International universities and implications of internationalisation for minority languages: views from university students in Catalonia and Wales

European higher education institutions are in general highly committed to internationalisation, seeing it as providing ways into the global education market, as an indicator of academic excellence, and for generating income. In multilingual settings, minority languages are not always given adequate consideration in this process and may be a source of tension and ambiguities in the local communities and universities. We report on part of a study of international and home students at universities in two bilingual contexts: the University of Lleida (UdL, Catalonia) and Cardiff University (CU, Wales). The students were asked about their understanding of what an ‘international university’ is, and also, how they saw the implications of internationalisation in the universities for the respective minority languages. Results show some differentiation amongst the student groups. For example, in the case of the first item, the Lleida home students associated international universities with opportunities for travel, exchange and employment advantages far more than the other groups. And in the case of the second, international students at Lleida showed greater negativity and opposition towards the minority language than other groups. Findings highlight the difficulties in reconciling the vitality of minority languages with the forces and the priorities of internationalisation.


Introduction
Higher education institutions in Europe are strongly committed to internationalisation (Woodfield 2010, 170), viewing it as an opportunity to enter the global educational market as well as (1) to raise the status of the university and (2) in the case of non-EU full-fee paying students, for financial benefits. However, institutions do not always regard internationalisation as an opportunity to the same degree since they need to accommodate a new population arriving with specific expectations. One expectation has to do with language use and the sociolinguistic profile of the students' new institutional environment. This is especially relevant for universities in some bilingual communities, where international students within the academic community through offering Catalan and Spanish courses, which seems to embody a perception of a possible decrease in its presence in academic provision.

Cardiff University
Cardiff, the capital of Wales, had a population of around 325,000 in 2011, about 11% of the population of Wales. Numbers of Welsh language speakers in Wales declined considerably through twentieth century (54% of the population of Wales in 1891, 21% in 1971, 19% in 1981 and 1991). After some indications of a revitalisation (21% in the 2001 Census), numbers fell back to 19% in the 2011 Census. Cardiff is a relatively anglicised city, with about 11% of its population speaking Welsh, and the area around the university, where many students reside, having a lower percentage than the city average (Aitchison and Carter 2004).
As a Welsh public institution, CU has had to comply with the 1993 Welsh Language Act. This has involved developing and maintaining a Welsh Language Scheme, setting out actions to promote the implementation and development of Welsh language provision in many domains of use, e.g. recruiting Welsh-speaking staff, communications with the public, translation services, the 'public face' of the University (e.g. website, official notices).
CU had about 28,000 students at the time of data collection (2009)(2010). CU's internationalisation is particularly visible through its numbers of international students, some on exchange schemes allowing them a semester or a year at Cardiff as part of their studies, but far more taking CU degrees and so staying three or more years. In 2009-2010, around 1.5% of CU students were exchange students, and 10% were international students.
Welsh is not at all prominent at CU. In CU's 2006-2011 Strategic Plan, for example, there is only one mention of the Welsh language, expressing support for using 'the Welsh language in the University in accordance with the Welsh Language Scheme' (section 9.7). In the more recent -2014Strategic Plan (Cardiff University 2009, the Welsh language is mentioned once in relation to the provision of Welsh language teaching. There is bilingual signage across CU, and the CU website has a parallel website to access information written in Welsh. However information about the Welsh language itself on the university website is generally sparse (except, of course, in relation to the School of Welsh itself, and the language tuition provided in the Welsh for Adults Centre). On the CU website, four levels down from the home page, is a page called 'Welsh Language Information' (Cardiff University 2011), where website visitors are given reassurance that 'everybody in Wales speaks English, and all your lectures, seminars and coursework will be conducted through the medium of English. You will not be expected to have any knowledge of or learn Welsh (unless you want to)'. The bilingual context in which CU is situated has substantial institutional support both at governmental and at some grassroots levels, but English is dominant in both Cardiff and Wales as a whole. Balfour (2007, 37) mentions just 6% of staff and students at CU using Welsh in 2003 and concludes English is the default language at CU. As Balfour (2007, 45) notes, 'it is unlikely that students migrate to the UK to study Welsh'.

The present study
Part of our study involved inviting students to complete a questionnaire regarding their perceptions and attitudes relating to the notion of an international university and their self-reported language use in their daily lives inside and outside the university. In this paper, we focus on two items on the questionnaire addressing the following research questions: RQ1. How do students in the project universities in these two bilingual contexts conceive of an 'international university'? RQ2. How do these students see the implications of internationalisation in their universities for the respective minority languages?

Method
The participants were convenience samples of 'home' or 'local' students and international students, who completed questionnaires at the end of lectures or in free moments during laboratory sessions. The matching of student samples across universities is seldom 'neat', and there are contextual differences that need to be mentioned. First, while UdL draws its students from the immediate, and therefore Catalan-speaking, area, there is still a stronger tradition in the UK that undergraduates study at a university located away from where they went to school. Hence, the majority of CU students are not local, but rather many come from various parts of England, particularly the south and south-west. While most of the students who do come from Wales are not Welsh-speaking, it is a reasonable assumption that virtually none of those from England speak Welsh. This also raises a terminological issue, since the 'non-international' students at UdL are usually referred to as 'local students', whereas those at CU are usually referred to as 'home students'. In this paper, we use the term 'home student' for both.
Second, the participants were drawn from a range of disciplines to get a general student view from each university. Universities inevitably differ in the range of disciplines offered, and there is at least a possibility that views vary with the disciplines. However, for a comparison of universities, the range of different disciplines is an integrative dimension of their composition.
The first of the two questionnaire items we look at in this paper is an open-ended item that appeared at the beginning of the questionnaire, asking respondents to write down the first five things that come into their heads when they see the term 'international university'. The second is an open-ended item that appeared at the end of the questionnaire, requesting students to write a few lines about how they thought internationalisation affected the position of the minority language.
The data gathered from these two items differed. The second allowed longer responses, usually one or two sentences, whereas the first question required five quick responses, usually consisting of one (or two) lexical items. We call these 'keyword' responses. The methodological background of this approach has been covered elsewhere in relation to other studies (e.g. Garrett 2010). Suffice it to say here that keywords offer a kind of shorthand setting out the most immediate associations that respondents make with the concepts presented to them. The concepts of 'internationalisation' and 'international university' are likely to have been experienced and conceived of quite variably and to different extents in different parts of the world (see Egron-Polak and Hudson 2010) and in different universities and amongst the different groups of people working or studying in them. The keywords study seeks an emic view of the concept of international university, gathering perspectives from people with various kinds of involvement and experience of international universities in their studies or working lives. Keywords and the longer responses to the second question can serve the kind of discursive function that gives such concepts their shape. While there are, on the one hand, institutional or policy document representations of these concepts, our focus here is upon those whose daily lives are lived very much in their midst.

Keywords processing
Keywords analysis aims at a balance between reflecting the varied character of the responses while also reducing the data to manageable proportions to allow some broadly descriptive quantitative comparisons. Thus, content analysis was employed to organise the data into groupings. First, the relatively discrete ideas expressed in each answer ('referential units') were identified (see Krippendorff 2004). These were clustered into broader thematic categories ('thematic units'). Percentages were then used as a basis of comparison.
One common quality of this type of data, and folk comment generally, is that responses range along a continuum of general to specific in terms of detail (see Preston 1996). Some responses such as 'Erasmus' and 'exchange programmes' were easily gathered up into themes. But many comments contain less specific detail and placing them into discrete, mutually exclusive categories is obviously not an exact science. Furthermore, as Potter and Wetherell (1987, 41) have emphasised, one needs to guard against obscuring interesting patterns in the data through attempting to 'over-fit' the data into categories. Some items in our data were left ungrouped, but these were few in number. The groupings and their contents are shown below in the results section.
Processing of the open-ended responses to the second questionnaire item Responses to the second item generated a more elaborated discourse than the keywords. Students set out ways in which they thought internationalisation might affect the minority language. It was possible to place most responses into three primary categories, according to whether they were primarily positive about the impacts of internationalisation on the minority language, negative, or neutral (some were counted as neutral if they contained responses with both positive and negative sides). A further grouping was simply labelled 'Other'. Many of these 'Other' responses seemed not to be interpretable as answers to the question itself, even if they were about the same topic, and are not given much attention here. Within these broad groupings, the responses were examined to identify subgroupings according to their thematic properties and types of arguments presented.

Results and discussion
How the students conceive of an international university Given the large database, the home student samples were reduced through systematic sampling to subsamples of 150 in each university. The numbers of students included in the analysis were then as follows: UdL home students: 150 (from the total sample of 624) UdL international students: 75 (the total sample) CU home students: 150 (from the total sample of 417) CU international students: 61 (the total sample) The totals of keywords gleaned from the student samples, followed by the numbers of students who left the question blank (in brackets), were as follows: UdL home students: 609 keywords (15 students) UdL international students: 280 keywords (10 students) CU home students: 580 keywords (12 students) CU international students: 235 keywords (4 students) The groupings of keywords (as they are also listed in Table 1) are as follows: (1) LANGUAGEe.g. 'language', 'languages', 'multilingualism'. The number of mentions of English are given in brackets.

General overview of results
Keyword items generally include both descriptive and evaluative items. With individual lexical items, though, it is often not possible to be sure how respondents position themselves. Hence, with, say, 'multicultural', we cannot be sure whether the respondent holds an approving or disapproving stance to this, and it is important not to go beyond the data in our interpretations. Nevertheless, it is reasonable to say that there is generally little overt negativity (a few instances of keywords such as 'difficult', or 'language barriers' might be fairly seen as negative), and if we can reasonably view items such as 'education' 'fun', 'exchange and travel', 'opportunity' and 'experience' as favourable associations with the concept, then there is more overt positivity linked with international universities across the student groups.
The largest groupings overall in Table 1 are those for language (grouping 1) and culture (grouping 2), along with people (grouping 3). Within this pattern, there is some variation across the four groups. While language has a similar high level of salience across all groups, culture and people, though prominent for all groups, appear more so for the CU than the UdL students. This variation aside, though, it is clear that language, culture and people are particularly strong components of an international university for all these students.
Mentions of English differ across the groups. It is the UdL students for whom it has the most salience, accounting for 22.4% of the international students' 49 language keywords, and 22.1% of the 113 language keywords of the home students. The CU home students make least mention of it (2.8% of their 106 language comments). This is the respondent group for which English is the main if not only everyday language and which comprises the most English monolinguals. For the CU international students, it achieves more salience, accounting for 15.4% of their 39 language keywords, as doubtless one of their primary motivations for choosing to study in the UK. The high presence of 'English' in responses at UdL may reflect the low presence of English as a medium of instruction at UdL (approximately 3%), as well as indexing concern about the spread of English in the Catalan context, especially if internationalisation in non-Anglophone countries is expected to develop through the introduction of English as a language of instruction (see papers by Kuteeva, and by Mortensen in this issue).
The diversity grouping (10), slightly more salient for the CU students, is at first sight unintuitive in that one might expect diversity to be a strong characteristic of almost anything international. And, indeed, this is the case in our data. Grouping 10 contains only the residual, unqualified items, simply saying 'diversity', variety', etc., but many of the comments placed in the other groupings, such as 'multilingual' and 'multicultural' are clearly also about diversity but qualified in terms of language, culture, etc. Setting the responses in the context of the task of inviting associations with the term international university, many other items, such as 'languages', 'cultures', 'opportunities', arguably themselves carry a comparative plurality, and hence, diversity inhabits many of the groupings in Table 1. Discounting keywords that can be reasonably seen as not suggesting diversity (e.g. 'reputation', 'expensive', 'fun' and 'out of the country') diversity can be seen as the strongest underlying association with 'international university', accounting for 73% of keywords from UdL home students, 70% from UdL international students, 71% from CU home students and 66% from CU international students.
There are just two other areas of Table 1 in which CU students show a stronger association than the UdL students. One of these concerns the CU international students, who comment more on the status and the reputation of the university (grouping 9), and the other concerns the CU home students, who comment much more on the foreignness of an international university (grouping 8e.g. 'an overseas university' and 'out of the country').
Having looked at the stronger associations for the CU students, what are those for the UdL students? One area of higher salience concerns exchange and travel (grouping 5), and specific references to Erasmus. The Erasmus programme, along with exchange programmes in general, has tended to have a smaller uptake among the UK students than in other countries, which is doubtless why it features larger in UdL associations.
A second area concerns the 'learning' grouping (4), where there is more salience for both the home and the international students at UdL. This grouping contains items such as 'education', 'study', 'training', 'knowledge', 'development', and it is striking that these have less salience for the CU students.
A third area concerns 'experience and opportunity', where the UdL home students have more to say than the others. This may in part be connected with the tendency, mentioned in the introduction, to enrol at their local university rather than further afield, but it was also notable that the UdL home students said much more about jobs and job opportunities than the other students, and it is this that makes their percentage higher. Improving their employability is linked more potently to 'the international university' for this group than the others. Certainly, in their study of UK students, Brooks and Waters (2009) found some scepticism regarding claims of greater employability arising from studying abroad. Fourth, the UdL home students have stronger associations with the global reach of international universities and with cooperation and links (the 'World' grouping).
How, then, might we sum up the four 'shapes' that emerge from the keywords of the four groups? For all four groups, a strong feature of international universities concerns language, culture and people (students, new friends and people in general) and, indeed, linguistic, cultural and human diversity. And we concluded too that diversity is a particularly potent quality that is underpinning a great many of the comments across some of the keyword groupings, thus constituting the prominent category connected with internationalisation in higher education, with the emergence of cosmopolitan environments where people from different cultural and linguistic backgrounds live amongst each other.
Beyond that, we could say that, among the UdL home students, there is an image that the global reach of international universities affords opportunities for the kind of study and experience (through exchange programmes and travel) that would otherwise not be available to them, and that this too may help in gaining better employment. In this regard, one senses strong positive value placed on international universities. Among the UdL international students, there is a similar picture, but with less focus on the experience and opportunities or upon the international links and reach of such institutions. Among the CU home students, there is a picture of strong associations with culture. Alongside, this is a weaker association with exchange programmes and travel, despite these students having a stronger sense that an international university is located somewhere else (and perhaps, therefore, not seeing CU as an international university). Comparatively, links with learning and development and with experience and opportunity are also backgrounded, and the value they might place on an international university seems comparatively more concentrated into linguistic, cultural and human diversity than linked with formal education, experience and opportunities or the international links. Finally, the view of the international university among the CU international students stands out from the others in two ways. One is that they associate an international university much more than the others with better quality, prestige and reputation. The other is that, for them, an international university has even stronger connotations of human diversity and meeting and making new friends.
Keywords data can also be revealing as regards what is not mentioned and what is not prioritised in the minds of the students when they are confronted with the term 'international university'. One of our primary research interests is in the specific languages in these two contexts, and so it is striking that despite the high salience of language for all four groups, there is no mention of Welsh, Catalan or Spanish. There is a greater focus on linguistic diversity (albeit with some attention to a link between internationalisation and English), and seemingly no salience regarding the relations with, or further threats to, the other languages in their immediate environment. It is clear, then, that the concept of an international university does not bring thoughts of the implications for these languages to the fore. To study how they might see internationalisation affecting Catalan and Welsh, we now turn to the open-ended item at the end of the questionnaire.

How the students see the implications for the minority language
The second questionnaire item asked students in what ways they thought that internationalisation affected the position of the minority language 1 . As with the keywords, the large home student databases were reduced through systematic sampling, while all international student respondents were included. Table 2 shows the percentage of total responses, divided into positive, negative, neutral and other, plus recording those who left the item blank or simply wrote 'don't know'. An example of a 'neutral' answer would be that they saw no connection between internationalisation in the universities and the future of the minority languages, i.e. internationalisation has a 'neutral' role in the future of the languages.
It is notable that for all groups, there are relatively high proportions of students who did not complete this item. We can speculate as to why. The item came at the end of the questionnaire, and since it was an open-ended item requiring more thought and articulation, it might have been the victim of fatigue effects. Another possibility is that it was a question that the students had not really considered previously and on which they had no already-formed clear view, or that they found too complex to answer on the spot (perhaps alongside fatigue). It may, of course, be that different students or groups of students left it blank for different reasons. To sum up some of the main features in Table 2, there is strikingly more polarisation between the home and the international students at UdL than at CU; the UdL home students showing more positivity, contrasting with the UdL international students, who show more negativity (in each case, more than any of the other student groups). Alongside the UdL home students showing this comparatively high level of positivity, they show less neutrality than the other groups. In the case of the CU students, one might argue that the figures for positivity and negativity are more balanced, especially amongst the home students. Equally, though, one could see this in terms of the CU students being more divided amongst themselves, falling more evenly across the divide (again, especially the home students). Alongside the higher proportions of neutral comments compared to the UdL students, the CU picture does not suggest a greater strength of feeling one way or the other, in contrast to the UdL students.
Below, we set out the main qualitative themes that emerge from these open-ended responses. There is a degree of overlap in the kinds of responses given by the CU home and the international students and the UdL home students, and we report these three groups first.

CU home and international and UdL home students
Comments that were positive about the impact of internationalisation on minority languages focused primarily on the following: (1) Internationalisation will raise awareness of the minority language, give it more importance and encourage people to take it seriously. (2) It will encourage more people to learn it, regardless of other languages in the environment.
(3) The presence of so many other languages will generate more respect for diversity and languages, generally, and place the minority language alongside the many others. (4) The very presence of so many international students will favour the language because they are interested in it.
These themes are exemplified in the following comments: It gives Welsh more importance (CU home).
Celebrates it amongst others. Encourages its use. The more languages, the better (CU home).
By promoting internationalisation along with Welsh, I feel it raises the status of the language by having it taught at the same level as a number of the other more widely spoken foreign languages (CU international).
From my own experience, international students are far more interested in the Welsh language than the English/British students, so I do not think internationalism is detrimental in that sense (CU home).
I think it could be positive because international students can be interested in learning this new and unusual foreign language (CU international).
Fer conèixer al món l'existència de la nostra llengua [It makes the world know about the existence of our language] (UdL home).
Positivament que el català es doni a conèixer a la resta de països. [Positively, because Catalan gets known in the other countries] (UdL home).
Comments that were negative about the impact were of the following types: (1) Relatively unspecified comments along the lines that the minority language will not be promoted so much.
(2) The presence of so many other languages will lead to the minority language losing ground. (3) The presence of so many international students means fewer students using the minority language.
The following are illustrative examples of these themes.
Crec que el català pot sortir perjudicat. Si vénen estudiants de fora segurament sabran anglès i també castellà, però no el català. Això fa que s'hagi de canviar el català pel castellà i per tant el catatlà s'usa molt menys. [I think Catalan can be damaged. If outside students come, they will probably know English and also Spanish, but not Catalan. This means that we have to switch to Spanish, and therefore, Catalan is used much less.] (UdL home).
It reduces the chances for the Welsh language to be learnt and redeveloped with English being a primary language and alternative languages being the other major source of communication for international students (CU international).
Welsh is only one language, whereas internationalisation deals with many. This could cause Welsh to be slightly neglected (CU home).
Reduces the use of Welsh as the majority of foreign students do not speak it, which encourages communication with home students in English (CU home).
Comments expressing neutrality gave the following views: (1) Simple statements that it will not be affected.
(2) Internationalisation will make no difference because the minority language is already well-supported. (3) The minority language is already less and less influential, and the situation would be the same whether there was internationalisation in universities or not.
The following comments illustrate these views.
No creo que tenga apenas influencia, la gente que ya habla catalán, no dejará de hacerlo solo porque aprenda a hablar, por ejemplo, el inglés. [I do not think there is an influence; people who already speak Catalan will not stop doing it just because they learn how to speak English, for example.] (UdL home).
Welsh can be equally promoted and practised even if there are many international students who speak English (CU international).
I do not think that the Welsh language is affected as there are a large number of Welsh speakers who attend the university. Also, there are many students studying Welsh to maintain the language alive (CU home).
In my opinion, the Welsh language is becoming less influential because people in Wales are brought up speaking English. Internationalisation therefore doesn't affect the position of the Welsh language. If internationalisation doesn't occur in universities, the position of Welsh wouldn't change (CU home).

UdL international students
Of the 75 UdL international students, 41.7% left this question unanswered. Those answering seemed to interpret the question in various ways, with many of them not explicitly saying whether internationalisation was a good or a bad thing for the minority language, or whether they were positive or negative about the impact of internationalisation on its future (i.e. 'how does internationalisation affect Catalan?') but instead placing the focus on 'how does Catalan affect you?'. The importance of Catalan in their everyday lives is particularly obvious in their answers. They position themselves as those experiencing its impact at UdL, and this experience clearly underlies many of their responses, which suggest that their experience differs from their expectations of the sociolinguistic environment at UdL. With so many of their responses taking this kind of focus, we were inclined to see their responses not so much as 'not answering the question', but rather as responses in which they may indeed be regarded as in some sense representatives of internationalisation from the student perspective. Hence, emotional displays of a negative stance against Catalan can arguably be viewed as embedded in the processes of internationalisation in the case of these students and working against Catalan.
Only 8% of these students saw a positive relationship between Catalan and university internationalisation. This is the lowest of all the four groups. The main theme echoes that of other groups: that internationalisation will promote the local language through promoting its learning and placing it alongside other languages.
There were 25% of the respondents who were negative, clearly the predominant perspective of those who answered this question. The responses are seen as negative because they directly or indirectly indicate a preference for less Catalan. In terms of Risager's (2012) university language hierarchies and policies, this could be viewed as a preference for a bilingual policy based on international languages (here, Spanish and English) and as resistance to a trilingual policy of international and regional languages (here, Spanish, English and Catalan). There are two main themes in their comments.
The first is that the promotion of a minority language is incongruous with internationalisation in universities, with some students pointing to English and to Spanish. We can see this as a demonstration of Balfour's (2007) point that language policies aimed at promoting minority languages may be seen as an obstacle to the expansion of the institution. For example, the following: No me parece bien que en una universidad internacional se potencie tanto una lengua 'minoritaria' a nivel europeo. En una universidad internacional no debería ser así. [I do not think it is right that an international university promotes a 'minority' language so much at the European level. In an international university it should not be like this.] Es muy importante por universidades de ser International y de pueden hablar inglés para esto; la internacionalización de la UdL es muy mal, porque todos hablan catalán pero no inglés y un poquito de español pero no quieren hablar español, e inglés claro que no! [It is very important for universities to be international and that they can, therefore, speak English; internationalisation at UdL is very bad, because everybody speaks Catalan but no English and a bit of Spanish, but they do not want to speak Spanish, and English, of course not!] The second theme is that Catalan is not useful compared to other languages: Honestamente el catalán me parece un tanto innecesario ya que sólo se habla en una pequeña región del mundo, no hay lugares donde también se hable por lo que no sirve de nada. El español es mejor. [Honestly, Catalan seems to me a bit unnecessary since it is only spoken in a small region of the world, there are no other places where it is spoken, and therefore, it is not useful at all. Spanish is better.] There were 21% of the respondents who gave a more neutral response, saying that internationalisation will not affect Catalan because the local community will not change their linguistic habits, with some comments pointing to an unwillingness on the part of Catalan-speakers to learn or use other languages: La internacionalización y el catalán pueden ser compatibles si los catalanes se abrieran a todas las culturas del mundo. [Internationalisation and Catalan can be compatible if Catalans opened themselves to all the languages in the world.]

Conclusion
In their keywords data, we noted that the students appeared to overlook the place of the minority languages in international universities. To some extent, this may have been due to the question addressing an idealised notion of 'the international university' rather than in relation to their own university. The second question, bringing their own institution into the frame, has shown how the keyword views link with their own contexts. The keywords themselves revealed some marked differentiation amongst the respondent groups in terms of what was salient for them. For example, UdL home students made stronger associations with global reach and the better opportunities and experience offered by an international university than did CU home students. For CU international students, institutional prestige and reputation were more salient than for those at UdL. Beyond this differentiation, though, amongst the strongest perceived components of an international university for all four groups in the keywords data were language, culture, people from all over the world and, we maintained, diversity. And we can see how, with the exception of culture, these perceptions influence the interpretive repertoires in the responses to the second item. Since the item placed the focus on the minority languages, it is perhaps unsurprising that culture does not feature. Many of the responses consider the effects of a diversity of languages in the environment and of the presence of an international diversity of students. In many instances, these repertoires comprise paired oppositions of conclusions. Hence, particularly in the cases of the CU home and international, and the UdL home students, the presence of many other languages can lead to optimism (the minority languages will gain prominence and status) or underlie arguments to the contrary (the minority languages will be less visible, pushed aside, etc.). And in a similar vein, the presence of international students can generate positive arguments (there will be more interest in the languages, and more people learning them) and negative arguments (other languages will be their main means of communication, and they will strengthen the encroachment of Spanish or English at the expense of the minority language). The responses of the UdL international students, while perhaps not articulating these points as arguments, can be considered to embody them at times, in the negative category in particular, as more personal reactions to their own experiences. Table 2 and the individual group profiles within it have revealed some significant differences, as well as similarities, amongst the groups of students who answered this item. We can conclude that both home and international students at UdL have very different experiences of their minority language than those at CU. The strong positivity of the UdL home students places great emphasis on a boost to Catalan from foreign students having the opportunity to learn it (a view seemingly not greatly shared by the foreign students themselves). With Welsh being a far less prominent factor in Wales, there is less conviction about this, and a realisation that any strong expectations that students should be required to learn Welsh would be unrealistic. The polarisation between the home and international students at UdL might be viewed in terms of two kinds of 'psychological reactance' (Brehm and Brehm 1981). The home students tend to react against the language threat they perceive to Catalan from Spanish, pointing to the opportunities for additional speakers of the minority language. In contrast, the international students' reactions are more in alignment with those of majority language speakers, and indeed, given De Bres ' (2008, 476) claim that 'the umbrella category of "majority language speakers" can be defined in different ways at different times', they might arguably be seen as a part of that category. They tend to react against the 'constraints' of the minority language at the university, suggesting an appeal for freedom of choice (cf. May 2000, 119 on references to 'choice' and 'opportunity' amongst speakers and non-speakers of minority languages). They appeal for example to its inappropriateness in an international university and, perhaps, perceive a threat to their academic success arising from the restrictions on which languages they can use in their studies. In the case of CU, no one could view the far more limited presence of Welsh as placing any restrictions on non-Welsh speakers. It is rather the relatively weak position of Welsh that leads Welshspeakers to experience such reactance in terms of trying to ensure their desire to use Welsh is recognised, respected and protected and that policymakers ensure that Welsh language contexts increase. On the one hand, we see students in Wales who feel they experience little Welsh around them, and for whom Welsh has no, or need not have any place in their study environment, even where there are political initiatives to increase its presence. On the other hand, we see students in Catalonia where Catalan is much more pervasive, for whom Catalan has a central place in their study environment, with strong support from the home students but resisted by international students as problematic. Students in Wales are viewing internationalisation and the international university in a context where Welsh is being backgrounded and internationalisation, alongside English, has relatively free rein. In both contexts but in very different ways, then, the students' responses have highlighted the critical and difficult question of how to reconcile the presence and vitality of minority languages with the forces, priorities and pursuit of internationalisation.