two language s, French or English, exercises greater influence on Spanish. The results confirm that evidence of contact exists in the form of lexical borrowings, non established lexical items, semantic extensions, loan translations and morphosyntactic variation. In add ition, other innovations displaying varying processes and/or changes are also found. Finally, the analysis suggests that the dominating language of influence upon Montreal Spanish is French.
10 CHAPTER 1 INTRODUCTION 1.1 Introduction 1.1.1 Spanish in Montreal When w alking the streets of the Villeray St. Michel Parc Extension borough in Montreal, the Hispanic business signs display a fascinating array of linguistic combinations that characterize the community's contact with the city's Francophone and Anglophone commun ities. Figures 1 1 through 1 3 were taken on Chateaubriand and Blanger, two streets in the borough, and exemplify the written evidence of Spanish language contact in the city. Figure 1 1 depicts the signs and menu of the Mexican restaurant El Sombrero. Of particular interest is the chalkboard stating terraza ouvert as the Spanish terraza terrace' and the French ouvert open' have been combined into one single phrase. Figure 1 2 is a photo depicting the signs outside a lawyer's office and is exemplary the three languages coming into contact. Card' is expressed in English, French carte and Spanish tarjeta while other words such as mandats and assermentation seem to be expressed solely in French 1 and yet others such as poderes notariales only in Spanish. Figure 1 3 further illustrates the use of two languages in the same phrase. This time botanica used to denote an alternative medicine store, is written in Spanish while varit variety' is written in French, as seen by its two accents. Linguistic combi nations such as the ones expressed in these signs are not only evidence of language contact but also of linguistic creativity and ingenuity on the part of the speakers, as they incorporate elements from other languages into their native language and still maintain 1 The word "seem" is being used here to reflect t he fact that further linguistic analysis is needed before it can be determined whether these words are indeed being expressed in French or have been integrated as borrowings into the Spanish of Montreal.
11 fluidity and intelligibility in their communication. Such innovations are the focal point of this investigation. 1.1.2 Goal of the study Spanish language contact phenomena have been the focal point of many linguistic studies in the United States, Latin America, Europe and Africa; however, virtually no studies have addressed this topic in Canada, where prominent Spanish speaking communities exist As such, this study analyzes Spanish language contact in Montreal, Canada, in an attempt to discover the lin guistic innovations in a geographical area of Spanish contact of which very little is known thus far. When languages come into contact, the linguistic outcomes may be evidenced in one or more of the languages involved. However, as the dominant language(s) in the contact area are normally the one(s) that exercise most of the influence this study deals exclusively with the innovations appearing in the Spanish language, as it is the minority language and, consequently, the one that is likely to experience chan ges (Thomason, 1995). The numerous studies addressing this matter in other geographical areas have come a long way in documenting evidence of linguistic contact between Spanish as the first language (henceforth, L1) and a second language (henceforth, L2). They owe their success in part to large communities of bilingual speakers that exemplify such contact. For example, the United States features, among others, established communities of Mexicans, Cubans and Puerto Ricans. These have served to explain some of the linguistic changes that take place in Spanish English bilingual speech resulting partly from the imposition of the English language on long standing Spanish speaking communities (Gonzlez, 2001). In Spain, such studies are possible due to the contac t between Spanish and other languages spoken in the country such as Basque, Catalan, Galician and Valencian, and also due to contact with Arabic, spoken by the Moroccan refugees in
12 constant migration to the country. Furthermore, various language contact st udies have been conducted in the Americas for example Spanish Quechua contact in Peru and Spanish Portuguese contact along the border between Uruguay and Brazil and Africa, where contact is found in Spanish with Arabic in northern Morrocco and Spanish with aboriginal languages in Equatorial Guinea. In Asia, similar studies analyze the contact between Spanish and aboriginal languages in the Philippines. However, as stated above, there has been virtually no analysis of Spanish contact in Canada, though the c ountry has seen its fair share of Hispanic immigrants. Despite the fact that its waves of immigration are considerably reduced when paralleled with the United States, various metropolitan areas such as Vancouver, Toronto and Montreal as well as many small er cities are home to established Hispanic communities, flourishing in their hybrid Spanish Canadian or Latin American Canadian cultures. These geographical locations serve as optimal areas to investigate different facets of contact language research. Mon treal was chosen as the focus of the study for a number of reasons. In the first place, a geographical area as big as this city is likely to provide larger numbers of established Hispanic immigrants than the smaller cities, which translates into a more tig htly knit community and, subsequently, a greater chance that the native language will be maintained. This dynamic of language change and language maintenance is important when studying language contact, as the majority of linguistic innovations that typify a community of language contact will arise with prolonged contact between the languages, and not with the limited contact exemplified in short term cases where the L1 is usurped by the L2 (Sankoff, 2001) Furthermore, the city also lends itself well to th is study because immigrants are likely to come in contact with native speakers of Canada's two official languages that comprise the majority of the population: French and
13 English. Studies in three code Spanish language contact are scarce and as such, this investigation into Montreal's linguistic interactions is welcome on many fronts. In short, the purpose of the investigation is to further the knowledge acquired in studies that have been hitherto conducted on Spanish language contact by finding and docum enting linguistic innovations, trends and patterns in a relatively untouched geographical location of Spanish contact studies. Additionally, it seeks to gain new insight and contribute to the field of trilingual language contact. This last goal is of prime importance because it tackles the difficult task of identifying just which components of language are susceptible to influence from language contact and how this influence is exercised. By having two languages of differing grammar in contact with Spanish, as is the case in Montreal, the study is able to provide a comparison between the influence exercised from a grammatically similar language, as is the case with French and Spanish, and a dissimilar one, as occurs with English. Apart from the usual categ orizations of linguistic outcomes of language contact, the presence of the three codes in this particular study calls for further categorization when documenting which language effects a greater influence on Spanish French or English and on what level this influence is exercised, i.e. lexical, morphological, syntactical, etc. Given that the province of Qubec, of which Montreal is the largest city, has undergone various changes in language laws, different groups of immigrants likely had different amounts of exposure to either French or English, depending on when they arrived and the state of the language laws at the time. The data used for this analysis were collected from daily radio recordings of Montreal's ethnic radio station as well as from two local newspapers. These newspapers can be found in grocery stores and any Latin American shop and cater mainly to a middle class reader. While
14 spontaneous spoken speech is the common source of data for a study of linguistic evidence of language contact (Labov, 1981), the sources chosen for this study will prove to be more than appropriate for its purpose, as the implicit audience for both the periodicals and the radio shows is the Hispanic community at large. As such, it is being proposed that the linguistic inn ovations that make it past the medium of spontaneous speech and establish a presence either in formal speech or in the written medium likely typify, in part, how the community speaks. Seeing that this type of study of the community is considerably new, the present investigation has chosen a qualitative rather than quantitative methodology, in order to provide a panoramic view of linguistic innovations in the Spanish of this particular area. It cannot claim to be exhaustive given that new words, phrases and, possibly, grammatical elements are constantly being created, but as a first look at the linguistic picture of Spanish in Montreal, can provide an ample, descriptive analysis as a foundation for future studies. Before examining linguistic evidence of thi s contact variety of Spanish, it is necessary to examine briefly the immigration of Hispanics and the establishment of Hispanic communities in Canada and, more specifically, in Quebec and Montreal. Although in many ways the Hispanic ethnic picture in Canad a resembles that of the United States in terms of origins and historical reasons for flight, there are also some interesting differences worth reviewing, as these are the basis for the extensive research that has been conducted in Spanish/English bilingual ism in the U.S. versus the scarce research in Spanish/English/French bilingualism or trilingualism in Canada. 1.2 Hispanic Presence in Canada 1.2.1 Immigration to Canada Immigration to Canada in has been shaped by various conditions, some of which include chan ges in the country's immigration policy, the socio economic and political circumstances in
15 the immigrants' country of origin, and the state of Canada's economy and its needs. Over time, immigration patterns have shown fluctuations both in number of immigra nts and in their origin and class. In the 20 th and 21 st centuries, immigration in the Hispanic community has gained significance in Canada's ethno cultural blueprint, due to increasing numbers as well as the establishment of Hispanic communities in the cen sus metropolitan areas and in other small cities. Like the United States, Canada's Hispanic community consists of immigrants from both Latin America and Spain, though Latin American immigration is primarily responsible for the number of native Spanish sp eakers in Canada today. Mata (1985) outlines four main waves of Latin American immigration to Canada between 1965 and 1983: the Lead wave, the Andean wave, the Coup wave and the Central American Wave. The waves did not occur in discrete time periods, but r ather overlapped during the two decades in question. The Lead wave occurred from the late 1950s to the early 1960s and consisted of immigrants who were largely Latin American of recent German, Italian, Dutch, Polish, English, Jewish and Central European descent and who had lived in Latin America for some time. They arrived mainly from Argentina, but also from Mexico, Venezuela and Uruguay and comprised urban intelligentsia and blue collar workers. The Andean wave took place in the early to mid 1970s and consisted mainly of large numbers of skilled and unskilled laborers from Ecuador and Colombia. This wave was a result in large part of the 1973 Amnesty, which allowed the refugees to regularize their status in Canada. Overlapping with the end of the Andean wave was the Coup wave, which started in the early to mid 1970s and was spurred by the 1973 Pinochet coup, though the refugees were not only Chilean but also Uruguayan and Argentinian. They were a mixture of the intelligentsia, viewed as a threat to the c onservatives, and white and blue collar workers searching for a better life. Around 1980 began the Central American wave, a result of
16 the political turmoil propelled by the victory of the Sandinista revolution, in 1979. The immigrants were mainly Nicaragua ns, Salvadorans and Guatemalans, both skilled and unskilled persons from the rural middle class, the urban poor and the peasantry. This wave continues to bring Central American immigrants to Canada every year. In addition to the four aforementioned waves, Veronis (2006) identifies a fifth wave of Latin American immigrants that began in the 1990s and continues to the present day. Sometimes referred to as the IT wave (Veronis, 2006:19, footnote), these immigrants are primarily professionals from throughout La tin America and come in under Canada's 'skilled worker' and to a lesser extent 'business' classes of immigration. According to Statistics Canada, the majority of Hispanics in the country is of Spanish, Salvadoran, Mexican, Colombian, Chilean and Peruv ian and, to a lesser degree, Latin, Central or South American and Guatemalan origin (Statistics Canada, 2006 Census, Ethnic Origin). The designations "Spanish" and "Latin, Central or South American" are somewhat problematic in that the former is subject to the responders' interpretation of their ancestry and the latter does not specify the actual country of origin. Nonetheless, the origins of the majority of the Hispanic immigrants and their children reflect the waves described above. What follows is a brief overview of the main Hispanic groups in Canada according to the information provided by the website Multicultural Canada (www.multiculturalcanada.ca). Taking a closer look at some of the prominent groups of immigrants reveals the forces that brought this community into being. Argentinians Argentinians' immigration to Canada can be traced to the beginning of the 20 th century and came about primarily due to political and economic reasons. Most of the Argentinian immigrants were intelligentsia or highl y trained individuals of direct European descent, a reflection of Canada's 1952 Immigration Act, which covertly favoured white
17 immigrants and discriminated against many minority groups. Between 1964 and 1972, the growth in Canada's economy coincided with t he decline in Argentina's, thereby further fostering an influx of immigrants to Canada. The numbers increased again between the years of 1973 and 1983 thanks to the military overthrow of Pern's government and resulting political instability and oppression a high level of inflation and terrorism continued to contribute to the country's emigration. By 1983, democracy was reinstated in Argentina and immigration levels dropped; some Argentinians even returned to their country. Of those that remained, 2800 peo ple (more than half) settled in the province of Ontario, mainly Toronto, about 1520 settled in Quebec, mainly in Montreal, and the rest settled in British Columbia (Vancouver) and Alberta (Edmonton and Calgary). Chileans As discussed previously, Pinoche t's 1973 coup was the spark that sent many Chileans fleeing the torture, repression and death in their country, though a small number of immigrants that objected to President Allende's socialist government already resided in Canada. Multicultural Canada ou tlines four phases of Chilean immigration to Canada between 1974 and 1992 that show a fluctuating number of arrivals. It was during the violence of the 70s that Canada created an immigration program that facilitated the immigration of Chileans for politica l persecution. The 1991 Canadian census statistics reveal that Quebec had the highest percentage of Chileans at 36.3% while Ontario had the second to highest at 28.4%, followed by Alberta and British Columbia. According to Grmela (1991), they have achieved the highest level of economic integration of all Latin American immigrants (273). Ecuadorians Ecuadorians' earliest phase of immigration to Canada can be traced back to the 1950s and 1960s from the highlands of Azuay due to an economic crisis in the a rea. During the 1960s and 1970s thousands of Ecuadorians with education and money arrived for financial
18 reasons, due to a lack of economic opportunities in their homeland. 1975 showed the highest volume due to a relatively open immigration policy that, whi le favouring educated and skilled people, eliminated the racial categories that had until then discriminated against certain ethnic origins. In contrast to the situation in the United States, where many of the Ecuadorians have left to go back to their coun try, many Ecuadorians that came to Canada have remained. Ontario is home to the majority of the Ecuadorians in Canada, followed by Quebec, Alberta and British Columbia. Guatemalans. Guatemalans left their country on account of the political violence from the genocidal wave of state terrorist activity in the late 1970s and then due to the resulting economic crisis that followed. The increase in the number of Guatemalans to Canada was facilitated by the following: First, Canada's 1976 Immigration Act now inc luded refugee' as a category for immigration, second, Guatemalans could receive this status in the Canadian embassy in Guatemala and third, the United States tended to reject refugee claims from Guatemala owing to its support of the Guatemalan government. The 1991 census indicated that Quebec had the highest concentration of Guatemalans, followed by Ontario, British Columbia and Alberta. These immigrants were white collar as well as blue collar workers. Mexicans Mexicans have a short migration history t o Canada, and come mainly from middle and upper class backgrounds, either students or professionals. As such, they do not live in segregated or concentrated enclaves and the majority arrives as legal immigrants. Furthermore, the Mexicans immigrating to Can ada do so to improve an already decent economic position, a very different situation than is seen with in the U.S. There are four main groups of Mexicans in Canada: 1) middle and upper class professionals with children who came for economic reasons; 2) mid dle class individuals with professional degrees, technical training or certification who do
19 not have children and do not speak much French or English; 3) Mexican Mennonites 2 ; 4) spouses of Canadian citizens. While the immigrant numbers were relatively low in the 1950s and 1960s, they did see a sharp increase in the early to mid 1970s and then a rise again in the late 1980s. The largest concentrations of Mexicans are in Ontario, British Columbia, Quebec, Alberta and Manitoba. Nicaraguans. As discussed in t he Introduction, Nicaragua's violent political history has caused many of its citizens to flee in search of a better life. Specifically, the four principle reasons for Nicaraguan immigration to Canada are: 1) the economic deterioration caused by the Contra war; 2) the United States' embargo against Nicaragua; 3) the discontent with the way the Sandinista government chose to run the country and 4) the drafters wishing to escape compulsory military service. Most Nicaraguans arrived to Canada in the 1980s via a third country such as Honduras, Costa Rica or the United States. Again, the Immigration Act of 1976 contributed to make Canada inviting. In Canada, the Nicaraguan population is most numerous in Ontario, followed by Alberta, Quebec and British Columbia. Peruvians Peruvian immigration to Canada began with a small group of Euro Peruvians in the late 1960s that came about because of Canadian economic expansion and the need for professionals and skilled workers. In the late 1970s, violence between leftist te rrorist groups and right wing governments in Peru led to the continuing poverty and guerrilla war that lasted nearly ten years and was responsible for much of the immigration of Peruvians in the 1980s. In the 1970s, immigration began by independent applica tions and was later followed by sponsorship of spouses and family members. Ontario is home to the largest Peruvian community in Canada, followed by Quebec and British Columbia. 2 The Mennonites migrated to northern Mexico from Ca nada in the 1920s. Most of the immigrants in this group are the Mexican born children of the ones that left. They tend to be less well educated and hold agricultural occupations.
20 Salvadorians. Salvadorians are among the newest immigrants to Canada. As attes ted earlier, the civil war of 1979 to 1992 was the biggest cause for migration to Canada, bringing in thousands of refugees. Several changes in the Canadian immigration policy served to facilitate this migration and that of other immigrant group of the sam e time. Once more, it is necessary to mention the Immigration Act of 1976; however, also worthy of mention are the implementation of special measures that allowed permanent resident status on humanitarian grounds for those who had relatives in Canada, the newer and more proper interpretation of "well founded fear of persecution" in the policy, the establishment of a subclass of Salvadorian refugees in 1983 that allowed them to obtain refugee status in El Salvador and the cessation of deportation of refused applicants in 1993. Immigrants vary between skilled and unskilled workers in the white collar and labour force classes, with the majority inhabiting Ontario. Quebec, Alberta and British Columbia are the other three provinces that house the majority of the remainder. Spaniards Spanish presence in Canada can be traced back to the 16 th century with Basque fishermen and whalers in Newfoundland. There was also a small presence at a military post on Vancouver Island in the late 1700s but the residents chose not to inhabit the island. Modern Spanish immigration, however, occurred in the 20 th century and can be divided into three periods. The first period consists of the years until 1957 in which the arrivals were in relatively small numbers. The second period beg an in 1957, the year when Spain and Canada signed an agreement to bring Spaniards to Canada. This resulted mainly because, while Spain was concerned about excess population (especially from the countryside), Canada was seeking immigrants. The agreement, wh ich ended in 1960, led to three expeditions: 1) 125 couples from all regions of Spain; 2) 98 single men, mostly farmers from Navarra and 3) 50 single women to work as domestic servants. Finally, the third period consists of a steady but small flow of
21 immig rants after 1960. The distribution of Spaniards in Canada is divided almost equally between Ontario and Quebec. The following section addresses Hispanic immigration to the province of Qubec in order to clarify why Montreal, the largest city in the provin ce, is the ideal setting for this investigation. 1.2.2 Hispanic Immigration to the Province of Quebec Figure 1 4 presents a map of Canada that will assist in locating the different provinces that will be referred to within their national framework. Accord ing to Boyd & Vickers (2000), for most of the 20 th century, most immigrants settled in the provinces of Ontario, British Columbia and Quebec. The authors note that, "since the 1940s, a disproportionate share has lived in Ontario and the percentage has cont inued to rise over time. By 1996, 55% of all immigrants to Canada lived in Ontario, compared with 18% in British Columbia and 13% in Quebec" (2000:10). The remaining 14% was primarily concentrated in Alberta and Manitoba. The difference between Ontario and Quebec's immigrant population was 42%. While Ontario's immigrant population did greatly outnumber that of the other provinces, Quebec nonetheless boasted a noteworthy immigrant population. T he 2006 census confirms that Quebec has the second largest popul ation of Hispanics in Canada. Table 1 1, which shows the country's distribution of the population whose mother tongue is Spanish, shows that the number of residents in Quebec whose native language was Spanish was 108,790. With the second largest population of native Spanish speakers in Canada, Quebec's numbers display a 14% difference with Ontario, a much smaller number than the previously noted 42% difference in total immigrant population. Figure 1 5 is a map of the southern part of the province of Qube c, displaying the main cities that are home to native Spanish speakers. The census metropolitan areas in the province are Saguenay, Qubec, Sherbrooke, Trois Rivires, Montreal and Ottawa Gatineau. With a Spanish
22 speaking population of 90,105, Montreal co ntains the greatest number of native Spanish speakers in Quebec. The number of such speakers in the remainder of the cities in Qubec is far from approaching that of Montreal: Ottawa Gatineau (a city on the border of Qubec and Ontario) has 12,710, the cit y of Qubec has 4410, Sherbrooke has 2305, Trois Rivires has 665 and Saguenay has 380 (Statistics Canada, 2006 census). Clearly, Montreal has the biggest and most established Hispanic community in Qubec. Boyd & Vickers (2000) note that: Most immigrants l ive in Canada's big cities, with the largest numbers concentrated in the census metropolitan areas (CMAs) of Toronto, Montral and Vancouver Proportionally more immigrants than Canadian born have preferred to settle in urban areas, attracted by economic o pportunities and by the presence of other immigrants from the same countries or regions of the world (9 10). The (Quebec) Institut de la Satistique 's 2007 list of the 15 top countries of birth of Qubec's immigrants shows that Colombia occupies the fourt h place on the list; Mexico occupies the eighth and Peru the twelfth. These numbers are telling, as they are evidence that Hispanic immigration to the province is very prominent. However, since the language situation in Quebec is particular as compared to the rest of Canada, immigrants to the province have been exposed to different amounts of French and English, depending on their time of arrival, as discussed in the following section. According to these figures, the city of Montreal is the ideal place fo r this study. In the first place, it is located in a province that features the second largest Hispanic population in the Canada, which assures that the community is among the most prominent in the country. Secondly, this is supported by the fact that the city claims the largest Hispanic community in the province, thereby confirming its position as home to a well established community. Thirdly, as the city is located in a predominantly French province but also boasts an important Anglophone community, it pr ovides the added advantage of two languages, French and English, with which
23 the Hispanic community is in contact. In the following section the history of the province's language laws is outlined in order to demonstrate their effect on the amount of contact that Hispanic people have had with either language. 1.2.3 Quebec Language Laws Over the last 40 years, Quebec has witnessed a turbulent history of language laws that have caused conflict between its Anglophone and Francophone communities and created hurd les for the ethnic communities. A series of bills have been passed over the years in an attempt to simultaneously conserve the French language in Quebec and protect minority rights; however, all bills have affected the province's ethnic communities, who be long to neither the Anglophone or Francophone communities. Bill 63 was issued in 1969 to guarantee Quebec residents the right to choose the language of instruction for their children. It promoted the teaching of French in English schools and made French cl asses available to immigrants. Many people from the Francophone community were displeased, apparently wanting the freedom of choice for instruction to be eliminated altogether and all immigrant children to be forced to attend French schools (Blanger, 2000 :para.1). The consequence of the bill was increased allophone 3 integration to the Anglophone community, thus threatening the position of French in the province. In 1974 the National Assembly of Quebec adopted Bill 22. This bill proclaimed that French wa s the official language of Quebec, set up a board to supervise its implementation, forced all public institutions to address the public administration in French, required corporations to change their name to French, declared French the official language of contracts, had corporations advertise primarily in French and required them to obtain a certificate of 3 I n Canada, allophone means a person whose first language is neither of Cana da's official languages of English and French.
24 francization that declared their competency to function in French. While freedom of choice for the instruction of children was maintained, children had to prove through a test that they had previous knowledge of English. The bill also preserved English by maintaining the English language sector of education and allowing contracts to be available in English if both parties agreed (Blanger, 2000:para. 2). Again, there were protests from both the Fancophone and Anglophone communities, the former arguing that insufficient steps had been taken to protect and promote the French language and the latter arguing that the bill went too far. In 1977, Bill 101 was passed, making French the official language of Quebec in all aspects. English education was to be restricted mostly to those already in the system, their siblings, those temporarily posted in Quebec or whose parents had themselves received an English elem entary education in the province" (Blanger, 2000:para. 3) and francization programmes were only required for international and national corporations of over 50 employees. While provincial laws were to be ratified in French, an English translation was also to be made available. Outside of the three aforementioned aspects, French was to be the language of use in all facets of daily life in the province. Since then, some changes have been made to the bill to rectify violations of the Constitution (now called the Canadian Charter of Rights), others willingly such as the incorporation of English in commercial signs so long as French is given priority; or access to English schools by all who received English education in Canada. Nonetheless, it goes without sayin g that these bills, and specifically Bill 101, have radically altered the linguistic reality of Quebec's residents. Among those affected by these policies were, of course, the immigrants that came to the province during these years. Those immigrants that arrived in earlier, such as the 1950s and 1960s, did not experience this linguistic conflict in its full explosion at that time. Conversely, the
25 immigrants that arrived in the 1970s onward found themselves in the eye of the storm and subject to the laws th at were beginning to form the linguistic picture that is now the province of Qubec. While those first immigrants would have likely received their education in English and had a relatively equal amount of contact with English and French, the ones who arriv ed in the 1970s onward have probably had a greater amount of contact with French, including their education. Consequently, the earlier immigrants' contact with a more evenly distributed amount of English and French than the later immigrants' probably led t o a higher level of linguistic heterogeneity in the Spanish of the Hispanic communities in Quebec. That is, the earlier immigrants' speech is likely to display more influence from English than the later immigrants'. Statistics Canada's 2006 census reveals that, of Quebec's census metropolitan areas mentioned above, Ottawa Gatineau is the only city in which native Anglophones outnumber native Francophones. Of the remaining cities, Montreal has the closest ratio of Anglophones to Francophones. With a ration o f 2 Anglophones to 11 Francophones (2:11) immigrants' chances of contact with Anglophones in Quebec are highest in Montreal. This fact, coupled with the likelihood that different sets of Hispanic immigrants have experienced diverse linguistic realities in accordance with their time of arrival and with the inherent diversity of the community due to its members' national multiplicity, most likely yields a fairly diversified Spanish variety. The following two sections take a closer look at the Hispanic communi ty in Montreal and exemplify the Spanish French English dynamic in the city. 1.3 The Hispanic Population in Montreal Montreal is a metropolitan city in the province of Qubec with a population of over 3,350,000 people in the Greater Montreal Area and ove r 1,850,000 people living on the island itself (Statistics Canada, 2006 Census). The city is a major geographical area of settlement for immigrants from many different parts of the world. As stated in the previous section, it is the
26 main point of settlemen t for Hispanic immigrants in the province as it offers the best possibility for social and economic integration. It is the place with the best opportunity to find employment a main source of concern for newly arrived immigrants of lower social classes and offers organized groups composed of fellow Hispanic immigrants for moral support. It has an established and vibrant Hispanic community that offers its immigrants and their children a great variety of Hispanic culture: food, books, periodicals, music, resta urants, radio shows, television channels, festivals, parties and businesses and services in all fields (real estate, accounting, medicine, esthetics, automobile industry, etc.). Figure 1 6 is a map of the island of Montreal, divided into its current boro ughs. Groups of immigrants from many countries have established communities in the city and are concentrated in specific boroughs, or areas in one or more of the boroughs. Immigration is a very important factor of the city's growth. In 2001 there were appr oximately 491,000 immigrants in the city, close to 28% of its population. The 2006 census showed an increase to more than 560,000 immigrants, close to 31% of the total population (Institut de la Statistique, Population et dmographie, 2007) The Hispanic p opulation is an important part of these numbers due to the prominence of the Spanish language. Although the Latin American countries and Spain do not figure into the top five countries of birth of immigrants in the city, Spanish is the third most popular l anguage spoken at home. French (53%) and English (24%) make up 77% of population that speaks them at home; the non official languages make up 19% and approximately 5% speak more than one language at home (Insitut de la Satistique Population et dmographie 2007). Of the non official languages spoken at home, Spanish is the most popular, with 2.4% of the population speaking it at home, followed by Italian (2.3%), Chinese languages (2.2%) and Arab languages (2.2%).
27 Figure 1 7 shows the Latin American popula tion distribution in the city in 2001. Percentages are given relative to the minority population in the city. Unfortunately, these statistics are not available for the population of Spaniards. Certainly, the Hispanic population has grown in the last seven years since the publication of these numbers, and with it, the use of the language. As is evident from the darker gray coloring, most of the population is concentrated in the center to northeastern parts of the city. These are popular areas of settlement f or new immigrants. The bigg est concentrations are found in Plateau Mont Royal and Rosemont La Petite Patrie, where they constitute between 36% and 89% of the visible minority populations, though Hispanics are dispersed almost everywhere in the city. It is necessary to highlight how diffused the Latin American and Spanish community in Montreal is. The city's 90,105 native Spanish speakers translate into approximately 2.6 % of the population. While this figure may seem low, it constitutes close to 12% of 2 1% of the speakers of native languages other than French or English. When one considers the magnitude and diversity of immigration to Montreal, it is evident in figure 1 8 that Spanish speakers are an important slice in the population pie of native speaker s of languages other than French or English. This chapter has provided an introduction to the Hispanic community and its language in Montreal. It has shown that Hispanic immigrants have been arriving to the city for almost half a century, during which they have established a widespread and tight knit community. A community of this significance undoubtedly displays remarkable evidence of language contact with the Francophone and Anglophone communities. In the following chapter the theoretical framework for t his study and some research on Spanish in contact is presented in order to provide a solid basis for the development of the investigation.
28 Figure 1 1. Mexican restaurant sign
29 Figure 1 2. Lawyer's office sign Figure 1 3. Alternative medicine store sign
30 Figure 1 4. Map of Canada; Source: Wikiped ia ( http://en.wikipedia.org/wiki/Image:Map_Canada_political geo.png )
31 Table 1 1. Population in Canada: Spanish mother ton gue Number of Spanish mother tongue speakers Percentage of total Spanish mother tongue speakers in Canada Canada Total 345,345 100.00 Ontario 160,275 46.41 Qubec 108,790 31.50 British Columbia 34,075 9.87 Alberta 29,125 8.43 Manitoba 6,850 1.98 Saskatchewan 2,735 0.79 Nova Scotia 1,305 0.38 New Brunswick 1,040 0.30 Newfoundland and Labrador 670 0.20 Prince Edward Island 220 0.06 Yukon Territory 130 0.04 Northwest Territories 90 0.03 Nunavut 30 0.01 Source: Statistics Canada, 2006 Census ( http://www40.statcan.ca/l01/cst01/demo11a.htm ) Figure 1 5. Map of Qubec; Source: Investissement Qubec ( http://www.inv estquebec.com/en/index.aspx?page=1542 )
32 Figure 1 6. Map of Montreal; Source: Le portail official: Ville de Montral (http://ville.montreal.qc.ca)
33 Figure 1 7. Distribution of Latin American Population, 2001; Source: Le portail official: Ville de Mont ral (http://ville.montreal.qc.ca)
34 Figure 1 8. Native language distribution in Montreal
35 CHAPTER 2 THEORETICAL FRAMEWOR K AND REVIEW OF THE LITERATURE 2.1 Linguistic Outcomes of Language Contact When languages come in contact, bilingual speakers dev elop strategies to lighten the cognitive load of having to use two linguistic systems. According to Sankoff (2001), linguistic outcomes of language contact have been reported in the levels of phonology, lexicon, syntax and discourse pragmatics, and morphol ogy and grammatical categories (5). At the lexical level, the main linguistic outcome of language contact is evident in the form of borrowing most of which has taken the form of established loanwords. Poplack & Meechan (1998) confirm that those most like ly to be borrowed are "major class content words such as nouns, verbs and adjectives" (127). Nonce borrowings, not widespread or attested, have also been reported as a lexical outcome of language contact. This has sparked some debate on the differentiation between borrowing and code switching, which will be discussed further below. As this study consists of mainly written material, the nature of the data will likely lead towards a focus on the lexical level. This is because innovations and changes on this l evel are the first to occur and thus the first to be documented in writing, whereas phonological and morphosyntactical changes are more visible in spontaneous speech. At the morphosyntactic level, the research shows a different kind of debate. While the qu estion at the lexical level is centered on identifying what type of change is occurring (a switch in code or a borrowing), the main concern at morphosyntactic level deals with the language internal constraints and social influences on language change. Thou gh they do not discount the role of internal constraints, Thomason & Kaufman (1988) claim that the primary determinant of the linguistic outcome of language contact is the sociolinguistic history of the speakers (35). Furthermore, the authors use this basi s not only to explain language change but also to make
36 predictions (albeit limited), something they claim is lacking in the research. Depending on the direction in which the influence is taking place, Thomason & Kaufman outline two ways that lead to contac t induced linguistic changes: borrowing and interference through shift (37). They do not limit the term borrowing to lexical items alone but rather extend it to all elements from the L2 being incorporated into the L1. Interference through shift involves th e elements from the L1 influencing the L2. The two types of outcomes, however, may occur simultaneously. Borrowing is the outcome on which this study will be centered, since the analysis will be done on the L1 (Spanish). Along with internal constraints su ch as markers and typological similarity, these authors name one social influence upon which predictions can be made: intensity of contact. They predict that light contact will cause the borrowing of only basic vocabulary while intensive contact will cause greater lexical borrowing and moderate to heavy structural borrowing (phonology, morphosyntax), which may lead to massive grammatical replacement or, in extreme cases, language death. This prediction caries two implications: first, internal constraints ar e only a secondary factor in contact induced morphosynactic innovations; second, given enough social pressure (external force) a change in the language may occur at all levels. Thomason & Kaufman's notion of intensity of contact plays an important role in the Hispanic community in Montreal, specifically in the changes it has undergone over the course of the last 40 years. In the 1960s, when the community was smaller, the intensity of contact between Spanish and French or English was probably much higher tha n it is today. With the immigrant numbers and the community as a whole being younger and smaller, a greater amount of contact was present with the Francophone and Anglophone communities. However, the passing of the years has introduced larger numbers of im migrants and a growing community
37 whose members now own businesses and provide every possible service in Spanish. A bigger, more tight knit and self sufficient community now translates into less intense contact with the exterior Francophone and Anglophone c ommunities. On the other hand, it must also be noted that the members of the second and third generations are now receiving education in French, which exposes them to a level of contact that their parents and grandparents did not have. With these factors t aken into consideration, the Montreal Hispanic community as a whole is probably located in the middle of the intensity of contact scale. The amount of borrowing (of all language elements) to be expected from this situation is equally moderate, with more in the lexical area and little in the morphosyntax area. Sankoff (2001) believes that language internal constraints carry more weight than Thomason & Kaufman account for in their predictions and that language internal changes due to contact are difficult to come by. The author does not deny the role of social factors and sociolinguistic history of the speakers in the outcomes of language contact; however, in stating that morphology and syntax are the areas least likely to be influenced by language contact, sh e also claims that, "linguistic structure overwhelmingly conditions the linguistic outcomes" (19). This means that the structure of a language imposes certain constraints that make it difficult to yield to structural change, regardless of the intensity and type of contact. Moreover, Sankoff (2001) suggests that the specific types of socio historical situations that have produced diverse linguistic outcomes should also be explored, along with "inter individual [variability] within bilingual communities" ( 4). In other words, contact situations of social inequality such as war, conquests, colonialism, slavery and migrations, as well as those of equality such as urbanization or trade, must be investigated in order to determine their influence on linguistic ou tcomes. The author distinguishes between two major social processes: conquest
38 and immigration. She argues that conquests typically lead to slow language shift and several generations of bilinguals, which in turn lead to substratum influence this last term carrying the same meaning as Thomason & Kaufman's (1988) interference through shift Furthermore, this type of stable bilingualism is most likely to lead to structural convergence. Conversely, contact situations due to immigration lead to rapid linguisti c assimilation of newcomers due to a short duration of contact, which in turn leads to borrowing rather than substratum influence Although Sankoff does mention that some structural changes have been documented in immigrant communities that have been exist ing for many generations, the immigrant language varieties are short lived and the long term effects modest (5). In addition, the author suggests a heightened focus on the individual speaker's practices and strategies as these eventually lead to a change i n the community's practices and strategies. Given that Montreal's Hispanic community has come to be due to immigration, Sankoff's (2001) theory confirms that the language contact effect to be expected (and researched) in this community is that of borrowing from French or English into the Spanish language, as was also previously determined with Thomason & Kaufman's (1988) theory. Furthermore, structural changes will be very scarce, if existing at all. Silva Corvaln (1994) theory complements both above ment ioned theories. She believes that external influence does not radically change the structure of the L1 by introducing a new structure from the L2. Rather, the permeability of a language to external influence depends on the existence of superficially parall el structures in the L1 and L2. Generally, the author finds that any language contact outcome at the syntactic level is an existing intra linguistic process that was merely accelerated by language contact, otherwise known as the snowball effect.
39 In her b ook, Language Contact and Change Silva Corvaln (1994) outlines five strategies used by bilinguals to lighten the cognitive load of using two systems: 1) simplification of grammatical categories and lexical oppositions, 2) overgeneralization of forms, fr equently following a regularization pattern, 3) development of periphrastic constructions either to achieve paradigmatic regularity or to replace less semantically transparent bound morphemes, 4) direct and indirect transfer of forms from the superordinate language and 5) code switching, which involves the use of two or more languages by one speaker in the same turn of speech or at turn taking points (6). Simplification. The author describes simplification' as a process that results in the reduction of t he inventory of linguistic forms, semantic range, or language functions and the elimination of alternative structures at certain levels (3). In the case of language contact, this can be due to incomplete learning and/or language attrition. In language attr ition, it is a gradual process that includes the use of all available forms followed by a gradual show of preference (on the part of the users) for one alternative. An example of this is clearly seen is the Silva Corvaln (1986) study on the extension of t he verb estar in Los Angeles Spanish. In her study three factors influence the generalization and expansion of estar in the speech community under examination: 1) language internal factors; i.e. the opaqueness of the verb's semantic content, 2) individual speaker factors; i.e. the speakers' interpretation of the verb's definition and use and 3) external factors; for example, language contact may act as a catalyst for accelerating a change already in progress. This notion of the role of contact as a catalyst for an already existing change is pertinent to this stidy in that changes may be found that are not entirely new to the language but rather very uncommon. In cases such as these, the contact situation is not the initiator of a brand new change but rather the accelerator of an existing change. Overgeneralization. While simplification' deals with contraction where a competing form exists, overgeneralization' may affect contexts where no corresponding or competing form
40 exists. Thus, an element may be remov ed or added to an existing form. Again, Silva Corvaln (1994) shows this process in the use of pronoun clitics extended from reflexive constructions to verbs otherwise not appearing in reflexive forms in the Spanish of Los Angeles. According to the author, this parallels the diachronic development of many verbs in Spanish. Transfer. The concept of transfer also warrants further discussion. Essentially, transfer involves the incorporation of language features from one language (usually the L2) into another (usually the L1) with the subsequent reformation of the subsystems involved. Silva Coravaln (1994) outlines two types of transfer: direct and indirect Direct transfer is easily identifiable at the phonological, lexical and morphological levels; however, at the syntactic level it is very difficult to identify due to the difficulty in proving the permeability of the L1 grammatical system to influences from another system. As the author notes, a sentence such as Creo te hubiera gustado I think you would h ave liked it,' which demonstrates the non expression of the pronoun que is not entirely unquestionable in Spanish (4), as the form is not completely inexistent in monolingual Spanish but very uncommon. Therefore, it is difficult to identify if the form in question presents a case of direct transfer from the L2, as English does not need the expression of the pronoun that.' She traces four phenomena that lead to transfer; the first two lead to direct while the latter two lead to indirect transfer: 1. The repla cement of an existing form in L1 with the equivalent form of L2 or the incorporation of a form from L2 into L1 that had no L1 equivalent. For example, using bye in Los Angeles Spanish instead of the monolingual variant adios On the lexical level, this is also known as borrowing and results in the inclusion of established loanwords or nonce lexical borrowings. 2. Semantic extension: The incorporation of the meaning of a form from L2 to a structurally similar form in L1 that did not have that meaning, while L1 has its own existing form with that meaning. Silva Corvaln gives the following example: registrarse incorporates the meaning of to register' (in school), thus making obsolete the Spanish words matricularse/insribirse to register in a school/for a cour se'" (4). This type of modeling may also occur over entire phrases, commonly known as loan translations or calques
41 3. The higher frequency of use of a form in L1 in contact with L2, in comparison with a lower frequency of the form in L1 not in contact, due to a partially corresponding form in L2. For example, a higher frequency of use of the present progressive is found in contact Spanish due to a preference of the form in English in comparison with a lower frequency of use in monolingual Spanish. 4. The loss o f a category or form in L1 which does not have a parallel category or form in the system of L2; for example, the loss of adjective gender marking in L1 due to the absence of this grammatical feature in L2. As mentioned, calques and loan translations have often been cited as linguistic outcomes of language contact. In his study on the identity of the so called loan translation, Otheguy (1993) challenges just how linguistic the changes are. Whereas the concept of a loan translation has been taken, since Wein reich's (1953) original definition, as a linguistic outcome of language contact, Otheguy argues that many times this is, in fact, a cultural outcome, stemming from a re conceptualization of existing ideas and concepts or from a new concept altogether. The author posits that linguistic interpenetration is highly overstated and that loan translations are either, a) contact induced linguistic changes that were already existing or encompassed under other constructs; or b) not linguistic changes as they are clai med to be. The author's view is that many changes are not of a linguistic or structural nature; rather, they are cultural changes; i.e. they reflect a new way of perceiving something. Otheguy bases his argument on Weinreich's original 1953 definitions (up on which succeeding linguists built their studies): transferring involves additions to the L1 inventory; modeling involves alterations to the existing inventory. The modeling of a single word is known as a semantic extension while that of entire phrases is known as a calque or loan translation. According to Otheguy (1993) many items previously labeled loan translation/calque need be redefined by means of three assumptions of contact literature: 1) Different communities have different conceptual inventories which may or may not overlap; 2) Under contact pressure, many concepts that do not coincide will converge; and 3) Conceptual convergence is different from
42 and can occur independently of linguistic convergence. This means that phrases that have been taken as loan translations are indeed semantic extensions in phrasal contexts and phrases that reveal adaptation or change in cultural and conceptual patterns expressed though language, not in linguistic structure. The author argues that contact varieties a) un dergo semantic extensions or b) undergo "foreign inspired" grammatical changes. These grammatical changes are not loan translations/calques. He goes on to explain the creative processes and give several examples of Spanish phrases spoken in the U.S. that, although labeled as loan translations, show cultural, not linguistic, contact. When languages come in contact, reference is made to an object that either is not known in the contact variety or is known but is conceptualized differently in the non contact v ariety. Two examples include el Da de dar gracias Thanksgiving Day' and mquina de contestar answering machine.' In the first case the change is cultural or conceptual (of an unknown concept) since it is clear that the structure is the same as is used f or Christmas Day' in Spanish, i.e. El Da de la Navidad In other words, if this were an actual grammatical or structural innovation, then the term would look something like "Graciasdando." The second example, mquina de contestar is simply a different wa y of viewing the concept of the answering machine. While in Spain it is a contestador automtico denoting that it is viewed as something that answers automatically, the U.S. concept is that of a machine whose function is to answer. As such, it is the conc ept being modeled, not the structure. This is not to say that all cases called loan translations are actually examples of cultural modeling. Otheguy (1993) does bring up the case of the verb calificar used only transitively by monolingual Spanish speakers for example Calific a Carlos incapaz de desempear ese trabajo He judged Carlos incapable of occupying that position.' In the U.S. this verb is used intransitively as well,
43 for example, Carlos no califica para ese trabajo Carlos does not qualify for t hat position.' According to the author, this is a true case of linguistic modeling. He further argues that the popular U.S. Spanish construction VERB + para atrs for example, llamar para atrs to call back,' is actually a case of conceptual modeling, oc curring due to a new conceptualization of the meaning of para from the realm of space (physical) to the realm of time (abstract). In other words, where para was once used to indicate directionality in space, its concept has been extended to include directi onality in time. Otheguy's (1993) redefinitions of language contact outcomes conclude that semantic extensions reveal true linguistic modeling while loan translations reveal cultural or conceptual modeling (36). The present study will make use of Silva Co rvaln's (1994) first phenomenon leading to direct transfer, in which a form from L2 replaces an existing form in L1, to create a category of established lexical borrowings found in the data. Again, these consist of the replacement or incorporation of lexi cal items in Spanish with an equivalent lexical item in French or English. While Silva Corvaln's (1994) definition also includes nonce lexical borrowings, a discussion on lexical borrowing versus code switching further below will explain why these have be en placed in a different category in this study. Aside from Silva Corvaln's (1994) definition of lexical borrowing, this study also uses Otheguy's (1993) distinction between semantic extension and loan translation in its analysis. Therefore, a semantic ex tension is one in which linguistic modeling is observed, while a loan translation reflects a cultural or conceptual modeling but no linguistic innovation. Code switching. The last of Silva Corvaln's (1994) five outcomes of language contact to be discusse d is code switching Closely related to borrowing, Sankoff (2001) states that "switching [is] the royal road to borrowing" (12) but points out that researchers have long
44 debated the differentiation between a code switch and a borrowing. In the absence of c ommunity wide ratification or legitimation of a single word, researchers could not distinguish between a nonce (one time) borrowing and a code switch, regardless of the attempts to find suitable criteria to differentiate the two. Criteria such as phonologi cal or morphological integration and attestations of use by a wider community of speakers were unsatisfactory due to the variability in the data (12). In order to settle the issue of distinction between nonce borrowings and code switches, Poplack and Mee chan (1998) devise a quantitative methodology that seeks to identify and distinguish between nonce borrowings and code switching. First and foremost, the authors insist on two methodological imperatives: 1) the identification and language characterization of a true speech community that employs code switching and 2) "a sufficiently large sample of sustained discourse" (128) upon which to conduct the study. Their argument for a need to distinguish the two focuses on research findings that demonstrate that "l oanwords pattern according to recipient language syntax and code switches pattern like the donor language" (132). Poplack (2001) further analyses the nature of code switching and its distinctiveness from borrowing. Her interest in this particular theoreti cal discourse is intrasentential codeswitching that which occurs within the confines of a sentence, constituent or word. She argues that lexical borrowing constitutes lone lexical items, assumes the phonological, morphological and syntactic identity of th e recipient language, is recurrent in individual speech and widespread across the community, is available to monolingual speakers of the recipient language and displays no involvement of the phonology and syntax of the donor language. On the other hand, no nce borrowing is distinguished because the social characteristics of recurrence and diffusion are not always satisfied. It resembles established borrowing in that it assumes the phonological,
45 morphological and syntactic identity of the recipient language a nd does constitute lone lexical items; however, it differs from it because it is not recurrent, not widespread and it requires bilingual competence. Methodologically, nonce borrowings and code switches are difficult to distinguish, particularly when they a ppear bare, i.e. as members of neither language (3). According to Poplack (2001) the analysis of code switching requires access to contact languages as they are spoken (i.e. with variability) and quantitative data. The framework for the methodology is ling uistic variation theory (4), devised by Labov and amplified by the author herself, as it accounts for inherent variability. The methodology consists of the variationist comparative method of quantitative analysis. The lone item in the discourse of the cont act variety (contact L1) is analyzed within its surrounding grammar by means of its patterns of usage and grammatical constraints. This is compared to the same analysis from data of the monolingual variety (non contact L1) and donor language (L2) alike. If they rate and distribution are quantitatively parallel to its counterpart in the monolingual recipient language but display different patterns than the donor language, the lone item is a borrowing (loan). If, however, the pattern is parallel to its counte rpart in unmixed donor language (L2) and different from the monolingual variety (non contact L1) then it is a code switch. One of the studies that makes use of this methodology is Torres Cacoullos & Aaron's (2003) analysis of bare English origin nouns in discourse from the Spanish of New Mexico. They predict that the bare nouns will exhibit patterns matching established loanword data and be at odds with English and code switching data. Their data reveal that: a) bilingual speakers in New Mexico borrow and grammatically integrate by zero marking English origin nouns in forming Spanish predicates; b) when bilingual speakers use single English origin nouns they
46 make them grammatically Spanish and; c) the English origin nouns exhibit patterns matching the estab lished loanword data and not the English and code switching data. While the quantitative methodology described above will not be undertaken in the present study, the ambiguous nature of the items and the proper way to disambiguate their identity must be k ept in mind when classifying them. If, as Torres Cacoullos & Aaron (2003) show, bilingual speakers make single English origin nouns grammatically Spanish, the possibility of this occurrence in bilingual or trilingual Hispanics in Montreal cannot be ignored As a distinction between the two is not possible in this study, one category that includes those items that may be either non established (nonce) borrowings or code switches is created in order to comply with the aforementioned theoretical basis. 2.2 Re search 2.2.1 Research on Spanish in Quebec With regards to the linguistic outcomes discussed in this section as evidenced in Montreal Spanish thus far, Godenzzi (2006) conducts a sociolinguistic analysis of Spanish in Quebec in an attempt to discover some aspects of Spanish in contact in the province. His focus is on lexical borrowings, calques, transference and code switching. He analyzes data from 44 bilingual interviewees, divided into four groups: native speakers of Quebec French; native Spanish speaker s born in Quebec or who arrived between the ages of 0 and 11; native Spanish speakers who arrived when older than 11 years of age and who had more than 10 years of residency in Quebec; and, finally, native Spanish speakers who arrived when older than 11 ye ars of age and who had less than 10 years of residency. The author finds several examples of calques and code switching (both fluid and with hesitation). An example of a calque is magasinear to shop' from the French magasiner as compared to the monolingua l variant ir de compras While the study is a good attempt at a first step towards analyzing Spanish in Quebec, the author limits himself to a
47 short explanation of each example but gives neither extended discussion nor a conclusion to his study. Furthermor e, the inclusion of the first group is not necessarily valid because no knowledge is known of the native Quebec French speakers' knowledge of Spanish and no discussion of their possible influence on Quebec Spanish is given. Moreover, though the interviewee s are divided into groups according to their sociolinguistic profiling, the role of those sociolinguistic factors is not explicitly discussed. Finally, while the author mentions that some of the interviewees also speak English and other languages such as P ortuguese and Italian; however, this is not factored into his study. Nonetheless, while Godenzzi's (2006) investigation is not comparable in methodology to the present study, it does provide some insight into the variation that occurs in Quebec Spanish due to its contact with French. Furthermore, his use of spontaneous speech samples offers a first glimpse into the spoken Spanish language in the province of Quebec. 2.2.2 Research on Trilingualism and Multilingualism Research that is focused on the outcomes of language contact in a three code contact situation involving Spanish is scarce. The main studies discover the prevalence of use of two out of the three languages. Stavans & Swisher (2006) conduct a study of code switching in two trilingual children pr oficient in English, Hebrew and Spanish. They discover that at the lexical, morphological and sentential level, most switches involve only two out of the three languages. However, at the discursive and pragmatic level, the switches involve all three langua ges (207). Furthermore, the data reveal that some switches do violate morphosyntactic boundaries. However, it must be kept in mind that the subjects for this study were children and that results may vary according to age and language competence of the tril ingual speakers. Although code switching is not the focus of the present investigation, Stavan & Swisher's (2006) study provides useful insight on some of the dynamics present in language contact
48 involving three codes. Specifically, their results concerni ng the languages involved in switching at the lexical, morphological and sentential level all of which can be applied to this study in much greater proportion than the discursive and pragmatic level suggest that, of the two languages exercising influence o n L1, one dominates and exercises a greater influence. When taken in conjunction with the previous discussion on the history of language laws in Quebec, this fact clearly leads to the question to be amplified in the next chapter of which language will prov e to exercise more influence on the Spanish in the present study. Another study on trilingualism that involves Spanish is done by Clyne & Cassia (1999). The 36 informants, located in Australia, were proficient in Italian, Spanish and English, though the L 1 was not the same for all. The authors note that if two languages share a specific feature, speakers will tend to extend that feature to the third language. This happens on the lexical, syntactic and semantic levels. Lexical borrowings are also found freq uently as well as switching between the three languages. Another characteristic of this group of trilinguals is a large amount of ellipsis, particularly at switching points in the discourse, which the authors partially explain as occurring due to a low pro ficiency in English. The authors also speculate that the similarity between a southern regional Italian dialect, which some of the speakers knew, and Spanish plays a role in the choice of Spanish verbs over Italian verbs. Interestingly, the researchers fin d that Italian and Spanish become a "kind of merged system with items from both converging" (65). This last observation aligns itself with Silva Corvaln's (1994) theory that a language's permeability to external influence depends on the existence of super ficially parallel structures in the L1 and L2. Since Spanish and Italian are typologically more similar to each other than is English to either of them, it follows that convergence would more likely occur between the two romance languages. This has importa nt implications for the present investigation because, of the
49 three codes involved in the contact Spanish, French and English Spanish and French are the two most structurally similar to each other. Consequently, French would be the one to influence Spanish more than English would. 2.3 Concepts and Definitions The present study will make use of several concepts and definitions from the theory and literature reviewed above. It will deal exclusively with the concept of general borrowing as a language contac t outcome outlined by Thomason & Kaufman (1988), in which the directionality is from the L2 into the L1. In other words, the study will look at the influence from French and English (both acting as L2) onto Spanish in Montreal by taking into account the in tensity of contact experienced by the community in the predictions. Likewise, Sankoff's (2001) theory that the socio historical situation be considered when making predictions will also apply here. According to the author, the contact situation brought abo ut by immigration leads to rapid linguistic assimilation of the newcomers. Quebec's language laws are a second aspect that must factor into the expected linguistic outcomes. Stavans & Swisher's (2006) study suggests that one of the languages with which Spa nish is in contact will be the dominating one with respect to exercising influence. Clyne & Cassia's (1999) study in which convergence between two out of the three languages was found, in conjunction with Silva Corvaln's (1994) theory of superficially par allel structures in the L1 and L2 deciding the permeability of language to change, will apply in the prediction of which language French or English will have the dominating effect on Spanish. Furthermore, the data will be placed into specific categories based on the following: Silva Corvaln's (1994) theory of direct transfer in its definition of established lexical borrowings, Otheguy's (1993) definitions of semantic extension and loan translation, and Poplack's (2001) method of distinguishing between no nce borrowings and code switches. The
50 following chapter outlines the methodology employed in carrying out this investigation, along with a description of each category used to classify the data.
51 CHAPTER 3 METHODOLOGY 3.1 Research Objectives This study consists of the qualitative analysis of innovative and competing variants in Montreal Spanish, with the aim of providing an overview of the linguistic picture of Spanish in the area. The Montreal Hispanic community is particular in that it is quite establi shed yet under researched. While spontaneous speech samples are the best way to gain access to non mainstream linguistic innovations, so little is known about the linguistic reality of this particular community that a solid foundation of widespread innovat ions is needed as a basis upon which to build future studies. Such is the aim of this study: to investigate the evidence of mainstream linguistic innovations present in Montreal Spanish due to its contact with French and English in the region studied. Con sequently, the data for this study have not been extracted from spontaneous speech; rather, they are taken from weekly periodicals and daily radio shows in Spanish. As vernacular dictionaries are not widely available and standard dictionaries require wides pread usage of new terms before including them as entries, it stands to reason that widespread innovations would make their way into mainstream means of communication before being documented in dictionaries and, in cases of structural changes, grammar book s. Hence, the periodicals and radio shows are an intermediary step between individual innovations appearing in spontaneous speech that may or may not characterize the community and established innovations that have become widely accepted by the community. By tapping into these media, the aim is to find those innovations that have not yet become widely established but that are being documented in writing and recording and can be said to characterize the community.
52 In light of this, the investigation asks the following: Question 1 What linguistic evidence exists in Montreal Spanish that exemplifies its contact with French and English? Question 2 Subsequent to having provided an inventory and proposed categorizations, what distribution patterns are present in the oral and written data analyzed? Question 3 Which of Montreals two main languages, French or English, exercises a greater influence on Spanish? 3.2 Sources There are two sources of data for this study: printed and oral. The printed data were selected from two weekly periodicals, El Chasqui Latino and Journal lAlternativa The oral data come from the daily Hispano Latino show on CFMB Radio Montral, Montreals multilingual radio broadcast sta tion on 1280 AM. In total, six periodicals and six radio shows were analyzed, their dates ranging between July 2006 to May 2007. One periodical per month from each of the periodical titles was randomly selected for analysis during a period of six months. T he issue dates chosen are shown in Table 3 1. Since twelve months of radio recordings were available, in order to choose six, the selections were made for every second month in order to ensure as wide a time span as possible. The dates chosen are shown in Table 3 2. The data sources chosen for this study are meant to illustrate the nature of Montreal Spanish in the media that reach beyond spontaneous speech. As an added advantage over spontaneous speech, these periodicals eliminate the observers paradox 1 an inherent limitation of spontaneous speech recordings. While one may argue that the data is generated with an audience in mind, which contradicts the elimination of the observers paradox, Bells (1997) model of 1 The Observers Paradox is a phenomenon coined by William Labov used to describe the inherent paradox of the sociolinguistics field, which seeks access to peoples way of speaking when they are not being observed. The paradox is that the only way to access such speech is by observing the speakers.
53 audience design supports the validity of the data sources. Developed to account for both face to face as well as mass communication, the model states that speakers regulate their speech in accordance to that of their audience in order to express solidarity or win approval (244). Additionally, the model suggests that this style shift can be both responsive and initiative. It is responsive when a speaker shifts style to be more like that of the audience; it is initiative when the style instigates a change in the situation rather than resulting from such a change (247). Bells model of audience design suggests that the data in the sources is produced in a style that seeks to emulate that of its implied audience/readers and therefore strengthens the claim that the language used approximates that of spo ntaneous speech. 3.3 Data Selection This section explains the process by which data were selected for analysis. Being that not all the material encountered was for inclusion in this investigation, a detailed set of methodological decisions were made in o rder to ensure that the selection process encompassed all pertinent material. 3.3.1 Qualitative Analysis With respect to the written material, both articles and advertisements were included in the selection process. Given that the study investigates the evidence of language contact in Spanish, all monolingual French and English and bilingual French/English advertisements were excluded from the data selection process. The advertisements that were included were written either all in Spanish or mixed Spanish /French/English. In the case of multilingual advertisements, only those containing at least 50% Spanish usage were included. The minimum 50% rule was created because some advertisements contained little Spanish but much French or English. In this case, is impossible to claim that an item of French or English origin is a (nonce) borrowing or a code switch, since the advertisement is mainly in that second
5 4 language. Appendix A provides two figures as examples of advertisements that were not selected for analys is. Figure 3 1 represents the type of advertisement that was considered for analysis given that its main language is Spanish. This is quite obvious not only because the majority of the advertisement is written in the monolingual variety of Spanish, but al so because some of the elements that show evidence of influence, such as the noun especiales specials, have become integrated into Spanish, as is demonstrated by the use of the plural definite article los the. In an advertisement such as this one, wher e the main language is Spanish, the item dmarreur distance automatic car starter, although written in French, is considered part of a Spanish communication. Figure 3 2 is a further example of the types of data that were considered and extracted for an alysis. The advertisement shows borrowed or code switched items into Spanish. Figure 3 3 is an example of data extracted from an article. Given that the periodicals are in Spanish, all articles are written in Spanish and thus the 50% rule that was establis hed for advertisements did not apply for articles. As such, all articles were considered for analysis. However, many details that arose as anomalies in the Spanish language were nonetheless rejected. Orthographic accent marks were excluded from any analysi s due to the impossibility of determining the origin of the anomalies, as they showed influence from neither French nor English. The periodicals do not pay particular attention to the use of accents; many Spanish words that normally require written accents were found to contain no accent, while others that do not require written accents were found to contain one. Furthermore, Spanish stress is only indicated by use of the acute accent while French words may have a variety of accents to indicate the pronunci ation of a given phoneme, including the acute accent, the grave accent and the circumflex. Some Spanish words in the advertisements were written with a grave accent instead
55 of an acute accent. These anomalies were not included in the data selection process because little can be said about their effect given the lack speakers pronouncing the words in this investigation. Likewise, the unconventional use of capital letters was also excluded from consideration for analysis. Many advertisements capitalized the d ays of the week and the months of the year, something that is not normally done in monolingual Spanish. This type of change was considered to be of a stylistic rather than linguistic nature and was thus excluded. Similarly, words that are spelled the same way in French or English as they are in Spanish whose linguistic origin was not discernible were also rejected for analysis. Finally, words of indiscernible origin were also excluded from analysis. These are words that exist in Spanish and either or both F rench and English and have the same spelling. Appendix A contains an example of such occurrences. With regards to the oral data extracted from the radio recordings, many of the decisions of inclusion or exclusion discussed above do not apply. In view of t he fact that the Hispano Latino show is spoken only in Spanish, there was no question as to what was the main language. However, any monolingual French or English commercials were not considered for the data selection process. The shows lasted 30 minutes e ach; all linguistic evidences of contact such as lexical borrowings, calques and morphosyntactic innovations as well as code switching were extracted and tabulated. 3.3.2 Quantitative Data In addition to the data extracted for the qualitative analysis, so me frequently occurring competing variants were used in a small quantitative analysis aimed at further investigating the extent to which the innovative forms are integrated in the Spanish of the periodicals and the radio shows. The pairs commonly found to be in variation are listed in Table 3 3. These items were extracted from all the radio recordings and from all the written material considered for analysis. Their purpose is to determine if there is a preference for a particular
56 competing item and if so, w hat that preference is and what is its language. In doing so, the competing variants standing and the innovations integration in Montreal Spanish will be more evident. 3.4 Data Organization and Analysis 3.4.1 Cross References All data were cross referen ced with the use of the following five dictionaries: The Dictionary of Chicano Spanish (henceforth, Chicano) A Dictionary of New Mexico and Southern Colorado Spanish (henceforth, New Mexico ), Diccionario del espaol actual (henceforth, Actual ), Nuevo dicc ionario de americanismos e indigenismos (henceforth, Nuevo ) and Diccionario del uso del espaol de Amrica y Espaa (henceforth, Uso ). The purpose of the dictionaries was to properly identify established borrowings, semantic extensions and loan translation s. Dictionaries were referenced in order to ensure that certain terms thought to be innovative did not exist already. In order to ascertain that a Spanish monolingual variant of an innovative type does exist, the definitions for the Spanish monolingual var iants were taken from El diccionario de la Real Academia Espaola As a tool to assist in determining the direction of influence for some innovative types, the Quebec Bureau of the French Language (Office qubcois de la langue franaise) online dictionary Le grand dictionnaire terminologique was consulted. Rather than use a standard French dictionary, the use of dictionary of Quebec French provided insight into how some terms are used in the province. 3.4.2 Spelling Variations It is important to note her e the importance of considering entries that may reflect the L1 phonetic spelling of the L2 word. For example, one of the findings was the French origin word livraison delivery, pronounced [li.v e.z ] in French. However, if the word is being used often,
57 there may be a possibility that the spelling has been changed to reflect the Spanish pronunciation of the word. Hence, in Spanish one possible spelling that would approximate the original French pronunciation is the hypothetical livreson. For all words, it was necessary to come up with all the spellings that could represent possible Spanish pronunciations of the word in order to be as certain as possible that the word had indeed not made its way into the dictionary. Another example of this is the word s ticker. Since Spanish words do not begin with the letter s followed by a consonant, epenthesis occurs with many borrowings of this sort and the vowel e is added to the beginning of the word to account for the phonotactic necessities of the language. As we ll, the sound [ke ] in Spanish can only be spelled with the sequence of letters quer. Taking this into account, the word that was searched for cross reference was not only sticker but also estquer estker and stquer of which estquer was found in one of the dictiona ries as an established lexical borrowing from English. 3.4.3 Categorization Finally, all data were then sorted into the categories in Table 3 4. The Established Lexical Borrowings category contains those lexical items that show, by Silva Corvalns (199 4) definition, direct transfer from either French or English into Spanish and that have replaced or are competing with a Spanish monolingual variant. To be considered an established lexical borrowing, the form must be documented in at least one dictionary. For example, parking is attested for in a dictionary and is found to be competing with estacionamiento Both the English origin and French origin Non established Items categories contain the lexical items that are not found in any dictionary. These items may possibly be nonce borrowings, which are non established lexical borrowings, or they may be code switches. Given that the nature of this study does not permit the differentiation between the two, these categories will contain those
58 items that may be either possibility. For example, remorcaje towing is not found in any dictionary; therefore, it belongs in the category of non established items. The Semantic Extensions and Loan Translations category consists of existing items in monolingual Spani sh but whose meaning or arrangement is particular in the bilingual Montreal variety. Semantic extensions are single word items that existed before contact but have adopted a new meaning from French or English. An example of a semantic extension is the item carta card. While its monolingual meaning is that of letter, it has acquired a new meaning in the contact situation. This is a linguistic innovation due to language contact. Loan translations are those phrases that consist of existing words in monolin gual Spanish but whose sequence in a phrase is unique to the contact situation. Since there is no change in the actual meaning of the words, Otheguys (1993) view will be taken that the phrases reflect cultural contact, not linguistic. The Morphosyntactic Innovations category is made up of items, either one word or a string of words, which show a structural innovation either at the morphological or syntactic level. For example, mueblado furnished is showing the omission of the prefix a with which it i s written in monolingual Spanish. The Additional Changes/Processes category contains those items that display more than one process of change or those sets of items that jointly show more than one change. Some of the words appearing in Additional Chang es/Processes do demonstrate only one change; however, they belong to a whole set of items that belong in this category and were therefore placed it so that they may be discussed with their variants. Likewise, those items that display changes that do not f it the aforementioned categories are also placed in this one. For example, pharmaceutico pharm aceutical is normally spelled farmacutico in Spanish and thus this word displays a spelling change due to influence from either French or English.
59 Finally, th e Quantitative Data category displays the competing variants mentioned above that were counted and tabulated. This category will also display the percentage of use of the variants. Its purpose is to assist in giving a clearer picture of the language pref erences for particular variants over others and how integrated some borrowings are in the Hispanic community. 3.5 Hypotheses Based on the theoretical basis and literature review for this investigation, the following predictions are made for this study. Gi ven Sankoffs (2001) speculation that the contact situation brought about by immigration leads to rapid linguistic assimilation of the newcomers, it is expected that the majority of the findings will be focused on lexical items, either borrowed or code swi tched. This hypothesis is supported by the fact that the data comes from written material that consists of few articles and mostly advertisements and that the oral data comes from a radio shows. These are media intended for wide audiences and thus writers and speakers do attempt to standardize their speech. As such, since lexical borrowings and code switches are the first indications of language contact, they will probably be the most numerous outcomes in the data. Similarly, Sankoffs (2001) theory that the socio historical situation be considered in language contact studies plays an important role in this study of three codes. Whereas the early immigrants probably had more exposure to English, those that arrived or were born after Bill 101 passed in 1977 have been in contact with more French than English. Furthermore, Silva Corvaln claims that a languages permeability to change depends on the parallel structures of the L1 and the L2. In addition, Stavans & Swisher (2006) find a dominant language of infl uence in a trilingual setting and Clyne & Cassia (1999) find convergence of two similar languages in a trilingual setting. The theory and findings above lead to the hypothesis that French will dominate in influence on Spanish because a) it is the one to wh ich the majority of Hispanic immigrants and
60 Montreal born Hispanics have been exposed to most and b) French is, like Spanish, a romance language and is, as a result, structurally more similar to Spanish than is English. However, language maintenance is als o to be expected as a result of Thomason & Kaufmans (1988) predictions regarding intensity of contact. As has been shown, the Hispanic community is very tightly knit; when a large community such as this one is established and has its own churches, social gatherings, businesses, health care workers, etc., much of the communication will be done in the L1 and therefore it may resist the rapid linguistic assimilation described by Sankoff (2001). Consequently, it cannot be expected that French or English have g one a long way in permeating Montreal Spanish. In short, it is expected that, on the continuum of linguistic outcomes, this particular situation will yield ample evidence; however, most of it is expected to be lexical, due partly to the nature of the sou rces and partly to language maintenance. Furthermore, French is expected to have a stronger influence on Spanish than English.
61 Table 3 1. Periodical dates LAlternativa El chasqui latino 08/08/06 10/19/06 09/19/06 11/30/06 12/12/06 01/11/07 Table 3 2. Radio recording dates 7/10/06 9/29/06 11/07/06 01/18/07 03/27/07 05/01/07 Figure 3 1. Analyzed bilingual advertisement 1
62 Figure 3 2. Analyzed bilingual advertisement 2 Figure 3 3. Data in an article Table 3 3. Competing variants S p. este and Fr. est east Sp. norte and Fr. nord north Sp. oeste and Fr. ouest west Sp. oficina Fr. bureau and Eng. office Sp. esquina and Fr. Coin corner Sp.= Spanish; Fr.= French; Eng.= English
63 Table 3 4. Categories Category Example Establi shed Lexical Borrowings parking English origin Non established Items tax French origin Non established Items remorcaje towing Semantic Extensions and Loan Translations carta card Morphosyntactic Innovations mueblado furnished Additional Changes/ Processes pharmaceutico pharmaceutical Quantitative Data esquina vs. coin corner
64 C HAPTER 4 RESULTS, ANALYSIS AN D DISCUSSION 4.1 Collective Findings This chapter presents the analysis of the data. The findings are categorized under the categorizations described in the previous chapter: Established Lexical Borrowings,' English origin No n established Items,' French origin Non established Items,' Semantic Extensions and Loan Translations,' Morphosyntactic Innovations,' Additional Changes/Processes' and Quantitative Data.' Table 4 1 provides an overview of the data collected for this investigation. It presents the types and tokens of the items in each section to be discussed. A type' is each separate word or group of words as they appeared, while a token' is counted for every occurrence of a given type. The items total 708, with onl y 17 originating from the radio recordings. Given that this number is quite low, there will not be a separate discussion for the oral data and the written data. Rather, the oral data will be included with the written and together they will be treated as a whole. In this sense, the oral findings are considered as further support of the patterns found in the written data. The table shows that the least number of documented innovations comes in the form of morphosyntactic changes, a fact that was to be expecte d according to Sankoff (2001). Conversely, the most numerous tokens are of French origin, which also leads to an initial speculation that French exercises a stronger influence on Spanish than does English. A more detailed look at the results as well as an analysis of the quantitative data in variation will provide further clarification to this question. The data for each category is presented in tables divided to provide detailed information for each item. The type is listed in the first column while the n umber of tokens for the given type is listed in the second column. The third column presents the type's rank according to its number
65 of occurrences. Ranks were assigned in the manner shown in Table 4 2. The fourth column, Unique Appearances,' gives the nu mber of unique places in which a given type appears. For example, if a given type appears recurrently in a reprint of the same advertisement, it has a unique appearance of 1, regardless of how many tokens there are. Different advertisements from the same b usiness were counted as unique places. With regards to the articles, if the type appeared several times in the same article, then its number of unique appearances was 1. However, if it appeared in two different articles, its number of unique appearances wa s 2, regardless of the number of tokens. Finally, the oral data were assigned unique appearances in accordance with the subject matter in which a given type appeared. If the type appeared several times in the same commercial or news story, its unique appea rance was 1. With every change in news story or commercial, a new unique appearance was assigned. The column of unique appearances is useful in demonstrating how widespread a given type is. In other words, if a type contains many tokens but only one unique appearance, it cannot be said to be extremely widespread. Each table also displays a breakdown of those types that appeared both in oral and written data. The division is provided in the column of tokens, with O' used to indicate the number of oral token s, W' the number of written ones and T' the total tokens of that given type. As the spelling of the oral data is not known, it has been given in its language of origin. For each section, a detailed analysis for all items in the highest rank is done, whil e selected items in the lower ranks are analyzed based on the complexity of the changes they have undergone. That is, the ones displaying the most complex changes are discussed. Throughout the course of the analysis and discussion, all foreign words are i talicized, all English glosses are given in single quotation marks and all items appearing in the data are underlined. If an item is foreign and also appears in the data, it is underlined and italicized.
66 4.2 Established Lexical Borrowings Table 4 3 provi des a detailed overview of the established lexical borrowings that appear in the data. In addition to the aforementioned columns, this table also provides the names of the dictionaries in which entries for the types were found so that they may be classifie d as established. The highest rank in this section is 4 and it contains two types. Mouse refers to the peripheral apparatus used in computers to direct the c urser. Commonly referred to as ratn in Spanish, this borrowing clearly belongs to a long list of borrowings from the field of technology, many of which never acquire a monolingual c ounterpart such as ratn As indicated in the methodology section, both mouse' and maus were searched in the dictionaries because many of the borrowings are found as entri es spelled to reflect their Spanish pronunciations. Only one dictionary ( Uso ) had mouse' as an entry. The use of this borrowing is particular in that it appears in the advertisements as teclado y mouse or mouse/ teclado keyboard and mouse' and mouse/keyb oard' respectively. With the existence of ratn the adverti sements could easily have read teclado y ratn or a similar variation; however, it would seem, at least in this corpus, that the borrowing is preferred to express mouse while the monolingual var iant is preferred to express keyboard,' as n o occurrences of keyboard' or qubor or any similar variant were found. Both parking and parqueo parking' occur in the dictionaries and are used to express the English noun parking, many times employed as a shortened version of parking spot. A common advertisement read parking disponible para nuestros clientes parking avail able for our clients.' Many other derivatives appe ar in the dictionaries such as parqueadero parquiadero and parquera though these we re not found in the sources from which the data for this project were extracted. Show is an entry in all but one of the dictionaries, which also c ontain variant entries such as cho or sho
67 Many of the remaining established borrowings are easily recognizable as they are of English origin and appear in the dictionaries either in their original English spelling or as variants spelled in Spanish such as muffler which a lso has the corresponding entries mofla mofle and mufle and ticket which has the corresponding entries tiquete tquete tquet and tique Of all the English origin items in this list, the following are also documented entries in the French dictionary, Le grand dictionnaire terminologique (Office qubcois de la langue franaise), either as established borrowings or as terms to avoid: parking, show, scanner, stock, cake, cash, ticket, gadget, feedback, stress. In either case, their presence in the dictio nary is an important finding, as it nullifies the assumption that the items in the list are borrowed into Montreal Spanish directly from English. The fact that they are documented in an official Quebec dictionary proves that the items are also used in Mont real French, which means that they could be borrowed into Montreal Spanish through contact with French, rather than English. Given that the socio political situation in Quebec urges for a greater amount of contact with French, it is highly probable that th e aforementioned items have entered Spanish through French. Coupled with the items discussed above, those established borrowings of French origin are of particular importance to this study because they represent a rare situation in Spanish language contac t studies. With entries only in the Actual and Uso dictionaries, it makes sense that these items are found only in Spanish dictionaries from Spain, as this would be the only geographical location outside of Quebec and Equatorial Guinea where one might find contact between Spanish and French. This presents a unique and under researched linguistic situation where studies of Spanish in contact are concerned. The established lexical borrowings of French origin are bifteck stage Noel and mis en plis Bifteck is spelled as it is in French and also appears in Uso as biftec which reflects a
68 S panish spelling. Both borrowings are different than bistec the word used to denote steak by speakers of Spanish. Stage pronounced [sta # ], is used to signify a vocational training or internship as indicates its meaning in French while Noel found only in Actual is used to express Navidad Christmas.' Mis en plis may reflect further integration in the process of borrowing because part of its spelling has changed in accordance to a Spanish spelling of t he pronunciation. In French, a mise en plis is a set, a coif that women may typically get done at the hairdressing salon. While the version that appears in the corpus is almost identica l to the French one, the word mise has lost its final e There are two options for this: either it is a typing error on the part of the periodical or it may reflect the Spanish spelling of its pronunciation. What does challenge this second possibility is that the same type of spelling change has not taken place in plis which in French is pronounced [pli]. Had the word ap peared in the advertisement as mis en pli then the argument that the spell change was made to reflect the pronunciation of the word in Spanis h would have been more compelling. Being that as it may, since the borrowing is established as per Actual it has been classified as such in the corpus of this study. The French origin established borrowings found above give a first look at the possibiliti es of linguistic outcomes when two structurally similar languages such as Spanish and French are in contact. With so much research on Spanish/English contact having been conducted in the United States due to its large communities of Hispanics, Spanish/Fren ch contact studies are lagging behind and more research is needed in order to examine trends and patterns that evolve from this contact situation. The following two sections will examine the non established items found in the investigation. As will be seen the French origin items are more numerous than the English origin ones, which leads to the preliminary deduction that French is probably more influential in Montreal Spanish than is English.
69 Furthermore, it must be noted that tracing the path from a cod e switch is much easier than tracing one from a borrowing. Since a code switch entails beginning a sentence in one language and switching codes before it ends, the only question pertinent to a thorough analysis is precisely a study of the grammar surroundi ng the borrowed or switched element in order to determine whether it appears embedded in the grammar of the user's native language or whether the grammar is that of the second language. Identifying the path of a borrowed element in a setting where three co des are present, as is the case in this study, becomes slightly more complicated as it is difficult to determine whether the item is borrowed directly from its own language or whether it enters Spanish through a second language that has itself borrowed it. In Montreal, for example, French employs many lexical borrowings from the English language, as does English from the French language. If Spanish borrows a particular lexical item of English origin there is a possibility that it may be due to the contact b etween Spanish and French, with the item having first been borrowed by French (from English) and then by Spanish (from French). In order to determine the path taken by the borrowed item, two elements must be determined: the variety of French in contact wit h Spanish and the extent of contact with French in comparison with English. The former serves to determine whether the item is present in that particular variety of French as well, while the latter attempts to best trace the path undertaken by the borrowed item. If, for example, the item is an established loanword in the French variety in contact with Spanish and Spanish has limited contact with English, then there is a likelihood that the borrowed item came from the contact between Spanish and the given va riety of French. In this study, items of English origin that are known to be used in Montreal French even if not documented in Le grand dictionnaire terminologique will be duly reported 1 1 A native speaker of Montreal French has reviewed the results of this investigation and identified such items.
70 4.3 English origin Non established Items As this study does not atte mpt to distinguish between nonce borrowings or code switches the items that may belong to either of these categories are referred to as non established. The analysis of this data reveals what types of French origin and English origin items are being employ ed in Spanish and how they are being used. Table 4 4 presents the English origin non established items found in the corpus. As can be readily seen, the item with the greatest number of occurrences and the only one with a rank of 1 is tax or tx Although this type is not established, a variant was found in one dictionary ( Chicano ) as taxa The item is probably on its way to being established in Montreal due to the large number of tokens of this type in this study and the unique appearances of 15. Addition ally, a very similar variant has made its way into a dictionary of spoken Spanish in the United States, which suggests that the type may be in the process of becoming established in other geographical areas of contact as well. Ranking 3 and 4 respectively, flyers and banners are either being borrowed in Spanish and replacing the monolingual variants folletos and pancartas or they are being commonly used as code switches. Steackers is a recurring borrowing in the ads of one particular company and is an alte rnate spelling of the English word stickers.' New Mexico contains the entry estquer a variant that shows a high level of integration into Spanish by the epenthetic vowel e Actual and Uso contain an entry for stick' while Nuevo contains one for estic all referring to the noun stick.' While a stick is not related to the verb to stick,' from which sticker' is derived, the fact that stick' and estic' are both to be found in two of the dictionaries while estquer is already in one of them does sugg est that the term (either sticker' or estquer ) may eventually make its way into more dictionaries of the spoken Spanish language.
71 As mentioned in the methodology section, the words were not only searched with their spelling from the advertisements but a lso with other possible spellings that would match other patterns found in borrowed or code switched words in the Spanish language. For example, since words cannot begin with an s followed by a consonant in Spanish, many established borrowings show that a vowel has been added before the s to maintain Spanish phonotactic rules. An example of this phenomenon is seen in the established borrowings estquer and escner both of which show the epenthetic vowel e ([ $ ]). Similarily, a word such as scrap could have an entry spelled escrap although both scrap' and escrap were searched and found to not be established borrowings. In the advertisements used to create this corpus, this particular term refers to the leftovers from automobiles found in a junkyard. The te rm commonly used in Spanish is chatarra ; however, scrap is either beginning to replace this term in Montreal or being used as a code switch. This term was identified by the native Montreal French speaker as a common lexical borrowing in Montreal French. Th erefore, although it has been categorized as an English origin item, there is a concrete possibility that it has entered Spanish through the influence of French, a possibility that must be kept in mind when gauging the influence of French versus English on Spanish. Switch and wave(s) are both ranked 5 in the list. In this particular study, all tokens of switch were used in the context of computer or technology while w ave refers to a type o f hairstyle. Both switch and suich as well as weiv geiv and hueiv were all searched as possible entries in the dictionaries. Their low ranking and their unique appearances of 1 both suggest that these types are not very diffused in Montreal Spanish. Of all the types in this section, sticker scrap and bus bo y are found in Le grand dictionnaire terminologique This shows that these three items, along with others that may not
72 yet be documented in the dictionary, may have found their way into Montreal Spanish through French, not English. Nonetheless, there are fewer items from this list documented in Le grand dictionnaire terminologique which also points to a certain influence from English, given that only three types out of eighteen are documented in French usage. 4.4 French origin Non established Items Table 4 5 below presents the list of French origin non established items. With a total of 272 tokens, this list is lengthier that its English origin counterpart and shows higher numbers of unique appearances. This in turn, suggests a greater degree of diffusion of the typ es in this table as compared to those in the previous one because a greater amount of places in which a given type appears indicates that the type is used on a wider scale. If a type has many tokens but only 1 unique appearance, the main claim to be made a bout it concerns its degree of exposure to the community, i.e. the community's frequent exposure to said type may cause it to begin being used in more places. Conversely, a type that has many unique appearances shows not only a large exposure to the commun ity through frequent contact with the item but also translates to a more extensive use by diverse businesses or writers, reporters, etc. The two types in the highest rank are the cardinal directions est east' and nord north.' It is not surprising that c ardinal points have a high number of tokens as nearly every business advertisement has its address listed and that is where all tokens were found. Likewise, it is not surprising that est is the type with the greatest amount of tokens because the street Be langer est' is the location for most Hispanic businesses. Both est and nord are in variation with their Spanish monolingual counterparts and will be analyzed for such variation in the quantitative section further below. Their respective unique appearances of 70 and 9 are indicative of widespread usage, though est is the type that displays this to a greater degree. However, it is also possible that the items are in reality being used as proper nouns; that is, they are acting as part of a
73 compound word togeth er with the street name. If this is the case, the presence of the monolingual Spanish variants with which they compete is indicative of language maintenance, because it shows that a compound word that is a proper noun becomes separated in order to have par t of it replaced with an equivalent word in the L1. Canet(s) is the only type in rank 3; however, its unique appearance of 1 does not show very perv asive usage. In Quebec French, canette ([k.n $ t]) is the term used to denote a can,' the metal container used for many different types of drinks (Office qubcois de la langue franaise). All tokens of canet(s) were found in this context and display a loss of the French ending ette to reflect the S panish spelling of the pronunciation. Coin corner' has the greatest number of tokens in rank 4 and will also be analyzed for variation in the quantitative section. Its high number of unique appearances is to be expected as all tokens were found in addres s listings, just as with the two cardinal points mentioned above. Ouest west' as well shows a high number of unique appearances (4) with respect to its number of tokens (7), thereby furthering the aforementioned assumption that cardinal points and types r elated to address listings will show a high number of tokens. These types are those that will likely experience a greater degree of diffusion due to both a greater amount and prolonged exposure. Demarror shows a greater integration into Spanish for two rea sons. First, its number of unique appearances goes beyond 1, which indicates greater degree of proliferation and suggests a tendency to continue the propagation of the type. Secondly, it displays a spell change that reflects the Spanish spelling, indicatin g that the word is being integrated i n writing as well. In French a dmarreur is a car starter,' known in Spanish as motor de arranque However, the monolingual variant is not employed probably because dmarreur lends itself well to being employed in Span ish. As Silva Corvaln (1994) theorizes concerning parallel structures between
74 the L1 and L2, the structural similarity between the French suffix eur and the Spanish or allow the root dmarr to be more readily borrowed. Remorcaje has a unique appearanc e of 2 and displays a similar process as the discussed above. In mo nolingual Spanish, the word is remolque and it refers to the noun tow(ing). Quebec French uses a similar remorquage The innovative term has been classified in this category because the w ord is strikingly similar to its monolingual French counterpart but for the changes that have been made to incorporate it into the Spanish language. The sequence qua has been changed to ca to reflect the Spanish pronunciation [ka] and the French suffix age has been replaced by its Spanish equivalent aje The replacing of the suffix potentially changes the pronunciation of the French word and shows added phonological integration of the type into Spanish. Bureau/bur. office' and revenu revenue' con tain the greatest number of tokens in rank 5. Bureau/bur. falls in the list of tokens found in addresses along with the cardinal points and coin and is also used in the forthcoming quantitative section analysis. Revenu appears in the advertisements of a lo aning agency in the expression sin verificacin de revenu without verification of revenue.' This item is of French origin a nd has a Spanish equivalent of renta that seems to be losing its original meaning Spanish and acquiring the new meaning of rent, a s will be seen further below. Of the remaining items, asermentada (asermentar) and depanneur allow for more elaborate linguistic discusson The verb adjective asermentada is derived from the verb asermentar taken from the French assermenter Le grand dic tionaire terminologique of the Quebec Bureau of the French Language (Office qubcois de la langue franaise) defines the word as "to administer the oath However, it also contains a separate entry for assermenter un
75 document to certify a document,' in w hich it is spec ified that the use of the verb assermenter in this sense is incorrect and that the correct ver b to be used in such a case is attester to certify.' The entry further states that assermenter may be used a transitive verb, whose direct object can only be a person, not a thing. The existence of this entry in a dictionary whose purpose is to maintain the French language suggests that in Quebec assermenter is also being used to denote to certify (a document). The adjective asermentada certified is found twice in this corpus, both times referring to documents. In this case the verb is not only being used in Montreal Spanish as it is in Montreal French, but it has also undergone the necessary changes needed to integrate it into the Spanish langua ge such as the loss of one s and the change from the French infinitive ending er to its corresponding Spanish ar By doing this, the verb can be conjugated as needed and adjectives can be formed with it, as seen in asermentada Finally, depanneur is im portant because it recalls the aforementioned process in demarror whereby the root demarr is borrowed from French and the Spanish suffix or replaces the French eur While this change is not reflected in depanneur in this corpus of data, future studies may reveal the noun to be spelled depanor In the numerous tokens that are found in this section of non established French origin items, the following patterns are salient. The French root of a word is used with the Spanish suffix that corresponds to the s uffix in the original French item. The two Spanish suffixes that are found in this corpus thus yet are aje and or They replace the French age and eur Finally, among the types that have the highest number of tokens and unique appearances are those app earing in address listings, particularly cardinal points. 4.5 Semantic Extensions and Loan Translations As discussed in the theoretical framework above, this section will make use of Otheguy's (1993) definitions for semantic extensions and loan translatio ns. Therefore, a semantic extension
76 is one in which linguistic modeling is displayed while a loan translation will reflect a cultural or conceptual modeling but no real linguistic innovation. Table 4 6 below outlines the semantic extensions and loan transl ations found in the collected data. Some are already known in U.S. Spanish. The only type in rank 1 is carta taken either from the English card or the French carte Since the Spanish equivalent carta (letter) already exists, this innovation consists mai nly of the semantic extension of the word. Some of the phrases in which it has appeared are carta de llamadas internacionales cartas de negocios carta de crdito and carta de descuento In monolingual Spanish, a ll phrases would take the word tarjeta car d.' The French equi valents of the expressions are cartes d'appel internacionales cartes d'affaires, carte de crdit and carte d'escompte Due to the resemblance of carta with the French carte or with the English card' (or perhaps both), the term is now b eing extended to mean card.' The phrase carta de crdito already existed; however, Uso defines it as [d]ocumento que un banco u otra empresa expide a favor de alguien para que le sea e ntregada una cantidad de dinero thereby denoting a letter of credit .' While it may be obvious that a credit card is indeed a type of letter of credit, the fact remains that monolingual Spanish speaker s express card' with the word tarjeta and reserve carta str ictly for letter.' The use of carta to express card' not only in carta de crdito but also in the remaining aforementioned phrases strongly suggests that carta de crdito credit card' is an extension of the original meaning propelled by contact. Bearing this in mind, carta de crdito can now be looked at in two way s. Either it is a semantic extension of the word carta as determined above, or the whole phrase (which existed already) is a semantic exte nsion modeled after the French carte de crdit and/or the English credit card.' However, the most probable cause for this semantic extension is the contact with French, given that no other dictionaries
77 ( Chicano included) document this innovative use and that Qubec is predominantly a French speaking province. Furthermore, this speculation can be further supported by the fact that French utilizes the same syntax as Spanish to express the item in question, namely noun preposition noun while English employs adjective noun thereby making carta more easily modeled in its compound noun. El especial/los especiales is the onl y type in rank 2 and is of particular importance because its number of unique appearances matches its number of tokens. This indicates a high diffusion of the item in Montreal Spanish. This specific term has been defined in all dictionaries consulted as th e adjective special.' Its reason for being on this list as a semantic extension, however, is its usage in this corpus as a noun, to denote a special' in a business, an article being offered at a reduced price. The most common Spanis h monolingual word for this is rebaj a while English in Montreal has a special' or a sale.' French has the standard rabais solde and aubaine ; however, Le grand dictionnaire terminologique (Office qubcois de la langue franaise) does contain an entry for the noun spcial i ndicating that its use to denote a sale is incorrect. This implies that the item also exists in Montreal French as a semantic extension. Given the strong resemblance of especial wit h English special' and French spcia l it is impossible to rely on a purel y linguistic analysis to determine whether this semantic extension in Montreal Spanish is due to its contact with French, English or both. However, the predictions based on the amount of contact due to Quebec language laws suggest that it is probably mainl y due to the influence of French, while the possibility of influence from both languages simultaneously is not to be discarded. Tiempo pleno is the only type in rank 3 and is particularly interesting as it is an example of what Otheguy referred to in his debate on linguistic versus cultural modeling. In this corpus,
78 tiempo pleno refers to full time' when speaking of employment. The Spanish equivalent for this is tiempo completo However, it can be easily noted that tiempo pleno bears a striking re semblanc e to the French phrase temps plein In this instance, no linguistic alterations a re being made, given that both tiempo and pleno exist and are being used without any modification to their original meanings. What is being modeled is the Qubec French view o f the concept of time (when employed) as full instead of complete. For this reason, the phrase is a loan translation as defined by Otheguy (1993). Aplicar/aplicacin is not unique to Montreal Spanish. Existing in the U.S. as well, it consists of semantic exten sions of both the Spanish verb aplicar and the noun aplicacin While aplicar in monolingual Spanish refers to the application of one thing onto another' and aplicacin means the implementation of rules and laws,' the verb has incorporated the meani ng in to apply' (for a job), while the noun has incorporated the meaning job application.' Monolingual Spanish has the verb solicitar and the noun solicitud reserved for the aforementioned meanings. With six cumulative tokens, this type has a relatively high unique appearance number of five, which indicates a greater diffusion. While it is most probable that French is responsible for causing the semantic extension of aplicacin the same cannot be said for aplicar The reason for this is that Le grand dic tionnaire terminologique (Office qubcois de la langue franaise) contains an entry for application again stating that the use of the term in reference to a job application' should be avoid ed and the preferred phrase is demande d'emploi This suggests t hat application is used in Montreal French when referring to a job application.' Conversely, no entry of the French appliquer (to apply) in the dictionary suggests an erroneous use of the verb in Montreal French. This fact, coupled with the existence of a plicar in U.S. Spanish, implies that the origin of the semantic extension of the verb in Montreal Spanish is not
79 very clear. It may be that appliquer is actually being used in Montreal French to denote to apply for a job and that, while this added meaning has not reac hed the dictionary yet, its use is indeed affecting the meaning of its Spanish counterpart. It may also be that the semantic extension of aplicar is due to influence from English. Further yet, it may be that the semantic extension of aplicar was caused by the semantic extension of aplicacin in Montreal Spanish itself. That is, aplicacin may have been the first to incorporate the new meaning (due to influence from French) and then passed the added meaning onto the verb aplicar This last scenario is the on e most plausible because all the analysis thus far points to a greater influence from French than English, not only through the number of types originating from each language but also because some of the English origin items are already used in Montreal Fr ench. Renta rent and rentar to rent are common semantic extensions in U.S. Spanish as well. While both terms exist in monolingual Spanish, renta is used by monolingual spe akers to refer to income and rentar means to produce a profit. The monoling ual Spanish equivalent s for their innovative use are alquiler (noun) and alquilar (verb). These two semantic extensions come straight from English as their French counterparts are not similar: loyer rent and louer to rent. As they are attested, there i s a possibility that their existence in Montreal Spanish is not due to linguistic contact in Montreal but rather that they were part of the vocabulary of the speakers before arriving to the city. However, since this possibility cannot be proven in this stu dy, as this would require extensive knowledge about the origin of the speakers using these terms, the possibility that they arose due to contact with English in Montreal must also be considered. Furthermore, it is quite probable that the innovative m eaning s of Spanish monolingual renta and rentar have incorporated the new meanings due to metonymy, as rent can be a form of income and to rent can also produce a profit. In this sense, renta and rentar may
80 represent changes already in progress i n monolingual Spanish that were simply accelerated due to contact with English, as suggested by Silva Corvaln (1994). Calificar is another of the items that comes up in Otheguy's (1993) article. He claims that its use as an intransitive verb in the Span ish of the U.S. makes it a case of linguistic modeling (32). For example, monolingual Spanish speakers would generally not use the verb in a sentence such as Carlos no califica para ese cargo Carlos does not quali fy for that position' in which calificar d oes not have a direct object. Conversely, they would say Calific a Carlos incapaz de desempear ese cargo He judged Carlos incapable of occupying that position.' In this corpus, the verb is used intransitively all three times, for example Todos califican con nuevos programas Everyone qualifies with new programs,' in agreement with English to qualify.' French qualifier is also intransitive; however, it is mainly used in its reflexive form se qualifier Therefore, it is probably not the origin of influenc e. Regardless of its origin, this innovative use is classified as true linguistic modeling, in accordance with Otheguy's (1993) definition. Mquina contestadora and respondedor are discussed together since they both relate to Otheguy's (1993) debate as w ell. He has stated that the conceptualization of the answering machine as a machine is what makes the U.S. mquina de contestar a cultural modeling, as opposed to the monolingua l Spanish conceptualization of contestador automtico as an automatic answerer The first term, mquina contestadora seems to align itself very well with that description, as the only difference from its U.S. counterpart is the adjective contestadora used in place of the prepositional constructi on de contestar This may very well be due to the Spanish monolingual contestador automtico still exercising some kind of resistance to change or it may be due to the French rpondeur having a the corresponding ending eur exercising its own influence. In any case, the first innovation models the English view of a machine that
81 answers and is classified as a loan translation that displays only cultural modeling. The second term, respondedor must n ecessarily come from French as respondedor already exists in monolingual Spanish but is used as an adjective to denote that which answers. Based on Otheguys (1993) definitions, this second type is also classify this as an example of a loan translation, showing only cultural modeling, as no true linguistic changes seem to have taken place. Reportar does exist in Spanish though monolinguals tend to use it to mean to inform, though generally it is limited to a situation which does not include the reporting of a lost or stolen item. However, in this corpus, the item is used with such a meaning, a use that is also attested in New Mexico and Nuevo and thus not unique to Montreal Spanish. As is the case with renta and rentar this study cannot answer the question of whether these English influenced semantic extensions existed in the speakers repertoires before arrivi ng to Montreal, or whether they have arisen in Montreal and simply parallel their equivalents in U.S. Spanish. The majority of semantic extensions and loan translations in this section are of French origin while some of the ones from English origin, such as aplicacin and el especial were found to exist already in Montreal French as semantic extensions due to influence from English. Consequently, it is found in this section that French exercises a stronger influence in the formation of semantic extensions and loan translations in Montreal Spanish. Nonetheless, there remains one unanswered question regarding those types of English origin renta rentar and reportar that are not documented in Montreal French: Were they generated in Montreal perhaps by the imm igrants that arrived before Quebec language laws imposed French on newcomers or did they exist already in the immigrants repertoires? As the methodology in this study cannot answer the question, it will be left for future studies to address.
82 4.6 Morphosyntactic Innovations This section discusses the types found in the corpus that display morphosyntactic innovations in Spanish due to influence from contact with French or English. Table 4 7 displays the results. As was to be expected (Sankoff, 2 001), there are few documented linguistic innovations at the morphosyntactic level. Mueblado though only having been used once in the whole corpus, is interesting due to the loss of the Spanish prefix a with which the mono lingual adjective is composed ( amueblado ). The adjective is found in reference to an apartment for rent and signifies furnished.' The loss of the prefix is probably due to influ ence from the French adjective meubl which does not contain any prefix. While the number of tokens is too low to make any certain claims, this change does imply that Spanish verbs or adjectives that have the prefix a whose French like counterparts do not contain such a prefix are influenced to lose the a For example, the Spanish verb apadrinar to be godfat her/godparent to' is parrainer in French. If these two terms were to come in contact often enough, apadrinar may be influenced to become padrinar The second documented morphosyntactic change is the phrase esperamos regreses we hope you return. This sequence of verbs represents a possible syntactic change given that the first one is a verb that indicates wish, which in Spanish calls for the second verb to be conjugated in the subjunctive. Spanish, like French and unlike English, normally requires this sequence to be divided by the conjunction que which does not appear in the phrase. Though the conjunction may also be omitted in French, this use is quite infrequent. Monolingual Spanish speakers would most l ikely express the above as esperamos que regre ses As such, it must be speculated that if there truly is a syntactical change occurring, it is probably due to the influence of English. Silva Corvaln (1994) has analyzed the non expression of que in Los Angeles Spanish and suggests that it may be due t o an extension from the written or formal register (in which que is sometimes
83 omitted) into the oral conversational style provoked by the contact situation. Of course, this conclusion may not seem to apply here because the token was found in the written da ta. However, the linguistic changes discussed above reflect changes that are not present in the formal or written register of Montreal Spanish. Hence, it is likely that this phenomenon as well belongs to a more colloquial register. At the moment, with only one token in the whole corpus, it is difficult to make any claims as to whether this represents an actual change that is occurring in the Spanish spoken in Montreal. This would, of course, require further studies with appropriate methodologies. The two changes discussed in this section, while limited in tokens, give a general picture of the kinds of morphosyntactic changes that can be expected in this language contact situation. While mueblado is a clearer case, esperamos regreses requires spontaneous Montreal Spanish speech and a measurement of the intensity of contact with the Francophone vs. Anglophone communities, along with the use of the subjunctive in the contact French, to establish if there is indeed a change that is occurring. 4.7 Additional Changes/Processes This section contains words and/or phrases that do not neatly fit in any of the categories discussed above because more than one simple process has led to the formation of the linguistic innovation. It also contains some words or phrases that can neatly fall into one of the categories above but that appear as variants of other terms that fall into this category. As such, the variants have all been grouped together in this section so that they may be analyzed collectively. Table 4 8 lists all types and tokens, with the ones to be discussed together grouped as a set. The first set is composed of the compound sous sol and all its variants. Before any analysis is conducted, i t is necessary to specify that sous so l in French has various meanin gs, among which is basement. In the sources for this corpus, sous sol and its variants all refer to
84 the basement of a home. The variant with the most tokens is the French monolingual sous sol With 16 tokens in all, it is a nonce borrowing or code switc h spelled exactly as it is in French. The next variant is subsol(es) or its hyphenated version sub sol(es) With a cumulative total of 13 tokens, this term presents both calquing and borrowing. The second word in the sequence, sol means ground,' among o t her things. This explains why sous sol means basement,' literally under (the) ground.' Hence, in the second and third variants on the list, sol is borrowed from the French 2 and sub is the Spanish corresponding prefix for sous The calque occurs in the us e of the Spanish prefix coupled with the borrowing to form the full phrase. Sub suelo(s) and the non hyphenated variant subsuelo(s) cumulatively appear six times in the corpus. These are semantic extensions of the existing monolingual word subsuelo While the monolingual term refers to subsoil the innovative use in Montreal has extended to include the inhabitable room beneath the ground floor of a building. The next six types on the list are either used as adjectives to describe a person or thing that or iginates from the province of Qubec or as nouns to refer to people from Qubec. The main feature that stands out in both the noun and adjective forms is that there seems to be no consensus at the moment on how to end the root quebec For the noun, both l os quebecois/los quebecuas 3 and los quebequeses are used. The first variant is taken directly from the French qubcois,' while the second has maintained only the root of the word and has substituted the Spanish suffix s (sg.)/ eses (pl.) for the French ois Similarily, the adjectives also feature two possible suffixes, either the French ois (or its Spanish spelling ua ) or the Spanish suffix ense commonly used for adjectives of nationality or 2 Although sol does exist in Spanish and means sun ,' it is being assumed here that sol is a lexical borrowing and not a calque of the Spanish form, else the term subsol/sub sol would mean under the sun .' 3 The form spelled quebecua(s) is present so that it may reflect the possible spelling of the word in Spanish. As the word was encountered in the radio recordings as well, it cannot be assumed that it is always spelled in the French form: qubcois
85 origin. What is noteworthy about the adjectives with ense is that their spellings all reflect different pronunciations for the word. Quebequense is pronounced [k $ $ .k $ n.s $ ] and retains the voiceless velar plosive with which the province's name ends while quebecense is pronounced [k $ $ .s $ n.s $ ] if read in Spanis h and therefore loses the [k] to the voiceless alveolar fricative [s]. The sound of the voiceless alveolar fricative is caused by the sequence ce however, it can be speculated that the letter c is maintained in the adjective precisely because Qubec ends with a c which would make the change in pronunciation unintentional. This cannot be determined until a pronunciation in order to ascertain how the word is being pronounced. For the present moment, the only claim that can be made is that there seems to be indecision on how the word should be spelled. Finally, quebeqense is pronounced [k $ $ .kw $ n.s $ ]. This does seem to be an intentional spelling given that the only reason to use the trema above the u in Spanish is to indicate that the letter must be pronou nced. If such is the case, this may have been done under the influence of the French qubcois pronounced [ke.be.kwa]. In this way, although both quebecois/quebecua and quebeqense have different suffixes, quebeqense can retain the segments [ ku ] as dis plays the French adjective. The next set of types to be discussed have been generated due to the influence of the French noun dfrisage in Montreal, used by the hairdressing industry to denote a straightening of curly hair. The closest term in monolingu al Spanish is desrizado literally an uncurling.' Defrisado and desfrizado are the most frequently occurring terms, appearing in the corpus four times each. The first one, defrisado shows a morphological change of the French suffix age in the word dfris age to the Spanish ado One may question as to why aje was not used instead, as is the case in one of the following words to be analyzed. One possibility for this might be that the Spanish desrizado has retained some influence where the suffix is concern ed. Desfrizado is
86 simply one letter away from its Spanish monolingual corresponding item; a letter f has been inserted, probably due to influence from the French term. Defrisage shows up three times in the corpus and can be only analyzed as either a nonce borrowing or code switch from French. Defrizage is almost identical to defrisage but has the letter z in place of the s This is probably due to influence from the Spanish monolingual variant but can still be classified as either a nonce borrowing or a cod e switch. The last of the group is defrisaje Remarkably close to the aforementioned defrisage this term occurs but once in the corpus. Again, it seems to be either a borrowing or code switch but with an added morphological change that shows further integ ration of the type into the Spanish language of Montreal: the French suffix age has been replaced by the Spanish aje The next four types all display a spelling change from their Spanish monolingual counterparts under the influence of French or English Retroprojector injeccin projeccin and projecto(s) are spelled retroproyector inyeccin proyeccin and proyecto in monolingual Spanish. Taking into account dialectal variation, Spanish speakers generally pronounce y as [j], [ % ], [ ], [ # ], [ & ] or [d # ] while j may be pronounced [x] or [h], though they are not in free variation and speakers have only one allophone (Guitart 2004, 107, 119, 122, 140, 162). In French, j is generally pronounced [ # ] while in English it is normally pronounced [ ] If the words were to be read with Spanish phonetic allophones of /j/ in their new spelling they would be pronounced differently than would be their Spanish monolingual equivalents. Hence, two possibilities arise: Either the letter j in Montreal Spanish is acquiring th e allophones that currently correspond to the letter y in monolingual Spanish and the words are being pronounced exactly as they are in monolingual Spanish, thereby making the change purely a written one, or the letter j in Montreal Spanish is acquiring a new allophone, one that comes closest to Spanish
87 speakers' pronunciation of [ # ] or [ ] Again, further studies are needed in order to determine exactly which process is occurring and which language, French or English, is exercising more influence. The nex t group of words and expressions reprise(s) reprise(s) de banque reprise de finances reprises bancarias and tisaje will all be discussed together given the unique process that is the common denominator among them. All reprise and tisaje are established lexical borrowings from French that have undergone semantic extensions in their use in Montreal. Although reprs / reprise is found in both Actual and Uso the definition of the established borrowing does not match that with which it is used in this corpus. U so defines reprs/reprise as "[c]apacidad del motor de un automvil para acelerarse mucho en poco tiempo" while Actual contains an entry with the phrase reprise bancario However, this turns out to be but a metaphor of the aforementioned definition: "Un re prise bancario El sector bancario, cuyo ndice avanza vertiginosamente hacia el 300 por 100." It is clear that in this sentence, reprise bancario is being used to describe a pickup or rapid acceleration in the bank's index. In this corpus, all references to reprise(s) are to indicate property that has been repossessed by a financial institution. All advertisements that contain the phrases in question are from real estate companies selling repossessed homes. Of the lot, reprises bancarias is the type that o ffers two possibilities of interpretation since the phrase can either be considered as two separate words with reprises being either a nonce borrowing or a code switch that can be applied in other cases as well and bancarias being the adjective in Spanish or it can be considered a compound word that may or may not become a fixed expression. It is also important to note that reprise de finances can potentially be of French or English origin given that finances' is a word on both languages. However, it is be ing assumed in this analysis that the item is probably of French origin because
88 the items reprise and reprise(s) de banque in the same group of types come from French. Furthermore, given that the first word of the phrase reprise de finances is of French or igin, the whole phrase is probably of the same origin. Tisaje from the French tissage shows a spelling change from the original French word that indicates a change in pronunciation as well. Originally used in the textile industry, this term appears in t he advertisements of hair salons in Montreal to denote a hair weave' or extensions.' While Spanish does have its monolingual variety, namely tejido this word has proven remarkably easy to borrow, given the written similarity of the French suffix age wi th the Spanish one aje While the suffixes are corresponding, their pronunciations are different, the French being pronounced [a # ] and the Spanish [ax $ ] with some dialectal variation in the fricative consonant. The change in pronunciation that accompanies the change in spelling reflects a higher level of integration of the type. The influence of this French was previously seen in remorcaje and continues to be explored below. The next set of tokens garage etiquetage maquillage and despistage consists of established lexical borrowings of French origin whose only change in this corpus is in the spelling of the suffix. While all words are spelled with the suffix aje in non contact Spanish, here they are seen to have the French suffix age However, no claim s can be made on how these words are actually pronounced by Spanish speakers in Montreal, since the sequence age in Spanish is roughly pronounced [a.x $ ], as is the sequence aje Thus far, the change can only be said to be in the written form. Immobiliar io/a and immigracin share the same spelling change and are therefore discussed together. Immobiliario/a are adjectives, masculine and feminine respectively, used to signify real estate.' Immigracin immigration' is a noun. It is important to note that t his change
89 is being analyzed for written components only, as this phenomenon almost certainly does not affect the pronunciation of the word. All words in non contact Spanish are spelled with the prefix in : inmobiliario/a and inmigracin The spelling cha nge, therefore, is from the prefix in to im which is the prefix used in the French immobilier (adj.), immobilire (adj.) and immigration and the English immigration.' As [m] and [n] have a very slight difference in point of articulation, inm is prono unced [imm] in standard monolingual Spanish due to assimilation As no claims can be made about the pronunciation of these innovative spellings, the analysis will limit itself to the directio of influence. While French seems to be executing some influence on the spelling of Spanish words that begin with inm causing them to be spelled with imm it must also be noted that immigracin is also very similar to English immigration' and so for this type the influence may be coming from both languages. Immobili er the last of the types in this group, is spelled exactly as it is in French and is therefore not showing any particular spelling changes. It is therefore a non established item of French origin that shows variation with immobiliario/a Hypoteca mortgag e' (noun) and hypotecario/a mortgage' (adjective) are spelled hipoteca and hipotecario/a respectively in monolingual Spanish. The likelihood for these two innovative terms is that the monolingual Spanish pronunciation is still being retained in the new sp elling as the first syllable of French equivalents hypothque and hypothcaire is pronounced [i] exactly as it is in Spanish. As a result, French is most probably only exercising an orthographic influence in this particular case. Livraison delivery' is t he next type on the list and is spelled exactly as it is in French. The word appears in two different phrases : Livraison comercial y residenciales and Servicio de livraison In both instances, it is the only word written in French while the remainder of th e
90 phrase is written completely in Spanish. The first phrase calls attention because it is used to mean commercial and residential delivery,' however, this phrase may be a modeling of either the French or English equivalents. The Spanish equivalent for de livery' is entrega and the phrase normally used to indicate home/residential delivery' is entrega a domicilio With regards to the French way of expressing this, the phrase livraison domicile is commonly used, which resembles the monolingual Spanish for m, save for livraison It would seem that livraison commercial y residenciales consists of the borrowing of livraison from French and a loan translation of the English phrase using the existing Spanish word residencial However, if the phrase is also used in spontaneous French speech, in place of livraison a domicile then certainly this constitutes added pressure for the formation of the borrowing/loan translation. In order to be able to make such claims, it must first be determined if the expression livra ison commerciale et rsidentielle exists in spontaneous speech. While Le grand dictionnaire terminologique (Office qubcois de la langue franaise) does not document any other use save for livraison domicile it is possible that the phrase is being used without having yet been documented. The type livraison alone appears in servicio de livraison and consists of a non established French origin item. While it only contains one token and no spell change to reflect a Spanish pronunciation, for example livre son if it is considered together with its use in livraison commercial y residenciales and the two cumulative unique appearances, this type will probably become more diffused in the future. Desabollaje, debosselage and debosallador consist the next set of types to be discussed. The noun for dent removal' in French is dbosselage from which debosselage clearly originates. However, desabollaje presents a more intricate process of formation. While monolingual Spanish does have its own noun, desabolladura it seems that this innovation contains the prefix
91 and root of the Spanish noun, namely desabolla coupled with the Spanish equivalent of the suffix in the French noun dbosselage aje In essence, the Spanish suffix dura has been replaced by another Spani sh suffix, aje under the influence of the French noun dbosselage A similar process does occur again with the type debosallador Again, a term employed in the automotive industry, its French equivalent is dbosseleur and is used to designate a person wh o repairs and/or removes dents from the body of a car. Interestingly enough, the Spanish word that comes closest to the French is desabollador and is defined in Diccionario de la Real Academia Espaola as "instrumento que emplean los hojalateros para quita r las abolladuras de las placas metlicas." While the two terms do not mean exactly the same thing, it seems plausible that desabollador may have played a role in the formation of the word debosallador due also to the corresponding endings (Fr. eur / Sp. or ) mentioned above. Indeed, the word seems to be some hybrid of the beginning of the French dbos a nd the Spanish llador with an a replacing the e in the French word. It is not surprising that this type of word formation take place given that debossela ge the French equivalent of desabolladura is also present in this list, with two tokens. Centre ville is downtown in French and is called el centro in Spanish. Not only does this non established item appear twice in the corpus but it gains importance wh en coupled with centro de vil(le) a s both may begin competing with their Spanish monolingual counterpart. Centro de vil(le) is placed in this category is because there is more than one process occurring in its formation. In the first place, the term appea rs in the radio recordings, which is why the spelling of vil(le) cannot be determined. In French, ville means city,' which makes centre ville downtown' literally city center.' In centro de vil(le) the process consists of a borrowing of the word ville an d a loan translation of the whole phrase centre ville Had it been a true loan translation with no borrowing, the phrase would probably have been centro ciudad ; however,
92 since vil(le) has probably been borrowed and there is the presence of the preposition de the actual classification of the phrase becomes blurred. The next two types on the list consist of Rivera sur and Riviera del sur de Montreal Both phrases refer to one of Montreal's suburbs located off the island, commonly called the South Shore in E nglish and the Rive Sud in French. The two Spanish words that approximate in meaning are ribera which means shore,' and rivera which means brook' or stream.' It seems that in order to obtain a literal translation of South Shore' the word ribera shou ld be employed. It may be that rivera is simply a misspelling of ribera as both are generally pronounced the same way since both intervocalic /v/ and /b/ are pronounced ["]. Riviera de sur de Montreal is a little more difficult to explain as riviera is no t a word in Spanish and the only use is to refer to proper names such as the Italian Riviera, the Mayan Riviera, etc. The only possible influence can be the French rivire river;' however, the term is not used in Rive Sud. At this time it is difficult to determine the origin of Riviera though with only one token, its presence does not carry too much weight. Regarnizado appears twice in the corpus from two advertisements of the same dental company. The term is used to refer to a relining' of one's dentur es (adding material to adjust the fit) and is commonly referred to in Spanish as a rebasado Qubec French uses both regarnissage and rebasage Regarnizado seems to consist of the root of the French regarnissage ( regarnis ) and the suffix of the Spanish re basado ( ado ). Feta canadien is a type of cheese which, if looked at as a compound word would be considered a possible (nonce) borrowing or code switch from French. However, if analyzed as two separate words, feta remains a proper name for the cheese whil e canadien is the French adjective for Canadian In Spanish, if the phrase were to be treated as two separate words, it
93 would be called feta canadiense using the Spanish adjective for Canadian.' The issue, then, is whether canadien is simply a French ori gin non established adjective being used to describe the cheese or whether there is another process occurring with adjectives that are similar in French and Spanish, such as canadien (Fr.) and canadiense (Sp.) whereby the French ending en is exercising an influence over the Spanish ense Examples of such adjectives would be those of nationality such as Spanish costarricense Costa Rican' and nicaragense Nicaraguan ,' whose French equivalents are costaricien and nicaraguayen Given that there is only one token in the whole corpus, no definite conclusions can be made; however, future studies should be conducted with this possibility in mind. Pharmacutico has been placed in this category for the simple reason that, while spelled pharmaceutique in French an d pharmaceutical' in English, it is spelled farmacutico in Spanish. That is, the first phoneme [f] is represented by the letters ph in English and French but by the letter f in Spanish. Although Old Castillian did use the digraph ph to represent the phon eme [f], this has not been the case since 1803, which leads to the likelihood that this new spelling is due to the influence of either French or English. Any future studies conducted on the written forms of Spanish in contact should also look to see if thi s phenomenon is extending beyond the one token found for this study. This section has discussed a variety of tokens in which some cases represented simply one letter spelling changes, such as hypoteca or immobiliario/a while many others showed more comple x innovations that consisted of suffix replacements for Spanish words under the influence of existing similar French words, for example, defrisaje and desabollaje These produced the generation of brand new words. Though there are a variety of changes docu mented here, is must be noted that the majority was due to influence of French.
94 4.8 Quantitative Analysis This section presents the quantitative analysis conducted on certain words that were found to be in variation with their counterparts. The goal in thi s analysis is an initial attempt to establish which language exercises more influence on Montreal Spanish. Furthermore, once the variants themselves are discussed, the frequency and percentages will show where the preference lies, either with the innovativ e form or the monolingual one. This will help in assessing how strong the influence really is. The sets of items represent directions that are used in addresses. Jarry Est vs. Jarry Este or Montreal Nord vs. Montreal Norte are common exemplars. Since the o fficial language of Qubec is French, addresses are usually in the official language, which explains why no variation was found with any English items. Table 4 9 presents the tabulated results. The variants above show a predominant influence from French. Indeed, of all sets of variants, only one contains an English origin type: set 4. The remaining variants that compete with the Spanish monolingual ones are all of French origin. This fact further supports the results discussed in the previous sections, sho wing a tendency of greater influence from French either by the sheer number of tokens or in the analysis themselves. With regards to the numbers found in the table above, sets 1 and 2 show a slight preference for the French variant while set 3 shows equal preference. Conversely, sets 4 and 5 show a definite preference for the Spanish monolingual variants. These numbers lead to some interesting interpretations. In the first place, influence from French is found, not only in the form of variants that compete with the monolingual ones but also in the slight preference of those French origin variants over the Spanish monolingual ones, which show a slight succumbing to this influence. However, the slight preference in sets 1 and 2 is challenged by sets 3 through 5, in which the last two show a very high preference for the Spanish monolingual variant. What this indicates is that, while there
95 is influence from French and while there is also some surrendering, there is also a resistance to such influence. In essence, the variation in itself shows the influence from French (as variation can only occur where change is taking place), the slight preference of sets 1 and 2 show some type of yield to this influence, and the high predominance of the Spanish monolingual varia nts in sets 4 and 5 demonstrates the resistance to such influence. If taken as a microcosm of the forces at play in Montreal Spanish, the variation above shows not only language change but also language maintenance. In general, the findings analyzed in t his chapter display linguistic creativity on the part of the users, specifically with the innovative words and phrases found in the Additional Changes/Processes section. They also indicate a generally stronger influence from the French language than from the English. Certain types show more diffusion through a higher number of unique appearances while others display greater integration into the Spanish language through morphological adaptations such as replacing the French suffixes age and eur with thei r Spanish counterparts aje and or With respect to the quantitative analyses, it can be concluded that Montreal Spanish not only displays language change due to contact but also shows language maintenance. The next chapter uses these findings to propose answers to the initial questions that drove this investigation, as well as provide directions for future studies.
96 Table 4 1. Total data Established Lexical Borrowings English origin Non established Items French origin Non established Items Semantic Exte nsions and Loan Translations Morpho syntactic Innovations Additio nal Changes /Process es Total Tp. Tk. Tp. Tk. Tp. Tk. Tp. Tk. Tp. Tk. Tp. Tk. Tp. Tk. Oral 2 2 0 0 2 2 1 5 0 0 6 8 11 17 Written 15 42 18 145 35 259 10 80 2 2 48 157 128 685 Total 17 44 18 145 37 261 10 1 85 2 2 48 165 132 702 Tp.= Type; Tk.= Token Table 4 2. A ssignment of ranks # of Occurrences Rank > 20 1 16 20 2 11 15 3 6 10 4 1 5 5 1 When the total number of types does not reflect the mathematical total of the addition of oral and written types, it is indicative of overlap in types; i.e. a particular type has both written and oral tokens.
97 Table 4 3. Established lexical borrowings Type Tokens Total: 44 Rank Unique Appearances References mouse 8 1 Uso parking 7 4 3 Actual, Uso show 5 1 Actual, Nuevo, New Mexico, Uso parqueo 5 2 Nuevo, Uso muffler 2 1 Uso scanner(s) 2 2 Uso bifteck 2 1 Actual, Uso stage 2 1 Actual, Uso Noel 2 2 Actual mis en plis 2 5 1 Actual stock 1 1 Actual, Nuevo, Uso cake 1 1 Nuevo, Uso cash 1 1 Actual t icket 1 1 Actual, Nuevo, Uso gadget O: 1 T: 1 1 Actual feedback O: 1 T: 1 1 Actual estress 1 1 Actual O = oral tokens; T = total tokens
98 Table 4 4. English origin non established items Type Tokens Total: 145 Rank Unique Appearances tax/tx 91 1 15 flyers 15 3 1 banner(s) 10 4 1 steackers 6 1 el/la scrap 4 2 switch 3 1 wave(s) 3 1 bachelor 2 2 twist 2 1 biper 1 1 buss boy (girl) 1 1 guestlist 1 5 1 jacket(s) 1 1 no cover 1 1 off. (office) 1 1 line up 1 1 tip 1 1 website 1 1
99 Table 4 5. French origin non established items Type English Gloss Tokens Total: 261 Rank Unique Appearances est east 143 1 70 nord north 21 9 canet(s) can 13 3 1 coin corner 9 6 site webs websites 8 1 demarror car starter 7 4 3 ouest west 7 4 remorcaje towing 6 2 bureau/bur. office 5 3 revenu revenue 4 1 brochure(s) brochure 3 1 tress & tissage tress & weave 3 1 saln de coiffure hair salon 2 2 dosier file 2 1 asermentada (asermentar) certified 2 2 march (latin o) store 2 2 placement gratuit free job placement 2 1 termopon heat pump 2 1 depanneur/depaneur corner store 2 2 dmarreur distance car starter 1 1 ecran screen 1 5 1 el abattoir slaughter house 1 1 commis de bureau office clerk 1 1 curri er courier O: 1 T: 1 1 el semanario latino de mayor circulacin au (en el) Canad the most circulated weekly periodical in Canada 1 1 pulet (poulet) chicken O: 1 T: 1 1 emploi job 1 1 gateaux cakes 1 1 horaire (de apertura) opening hour s 1 1 argent comptant cash 1 1 casa jumele semi detached house 1 1 patisserie bakery 1 1 logement dwelling unit 1 1
100 Table 4 5. Continued epicerie grocery 1 1 Prepose aux Beneficiaires orderly 1 1 admirer admire 1 1 decoration decoration 1 1 O = oral tokens; T = total tokens
101 Table 4 6. Semantic extensions and loan translations Type Tokens Total: 85 Rank Unique Appearances carta(s) 40 1 10 especial(es) O: 5 W: 11 T: 16 2 16 tiempo pleno 10 3 6 aplicar/aplicacin 6 4 5 renta (noun) 3 3 rentar (verb) 3 3 calificar 3 2 mquina contestadora 2 5 1 reportar 1 1 respondedor 1 1 O = oral tokens; W = written tokens; T = total tokens Table 4 7. Morphosyntactic innovations Type Tokens Total: 2 Rank Unique Appearances mueblado 1 5 1 esperamos regreses 1 1
102 Table 4 8. Additional changes/processes Type English Gloss Tokens Total: 165 Rank Unique Appearances sous sol basement 16 2 8 subsol(es) basement 7 4 4 sub sol(es) basement 6 3 sub suelo(s) basement 4 2 sobsuelo(s) bas ement 4 5 2 subsuelo(s) basement 2 1 s/sol, s sol basement 3 3 quebecois/quebecua (adj.) Quebec (adj.) O: 2 W: 3 T: 5 4 quebequense (adj.) Quebec (adj.) 4 3 los quebecois/quebecua(s) (noun) the Quebecer(s) O: 2 W: 1 T: 3 3 quebecense (adj. ) Quebec (adj.) 2 5 2 quebeqense (adj.) Quebec (adj.) O: 1 T: 1 1 los quebequeses (noun) the Quebecer(s) O: 1 T: 1 1 defrisado hair straightening 4 2 desfrizado hair straightening 4 1 defrisage hair straightening 3 5 1 defrizage hair straigh tening 2 1 defrisaje hair straightening 1 1 retroprojector projector 4 1 injeccin injection 4 5 1 projeccin projection 2 1 projecto(s) project 1 1 reprise(s) repossession 3 1 reprise(s) de banque bank repossession 3 1 reprises bancarias b ank repossession 3 5 3 reprise de finances repossession of finances 1 1 tisaje weave 2 1 garage garage 11 3 3 etiquetage ticketing 2 1 maquillage make up 1 5 1 despistage preventive medical exam 1 1
103 Table 4 8. Continued immobiliario/a real e state (adj.) 19 2 6 immobilier real estate (adj.) 2 5 1 immigracin immigration 7 4 4 hypoteca mortgage (n.) 4 5 2 hypotecario/a mortgage (adj.) 4 1 livraison commercial y residenciales commercial and residential delivery 3 5 1 livraison delivery 1 1 desabollaje dent removal 3 1 debosselage dent removal 2 5 2 debosallador dent remover 1 1 centre ville downtown 2 2 centro de vil(le) downtown O: 1 T: 1 5 1 Rivera sur South Shore 1 5 1 Riviera del sur de Montreal South Shore 1 1 regarniz ado relining 2 5 1 feta canadien Canadian feta O: 1 T: 1 5 1 pharmaceutico pharmaceutical 1 5 1 O = oral tokens; W = written tokens; T = total tokens
104 Table 4 9. Quantitative data Set Type Frequency Percentage este 95 40 1 est 143 60 Total 238 1 00 norte 19 47.5 2 nord 21 52.5 Total 40 100 oeste 7 50 3 oest 7 50 Total 14 100 oficina/ofic. 25 81 4 bureau/bur. 5 16 off. 1 3 Total 31 100 esquina/esq. 72 89 5 coin 9 11 Total 81 100
105 CHAPTER 5 CONCLUDING REMARKS 5.1 Summary of Findings The purpose of this study was to investigate the variety of Span ish used in Montreal in an attempt to answer the following questions: Question 1 What linguistic evidence exists in Montreal Spanish that exemplifies its contact with French and English? Question 2 Subsequent to having provided an inventory and propose d categorizations, what distribution patterns are present in the oral and written data analyzed? Question 3 Which of Montreals two main languages, French or English, exercises a greater influence on Spanish? After having gathered, cross referenced, clas sified and analyzed all the data, it becomes obvious that there is definitely evidence of linguistic contact in the Spanish of Montreal that reflects its contact with French and English. In reference to Question 1, the results show that linguistic contac t in Montreal Spanish is evidenced in the forms of lexical borrowings (established or not), semantic extensions, loan translations, morphosyntactic changes and other outcomes that display more than one process of change. As was determined earlier, the smal l number of oral findings (n=17) as compared to written ones (n=685) has led to their analysis as support for the patterns in the written data. With regards to Question 2, the results confirm the hypotheses. The majority of the data is in the form of lexic al items, established or not, representing 64.1% of the data. The items in the Additional Changes/Processes category amount to 23.5%; however, when one considers that some of the items in this category are lexical items that display more complex changes or processes, the percentage of lexical results increases slightly. Semantic extensions and loan translations constitute 12.1% of the data and morphosyntactic innovations make up, as expected, the least amount of innovations, making up 0.3% of the data.
106 I n answer to Question 3, some facts must be considered. Spanish in contact has a longer history in the United States that it does in Montreal. Mexicans, particularly, have been the focus of many linguists studies such that dictionaries, i.e. Chicano and Ne w Mexico have been created reflecting their particular variety of spoken Spanish. Although Mexican immigration patterns are different in Canada than in the U.S., it is noteworthy that several of the terms and phrases found in Montreal Spanish exist already in Chicano Spanish. This clearly indicates that assumptions made about English influence in Montreal Spanish must be done carefully, as some items may have belonged already in the repertoires of the immigrants before arriving to the city. Furthermore, wit h the language of commerce and globalized media being English, it is very difficult to establish the extent to which the quantity of contact with English in Montreal and how much effect it has on Spanish in the city. French, however, presents a different c ase due to lack of contact between the two languages elsewhere in the world except for limited parts of Spain and the Caribbean. The French origin findings, therefore, reflect the particular contact between Montreal French and the Spanish language spoken i n Montreal by Hispanic immigrants and their children. The data also show that the language of major influence on Spanish is French. This is seen throughout the analysis. Of the total eighteen types in the category of established lexical borrowings, five a re of French origin while thirteen are of English origin. Nonetheless, of those thirteen, ten are items that appear in Le grand dictionnaire terminologique This shows that even in the category of items that should show unique influence of English, there i s still a strong possibility of influence from French. This same phenomenon is observed with the category of English origin non established items, albeit only three of the eighteen types in that category appear in Le grand dictionnaire terminologique The French origin non established data presents
107 all items of French origin and contains, as stated above, the greatest number of tokens. The semantic extensions and loan translations category shows that five out of the ten types are of French origin while the origin of the other five cannot be determined in this study, while one out of the two types in the morphosynyactic innovations is influenced from French. Finally, the Additional Changes/Processes category shows an overwhelming influence from French, with thirty eight of the total forty three tokens displaying a French origin. The data also demonstrate various trends, one of which demonstrates a heightened level of integration of some of the non established items into the Spanish language. For example, ma ny of the French origin items that are spelled with the suffixes age and eur in French are being spelled with the Spanish corresponding suffixes aje and or Other examples show the opposite trend, though these consist of items that exist already in th e Spanish language, whose spelling changes to mirror that of French or English. This is shown in words that end with aje in Spanish and whose spelling is changed to age (French suffix). Others include those spelled with the letter j as in French or Eng lish, as opposed to the letter y Two examples are projecto and injeccin Other important findings are those of the semantic extensions and loan translations, where true linguistic evidence of contact was found through semantic extensions but cultural con tact was also found in the loan translations that displayed no linguistic innovation. The quantitative analysis served to determine two things. First, it reinforced the findings in the qualitative analysis that showed a majority of influence from French, a s the variants in competition with the Spanish monolingual counterparts were mainly of French origin. Secondly, it also displayed a certain amount of language maintenance through a strong preference for the Spanish monolingual counterparts in some cases. T his finding parallels Lamarre & Dagenaiss (2004) findings that immigrants and second generation youths value the maintenance of a
108 heritage or minority language: by maintaining Spanish, youth may claim membership not only in the Latino communities of Montreal or Vancouver, but also in language communities in their country of origin and other Spanish speaking nations (56). Furthermore, Meintel (2002) finds that even in mixed unions in Montreal, many parents make it a priority to transmit pluralism t o their children, which includes placing emphasis on a language other than French and English. Regardless of the fact that this investigation did not interview actual speakers, the sources reflect a shared repertoire of the Hispanic community in which pref erence for some monolingual Spanish elements is clearly displayed. Of particular interest are the variants est nord and ouest because, used after street or borough names, they serve to support one of two possibilities. If viewed as discrete types, their h igh number of occurrences supports the dominance of French influence as compared to English. However, if viewed as parts of proper nouns, the presence of Spanish monolingual variants used in their place, and sometimes preferred, shows an effort to countera ct the influence and maintain the Spanish language. 5.2 Limitations of the Study and Directions for Future Studies 5.2.1 Limitations Some methodological limitations in this study act as obstacles to more definitive conclusions. The principle limitation o f this study is the lack of spontaneous speech, which would most likely have yielded more numerous and varied results, specifically of the calque and morphosyntactic natures. The radio recordings were initially selected as a source with the hope that the s hows would feature call in speakers, which would allow the analysis of spontaneous speech and possibly provide evidence of more complex morphosyntactic innovations. Unfortunately, the radio did not include a listener call in segment and thus the findings w ere not numerous or varied as originally hoped. Their small number features a more limited variety of occurrences than did the periodicals; however, they support the findings in the print data.
109 Another drawback that results from the use of periodicals is the lack of surrounding grammar in the advertisements from which the innovations were selected. Only two of the findings, tip and reportar come from articles, which means they appear in a complete sentence. The first is a non established item and the seco nd is a semantic extension. The remainder of the items comes from advertisements that contain isolated words or phrases. As discussed in the literature review, the presence of surrounding grammar allows comparisons to be made with monolingual varieties and determine if a given item is a borrowing or a code switch. The absence of surrounding grammar in this study has not allowed a distinction to be made between nonce borrowings and code switches. Finally, it must also be kept in mind that the nature of the periodicals, from which the majority of the results was extracted, does not allow for any background information of interlocutors to be collected. In essence, the absence of live speakers means that it is not entirely known to what extent Hispanics produce the material found in this study. Non natives who have knowledge of Spanish and may be using online translators could potentially have produced the results found in this corpus. However, it must be kept in mind that an editorial team runs the periodicals. Clearly the items making the most numerous appearances are being reprinted time and time again without any interceptions or corrections. The recurrent nature of these items in print increases the readers exposure to such innovations and, in turn, the p ossibility that the innovations will become part of the readers speech. 5.2.2 Directions for Future Studies The study of Spanish in Montreal is particularly challenging for a number of reasons. Montreals official language is French but it also has a la rge Anglophone community. Therefore, the Hispanic community is in contact with both French and English. Although it has been shown here that French influence is dominant over English, it is very difficult to determine just how
110 much contact with each of the languages a given speaker has on a given day, especially since the rest of Canada operates mainly in English. Another challenge to linguists studying this variety is the identification of the speakers native Spanish variety. In contrast to the situation in the United States, Hispanics in Montreal are not as separated into ethnic groups as the Puerto Ricans in New York, for example, or the Cubans in Miami, or the Mexicans in New Mexico. While still retaining their individuality in many ways, Hispanics in M ontreal are also very united and the city does not display discrete neighborhoods such as a Colombian neighborhood or a Mexican neighborhood. Although Hispanics tend to concentrate in specific areas in the city, they are heterogenic among themselves; that is, their areas of dwelling feature Hispanics from a variety of different countries. While their speech often reflects their origins, contact among themselves may alter their spoken Spanish. This presents an added challenge to linguists because this situat ion makes it difficult to identify exactly which variety they speak. Hence, one important question for future studies in this Spanish variety is whether Hispanics in Montreal constitute a speech community in themselves or if studies should attempt to separ ate Hispanic speech communities in the city according to their country of origin, place dwelling in the city, etc. The year of arrival to the city will be the determining factor in this question. Those born in the city of Montreal or those who have been in the city for a number of years constitute a speech community as they probably share features common to the community repertoire. On the other hand, recent immigrants belong to a different speech community because their reduced quantity of contact will not reflect the speech of the Montreal Hispanic community at large. Finally, once a properly outlined speech community has been defined according to place of birth and number of years in the city, detailed information about the speakers should be obtained a nd a study of their spontaneous speech should analyze what types of innovations are
111 taking place, specifically at the phonological and morphosyntactic levels. Again, the amount of French influence over English is of importance here because French, being a Romance language, has a similar syntax to Spanish and it may actually exercise less structural influence than does English. It is also important that the study have at least three added groups as points of comparison: a Montreal French monolingual group, a Montreal English monolingual group and a Montreal French/English bilingual group. The first two groups, however, may be difficult to find as Montreal mainly features French/English bilinguals of various degrees of linguistic competence. Be that as it may, further research in this culturally and linguistically fertile region shows great potential for future research. 5.3 Conclusions of the Study In spite of its limitations, this study has provided interesting data in its effort to investigate Spanish langu age contact in Montreal. Whereas linguistic innovations are normally employed and appropriated by the speakers for lengthy periods of time before making their way to a dictionary or printed material, this study has, in some ways, gone deeper than tradition al studies in that it has searched for evidences of linguistic contact in printed material, innovations that have made it past the speakers mouths and onto the written field. The importance of these innovations cannot be ignored, for written material carr ies not only permanence in its nature but also allows for more diffusion as it reaches many readers. This undoubtedly points to the importance of the frequently occurring items discovered in this study, particularly those of French origin. Furthermore, t he average reader of the periodicals may be working class or upper middle class and the periodicals are available for free in some grocery stores, convenience stores and Latin American food specialty stores. A similar study using literary magazines would c ertainly not have yielded the same results. Due to the absence of the observers paradox, the
112 periodicals used in this study are the best written representations of spontaneous speech aside from personal letters, emails, IM, text, etc. Their advantage over the latter forms of communication is that the frequency of occurrences and the variety of advertisements over which a given phenomenon occur, coupled with the intended ideal reader for which the periodicals are printed, all lend validity to the attestatio n of the frequently occurring items and their pervasiveness in the Hispanic community. The object of this study was to find linguistic evidence of Spanish in contact with French and English in Montreal. Without a doubt, it has succeeded in finding that t here is indeed evidence of such contact and that the evidence displays certain patterns of distribution. A comparison of the direction from which the influences come shows a higher proportion of French origins than English. The most frequent items found in the results are French origin non established items, while the least frequent are morphosyntactic innovations. However, even English origin items were found to potentially have entered Montreal Spanish through French. In addition, this study provides furt her concrete evidence of an added variety of Spanish in the world. With only one study having been conducted on Spanish in Montreal, this particular variety has been virtually untouched by researchers. This community has immigrants from various Hispanic c ountries and no longer consists of first generation Hispanics alone. The immigrants that arrived in the five main waves have begun to have children and the city now has second and third generation Hispanics. With a community this large, and growing, there is no reason to believe that these immigrants and their children would lose their native language, specifically not in a multicultural setting such as Montreal, where diversity is encouraged and celebrated. Rather, Spanish in Montreal is expected to endur e, albeit with its own characteristics. Herein lies the value of this study. It has
113 investigated a variety of Spanish of which next to nothing is known and has discovered many linguistic innovations, some of which are reflected in as spelling changes. This last detail is extremely important because studies involving spontaneous speech frequently overlook the written expression of a language in contact. Furthermore, the study shows promising results for future studies both in language change and in language maintenance due to contact Finally, this study provides further support of some of the hypotheses put forward by researchers of trilingualism. The findings in this investigation agree with Stavan & Swischer (2006), whose study shows that, in a trilingual s etting, one of the languages with which the L1 is in contact will dominate with respect to influence, in this case, French. Furthermore, the findings also support Silva Corvalns (1994) theory that languages permeability to change depends on the parallel structures of the L1 and the L2 and Clyne & Cassias (1999) study, which shows convergence of two similar languages in a trilingual setting In this investigation it is seen that, as Spanish and French are more structurally similar than Spanish and Englis h, French exercises greater influence on Spanish.
114 APPENDIX SAMPLE DATA NOT SELE CTED FOR ANALYSIS Figure A 1. Monolingual French advertisement not selected Figure A 2. Bilingual French/Spanish advert isement not selected
115 Figure A 3. Sample indiscer nible origin. Entre between exists both in French and Spanish with the same spelling and meaning; hence, there is no way of distinguishing any language change.
116 LIST OF REFERENCES B langer, Claude. (2000, August 23). The language laws of Quebec. Retrieved August 7, 2008, from http://faculty.marianopolis.edu/c.belanger/QuebecHistory/readings/langlaws.htm Bell, Allan. (1997). Language style as audience design. In N. Coupland & A. Jawor ski (eds.), Sociolinguistics: A reader and coursebook New York: Palgrave. 240 250. Boyd, Monica & Vickers, Michael. (2000). 100 Years of immigration in Canada. Canadian Social Trends Statistics Canada. Clyne, Michael & Cassia, Paola. (1999). Trilingua lism, immigration and relatedness of languages. I.T.L. Review of Applied Linguistics 123 124:57 77. Cobos, Rubn. (2003). A dictionary of New Mexico and Southern Colorado Spanish Santa Fe, NM: Museum of New Mexico Press. Real Academia Espaola. Diccion ario de la lengua espaola Retrieved August 7, 2008, from www.rae.es Diccionario de uso del espaol de Amrica y Espaa (2004). New York: McGraw Hill. Galvn, Roberto A. (1995). El diccionario del espaol chicano / Th e dictionary of Chicano Spanish Lincolnwood, IL: National Textook Company. Godenzzi, Juan Carlos. (2006). Aspectos sociolingsticos del espaol en Quebec. Tinkuy 3:7 18. Gonzlez, Juan. (2001). Harvest of empire: A history of Latinos in America New Y ork: Penguin Group. Grmela, Sonia. (1991). The political and cultural identity of second generation Chilean exiles in Quebec. In S.P. Sharma, A.M. Ervin, & D. Meintel (eds.), Immigrants and Refugees in Canada: A National Perspective on Ethnicity, Multicul turalism and Cross Cultural Adjustment. Saskatoon, University of Saskatchewan. Guitart, Jorge M. (2004). Sonido y sentido: Teora y prctica de la pronunciacin del espaol contemporneo con audio CD Washington: Georgetown University Press. Institut de la Statistique du Qubec. (2008). Immigrants selon le pays de naissance, Qubec, 2003 2007 Retrieved August 7, 2008, from http://www.stat.gouv.qc.ca/donstat/societe/demographie/migrt_poplt_imigr/603.htm Lamarre, Patricia & Dagenais, Diane. (2004). Langu age practices of trilingual youth in two Canadian cities. In C. Hoffman & J Ytsma (eds.), Trilingualism in family, school and community Buffalo: Multilingual Matters. 53 74.
117 Office qubcois de la langue franaise. (2002). Le grand dictionnaire termino logique Retrieved August 7, 2008, from http://www.oqlf.gouv.qc.ca/index.html Mata, Fernando G. (1985). Latin American immigration to Canada: Some reflections on the immigration statistics. Canadian Journal of Latin American and Caribbean Studies 10(20):2 7 42. Meintel, Deirdre. (2002). Transmitting pluralism: Mixed unions in Montreal. Canadian Ethnic Studies 34(3):99 120. Mornigo, Marcos Augusto. (1998). Nuevo diccionario de americanismos e indigenismos Buenos Aires: Editorial Claridad. Multicultural Canada. Encyclopedia of Canada's peoples Retrieved August 7, 2008, from http://multiculturalcanada.ca/Encyclopedia Otheguy, Ricardo. (1993). A reconsideration of the notion of loan translation in the analysis of U.S. Spanish. In A. Roca & J.M. Lipski (e ds.), Spanish in the United States. Linguistic contact and diversity Berlin: Mouton de Gruyter. 21 45. Poplack, Shana. (2001). Code switching (linguistic). In N. Smelser & P. Baltes (eds.), International encyclopedia of the social and behavioral science s Elsevier Science Ltd. 2062 2065. Poplack, Shana & Meechan, Marjory. (1998). Introduction: How languages fit together in codemixing. International Journal of Bilingualism 2(2):127 138. Labov, William. (1981). Field methods of the project on linguisti c change and variation. Sociolinguistic Working Paper Number 81 Retrieved August 25, 2008, from http://eric.ed.gov/ERICWebPortal/custom/portlets/recordDetails/detailmini.jsp?_nfpb=tru e&_&ERICExtSearch_SearchValue_0=ED250938&ERICExtSearch_SearchType_0=no &a ccno=ED250938 Le portail official: Ville de Montreal. Population et demographie, 4 dec 2007. Retrieved August 7, 2008, from http://ville.montreal.qc.ca/pls/portal/docs/PAGE/MTL_STATISTIQUES_FR/MEDIA/D OCUMENTS/09_POPULATION_04%20D%C9CEMBRE%202007_0.PDF Sankoff, Gillian. (2001). Linguistic outcomes of language contact. In P. Trudgill, J. Chambers & N. Schilling Est es (eds.), Handbook of sociolinguistics Oxford: Basil Blackwell. 638 668. Seco, Manuel, Andrs, Olimpia, & Ramos, Gabino. (1999). Diccionario del espaol actual Madrid: Aguilar.
118 Silva Corvaln, Carmen. (1986) Bilingualism and language change: The exte nsion of estar in Los Angeles Spanish. Language 62(3):587 608. _______ (1994). Language contact and change: Spanish in Los Angeles New York: Oxford University Press. Statistics Canada. Ethnic origin (247), generation status (4), single and multiple eth nic origin responses (3) and sex (3) for the population 15 years and over of Canada Retrieved August 17, 2008, from http://www.statcan.ca/bsolc/english/bsolc?catno=97 562 XWE2006015 Stavans, Anat & Swisher, Virginia. (2006). Language switching as a windo w on trilingual acquisition. International Journal of Multilingualism 3(3):193 220. Thomason, Sarah. (1981). Language mixture: Ordinary processes, extraordinary results. In C. Silva Corvaln (ed.), Spanish in four continents: studies in language contact and bilingualism Washington: Georgetown University Press. 15 33. Thomason, Sarah & Kaufman, Terrence. (1988). Language contact, creolization, and genetic linguistics Berkeley: U. of California Press. Torres Cacoullos, Rena & Aaron, Jessi. (2003). Bare English origin nouns in Spanish: Rates, constraints, and discourse functions. Language Variation and Change 15:289 328. Veronis, Luisa. (2006). Rethinking transnationalism: Citizenship and immigration participation in neoliberal Toronto Dissertation, U niversity of Toronto. Weinreich, Uriel. (1953). Languages in contact. The Hague: Mouton.
119 BIOGRAPHICAL SKETCH Maria Ida Fionda obtained her bachelor's degree with honors in Hispanic literature and a major in Italian literature, from Concordia Universit y in Montreal. Upon completion of her master's degree, she would like to continue the linguistic investigation of Montreal's Hispanic community as well as begin investigating the city's Italian community.