Perspectives of postgraduate professors and students on internationalization and English language use at a university in the south of Brazil

The aim of the present study is to understand how internationalization has been understood, lived, and discursively constructed by postgraduate students and faculty members at a publicly funded university in the south of Brazil. Through data generated using questionnaires and interviews (with 406 and 19 participants, respectively), we observed at this university that there are discrepancies between how student and faculty participation in the process of internationalization is perceived. We also identified inconsistencies in relation to the understandings of the role of foreign languages in this process. The results show that internationalization is conceptualized as the establishment of interpersonal, intercultural, and interinstitutional relationships (as defended by Martinez, 2017). They also point to two specific needs: a) for more student involvement in discussions over the internationalization process of the university; and b) for a plurilingual understanding of the status of English within this process.


Introduction
In recent decades, higher education in Brazil has undergone considerable transformations. One major change that has taken place is the implementation of new policies for the internationalization of Brazilian universities (Miranda & Stallivieri, 2017). According to Knight (2003:2), internationalization is "the process of integrating an international, intercultural or global dimension into the purpose, functions or delivery of post-secondary education". In the case of Brazil, the concept has been understood as a desire for academic exchange and cooperation with institutions and scholars from beyond the boundaries of Brazilian academia (Martinez, 2017), especially through mobility initiatives (Baumvol & Sarmento, 2016).
As explained by Guimarães, Finardi, El Kadri and Taquini (2020:1), motivations to internationalize in Brazil are to "contribute/promote social and cultural development, mutual and intercultural understanding, citizenship, social and community development". In addition, there seems to be a desire for an increase in the global competitiveness of Brazilian universities (Leal, Stallivieri & Moraes, 2018). Still, as Guimarães et al (2020:2) show, the process of internationalization remains confusing in this particular context, since the concept is "used as an umbrella (and often empty) term to describe anything remotely associated with the terms global, international, intercultural and the like".
Internationalization policies in Brazil have been promoted nationally and institutionally through programs such as 'Science without Borders 1 ', which sought to send Brazilian students/scholars to universities abroad, 'Languages without Borders 2 ', which attempted to increase access of Brazilian students/scholars to languages other than Portuguese, and, more recently, CAPES PrInt 3 (Institutional Program for Internationalization), whose objective was to reward universities that have implemented internationalization policies and practices that are considered successful. Through such programs, educational and linguistic actions have been increasingly encouraged and disseminated in Brazilian universities. These include the mobility of students and faculty members, the push for publications/presentations in languages other than Portuguese, especially English, the encouragement for students and professors to learn other languages, and the establishment of partnerships with international institutions (Jordão, Diniz de Figueiredo, Laufer & Frankiw, 2020). In addition, there has been an increasing push for the use of English as a means of instruction (Martinez, 2016) and for raising positions in international university rankings (Leal et al, 2018). These measures have been very active in the past ten years, which means that internationalization, and English as a means of instruction are considered quite new in the country. As explained by Martinez (2016), it has been common for scholars at universities to engage in internationalization and English as a means of SOTL in the South 5(1): April 2021 de Figueiredo, Jordão, Antunes, Emmerich, Cons instruction practices without any institutional policy in place at the universities where they work. This scenario has changed somewhat in the past few years, especially after CAPES PrInt, but policy implementation seems to have gained strength particularly in more reputable universities, rather than in less prestigious ones.
Despite growing interest in internationalization processes and policies, there are few studies investigating the perceptions and attitudes of agents involved in this process. These voices need to be more prevalent in discussions on internationalization, especially in understudied contexts like Brazil, since students and professors are those who actually experience (or are expected to experience) internationalization phenomena more directly.
Therefore, this study aims to understand how internationalization has been perceived, experienced, and discursively constructed by professors and students of a publicly-funded university in the south of the country. In particular, we explore discrepancies found in these discourses, as explained in more detail later (for other aspects of our data, see Jordão et al, 2020). Our contribution to current literature is that the study is one of the first in Brazil -to the best of our knowledge -to bring a combined account of discursive constructions by professors and students, at both a large scale (with descriptive quantitative analysis) and in more detail (through qualitative data).
We present data obtained through electronic questionnaires (answered by 123 professors and 283 students), and interviews with 19 participants (ten students and nine professors). We emphasize two issues in which divergences were observed, both between professors and students and between conceptual perspectives: a) participants' knowledge about internationalization and their involvement in internationalization actions; b) understandings about foreign languages as part of this process. We intend to contribute to the understanding of the phenomenon of internationalization in Brazilian higher education institutions, especially concerning the perceptions of the academic community about how this process reflects upon their activities as professors and students. The emphasis we give to discursive constructions and to languages reflects our locus of enunciation as applied linguists in Brazil, and the need to problematize issues of language in internationalization scholarship. We hope that the study contributes to the process of critical development of internationalization in Brazil and similar contexts in the global South.
The global South is defined here based on the understanding that there are contexts worldwide that face social, economic, racial, academic and other types of inequalities that are sustained by sociocultural and historical issues that stem from colonialism, capitalism and patriarchy (de Sousa Santos, 2018). The global South refers to these contexts that have suffered from such inequalities, while the global North refers to contexts that have been privileged by them. Following de Sousa Santos, the term global South refers to an epistemological South, composed of contexts with knowledges that are born in struggles and that are often neglected. Thus, we understand that inequalities and struggles experienced in the global South are not based on geographical issues only; still, we do see overlaps in terms of geographical location and socioeconomic inequalities, particularly in contexts like Latin America.
We also highlight de Sousa Santos' conceptualization of 'epistemologies of the South' (de Sousa Santos, 2018). These refer to the knowledges and representations of the world which are constructed SOTL in the South 5(1): April 2021 de Figueiredo, Jordão, Antunes, Emmerich, Cons by groups that are often silenced -i.e., those in the global South. Such a concept is important in this study because the terms that generally define what internationalization is, and how it should be done, come mostly from conceptions of academic knowledge that originate in the global North. Therefore, investigations on how global South professors and students understand and experience internationalization in their own contexts are crucial for discussions about the ways in which internationalization may be reshaped, as based on local knowledges and lived realities. Chowdhury and Le Ha (2014) investigate the internationalization of higher education as part of globalization. They argue that it is not possible to disregard the forces that play a strong part in the internationalization of universities, which include the construction of a desire for English and for being part of global North institutions. They also provide evidence that globalization can be interpreted from different angles. To this end, the authors rely on Bakhtin and Foucault, especially on the concepts of 'heteroglossia' (Bakhtin & Holquist, 1981) and 'archaeology of knowledge' (Foucault, 1984), to analyze reports from students inserted in contexts of internationalization in the global North.

Literature review
Chowdhury and Le Ha (2014) point out that English has received a prominent role in understandings of globalization and internationalization. The altruistic idea that English would be a bridge between various cultures is contested by the authors, who argue that the language often acts as an instrument to maintain power relations. They assert that there is a normalized discourse -mainly in higher education institutions in the global North -that international students are a form of capital. Moreover, many of these institutions see the need to become competitive in an internationalization 'market'. Chowdhury and Le Ha (2014) also show that international students are often positioned as deficient, as if they always needed the support of global North institutions. For the authors, there is a conflict between expectation and reality. On the one hand, the expectation would be that, in an international paradigm, there would be an exchange of knowledges. The reality, on the other hand, is that there is an economic interest in selling the English language and mobility as commodities. In this reality, international students end up experiencing xenophobia, feelings of inferiority, among other negative experiences.
In the Brazilian context, Baumvol and Sarmento (2016) explain that internationalization is commonly understood as mobility. Thus, there is an imbalance between contexts understood as belonging to the global North and others like Brazil (from the global South), since Northern ones receive many more foreign students and have more funding for internationalization. Additionally, mobility programs in many contexts are restricted to only a small portion of students, which highlights the difficulty of including students in what is perceived as internationalization processes.
Another factor that has received much attention in Brazil is the push for EMI (English as a Medium of Instruction), which has happened in many countries worldwide, from both the global North and global South (Martinez, 2016). As explained by Martinez (2016), the push for EMI is related to a desire by Brazilian universities to attract students from other countries, to prepare local students for a supposedly globalized market, and to raise institutions' positions in international rankings. Even SOTL in the South 5(1): April 2021 de Figueiredo, Jordão, Antunes, Emmerich, Cons though some benefits of EMI have been observed -such as the possibilities of developing EMI locally and incorporating it within local professor development -there is concern regarding the way it has been promoted and discussed within Brazilian higher education (Jordão, 2018). Martinez (2017), for example, questions the celebratory framework of discourses regarding internationalization and EMI, which does not address local characteristics. According to the author, internationalization and EMI policies, when based solely on onto-epistemologies of the global North and on competitiveness, are forms in which colonialism and neoliberalism materialize in contemporary higher education. The results obtained by Martinez (2017) show tensions between a neoliberal logic often operated in processes of internationalization and anti-hegemonic discourses, which seem to reveal local consciousness and a search for legitimizing local onto-epistemologies.
These results point to a need for epistemological change in unequal relations between the global South and North, and a deeper concern from Brazilian institutions as to what the author classifies as local internationalization; i.e., internationalization in which intercultural dialogues between different localities are constructed in ways that respect local particularities. Local internationalization is particularly important because it poses knowledges that are generally considered global -i.e., hegemonic ones -as also local, developed within particular contexts -which in turn enables intercultural dialogues that are respectful and inclusive.
Martinez (2017) also points to the importance of greater attention to issues of interculturality in understandings about internationalization, since cultural matters play an essential role in this process. For the author, any practice of internationalization "presupposes that one person needs another, that there is some form of meeting with the other, with whom to relate, exchange, conflict, coexist, disagree, fight" (Martinez, 2017:149). In this encounter with the other, there is a centrality of human relations. Martinez concludes that internationalization projects need to reflect upon what it means to be face to face with the other, and how interactions with this other can occur. In this aspect, it is worth highlighting the assertion by Gimenez (in Finardi, Gimenez & Lima, 2020), which suggests that the search for interculturality should be the focus of international exchanges; thus, internationalization should be the result of this search.
The works cited here problematize what it means to internationalize and how such meaning can be related to neoliberal discourses. They also bring perspectives on how the role of English in this process has been understood -sometimes with suspicion, and sometimes with possibilities of appropriation and destabilization of asymmetric relations. Finally, the studies lead to a reflection on how agents involved in these processes may build their understanding of internationalization and their relationships with each other. By emphasizing the importance of localizing processes of internationalization, these studies highlight the need to take into account the realities of each context.

Method
The study was carried out at a large, publicly funded university in the south of Brazil. We chose to conduct our investigation in this university for three reasons: a) because this is one of the most reputable universities in the country, and one of those that were awarded with a national grant for internationalization purposes; b) because of the increasing efforts the university has made to achieve SOTL in the South 5(1): April 2021 de Figueiredo, Jordão, Antunes, Emmerich, Cons international status; c) due to our access to its professors and students. We paid particular attention to postgraduate programs, which is where most internationalization efforts in this institution have taken place. In fact, the university's internationalization program is hosted by the Office of Research and Postgraduate Studies, which shows the increased focus that is placed upon internationalization at the postgraduate level. Nevertheless, the university has recently included more undergraduate actions as part of its internationalization plan.
The research -which followed a mixed-method, two-phased sequential design -began in 2016 and lasted until 2019. In 2016 (first phase), two online questionnaires were sent via e-mail to all of the university's 91 postgraduate programs. The questionnaires consisted of 25 questions related to the respondents' knowledge about the university's internationalization process. This was in addition to questions relating to the use of English and other foreign languages in this context. The forms were sent electronically through the Office of Research and Postgraduate Studies of the institution, and programs were asked to forward the questionnaires to professors/students. In total, 406 responses were obtained (283 from students, 123 from professors). This accounted for a response rate of 6% for professors and 10% for students, which is considered reliable for college research with a sampling frame as large as the one we had (Fosnacht, Sarraf, Howe & Peck, 2017). It is also relevant that we have data from more than half of the programs of the university (50 out of 91). The tables below show the postgraduate programs in which the participants of the questionnaire teach/study.  The analyses of the questionnaires (which took place in two phases) were carried out both quantitatively and qualitatively, through descriptive statistics and content analysis. For content analysis, we familiarized ourselves with the data and then coded it individually at first (using Quirkos, a qualitative data analysis software). Quirkos was used to simplify this process (in terms of tagging parts of the text, creating a visual interface where the salience of particular codes is observable, and understanding overlaps between codes). Codes were then refined collectively in group discussions.
We generated themes that we felt were most salient in the data; this process was finally followed up by a review and refining of themes. This choice was based on two factors: a) our need to have a general picture of how professors and students at the university perceived internationalization in that institution (which we achieved through the quantitative data); b) because qualitative data analysis would be crucial for us to dig deeper into our initial findings.
When performing such analyses, we found discrepancies between answers obtained among the groups of professors and students, and also internally within each group. These discrepancies, which are presented and discussed in the following section, led us to subsequently design specific semistructured interviews for professors and students of postgraduate programs.
For this purpose, two interview scripts were designed, one for professors and a similar one for students, each with nine questions. This was the second phase of the research, which was qualitative and weighed more in terms of our design and results. The questions asked referred to the quantitative results obtained through the analyses of the questionnaire, including questions about the discrepancies observed between the answers of professors and students, and the use of English and other foreign languages at the university. The interview participants were invited via e-mail. Invitees were professors and students who had participated in university initiatives (e.g., language classes, writing centre work) in which the researchers had also taken part. All of those who responded to our invitation were included in the study.
Participation in the previous phase of the research was not a criterion for choosing the interviewees. This might be understood as an inconsistency in the research, since we sought to better understand the discrepancies that we found in the empirical material generated through the questionnaires. However, our understanding is that the discrepancies we found in the questionnaire data are not individual to each participant, but instead are part of existing discourses in the current discussions on internationalization in Brazil. Moreover, as the questionnaires were anonymous, we did not know who had completed them. What could be considered a problem, therefore, was seen by us as an advantage. The possibility of interviewing professors and students, regardless of their participation in the questionnaire, allowed us to understand more clearly if the interviewees also perceived the discrepancies we found. Ten students and nine professors (out of 23 invited professors and 30 invited students) accepted our invitation and voluntarily participated in the interviews. The participants who were faculty members (n=9) were affiliated to programs in Informatics (1), Education (1), Linguistics/Literary Studies (1), Mechanical Engineering (1), Bioinformatics (1), Economic Development (1), Microbiology (1), and Molecular Biology (2). As for students (n=10), the number per program was Mechanical Engineering (1), Molecular Biology (1), Business Administration (1), Oceanography (1), Water Resources (1), Physiology(1), Electrical Engineering (1), Cartography (1), Nutrition (1), and Dentistry(1).

SOTL in the
All interviewees were given a pseudonym, in order to protect anonymity. Since the qualitative part of our research weighed more than the quantitative one, we treated reliability and validity as a combined construct -trustworthiness (Golafshani, 2003). We are aware that as researchers we "interpret the world" rather than having access to it "just as it is" (Gee, 2011: 122). Trustworthiness was thus achieved by: a) our triangulation of methods (which included checking our interpretations of survey results through interviews, as detailed earlier); b) the convergence we found in participants' answers; and c) the detailed analysis carried out by our research team, rather than a single researcher (Johnson, 1997). Since we come from an interpretive paradigm, we see our analysis as an interpretation, rather than a search for a single truth.

Results and discussion
Here follows a presentation and discussion of the results in two sections, namely the results of the questionnaire and the results of the interviews -each with relevant subsections.

Questionnaire results
In this section, we present the results of the empirical material generated through the questionnaire. First, we will discuss the results regarding internationalization actions, and the discrepancies related to how they were perceived by students and professors. Next, we will look in more depth at the results regarding foreign languages, particularly the discrepancies related to the use of English and other foreign languages in postgraduate programs. We highlight once again that we mainly present the aspects that seemed to show discrepancies between faculty and students. For this reason, we have not reported the data of all the questions in the questionnaire. The participants were identified according to their position as professors or students and with a number (for example, the tenth faculty member to answer was identified as Professor 10, while the tenth student was identified as Student 10). To reiterate, in total, 406 responses were obtained; 283 from students and 123 from professors.

Internationalization actions
The results obtained through questionnaires suggested that there are divergences in the knowledge about internationalization actions among professors and students in their programs. While more than 80% (n=100) of professors stated that their postgraduate programs have internationalization actions, almost half of the students said they do not know about the subject or affirmed that these actions do not exist (n=141). Also, while just over 50% (n=63) of professors said they are part of internationalization actions, almost 90% (n=244) of students said they are not part of such actions. ISSN 2523ISSN -1154 SOTL in the South 5(1): April 2021 de Figueiredo, Jordão, Antunes, Emmerich, Cons We listed two main reasons that could help us understand these results: a) the understanding of the concept of internationalization by professors and students, which could be divergent; b) the fact these actions may not reach the student body. Regarding the first reason, it seems likely that several of the responding students participated in actions such as 'Languages without Borders' or events that are part of institutional agreements, for example. However, they did not associate such actions with internationalization. As for the second explanation, we asserted that the actions were developed by and involved almost exclusively professors, or perhaps a much larger number of professors than students. These interpretive possibilities were explored with the participants of the interviews, as shown later.
Other issues caught our attention in the questionnaire answers. The first was that, although the percentage of professors who said they were part of internationalization actions was much higher than that of students, this number of professors (just over 50%, n=63) still seemed low to us. Another relevant factor was that in 2016, the year in which we administered the questionnaire, more than 11% (n=14) of professors in postgraduate programs affirmed they did not even know about the existence of internationalization actions within their courses. This lack of knowledge seemed intriguing to us in terms of the scope and dissemination of internationalization actions within the institution.
These figures reinforced our assertion that one central issue related to our data was the concept that each participant has about what internationalization is. This assertion was further strengthened by the analysis of the qualitative questionnaire data regarding internationalization actions within the participants' programs. The analysis of these data led us to the understanding that the subjects of the research (professors and students) understood internationalization mainly as: a) 'mobility'movement of people to/from universities abroad; b) 'affinity' -bonds of cooperation, or desire for cooperation, with individuals outside Brazil; or c) 'branding' -individual/institutional promotion at the international level. Moreover, it seemed to us that several participants valued issues related to the teaching-learning of foreign languages, a theme that we will further explore later. We categorized our qualitative questionnaire data based on these four themes (mobility, affinity, branding and languages).
These categories demonstrate the complexity of the term 'internationalization' and its understandings. As previously discussed, there seems to be tensions between desires to construct academic exchange and cooperation with institutions and scholars from beyond the boundaries of Brazilian academia, and discourses that conceptualize internationalization in terms of competitiveness, ranking positions and a need to align with onto-epistemologies of the global North (Martinez, 2017).
When asked about the fundamental actions in the internationalization process of the university, for example, Professor 14 highlighted "teaching missions among partner institutions", which we classified as a reference to mobility (missions) and affinity (partners). The same occurred in the case of Student 200, who emphasized the importance of "sending students and professors to missions in other countries" as a fundamental action to be sought. Regarding the item in the questionnaire that asked about the interest in participating in internationalization actions of the university, some professors (and also students, but in smaller numbers) reported previous participation in international cooperation projects (which we understood as referring to affinity). For this same item, we had replies from professors who mentioned the importance of internationalization "as a means to seek resources that make the university more competitive/innovative" (Professor 12), which we understand as a reference to branding.

SOTL in the
The language category included explicit mention of language issues, such as the answer of Professor 95 when writing: "I consider it essential to be prepared to teach courses in other languages and receive foreign students in our university". In the case of students, what seemed to be prominent was the perception of a need for language courses that should be offered by the university.
Our categories illustrate the multifaceted nature of internationalization, and represent the multiple understandings of the phenomenon by subjects within such a large and diverse university. The divergences between professors and students regarding the visibility of internationalization actions seemed to us a conceptual issue. These conceptualizing differences reflect such complexity and characterize internationalization as a process that takes place in several dimensions. We think, therefore, that we would only realize this conceptual entanglement if we talked personally with professors and students of postgraduate programs of the university where the study was conducted.
Before moving on to the results of these interviews, we will explore another discrepancy that seemed relevant to us in the questionnaire data; this discrepancy concerns understandings about the use of foreign languages as part of internationalization processes.

Foreign languages
The main divergence regarding the use of foreign languages in internationalization processes was the contrast between English and other languages. While English was pointed out by both professors and students as being the most important language for the research developed by the participants themselves and for their programs as a whole, other languages (mainly French and Spanish) were pointed out by both professors and students as being the ones they use most -both inside and outside the university.
The most interesting aspect for us in this case, was that, even within the university environment, where English is positioned by the participants as the most important language, other languages were considered by them to have greater use in their academic practices. In raising assertions about the reasons that may have led to this divergence, we consider: a) the fact the discourse of English as 'the' language of internationalization is strong; b) the fact that what participants understand by 'use' may not reflect the ways in which they 'actually use' the languages in question 5 , since informal situations of contact with languages, such as listening to music, for example, may not be considered language practices or even noteworthy practices in an academic questionnaire.
In both cases, it seemed clear to us that discrepancies reflected the language conceptions of the participants who answered the questionnaire. Our first assertion concerns the imbrication that seems to exist, in the understanding of the role of languages in internationalization processes, between English and the act of internationalization itself. This refers not only to conceptions of what language is (for example, an externally given instrument, or a constitutive dimension of subjects) but also to the understanding of the English language as a concept -that is, to discourses related to this language in the international and national scenarios (Chowdhury & Le Ha, 2014). Our second assertion, on the other hand, brought discussions about the possibility of participants considering 'use' as merely moments of spoken and written production within formal settings of language. In this conception, they fail to take into account, for example, the fact that they read several texts in English inside and outside the classroom.
As already mentioned, we felt the need to explore these divergences in more depth through individual interviews with professors and students who are part of postgraduate programs in the studied context. We discuss these interviews in the next section.

Interview results
As explained in the previous section, the divergences we found in the results obtained through the questionnaire concern three specific assertions. These assertions are summarized below: • The interviews sought to explore each of these differences more specifically. The analysis of the empirical material generated in these interviews was explored within each of these concepts. After several collective readings of the narratives/responses of the participants in the interviews, we arrived at four main interpretations of what was said by them: • The confirmation of students' lack of knowledge about internationalization actions at the university.
• The understanding of internationalization as something occasional.
• Institutional partnerships as justification for the use of foreign languages (except English).
• The naturalization of English in internationalization and the consequent excluding potential that this language has -i.e., it can exclude people from participating in actions involved in this process.
We will discuss each of these points below. The first two -which concern the concept of internationalization -will be discussed together. The same occurs with the last two, which deal with language concepts and English as a concept.

The concept of internationalization in the interviews
Professors and students confirmed that (as also observed in the answers to the questionnaires) there was a lack of knowledge on the part of students about the university's internationalization actions. Even in cases where this knowledge seemed to exist, there was little involvement of the students in the internationalization actions in which the faculty members participated. The comments below illustrate this interpretation: . . . we have a colleague who's going to Portugal. . . I can't tell you how she's going to make this partnership, if she's going to be a university exchange student, why she's going to this other country. But we have never been offered any kind of opportunity, we are not acquainted with this, with these international partnerships, at least in our postgraduate program (Antônio, student).
I think it's because of a lack of understanding. Even me, I don't understand what part of the internationalization we live in, you know, in our postgraduate program. So, maybe the professors know, since they have meetings, information and are in direct contact with these initiatives (Cristiane, student).
. . . I have research projects and connections with universities abroad but there is no participation of my local students in all these activities. So if you ask me if I have internationalization activities, I do. Do they involve any students locally? No (Bruno, professor).
. . . this process is still very marked by personal relationships, and I think that . . . for the professors it is easier to have personal relationships because of their professional and financial condition and the amount of information that circulates in these circles (Daniele, professor).
It is noteworthy that the above reports emphasize that the lack of knowledge and/or non-participation by students in internationalization actions reflect issues of relationships between people and/or programs. Daniele, for example, highlights the importance of personal relationships for the internationalization process -which facilitates faculty participation, while making student involvement more difficult. The statements by Antônio and Bruno explicitly corroborate this understanding, since they refer to the direct relationships of professors, not only in the personal sphere (as highlighted by Daniele) but also in the institutional sphere.
Another important observation that can be made from the reports of participants, especially from the students, is that the relationships and initiatives established are usually occasional. The reports below highlight this understanding: . . . it seems to me that, regarding occasional initiatives supported by a professor, that professor . . . will speak more, and they will put the students more in contact with people who speak English. But again, it's not something that comes from the institution, it's not from the program, it's something occasional, that comes from the initiatives of some (Alessandro, student). . . . at least in our department there is always the promotion of university exchange scholarships when they are available, but it is basically this. Sometimes one or another international opportunity for the student to develop a project but, honestly, it is not discussed widely. Except for the necessity of studying English to write the articles, it is only these exchange scholarships from CAPES [Brazilian funding agency] that they give (Helen, student).

SOTL in the
In our understanding, narratives like these show that at least in the views of students, internationalization has not reached them because they are personal initiatives of each professor, and not institutional ones. Indeed, professor Bruno's speech, as exemplified above, emphasizes that he develops internationalization actions, which does not mean that these actions permeate his postgraduate program more broadly.
The empirical material generated by the interviews seems to evidence the understanding of internationalization as relationship building, which has been advocated by Martinez (2017). Participation or non-participation in actions related to this process is a consequence of these established relationships -many times by professors, and several times without the involvement of students. It seems to us that from the moment these relationships become institutional, that is, that they are conceived as practices of programs/institutions (and not just belonging to individuals), there will be more possibilities for students to actively participate in internationalization actions.

English vis-à-vis other languages
From the data, we came to the understanding that both professors and students have a perception of English as having a more prestigious status in postgraduate school. This result in itself is not surprising, since this language has been widely used in publications and presentations at events around the world. Moreover, English has also been considered by many as the lingua franca of academia, a notion that needs to be taken critically (Hamel, 2007).
Something that caught our attention was the analysis of how this specific perception of English relates to the participants' perceptions of the uses of other languages within their postgraduate programs. When we raised questions in this sense, we realized, at first, that our interpretations of the empirical material from the questionnaires were confirmed in the interviews. That is, despite the positioning of English as 'the' language of internationalization, the reports of the participants showed that there is much space for other languages in this process. The reports showed, in addition, that many times these other languages are more visible in some activities and within certain programs than English. This seems to be the case of programs such as Electrical Engineering, Mechanical Engineering, and Education, for instance, which receive students from countries where English is not the most spoken language.
What seemed particularly interesting to us, however, was that this differentiation between English and other languages used in postgraduate programs permeates the understanding of the relationships (personal and/or institutional) established by each faculty member and/or program. On one hand, languages such as German, French, and Spanish are used more widely than English by certain programs due to relationships that have already been established by them -that is, the use of these ISSN 2523ISSN -1154 SOTL in the South 5(1): April 2021 de Figueiredo, Jordão, Antunes, Emmerich, Cons languages is justified by prior existing relationships. On the other hand, in the case of the English language, its use and the emphasis given to it in the postgraduate programs seems not to depend on actual relationships between individuals and/or institutions, but on relationships that are imagined and/or desired by the participants and their programs. This seemed especially evident in comments by professors (but also by some students): I study English because, for God's sake, I need to learn this English (laughs), and I am studying German because of my master's degree scholarship. My professor is in Germany, and I am involved in this German environment (Carlos, student). Our interpretation of these reports and similar statements by participants was that while the use of other languages (including Portuguese) in the process of internationalization needs to be justified by 'real' relationships that are constituted in the programs, the use of English seems to be justified on its own -due to relationships that the participants and their programs intend to establish or foresee establishing. In other words, the desire for English, and even its use, actually reflect a longing to participate in an international academic community, even if that community is (still) on the level of the imagination for some individuals/programs.
While English is seen as having the potential to bring about a feeling of belonging to these communities, the language is also perceived as an excluding agent. For some of our participants, English seems to have a major role in deciding who will or will not succeed in postgraduate school. The comment below illustrates this possibility: . . . if I write there, in the text, that the student has to be prepared, let's suppose, I'm going to apply for a postgraduate course, ok? . . . it says 'the candidate must be prepared to listen and participate in seminars in English', I'm sure a large number of students will leave out of fear (Alessandro, student).
This understanding of English as a factor that can cause exclusion was familiar to us, since other studies have addressed it (see Rajagopalan, 2003). What seems worthy to stress here is that the desire and the imaginary associated with English are opposed to the fear or the rejection that it may cause. Thus, our data supports previous research that shows an ambivalent relationship with the English language on the part of Brazilian individuals and institutions (Rajagopalan, 2003). However, while Rajagopalan (2003) discussed this ambivalence as being part of a relationship of love and hate, we would like to emphasize here the onto-epistemological character of this relationship, which refers to English as being part of a desire versus rejection relation.
What is perhaps most important in the empirical material discussed here is the fact that English cannot be understood as the only language of internationalization, as it is commonly done. Rather, it needs to be considered within a plurilingual conception in the postgraduate field, where other languages may even have 'lower' status but are more frequent in academic practices. Although English often has the status of being 'the' lingua franca of academia, we have seen in our empirical material that this conception needs to be expanded to reflect what is happening in many of the relationships established between subjects and programs. This corroborates previous studies (e.g. Kubota & Mckay, 2009) that show that, despite the international status attributed to English, there are several other languages used as lingua francas in countless contexts -something that is established contingently by participants of each interaction.

Conclusion
In this study, we sought to understand how internationalization has been perceived, lived, and discursively constructed by professors and students of a publicly-funded university in Brazil. From the empirical material, it can be concluded that there are different experiences regarding how the process of internationalization is perceived and reported by each participant. Even so, at least some aspects seem to be common to several of our participants. Among them, we highlight: a) divergences between understandings and participation of professors and students in the process of internationalization; b) divergences of understandings regarding the roles of different foreign languages as part of this process; and c) the conception of internationalization as the establishment of interpersonal, intercultural and interinstitutional relationships, even if there is still a strong pressure from academic culture for understandings that privilege aspects such as mobility, branding and the preponderance of English.
We point out some practical implications of our results. Firstly, there seems to be room for and need for more dialogues and involvement with students regarding internationalization. We believe that this greater involvement can bring new possibilities for understanding the concept of internationalization itself. It can also establish new practices and increase the democratization of internationalization, so as to meet local needs and specific subjects in a more comprehensive way. As for the understandings about the role of foreign languages in the internationalization of higher education, we emphasize that the status given to English needs to be seen from a translingual conception, especially in view of the fact that other foreign languages are sometimes used more than English in different spaces, moments and contexts. This recommendation is in line with current discussions in the area of applied linguistics, where there is a tendency to increasingly see the importance of looking at subjects as practitioners of open and diverse repertories made up of varied semiotic resources (Canagarajah, 2013). If one of the main objectives of internationalization is the establishment of relations between different ways of building scientific knowledge (Martinez, 2017), then it makes no sense to plaster interculturality into the universe of English.
This translingual conception needs to go beyond postgraduate studies. While current language policies (e.g. the Brazilian Common Core Curriculum) exclude languages other than English from school contexts and academic spheres, the results discussed here show that the establishment of relationships at the international level takes place beyond the binary division between local versus global or between language X and language Y. Language teaching and language practices, therefore, would be better understood from a translingual conception -an ecology of languages and culturesin which English is perceived as only one of the possible options for the construction of intercultural relations. It seems fundamental to include the teaching-learning of other languages, not only in higher education but also in basic education and other contexts.
A final note we want to make is that despite the fact that a plurality of languages is highlighted in our data -and in spite of the multifaceted conceptualizations of internationalization we found in participants' answers -there seems to be little emphasis by participants on possibilities of learning from onto-epistemologies other than Northern, scientific knowledge systems. While some participants did mention possibilities of South-South collaborations, even in such cases internationalization and languages seemed to be mostly conceived on the basis of Northern, academic knowledges. We highlight the need for internationalization policies and practices to seek possibilities of engaging with ecologies of knowledge (de Sousa Santos, 2018), in which different, more symmetrical types of partnerships may be established.
It would be interesting for global South universities, such as the one investigated here, to engage with internationalization and language in ways that go beyond Northern understandings of higher education, and that instead build upon local realities/needs. Such endeavours may include not only searching for other ways of internationalizing, but also trying to redefine what internationalization is on a local level. Only then can we begin to understand internationalization more in terms of plurality and interculturality than in relation to a neocolonial hierarchy of onto-epistemologies. Future research may explore possibilities for developing such types of policies and practices in regards to internationalization in different institutions.