Passive Bilingualism in the Iberian Peninsula

Introduction

Authors such as Grosjean and Li (2013), Morgensternová, Sulová, and Schõll (2011) and Calvet (1993) started the discussion about bilingual competence by presenting numbers directly related to the current world language reality: approximately six thousand languages in about two hun dred sovereign states. Mathematically speaking, there are thirty languages for each of the two hundred countries. However, the result is not consistent with the current situation. In fact, the idea of such geographical multilingual units, especially in contemporary Europe, is mostly unrealistic. Taking this into account, we can say with certainty that bilingualism is not only a current, modern and highly studied phenomenon, but also widespread.

In this regard, we can define a seemingly endless number of bilingual environments around the world. Each of these territories is characterized by the languages in contact, which are used in everyday communication by its inhabitants. For example, a Czech language environment, which is officially represented by Czech, could be described as profoundly mono lingual. However, it was influenced by German and later by Russian (the languages of states that were once politically stronger). Moreover, we cannot overlook the role of other languages like Polish or Slovak (the languages of neighboring states) and Vietnamese or Romani (languages of minorities). As a result, we are talking about six languages that more or less influenced the Czech language and have been in contact with it. Still, it is Spain (which is unambiguously multilingual at first glance) will be the main focus of this article, and not the Czech Republic.

Although Spanish is the most widespread language (also referred to as "Castilian"), Catalan (Catalonia, Balearic Islands, Valencian Community), Basque (Basque Country), and Galician (Galicia) play an important role in the respective autonomous regions. As for the exclusive status of the official language and the geographical proximity, Portuguese is the most similar to Castilian. However, the Spanish linguistic and cultural identity was formed by other Romance languages (French, Italian) and, of course, Arabic. In the end, we can conclude that there are more or less eight languages with which Castilian is or was in contact with. The question, however, is how comprehensible they are for the present inhabitants of the Iberian Peninsula and how Spaniards understand it without ever learning them before, and how they could be used for teaching purposes; that is the purpose of this research.

Theoretical Background

Even before defining the term passive or, in other words, receptive bilin gualism, we believe it is necessary to stress on the fact that professional literature in this field of study is not too plentiful and that, to date, only limited attention has been devoted to this topic. However, such focus is also significantly limited to more specifically defined linguistic areas, especially the Germanic countries (Scandinavia, Germany, and the Netherlands). As for Hispanic countries, there has been no research on passive bilingualism in the Iberian Peninsula, and the only investigated area is in Latin America, especially in the contact regions at the Uruguayan-Brazilian border, which complicates the exact placement of the subject under discussion into the particular context of the Iberian Peninsula.

As outlined above, passive bilingualism is synonymous with receptive bilingualism, that is, for language competence, for which there are sig nificant difficulties in using or expressing the second language. Passive or receptive bilingual speakers are more likely to rely on the L1 knowledge and, based on their limited knowledge of the L2, conclude with possible interpretations of a conversation. Grosjean (2010) questioned the pas sive-receptive pair itself, arguing that passive bilingualism does not exist, since many kinds of cognitive processes are triggered by language inter action of any kind. Thus, this excludes any passivity of human perception. In the following paragraphs, we will only express this type of bilingualism as receptive and later justify our usage of the attribute passive.

Historically speaking, receptive bilingualism represented a natural state of interaction between two different languages and was not unusual: its deliberate suppression began about 200 years ago during the rise of national states and hence national identities (Rindler-Schjerve & Vetter, 2007). One of the pillars of nation-state formation was and still is a lan guage that represents national literature and traditions, the founding myth, and the idea of a unified and pure community altogether in timeless per ception (the five main elements of the nation according to Hall, 2006, pp. 52-56). In this "artificially created community," any presence of a heterogeneous language was undesirable. Those who defined themselves as other than monolingual were condemned and stigmatized by the rest of society. Standardization can also be described as the loss of contact with varieties, dialects and language interaction. However, productive bilingualism and receptive bilingualism are entirely different (see Table 1).

Table 1: Comparison of receptive and productive multilingualism

Source:Braunmüller (2007), p. 30.

If we summarize Braumüller's findings, we can conclude that receptive bilingualism is an informal communication that emphasizes the exchange of information at any cost. This communication is based on a great deal of knowledge of pragmatic practices and the situational context, while ignoring the other language components.

Speakers are introducing mutual accommodation agreements that sim plify the understanding of both parties, for example, when dealing with a difficult topic or a complicated situation. Zeevaert and Thije (2007), p. 4 set out two options: solving a problem that is based on slowing down the man ner of speech, more precise pronunciation, repetition, and reformulation, or employment of a so-called "let it pass" strategy. This implies that, in most cases, receptive bilingualism is the result of an agreement between the two parties involved and that only occurs randomly and spontaneously in a few cases (Braunmüller& Zeevaert, as cited in Ribbert & Thije, 2007, p. 78).

The arbitrariness of this type of communication can be gradual and consequently lead to productive competence in the L2, as well as to code-switching.

Receptive bilingualism and semi-communication

House and Rehbein (2004), p. 3 claim that the typological proximity of languages sets the conditions under which speakers can easily acquire receptive knowledge of a close language, or, in a case of a distant language, cannot acquire such knowledge at all. This criterion makes it possible to distinguish between receptive bilingualism and semi-communication.

Zeevaert (2007) introduces the following definitions, bearing in mind the typological proximity of languages:

Receptive bilingualism: A reasonable option of communication be tween languages that are unrelated or only remotely related - under the condition that all speakers involved are familiar with both languages, and provided that the speakers have only a passive competence at their disposal [...] or that the interlocutors prefer to use their own mother tongue in spite of an available active competence [...]. Receptive multilingualism thus provides the opportunity to avoid linguistic discrimina tion in officially multilingual countries such as Switzerland or Belgium.

Semi-communication: The mutual understanding of speakers of closely related languages. [.] Speakers are able to understand the language of their interlocutor due to the genetic proximity of the two languag es and the resulting large typological similarity. (Zeevaert, 2007, pp. 105-106).

Kloss's (1967) divisions on the so-called Abstand and Ausbau languages appear in older studies. Abstand languages are languages that are not typo-logically close and therefore do not show signs of mutual comprehension. Ausbau languages are ones that can be comprehensible to members of neighboring language communities. Thus, on an European semi-commu nication scale, the following languages can be pointed out: Scandinavian languages, namely Danish, Swedish, Norwegian (Haugen, 1966, p. 153), West Slavonic Languages, namely Czech and Slovak (Budovicová, 1987), and, last but not least, all the official and co-official languages of the Iberian Peninsula except the Basque language, since it is not typologically close to any Indo-European language. Hence, it falls into the category of Abstand languages and less possible receptive communication situation.

For a more thorough definition of receptive bilingualism and semi-com munication, we include Zeevaert's (2007) explanation in Table 2 below:

Table 2: Comparison of receptive bilingualism and semi-communication

Source:Zeevaert (2007), p. 109.

Before presenting the research itself, we would like to specify the terminology used, specifically the use of the adjective passive when refer ring to passive-receptive bilingualism. The research we conducted did not involve communication, but one-way passive bilingualism, since partici pants were not expected to have any knowledge of the L2. We also gave no space for any information exchange in a communicational sequence, which is crucial for the assessment of receptive bilingualism. In other words, for our research, it was important to take into consideration only the monolingual competence with the Castilian language as the L1 and, later, to demonstrate the degree of passive comprehension using passive linguistic competences, namely reading and listening. Galician, Catalan, Basque and Portuguese represented the L2.

Research Intent and Hypothesis

The subject of the study is to evaluate a questionnaire in order to answer the research question: whether the monolingual Spanish speakers are able to passively understand other official and co-official languages of the Iberian Peninsula. For a better understanding of the language situation in the context set by us, we will also list the four considered language units from the most to the least understandable, and also according to the results of the questionnaire. The expected results are as follows:

1. Galician (reading, listening)

2. Catalan (reading, listening), Portuguese (reading)

3. Portuguese (listening)

4. Basque (reading, listening)

5. Basque (listening)

The resulting degree of understanding depends mainly on the typologi cal similarity that the isolated Basque does not fully meet, and therefore we also evaluated it as the least comprehensible for any monolingual Spanish speaker. As for the other languages, Galician, Catalan, and Portuguese all fall into the Romance linguistic group. The exact distribution based on avail able linguistic descriptive information is, in our view, impossible, because it completely suppresses the subjective language skills of the speakers that are ultimately crucial and which we are fully aware of. However, from a purely linguistic point of view, we assume that the most comprehensible will be the written and spoken Galician, which, in terms of phonetic, morphological and lexical features, bears a large resemblance to Castilian. We consider the written and spoken Catalan and Portuguese to be less similar, however, much more present in everyday life: A significant part of the monolingual population spends holidays in the Catalan or Valencian coast or in Portugal, where languages come in contact. Also, Portuguese is the only official language of another political entity, and Catalan is currently the subject of continuous political discussions and debates. Lastly, we consider spoken Portuguese as the least understandable right before the Basque, due to the palatalization of the implosive /s/ and /z/. These changes are relatively recent (since the 19^th century), as Hricsina (2015), p. 166 states, and sharply distinguished Portuguese from Castilian.

Methodology

As previously stated, there are very few research projects that would be devoted exclusively to passive bilingualism; the vast majority of them were conducted in the area of Germanic languages, primarily German and Dutch. Researchers who focus on the subject use both qualitative and quantitative research methods.

Due to the extent and complexity of the study, we decided to use the quantitative method of our questionnaire, and the QuestionPro webpage was chosen to create and distribute the questionnaire. The project was assigned its own customized Internet address, available at https://bilinguismopasivo.questionpro.com. It was launched on Thursday, April 6, 2017 and closed on Tuesday, June 6, 2017; it was, therefore, freely accessible for two months. Students and academics from various Spanish universities participated in the survey.

Demographics

A total of 175 respondents took part in the survey during the given two-month period, 132 of which were women and 43 were men. This sum is just a fraction of the total number of participants (i.e., 13.6%). The respondents who did not complete the questionnaire were 1115. We expected such a small percentage of complete responses before the launch of the study, judging by the relatively vast extent of the submitted data. However, the overall evaluation application provided by the application reported an average time for completion of 14 minutes.

In terms of age, we decided to divide participants into five groups that respected the human ontogenesis given by the World Health Organization (Machová, 2008): adolescence (15-18), adulthood (19-30), young age (31-45), middle age (46-60), and old age (60-100). Each development period is typical of both growth and development. From a linguistic point of view, we assume that, despite the repression of Francoist Spain, older generations will have better passive language skills, determined by a longer potential contact with other official and co-official languages, which is proportionally determined by their older age.

However, the majority age group completing the questionnaire was the second, representing the full adult age of 19-30 years, with a total of 115 participants (66%). As for the rest, 27 of the participants (15%) were from the young-age group; 19 (11%) were from the middle-age group; 12 (7%) were from the adolescence group; and 2 of them (1%) were from the old age group.

These results are fully in line with our assumptions. As mentioned above, the questionnaire was distributed through higher-education insti tutions where academic staff forwarded the link of the questionnaire to their students. For this reason, we have also decided to omit the aspect of education, which is, presumably, mainly on the academic level.

The other two applied aspects, origin and residence, were chosen for two reasons. The first apparent objective was to obtain these two demo graphic data and compare them to each other. The second was to prevent the participation of inappropriate respondents; Catalonia, the Basque Country, Galicia, and Portugal were not included as options for choosing answers to these questions.

Most of the respondents were from Madrid (31 % were natives, with 54 respondents, and 36% were residents, with 63 respondents), Castilla y León (26% were natives, with 46 respondents, and 33% were residents, with 58 respondents), and Andalucía (18% were natives, with 32 respondents, and 17% were residents, with 30 respondents). Other places of origin and residence did not surpass 10%.

The following results are perceived as very positive: Madrid, the place of origin and residence of most respondents, is centrally located farthest from all bilingual regions and Portugal, and surrounded by monolingual regions. Castilla y León and Andalucía share the border with bilingual areas or Portugal, but no significant cultural metropolis with a bilingual language environment is in their immediate vicinity.

Instrument

We decided to use the QuestionPro web page. The questionnaire was divided into three blocks, the first consisting of detailed instructions addressed to all participants. Their primary purpose was to exclude inap propriate respondents who did not meet these requirements:

Este cuestionario acerca del bilingüismo pasivo en la península ibérica está destinado solamente a:

• los hablantes monolingües cuya lengua materna es español (cas tellano)

• los hablantes monolingües que viven en una región monolingüe en la cual se habla español (castellano)

• los hablantes monolingües que no entran regularmente en contac to con otras lenguas oficiales y cooficiales de la península ibérica (p.e. comunicación comercial frecuente con región bilingüe, visi tas regulares y repetidas de las regiones bilingües, etc.)

Si no cumple estos tres requisitos, no rellene este cuestionario, por favor.

The purpose of this questionnaire was to find out how the speaker estimates his or her passive knowledge of the languages under examination. It would be ideal to compare the results of this self-evaluation question with the average of the results of the three complementary survey questions, but the used software did not provide such complex operations that would lead to the desired outcome.

These survey questions were asked in order to verify the estimate given by respondents when answering the first self-evaluation question. They were made to be as clear as possible, and the answers could be traced directly in the text or recording; they were not of a meta-textual nature. Their purpose was, in frequent cases, to determine a specific time or correctness of the claim. While creating them, we often focused on words that are not or are a little similar in Castilian.

Examples:

1. Specific time (Basque - listening)

• El hombre dice que son las 9 de la tarde.

• Correct answer: NO (Gaueko hamarrak dira. "It's 10 o'clock in the evening.")

2. Correctness of the claim (Portuguese - listening)

• João quiere convencer a Pedro de que la astrologia predice el futuro.

• Correct answer: NO (Deforma alguma! A astrologia só aconselha, nada prediz! "No way! Astrology gives advice and does not predict anything.")

3. Lexical differences (Catalan - listening)

• La mujer bebe té.

• Correct answer: NO (No exageri, Sra. Sugranyes, vostè pren massa cafè, per què no passa a la tisana."Do not exaggerate, Mrs. Sugranyes, you drink a lot of coffee, why don't you have some tea instead?")

Results

In the following section, we will present the results of a linguistic nature. We will start with the dialogues read by the participants and continue with those they listened to.

Reading - Galician

Table 3: Results on reading a text in Galician

Source: Own elaboration.

As shown in table 3 we observe that the vast majority of the inter viewed respondents understand the text in Galician. A small decrease in the correct answers is observed in the third question. This is explained by a certain degree of ambiguity-some respondents could interpret it so that Helena's the father invites her on Friday for dinner and not for a dinner that will be held on Friday-as well as possible inattention when responding.

Reading - Portuguese

Table 4: Results on reading a text in Portuguese

Source: Own elaboration.

As shown in table 4 we can observe that written Portuguese is under stood by an overwhelming majority. The third question is the exception. As in the previous section, we are vigorously advocating for potential inattention on the part of the respondents, as the passage in which the answer to the question appears is not, in our opinion, morphologically or lexically incomprehensible to Spanish native speakers.

Reading - Catalan

Table 5: Results on reading a text in Catalan

Source: Own elaboration.

As it is shown in table 5 the results show the greatest difference between right and wrong answers, which we again explain by the lack of the respon dents' attention, as the answer to the second question is neither morphologically nor lexically incomprehensible. The owner explicitly says that he is looking for someone who would work with a flexible schedule, and Cristina herself informs him that she can work shifts. In the third question, we assume that roughly half of the speakers did not know the term demà mateix, which is far from the Castilian mismo mañana (See appendix). We consider this part to be the least balanced within the results of reading in general.

Reading - Basque

Table 6: Results on reading a text in Basque

Source: Own elaboration.

The results shown in table 6 are absolutely clear that the great typo logical difference between Basque and Castilian makes it impossible to understand the reading task. For a few right answers, we have several explanations: The respondents have either guessed or read the detailed instructions given at the beginning of the questionnaire distractedly and, therefore, had the active knowledge of the Basque. However, some of them could derive meaning after careful reading, as the Basque words in italics are very similar to those of Castilian (these are loans from Castilian): kotxe - coche, credit - crédito, maletan - en la maleta (See appendix).

Listening - Galician

Table 7: Results on listening a text in Galician

Source: Own elaboration.

Based on the presented results in table 7, we can assume that, as well as reading, the vast majority of respondents understand Galician. A significant decrease can be seen in the second question with a Galician word, which is considerably distant from the Castilian: louroin comparison with Castilian rubio.

Listening - Portuguese

Table 8: Results on listening a text in Portuguese

Source: Own elaboration.

As we expected, the specific pronunciation of Portuguese made it nearly impossible for the Spanish monolingual speakers to understand. The only exception was the word gozar ("make fun of someone"), which has a slightly different meaning in Castilian ("enjoy, look forward to"). In this case, there is a considerable shift from a relatively comprehensible reading task to a difficult and minimally understandable listening task.

Listening - Catalan

Table 9: Results on listening a text in Catalan

Source: Own elaboration.

In table 9 we see far better results compared to the reading compe tence task in the results presented: Catalan was understood by a remarkable majority of respondents. A smaller drop is evident in the second question, in which the lexical difference between Castilian (té) and Catalan (tisana) was again used (See appendix).

Listening - Basque

Table 10: Results on listening a text in Basque

Source: Own elaboration.

As with the previous reading task, we assume that the correct answers were either estimated by the respondents or that they had the active knowl edge of the Basque language. At the same time, however, we must say that the result of listening was better than reading, which is astonishing.

Discussion

To summarize and compare the results, we used the number and percentage of correct answers for each question from each part of the questionnaire. We averaged these numbers ([O1 +O2+O3]/3) and then compiled the resulting order for a part of the reading and listening part. We further averaged these results ([N+P]/2) and received a summary result containing both of the two competencies examined. These results are shown in Table 11 below.

Table 11: Summarized results of the questionnaire

Source: Own elaboration.

In the results for the reading part, we observe that most of the respon dents understood only Galician (82.09%) and Portuguese (68.57%). In Catalan, as mentioned above, the inattention of the respondents led to only about half (48.95%) of them understanding the written form of this language. As anticipated, success in the Basque language accounted for only a fraction (1.71%) of the total number of participants.

Meanwhile, the results show interesting information. Most of the respondents understood Catalan (76.38%) and Galician (75.04%), while Catalan, in a small difference, was more understandable than Galician. In the case of Portuguese, the already mentioned palatalization of the implosive /s/ and /z/ significantly complicated the understanding, and therefore only about 14.47% of the respondents succeeded in this part. The Basque was represented by only a minimum of correct answers (3.43%), yet listening was slightly more comprehensible than reading.

In the last part of the table (the final results) we confirmed the hypoth esis, which determined the subsequent order of languages (from the most comprehensible to the least): Galician, Catalan, Portuguese, Basque. The majority (more than 50%) understood only Galician (78.56%) and Cat alan (62.66%). Portuguese was successfully understood by about a third (41.52%) of the participants. The minimum of successful respondents in the case of Basque accounts for a small fraction of the total number of participants (2.57%).

After a detailed comparison of the results, we found that the respon dents repeatedly erred in the questions that contained:

1. Any verbal expressions or vocabulary that was not similar to Castilian

2. Exact (time) data

We are fully aware that passive comprehension is based primarily on searching for similarities between L1 and the target language. Therefore, it is perfectly logical that respondents were less successful in answer ing questions based on unrelated vocabulary than answering questions of a general nature. These were words such as: blonde (Galician louro, Castilian rubio), tea (Catalan tisana, Castilian té), make fun of somebody (Portuguese gozar, Castilian bromear, burlarse de), right tomorrow (Catalan demà mateix, Castilian mismo mañana).

The results of the questionnaire were also largely influenced by the inattention of the respondents. This was mainly reflected in the reading task in Catalan, but also in both the reading and listening tasks in Galician. As we have already mentioned, our intention was not to confuse the respon dents. We acknowledge that the answers to questions relating to the above parts may have been ambiguous to some of the participants. Nonetheless, the responses could be traced clearly in the text or the audio recording. We also consider it appropriate to specify the average time required to complete the questionnaire, which was estimated by the application tools to be 14 minutes. We believe that, during these 14 minutes, it is not possible to answer all 32 questions (4 questions per section) with care. The recording itself is a total of 4 minutes and 5 seconds (0:44 in Portuguese, 0:59 in Galician, 1:10 in Basque, 1:12 in Catalan), which means that the actual average time response time was somewhat about 10 minutes, i.e., 18.75 seconds per question.

On the other hand, respondents have done well with questions of a general nature that were related to the dialogue as a whole. Such questions were often answered correctly by a vast majority, which corresponds to the definitions of the passive bilingualism that underline the context com prehension as the key one.

As for the pedagogical aspects of the results, we suppose that Galician and Catalan have the potential to be used in the class as a source language since the final survey results showed that the majority understood them. From our point of view, Portuguese should be promoted more at schools in order to become more comprehensible for the student's possible future communication within the Portuguese territory, given the fact that Portugal was the only sovereign state represented by its language in the survey. Better knowledge of spoken Portuguese could also lead to a better intercultural exchange, which is, due to current lack of linguistic clarity, not so often. In terms of Basque, we presume that the typological distance cannot be surpassed easily and the only aspect we could consider to be working for possible implementation of the language at the school is the student's own motivation to master it and understand it. Nevertheless, we strongly believe that all four of them should be present in any class in a purely linguistic form; that means, for example, without any radical political ideologies included.