Japanese school administrators’ experiences after the implementation of the Top Global University Project language policy

As a result of globalisation, many higher education institutions in Japan have made changes to their curriculum and administrative system to attract international students. The Japanese government has also carried out different policies to support universities across the country and accelerate their internationalisation process. The study focuses on the implementation of the 2014 Top Global University Project (TGUP) in 37 Japanese higher education institutions. School administrators were interviewed in relation to their experiences with policy implementation. After coding and analysing the interview data, three themes were selected for discussion a) the financial burden; b) the English policy for staff; and c) policy transmission within an institution. Bernstein’s notion of the pedagogic device was employed to analyse school administrators’ discourses concerning the policy. Findings revealed how the process of implementation has taken place and the issues affecting its proper execution. The study concludes that top-down relations of power and control, hierarchies, people’s ideologies, and management culture affect and delay the internationalisation process of higher education.


Introduction
In the 1960s and 1970s, Japan was one of the few non-Western countries that had achieved economic success without making major changes to its traditional language and culture (Kubota 1998). However, in the 1980s, Japan's Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT) started to reconsider its position on internationalisation. Initially, this was seen as a process to attain and assimilate Western universities' education style into the Japanese higher educational system (Ebuchi 1989). However, in the late 1990s, the term "globalisation" was used by governmental organisations such as the Ministry of Economy, Trade and Industry (METI) and MEXT as an approach to deal with international trends (Yoshida 2017). As a result of the internationalisation process, government policies have generated a number of changes in the internal structure of national and private universities. For instance, adaptation to a new organisational culture affected their traditional structures and internal modes of communication (Ogawa 2002). The use of English as the medium of instruction to attract more international students has been one of the main reasons for such changes, and this is also what MEXT requires as one of the main steps towards internationalisation.
As a response to this requirement, Japanese universities have made changes to their traditional curricula, such as the creation and development of degrees taught entirely in English, initially in graduate schools but eventually at some undergraduate levels. As a result, by 2013, 194 universities across the country offered full English-taught undergraduate programs (Brown and Iyobe 2014). Concerning the role of English as the medium of instruction in Japanese higher education institutions (hereafter, HEIs), many empirical studies, mainly in the field of applied linguistics, have explored and described the path of internationalisation in Japanese tertiary education and its relation to the use of the English language (Bradford 2016;Bradford and Brown 2018;Chin Leong 2017;Galloway et al. 2020;Macaro 2018;Rose and McKinley 2017).
Since the implementation of the Top Global University Project (TGUP) in 2014, MEXT has been attempting to promote internationalisation by offering English medium instruction (EMI) courses in higher education. The main aim of this policy is "to enhance the international compatibility and competitiveness of higher education in Japan….[to provide] prioritized support for the world-class and innovative universities that lead the internationalization of Japanese universities" (MEXT 2014). The TGUP has become a focus of interest since changes in Japanese higher education are an example of the government's efforts to internationalise Japanese society. The study of this policy is of interest not only to Japanese readers but also to non-Japanese readers interested in higher education internationalisation. As Rose and McKinley (2017) pointed out, exploring this policy could also provide an illustrative example of failures or successes in internationalisation and language planning in higher education.
This study aims to provide an analysis of school administrators' experiences after implementing the 2014 TGUP. Bernstein's pedagogic device (1996) was employed to analyse the TGUP's recontextualisation process at four HEIs in Japan. The pedagogic device can serve as an analytical tool to understand how the process of interpretation of TGUP policy takes place and its effect on the language policy implemented by the institution. As Bertram (2012) has pointed out, Bernstein's pedagogic device allows a researcher to go beyond the normative question of how faithfully curriculum messages are interpreted and implemented to describing in nuanced ways the substance and nature of the messages carried by the policy and the ways they are re-fashioned, recontextualised, and re-interpreted as it moves through various levels of the education system.

Language policy and the internationalisation of higher education
Attention to language policy and planning due to globalisation increased during the 1990s, partly caused by the English language's popularity in the business sector and the threat it represented to small language communities worldwide (Hornberger 2005). This latter issue was raised by applied linguists concerned with the dominance of English since its increasing position promoted the disappearance and eventual death of languages, especially those with a relatively low number of speakers (Kirkpatrick 2010). However, such concerns did not prevent the implementation of EMI in polities that sought economic and social development to meet the demands of globalisation. Moreover, as expressed by Doiz et al. (2012), "in the international arena, the role of EMI at HEIs has primarily shifted to economic concerns ahead of cultural and academic considerations" (p. 214). This surge of instruction through the use of English in different parts of the world led researchers to consider the new role of English in HEIs. According to Dearden (2014), EMI was then defined as "the use of the English language to teach academic subjects in countries or jurisdictions where the first language (L1) of the majority of the population is not English" (p. 4). Since then, studies about the role of EMI at the tertiary level have increased considerably (e.g. Dafouz and Guerrini 2009;Dearden 2014;Doiz et al. 2012;Fenton-Smith et al. 2017;Hino 2018;Kirkpatrick 2014;Macaro 2017Macaro , 2018Macaro et al. 2018;Margić and Vodopija 2017;Nguyen et al. 2016;Rose and McKinley 2017).
In Japan, MEXT has primarily encouraged the internationalisation of higher education representing that part of the government in charge of creating and disseminating national educational policies. MEXT's guidelines are followed by both national and private universities in the country. Although they have a certain degree of autonomy, Japanese national universities and all public education institutions have been affected by government decisions, giving the impression that all public institutions in Japan are under the control of MEXT.
In the last 20 years, MEXT has implemented different educational reforms accompanied by financial support to improve the educational system and pursue the internationalisation of higher education (HE). Examples include two of the most salient reforms concerning HE internationalisation, such as the Global 30 Project (2009 and its follow-up, the TGUP (2014-2023). These two reforms have brought with them more than guidelines and money; they generate changes in policy documentation, universities' departmental decisions, internal communication among faculty members, and classroom practices.

The TGUP selection process and its language policy
The process of joining the TGUP began with the selection of the seven former imperial universities founded by the Emperor of Japan between 1886 and 1939, four national universities, and two private universities for a total of 13. They were assigned to the Type A group. This selection was based on the potential these institutions had to join the world's top 100 ranked universities. They were also part of the TGUP's predecessor, the Global 30 Project (2009)(2010)(2011)(2012)(2013)(2014), which aimed to support 30 universities, but only 13 were funded due to the project's financial difficulties (Ata et al. 2018). However, considering that other universities, mostly private ones, were already contributing to the process of internationalisation, MEXT decided to invite them to apply for this funding and included them in the Type B group (also called Global Traction Type). Besides, every year, a total amount of US$ 4.3 million was supposed to be allocated to each institution in the Type A group, and US$ 1.7 million for those in the Type B group (MEXT, 2014).
Nevertheless, some studies have reported that current funding is lower than that reported by MEXT and that funds have been used to increase the international presence within the institutions rather than to support other projects included in their proposals, such as research activities (Shimmi and Yonezawa 2015). However, the TGUP did set aside a specific amount of money for Type A and B institutions. It also included a list of criteria that the participating universities had to meet and report on at specified periods of time. This means, changes in policy aspects will always come from top-down, and any modifications to the initial policy document will always reflect the control MEXT has over its implementation.
As noted earlier, before the TGUP, the government funded 13 universities for five years as part of the Global 30. One of the project's goals was to increase the use of the English language in education, focussing on academic staff and the creation/ increase of degree programs in English for both undergraduate and graduate programs (Shimmi and Yonezawa 2015). Therefore, participating universities sought to increase the number of international faculty members, as well as international students. However, with the implementation of the TGUP, the English language policy was extended to the administrative staff. English language proficiency became a requirement for those working on the internationalisation of HEIs. The assumption was that English would act as the default lingua franca between them and non-Japanese students, which would promote the recruitment of international students. The influence of MEXT's fundings has affected not only language policy but also the development and progress of the institutions' initiatives towards internationalisation.

MEXT's influence on HEIs
While the decisions regarding curriculum, research, and recruitment of staff are made internally by each institution, government involvement, usually in the form of long-term grants, makes the decision-making process of TGUP universities' initiatives less autonomous. These universities knew beforehand that on being successful with their applications, MEXT would be indirectly involved in the development and perhaps the execution of the initiatives. In addition, MEXT's main request was for the universities to prepare for internationalisation and promotion of English education for undergraduate and graduate levels. Although MEXT funds represented a kick-start for those universities pursuing internationalisation, MEXT requirements for change extended beyond money. The Japan Scientists Association had previously criticised MEXT's style of promoting educational reforms through grants, stating that "the government presents desirable models and directions for higher education, allowing universities only the 'freedom' to select from the range of options presented" (Japan Scientists Association 2005).
In the case of the TGUP, in order to be selected, many participating universities adjusted their proposals to obtain MEXT approval. For instance, according to the study participants, their internationalisation plans anticipated a future increase in the number of international students, faculty members, and TOEIC scores by administrative staff during the first half of the project. Highlighting these predicted outcomes in the proposal would improve HEI's chances of selection.
The question now is whether, to what extent and how the TGUP-promoted internationalisation process affected an institution's organisation. Thus, examining the TGUP funding decision-making process and how decisions are communicated to staff members should provide information not only on the project implementation but also on the existing hierarchies, as well as the relations of power and control within an institution. These aspects, which affect policy implementation, will be analysed using Bernstein's notion of the pedagogic device.

Bernstein's pedagogic device
The pedagogic device describes the general principles which underlie the transformation of knowledge into pedagogic communication (Bernstein 1996). The device consists of three rules and three fields (see Table 1) containing agents Table 1 Rules, fields, and contexts of the pedagogic device Source: Adapted from Maton and Muller (2009) with positions/practices seeking domination. The three interrelated rules are termed 'distributive', 'recontextualising', and 'evaluative'. First, distributive rules attempt to control access to the field of production. This field, commonly represented by the ministry of education, is where the central government makes concerning education and its production. Second, recontextualisation rules are projected in the discourse employed in this second field of the same name. For instance, at government departments or offices affiliated with the ministry of education in charge of distributing and communicating educational policies or other requirements from the top. Third, evaluative rules that shape any given context of acquisition in the field of reproduction, i.e. institutions with their classrooms, teachers, and students (Bernstein and Solomon 1999)..
The fields of production, recontextualisation, and reproduction are also hierarchically related, which means recontextualisation cannot occur without its production, and reproduction cannot occur without recontextualisation (Singh 2002). In the production field, new knowledge, discourses, and ideas are created and modified, usually by academics. The selection of knowledge from the field of production takes place in the recontextualisation field, which is composed of two sub-fields: the official recontextualising field (ORF) and the pedagogic recontextualising field (PRF) (Singh 2002). The ORF includes the "specialized departments and sub-agencies of the State and local educational authorities together with their research and system of inspectors" (Bernstein 1990, p.192). The PRF is comprised of 1) university departments of education, together with their research fields; and 2) "specialized media of education, weeklies, journals, and publishing houses together with their readers and advisers" (Bernstein 1990, p.192). The field of reproduction is where teachers and students engage in pedagogic and assessment practice and where the evaluative rules regulate what counts as a legitimate production (Bertram 2012).
Bernstein's pedagogic device allows researchers to describe levels of hierarchies within society and the relations of power and control generated within these levels, in particular those where symbolic control is established by the state. For instance, in the recontextualisation field, agencies that are part of the symbolic control, public or private, and directly or indirectly connected to the government, regulate specialised discourses of communication that operate as dominant discursive codes regulating social relations, consciousness, and ordering (Bernstein 2003). These dominant discursive codes can be found in the pedagogic discourse employed in each of the fields of the pedagogic device, for example, in the field of recontextualisation through written documents demanding the execution or implementation of a state's requirement or through the speech of people working in the field, i.e. government agencies/agents. However, the pedagogic discourse reflects not only the specialised discourse but also the ideological orientation of its agents. The latter can provide examples of how hierarchies, ordering, and beliefs affect social relations among the actors, which consequently reveals how policy implementation is recontextualised in each field and their actors. As defined by Bernstein (2003), "Pedagogic discourse is a principle for appropriating other discourses and bringing them into a special relation with each other for the purposes of their selective transmission and acquisition" (p. 159).
There are two forms of discourse within the pedagogic discourse: instructional and regulative. Transmission and acquisition of knowledge are part of the instructional discourse, and the constitution of social order, relations, and identity are features of the regulative discourse. The instructional discourse is embedded in the regulative discourse, making it the dominant partner. For this study, analysing the pedagogic discourse can provide information on how government's goals for internationalisation of HEIs have been put into practice by the institutions. The internationalisation of higher education can then be analysed through the different policy actors involved in the implementation of the policy, which is found in the fields of production, recontextualisation, and reproduction of the pedagogic device. The pedagogic recontextualisation field (PRF) was selected to describe how the TGUP recontextualisation process has taken place in four participating institutions. In addition, attention was given to the pedagogic discourse employed by school administrators when referring to the policy. Investigating and analysing this discourse should provide information on how successful or otherwise the policy has been since its implementation.

Study
This study reports on an ongoing qualitative investigation into the impact of language policy implementation on Japanese HEIs. A total of 16 TGUP participating universities' school administrators were asked, via email, to participate in the study voluntarily. However, only four replied to the invitation. Using an ethnographic approach, the study focuses on school administrators from the four TGUP participating universities and their experiences since the implementation of the policy. A semi-structured interview was used as the data collection method since it facilitates the use of policy-framed themes and questions that can be adapted or reformulated without affecting the course of an interview. Further, a review of policy-related materials from the participants' university website was performed prior to the interview to facilitate the interpretation of their comments and experiences with the policy. The interviews were conducted in English and transcribed verbatim.
Although described as school administrators, the participants were those staff members who were actually working with the TGUP policy and were thus recommended by the institution as appropriate people to engage in conversations about the policy and its implementation. Thoughts on policy implementation were elicited through a set of policy-framed questions. The research project was granted ethical clearance prior to conducting the interviews.
The data collection instrument contained two sets of questions. The first set was on the TGUP implementation and its language policy for administrative staff, while the second one concerned classroom language. Only results from the first set were reported in this paper (Appendix A). After coding the data using NVivo 12, three main themes were selected as examples of how the recontextualisation process took place among the participants' universities. The selection of the three themes was based on their direct connection with policy implementation and its effect on policy actors' relations of power and control within the institution. Financial burden, English policy for staff, and policy transmission are the three themes selected for this paper. The discussion of these themes is representative of how the recontextualisation process has affected school administrators' discourse and its effect on the implementation of the policy.

Financial burden
School administrators used words such as "mandatory", "drastic", "shocking", and "stressful" to describe the financial difficulties they have faced since MEXT's decision on annual budget cuts. These words described the impact caused by the cut after the TGUP implementation. Three school administrators mentioned that the cuts generated a series of problems, particularly for small-sized institutions running projects that relied on MEXT's funds. Such projects could not be easily stopped since they involved funded project items such as exchange programs, offices in partnership's institutions overseas, recruitment of staff and faculty members, and newly created language programs. Only one school administrator (D) from a prestigious and financially strong institution made no comments concerning the budget cuts. It was reported that universities with a strong net worth are usually prepared to deal with these cuts without making changes in study programs or research projects.
Therefore, applying the policy can generate a set of negative emotions among policy actors such as anxiety, stress, and pressure (Ball et al. 2011), which was the case for the school administrators in the study. Further, MEXT's budget cuts were described by School administrators A and B as unethical and weird, respectively.
School administrator A (A Public University): "It is something that we did not expect to happen. We feel we were kind of cheated, they gave us that big picture, you are going to get money for this and that, but then, sorry we can't give you this much, sorry".
School administrator B ( A Private University): "Japanese system is very weird. They have these automatic 10% cut every year which we [participating universities] didn't know before we got here. 1 Only when I knew after getting the first cut and this is a good practice I was really surprised… our partner institution was also shocked with this 10% reduction".
Every HEI participating in the TGUP had to have a plan to achieve internationalisation for its selection. However, based on the school administrators' comments, their institutions were not entirely prepared to survive the budget cuts. They also reported that MEXT would not accept the budget cut as an excuse to stop running their proposed plans. This requirement to continue was directly generated by people in the ORF, who pressured the institutions to carry out the initial proposal. Thus, it does seem MEXT's involvement in their internationalisation plan was to kick-start it, not to maintain it. According to school administrators, this problem during the implementation of the policy was encountered by several TGUP institutions. School administrator B described the budget cuts: "We have these drastic cut like a 37% across universities two or three years ago, and then, MEXT made every university to run the program as we promised. So MEXT say, yes because of the cut you can or need to eliminate a few programs but MEXT still wants you to do what you promised in the proposal".
The financial difficulties communicated by the school administrators in this study revealed how relations of power control are acted out between MEXT and the TGUP institutions. MEXT wanted these institutions to operate at the same level as previously even after the budget cuts. For public institutions, and most probably private institutions, the pressure to proceed with the internationalisation project forced them to search for other funding alternatives. School administrator C (A Public University) reported that they could recover some of the costs after facing the budget cut thanks to their connections and agreement with certain companies. Nevertheless, this school administrator also noted that the budget cut was 'mandatory' and 'stressful': "We are now, is struggling because we don't have the money. We do have the pressure from the ministry of education we need to carry on with our promise so it is a stressful but this is a project so no choice we have to follow because we're national university so we have to follow the minister [Ministry] of education, it is as very strong institution".
These are examples of how MEXT, as a government institution and thus higher in the hierarchy, exerts control over the TGUP. School administrators' pedagogic discourse also revealed their negative emotions regarding the policy implementation. As mentioned earlier, their descriptions included words such as 'pressure', 'stressful', 'unexpected', and 'shocking'. This discourse is also a representation of the specialised discourse governing relations promoted by MEXT through the policy implementation and how important it is for participating universities to continue with the institution's internationalisation.
The financial issue reported by these institutions was likely to have taken place among the remaining 33 universities part of this project. As one school administrator reported, other institutions had to take strict decisions such as termination of short-term employment contracts, suspension of research funds, and discontinuation of scholarships for international students to survive the budget cut.
School administrator B (A Private University) recounted: "We did not hire many people with this money, we put the money for programs and the impact was smaller, but still that made everything very, very difficult for us, yes".
Since each institution has different administration and internationalisation initiatives, more research concerning this topic would need to be conducted to determine the possible problems MEXT's budget cut generated in other TGUP institutions.

The English policy for staff
When asked how the English language requirement affected administrative staff, two institutions reported that the use of English was already implemented before the TGUP. For instance, the recruitment process required submitting a TOEIC score; however, this requirement varies in each TGUP participating institution. Two school administrators reported that its submission was necessary regardless of the score, and the remaining two reported that a score of 650 points or more was expected for certain administrative positions. School administrators described a lack of concern in relation to the score as "usual" since most of the administrative tasks are still conducted in Japanese. Also, the submission of a TOEIC score is to acknowledge that as a HEI they are aware of the importance of this language for globalisation.
However, after implementing the policy, the submission of staff's TOEIC score became mandatory since these numbers must be sent in MEXT's reports. School administrators C and D recounted the reason for an English proficiency score among administrative staff: School administrator C (A Public University).
"TOEIC was not mandatory, but for the Top Global University Project we have to submit the scores of the administrative staff, so for our university, the level is 550 TOEIC score so every year we report that. I know other national universities encourage their staff to get 550 which is the standard score for some universities. It takes time to get that ability so unfortunately all people cannot get it".
School administrator D (A Private University).
"As for administrative staff, although it is not necessary to submit TOEIC exam, all prospective employees must take the TOEIC exam prior to the employment start date. Once employed, all administrative staff must periodically submit their TOEIC exam scores. Even though there is no minimum required score, we hope that this will motivate English language learning and acquisition among administrative staff".
For school administrator D, the main goal is for all permanent administrative staff to have a TOEIC score of over 800 points (approx. a 7.5 in the IELTS score system) by 2032. The main reason for this is to handle the demands of student/ teacher mobility and the increase of EMI programs on campus. The assumption is that a high score in the TOEIC test would be equal to good English-speaking skills which would facilitate the communication process between international students and administrative staff.
The effect of the English requirement on administrative staff in school administrator B's institution was similar to what school administrator D described above: "Our commitment is, I don't remember the exact number but, staff member with a TOEIC score higher than 700 should be more than 50% of the staff. What we do, if your score is lower than the standard, you are supposed to take some classes and you are advised to participate in English training programs provided by the university until you reach that level".
It was also reported that cultural ideologies, such as the belief that the Japanese cannot speak English well, still affect their confidence and willingness to improve it. In terms of the future of the TGUP English language initiatives after its culmination in 2023, the four school administrators reported that their universities will continue their efforts to improve administrative staffs' English proficiency since they all consider it a vital factor in the internationalisation of the university.

Policy transmission to staff and faculty members
According to the participants, the policy implementation process and dissemination pattern still strongly reflect the traditional Japanese cultural management of communication. For instance, they described a decision-making process as a linear management style where each superior has subordinates in a department who receive instructions and general information directly from their assigned superiors.
Most decisions regarding language policy are discussed by selected committees organised within the institution to discuss different matters concerning the creation, promotion, or dissolution of projects that involve staff, faculty members, and students. These committees follow what school administrators described as a deeply rooted Japanese business culture. School administrator C described decision-making and dissemination as: "Most decisions come from one office, but these decisions are based on the course of action and these decisions are not independent. It is a very difficult question because it's not just the culture of a Japanese university, it happens also at any company in this country. We always have to follow whatever the top wants".
Decisions are communicated in Japanese to department leaders, and then they decide whether and how to share the information. School administrator B reported that in some HEIs, information is collected through informal events such as luncheons or drinking events, which represent an opportunity to ask for more details on work-related matters that were not transmitted during an office meeting. School administrator B described: "Here is more like a performance, as pseudo democratic party of decision making, a body of decision-making [lead by one member], making some decisions, and then, they are supposed to share those results with their colleagues but not in the traditional way [the use of a memorandum]…and this style has been working in Japanese organisation for many years. We know how to get the information…, we go out drinking or we go out for dinner, and we gossip at the same time and we share information so that's how we, the kind of communication we do".
Information is then transmitted to other faculty members and staff if the department superiors desire. For example, school administrator A reported that departments with a large number of foreign faculty members require the presence of at least one bilingual staff who, during a meeting, will translate and communicate in English new procedures about language policy using the phrase "the university decided" as the introductory phrase.
When asked about the role of international staff in the decision-making process, school administrator B recounted: "International universities, like here, we do have to use different forms [of communication] which we haven't been doing. We said that we want to do it, but we haven't done [it]. We have failed to do it on this Campus, [it] is more than just a language policy, it is more a communication policy or the management policy as an international university, but we have been or maybe we'll be like that for another 10 years".
Concerning how policies regarding the English language are communicated to faculty members, all school administrators reported that, during the hiring process, they are not told about the language policy of the institution. It is normally taken for granted that they will teach their classes in English. Also, faculty members from non-Englishspeaking countries are not required to submit English proficiency test scores. In most cases, Japanese and international faculty staff who obtained an overseas degree from an English-based program are welcome to join the university. Thus, faculty members' command of English is commonly judged by their educational experience abroad.
School administrator D ( A Private University) explained the language policy: "When hiring faculty members, domestic or foreign, the university as a whole has no specific rules or policies when it comes to English language use or its proficiency, but there are situations where being able to teach English could be a requirement. However, a proficiency score is not required".
Findings suggest that school administrators may not be paying enough attention to faculty members' English language proficiency or skills. This could be due to the assumption that faculty members from English-speaking countries, and those with a degree from overseas, are automatically capable of delivering English classes, which is another issue resulting from the internationalisation process of HE in Japan.

Discussion
The present study used Bernstein's pedagogic device to analyse school administrators' experiences on policy implementation in higher education institutions. The primary question was whether, to what extent and how the TGUP-promoted internationalisation process affected an institution's organisation. Therefore, the analysis focussed on the recontextualisation field of the pedagogic device. Three themes were selected from the interviews to discuss policy implementation. These were financial burden, English policy for staff, and policy transmission. Findings suggest that the policy discourse employed at TGUP universities can be affected by different factors such as hierarchies, ideologies, and management culture of those involved in the distribution, communication, and reproduction of the policy.
Bernstein's pedagogic device allows researchers to describe the levels of hierarchies within society and the relations of power and control generated within these levels. School administrators referred to their existing relationship with MEXT as a top-down relationship. Their general description of the TGUP was 1) a rule that eventually most Japanese universities would need to follow and 2) an indirect command from the government to promote what they think is best for the country's economy and its citizens. According to the interviewees, MEXT's top-down influence and control started even before joining the TGUP. Although participating universities proposed and developed their internationalisation plans, MEXT's selection criteria list demanded a future projection in relation to the expected numbers of international students, and staff's English proficiency scores after TGUP implementation. This resulted in school administrators anticipating high numbers in their projections to increase their chances of selection. After joining the TGUP, they realised that more funding and time was necessary to achieve those numbers. The government's annual budget cuts exacerbated this situation, which affected the development and continuity of programs designed to promote internationalisation.
In the recontextualisation field of the pedagogic device, agencies that are part of the symbolic control, public or private, regulate specialised discourses of communication that operate as dominant discursive codes regulating social relations, consciousness, and ordering (Bernstein 2003). For example, school administrators' discourse revealed that policy implementation had been a difficult task. The received funds were just a kick-start to implement government's ideas on internationalisation. Interviewees' description of the recontextualisation process included words such as "cheated", "stressful", "struggle", "pressure" which are examples of the burden posed by MEXT's higher education internationalisation plan and its effect on TGUP discourse.
Another example from the TGUP recontextualisation field and its influence on discourse was found on participants' pedagogic discourse concerning the English language policy on administrative staff. As explained in the literature section, the pedagogic discourse reflects not only the specialised discourse but also the ideological orientation of its agents. Findings suggest that ideologies and assumptions seemed to have affected the role of language in the internationalisation process. For instance, the implementation of the TGUP required the submission of English language proficiency scores, normally from a TOEIC test. School administrators reported that while English lessons or subsidies for private ones have also been provided, many staff members have not achieved the target score submitted in their initial proposals for internationalisation. School administrator C reported that it is a task that requires time and certain language abilities that "unfortunately, not every Japanese administrative staff member can achieve it". It was also reported that among Japanese people, there is a widespread belief that they cannot speak English well and that the use of this language is not part of their daily routine. Staff's English proficiency is reported to MEXT with a TOEIC score. This test has a strict focus on English writing and listening; however, the assumption is that a high TOEIC score, 750 points or more, is equal to a high level of English-speaking skills. In addition, school administrators reported that obtaining a high score should ease communication between staff and international students.
The third theme selected for discussion, policy transmission, provided examples of how hierarchies, ordering, and beliefs affect social relations among policy actors. School administrators' pedagogic discourse revealed how management culture could affect policy implementation. While some universities have an office, or even a website dedicated to explaining TGUP's goals and those in charge of its implementation, other universities have one person or a small group working on varied administrative tasks, one of them being the TGUP. Policy transmission from the departments, or the person in charge, to staff and faculty members was reported as 'unusual' in Japanese management culture. Decisions are taken by people in top hierarchical positions and then briefly communicated to the staff involved, which means other staff and faculty members are unaware of how policy implementation has occurred. Staff receives information on policy implementation through both direct and indirect sources. For instance, gatherings during non-office hours with other staff members from top positions who informally shared official or unofficial information concerning policy and its implementation. This management culture was described by school administrators as traditional, not only among Japanese universities but also in companies. For example, one school administrator acknowledged that, to be truly international, this current traditional communication style should not exist. Although the institution is aware of the negative effect this has on their internationalisation plans, they are likely to maintain it at least for another ten years or more.
Using Bernstein's pedagogic device, the study has provided an analysis of the pedagogic discourse employed by school administrators when referring to the recontextualisation field of the TGUP. The analysis showed that policy implementation has been affected by issues concerning hierarchies, ideologies, and management culture of those involved in the distribution, communication, and reproduction of the policy. While more research using Bernstein's field of recontextualisation is necessary to provide information on how successful or otherwise the TGUP has been since its implementation, this study provided significance information on aspects affecting and delaying Japan's higher education internationalisation process.

Conclusion
To conclude, HEIs in Japan working towards the internationalisation of education should consider that the presence of international students and faculty members generates changes in their management culture. While MEXT's control over the policy and its budget cuts have directly influenced the implementation and development of HEIs' internationalisation initiatives, school administrators' pedagogic discourse revealed that issues with policy implementation go beyond the financial ones. Moreover, the existing organisation structure and traditional styles of management and communication will directly affect their plans for internationalisation. The presence of foreign faculty members and students from different parts of the world in an educational institution with a complex hierarchical system, as well as its people's ideologies, cannot be an effective path towards internationalisation. Although more research on this topic could provide information on other issues affecting universities and their plans for internationalisation, the author of the present study argues that HEIs should re-evaluate the TGUP implementation and consider participation from their internal diverse population. Concentrating all decisions within a small group of people with traditional management ideas will slow Japan's higher education internationalisation process.

Supplementary Information
The online version contains supplementary material available at https:// doi. org/ 10. 1007/ s43545-022-00318-7 Funding Open Access funding enabled and organized by CAUL and its Member Institutions. Not applicable.

Data availability
The data sets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Conflict of interest
The author has no relevant financial or non-financial interests to disclose.

Ethical approval
The study gained ethics approval from Griffith University (GU Ref No: 2019/608) prior to data collection.

Informed consent Informed consent was obtained from all participants.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.