Introduction

The rise of world-class university (WCU) rankings reflects the growing recognition that international competitiveness and economic growth are increasingly driven by knowledge, and that universities can play a key role in this context (Salmi 2009). Many critics claim that WCU rankings have transformed the competitive, cross-national, higher education market since their introduction (Altbach and Balan 2007; Hazelkorn 2014; Marginson 2013). In response to this global phenomenon, many countries have introduced national policies that seek to establish their top research institutions as world-class universities. China, Finland, France, Germany, India, Australia, Japan, South Korea, Latvia, Malaysia, Russia, Singapore, South Korea, Spain and Vietnam, to name a few, have launched initiatives to create WCUs (Deng et al. 2010; Hazelkorn 2013). In a similar vein, the Taiwan Ministry of Education (MOE) introduced the Aim for the Top University project (ATU) in 2005, which has progressed to a second phase that ended in December, 2017. The ministry aims to improve the worldwide academic competitiveness of selected universities with the introduction of competitive funding and a periodical review process. To compete for this funding, HEIs were required to submit a written project proposal to a review committee that consisted of around 20 highly esteemed academics and experts from both Taiwan and other countries. Institutions selected for this project would be funded by block grants from the MOE (2006). Furthermore, the executive plan and ongoing performance of each recipient institution was evaluated annually, with the results determining future funding (Chang et al. 2009).

Although the international impact of WCU rankings is widely acknowledged and WCU policies have become a global phenomenon among national governments, the empirical effects and policy strategies outlined in the literature tend to treat the WCU issue rather simplistically and take it as a universal given. However, some empirical evidence demonstrates the important role that regional actors and contexts can play in shaping a local model of WCU. For example, Erkkilä’s (2014) discourse analysis of European policy concludes that the rankings discourse has acquired different national attributes across the EU, “echoing institutional traditions, public values, and historical narratives on education” (p. 95). Comparing WCU policies in Europe and East Asia, Deem et al. (2008) argue that the meaning of the term WCU is somewhat fluid. It is mediated by each university’s specific context and is not always related to its actual world university ranking. While we may summarise or reduce the characteristics of a WCU to some ‘key’ attributes, we must avoid over-simplification and the urge to apply a single model to universities in divergent contexts. Although there is some research on the impact of ATU (e.g. Chou 2014; Chou et al. 2013; Chang et al. 2009), the fundamental question of how the idea of world-class university is interpreted, shaped and mediated within the island and how the two phases of ATU had evolved have never been critically explored. While ATU is the strong state-initiated policy with the largest amount of funding ever in Taiwan, concern was also raised regarding the resources for the other institutions outside ATU would be reduced under the zero-sum balance of national annual budget and their students’ right of enjoying the same level of education quality would therefore be sacrificed (Guo 2009). Accordingly, a critical reflexivity on how this project shapes our understanding of higher education is necessary. Follows Erkkilä and Deem, Mok, and Lucas’ argument, this study aims to explore how the idea of WCU is constructed and presented in the ATU project. It also seeks to understand how the ministry’s understanding of WCU affects the governance of top universities in the Taiwan context. The specific research questions can be summarised as follows:

  1. 1.

    What are the management policies and strategies that have been adopted by the Taiwan MOE since the implementation of the ATU?

  2. 2.

    What are the fundamental assumptions regarding the idea of a world-class university and how should these institutions be governed within the ATU? How are these related to WCU or other regional and local contexts in Taiwan?

  3. 3.

    How and by whom are management policies and strategies supported and challenged?

Literature review

Foucault’s notion of power and its relevance in studies of higher education

In the higher education literature, many studies have applied Foucault’s notion of power to the study of knowledge management (ie., Gordon and Grant 2004; Messner et al. 2008) and national research evaluation (ie., Broadhead and Howard 1998; Harding and Taylor 2001). These studies have shown how Foucauldian power can be a useful concept for enhancing their understanding and analysis of higher education policy and management impact issues. There are also studies in higher education that adopt Foucault’s conceptual tools. For example, in Erkkilä’s (2014) discourse analysis of European policy, he came to the conclusion that the rankings discourse had acquired different national forms across the EU, “echoing institutional traditions, public values, and historical narratives on education” (p. 95) which is discussed in previous section.

In a similar vein, this study aims to explore how Foucault’s notion of power offers a useful perspective for examining Taiwan’s pursuit of world-class status in higher education. In Foucault’s perspective, truth and knowledge are strategically constituted as an outcome in the process of power struggles (Gordon and Grant 2004). Meanwhile Foucault (1995) contended that nothing has meaning apart from discourse; only through discourse can power relations and knowledge be established and sustained. In Foucault’s term, discourse is far more than language. Anything which contains certain knowledge could be seen as kinds of discourse like customs, rules and practice in human life. He treated it as “sometimes the general domain of all statements, sometimes as an individualisable group of statements, and sometimes as a regulated practice that accounts for a number of statements” (Foucault 1972, p. 8 as cited in Carabine 2002). Therefore, discourse defines a group of conditions of existence based on which people could understand and represent power/knowledge in a given moment. It represents the internal rules and practices implicitly hidden in power/knowledge. In other words, by means of discourse, imagination and social practices are constrained; the relation and function of power /knowledge are established and consolidated.

The global interest in WCU rankings has been a game-changer for cross-national higher education competition (Altbach and Balan 2007; Hazelkorn 2014; Marginson 2013). In Foucault’s term, the discourse of world-class university can be argued as another form higher education discourse with transforming power potentially shaping our knowledge and imagination about the university. In the following sections, this argument is further elaborated and how Foucault’s conceptual ideas can contribute to current study is summarised.

Transforming global higher education: the discourse of the world-class university

Globalisation in higher education has been described as an international tendency towards the homogenisation of education practices in a competitive global marketplace (Currie 1998). It is understood to be part of an effort to secure an advantage within what Fotopoulos has called ‘the market economy’s grow-or-die dynamic’ (Fotopoulos 2007). In this context, universities no longer operate in a vacuum, but are influenced by local, national and international context (Altbach et al. 2009).

One current example of these homogenising practices and how they feed into grow-or-die forms of competition in the education sector can be seen in the rise of global university rankings. Since the first publication of rankings by Jiao Tong University in 2003, global comparisons between universities have attracted a great deal of attention from universities, governments and public media worldwide. The popularity of global university rankings reflects the general recognition that, in the age of the knowledge economy, universities play a key role (Canto-Sperber 2009). Thus, the bond between the development of universities and the nation in which they reside is emphasised.

In other words, the global rankings have geographic implications, as they produce rankings not only of universities, but also of countries and regions (Erkkilä 2014). The European Commission has noticed the failing of European universities in the global rankings and has described this as a serious problem, an issue of regional competitiveness in Europe, needing immediate action. In this context, the U-Multirank was created as a tool that might ‘balance’ the other ‘problematic’ global university rankings that were accused of favouring private US universities (Marginson 2013, p. 2). Accordingly, the global ranking of universities is no longer merely a tool for determining education quality, but also arguably a device for judging a university’s, nation’s or region’s global competitiveness. This interpretation of the rankings has led to their use for the intensification of comparisons between nations and an increase in the number of national initiatives to improve the performance of universities in the global rankings. This raises questions about the potential impact of the rankings and the appropriate response to their growing influence. Does this game-changer form a new threat to the governments’ power? How do WCU rankings enhance global competition? Does this process intensify or weaken the globalisation of higher education, especially with respect to homogenisation and convergence in the idea of WCU? And finally, how should institutions seeking WCU status be governed?

The world-class university: an emerging/converging global university model accepted worldwide?

Traditional theorists tend to see power as an entity that belongs to certain groups of people. For example, Marx considered power as belonging to the dominant class who has the wealth/power as a means of production (Gordon and Grant 2004). It poses a fixed top-down relation between the oppressor and oppressed. In contrast, Foucault argued that power is not an entity which is “neither given, nor exchanged, nor recovered” (Foucault 1980, p. 89). Moreover, “power comes from below; that is, there is no binary and all-encompassing opposition between rulers and ruled at the root of power relations, and serving as a general matrix” (Foucault 1972, p. 94). Power, in Foucault’s eye, must be something acknowledged by others who meanwhile become the dominated. Therefore, power is a fluid concept which could be localised at any point of the social web. Meanwhile Foucault believed there are always resistances: “they (resistance) too (like power) are distributed in irregular fashion… spread over time and space at varying densities, at times mobilising groups or individuals…” (Foucault 1972, p. 96). Once these resistances grow large enough in a certain time and space, power relation could possibly shift and a new power relation could possibly emerge. Accordingly he also denied the possibility of absolute truth; all political and social factors are inevitably involved in the interplay of knowledge and power and thus determines what true is at a given time and space (Wetherell et al. 2001). Therefore “Each society has its regime of truth, its general politics of truth; that is the discourse which it accepts and makes function as true” (Foucault 1980, p. 131). In brief, the historical and social context are vital in his analysis of power and knowledge and this is also echoed by empirical evidence from the east and west.

In their comparative study, Altbach and Balan (2007) argue that, with the growing competition between nations, governments and universities in Asia and Latin America have tried to improve the quality of their education and research with a variety of methods. In general, South American countries aim to create their own models rather than imitate the so-called world-class universities. In fact, some Latin American state-building and flagship universities are not research universities at all, especially in Brazil, Mexico and Argentina. More specifically, Altbach and Balan (ibid) identify a range of contextual circumstances that could prevent the transformation of these institutions into research-oriented universities. In the first instance, there are no guarantees that faculty in the most prominent universities in the region will support such reforms, as changes may threaten the interests of staff. Another possible impediment is the size of the universities themselves, which in many cases are too large to fit the principles of economy of size. From this evidence, it would appear that universities and governments in South America are not particularly interested in transforming their leading universities into research institutions (ibid.). Indeed, Altbach and Balan (ibid.) argue that this is perhaps due to the historical context within which South American universities emerged. Unlike universities in Asian countries, which were largely influenced by Anglo-American traditions during their establishment and at key moments in their development, universities in South American countries have closer historical ties with Spanish and Portuguese educational models.

On the other hand, Asian nations, unlike their South American counterparts, have made a range of efforts to build world-class research universities. This is demonstrated by Chan and Lo (2008), who compares the internationalisation strategies of four Asian countries. He points out that these nations have organised their internationalisation strategies and practices according to opposing ideologies whose differences have resulted in the formation of two separate clusters. Chang (2010) also argues that Japan and Taiwan tend to emphasise domestic improvement as a means for providing better education environments. Their goal is to attract international students and export local educational institutions to foreign countries. Singapore and Malaysia, by contrast, engage more with international providers as a means of improving their national education sector. These findings are also echoed by Marginson (2013),who contends that different state forms and political cultures shape the distinctive roads that lead to the WCU. Moreover, it is noticeable that “the different roads (and systems) of higher education tend to be not so much national, as regional, or sub-regional, reflecting historical overlaps and clustered cultures” (ibid., p. 20).

In the west, one interesting European example has been provided in Erkkilä’s (2014, 2013) work. Erkkilä (2013) argues that, in general, global university rankings have led to a degree of convergence in higher education globally. He gives the example of the marketisation of higher education, which has come at the cost of diversity in the sector (Erkkilä 2013). However, in Erkkilä’s (2014) later discourse analysis of European policy, he comes to the conclusion that the rankings discourse has acquired different national forms across the EU, “echoing institutional traditions, public values, and historical narratives on education” (p. 95). He (ibid.) argues that the pressure to achieve high rankings has been felt most in small European countries, such as the Nordic countries. They now favour international journals and publishers over local outlets more than ever before, which has developed into “a language issue, as journals in the domestic languages are now struggling to survive” (Erkkilä’s 2014, p. 96). However, his analysis shows that the effects of global rankings in Nordic countries have not been without limitation; most notably, the Scandinavian welfare-state model and the commonly held belief that education is a public good have likely challenged WCU narratives. As these discourses compete for prominence, they are likely to “form different layers of old and new institutional forms that may exist side by side” (Streeck and Thelen 2005, as cited in Erkkilä 2014, p. 96). In contrast with the Nordic example, Erkkilä’s article (2014) indicates that the effects of global university rankings are least felt in the UK. These findings are echoed in recent research that highlights how the different effects of global rankings are felt across Europe. This may be due to the size of each HEI and their historical reputation (Marginson 2013). In another study that compares WCU policies in Europe and East Asia, Deem, Mok and Lucas (2008) argue that the meaning of the term ‘WCU’ is somewhat fluid, and is mediated by the contexts in which universities operate. As a result, the policies that emerge from national governance bodies do not always seek to improve the actual world university rankings of institutions under their control. While the characteristics of the WCU may be reduced to some ‘key’ features, they are likely to be over-simplified and should not be expected to be applicable to all universities in their divergent contexts.

In light of these insights, this study argues that the transformationalist perspective might offer a more convincing framework for explaining the changes in higher education brought on by the pursuit of a world-class reputation. The effect of global rankings is better understood as mediated by regional and local contexts that can best be ‘managed’ by states or regional/international organisations. The current study takes the ATU project as an opportunity to explore how the Taiwan MOE responds to the emerging influence of WCU rankings and how local or regional contexts mediate this process. More specifically this research would mainly focus on how the idea of a top university would be defined and how it related to specific managing strategies which are promoted at the institutional level in Taiwan. In the next section, we further discuss how Foucault’s conceptual ideas can serve as the theoretical lens for exploring the questions of concerned.

Using Foucault’s conceptual ideas as the theoretical lens for current study

In Foucault’s perspective, discourse embodies social relationships and meaning, and thus forming subjectivity and power relations (Ball 1990). A key to this process is that power/ discourse has to be recognised or accepted by the dominated which is closely connected to the identity-constitution or self-discipline of the dominated (Deacon 2002). Through these power/knowledge dynamic, people knows who they are, what they should do, what right and other knowledge about the world surround them. This is the process of ‘subjectification’ and the ‘embodiment of discourse’. On the other hand, Foucault (1972) further argued that “knowledge is (also) a power over others, the power to define others”. For example, Foucault’s work Madness and Civilisation demonstrated how we identify ourselves with and also become the object in the mainstream discourse. While we struggle in the power relationship and constitute ourselves through the exclusion of the others, in this case, the insane, we claim that we are ‘normal people’ and meanwhile we objectify other people by identifying them as ‘the mad’. In other words, in the process of subjectification, we are not only subject to the ‘truth’ but also objectify other people at the same time. It is through the continuous identity-constitution and objectification that people can make sense of their or others’ ‘right’ place. In this way the power relation is thus formed and also consolidated. In this process, it is emphasised that individuals are not merely passive receivers of discourses and discursive subject positions; they can also be the agents of discourses and struggles for power.

Foucault’s macro perspective of modern state power is another useful theoretical lens for current study. In his work Discipline and Punish: the Birth of the Prison, Foucault used the prison as an example demonstrating how disciplinary power works in which the relationship between the individual and the State in modern time is also explained. The success of disciplinary power lies on “technologies of power—simple instrument (quantitative management technique, such as the registration of the pathological), hierarchical observation (the observation of each subjects are done by rather than one single observer but multiplication of supervisory statuses which are hierarchically ordered), normalizing judgment (departures from what seen as correct would be punished) and...the examination (the ceremony of the objectification process)” (Foucault 1995, p. 170). All these save trouble for the central state to pose direct surveillance on individuals and induce in the prisoners the permanent visibility which enables the “automatic functioning of power” even if the surveillance is not in action all the time (Foucault 1995, p. 200). Moreover, people are not only objects but also vehicles/instruments of the exercise of power. They internalise the normalising judgement, becoming part of the permanent surveillance to ‘gaze’ at other people’s every move. It is with the recognition and acceptance of people from ‘below’ that the function of power is sustained (Foucault 1995).

While there are more and more research adopting the approach of critical discourse analysis (CDA); for example, Norman Fairclough is one of the most cited author, and it does have a more systematic approach to follow when conducting the analysis. However, CDA is closely related to a specific set of social theories about the nature of language and literacy (Lin 2014) which to some degree beyond the researcher’s scope of interests. Despite the broad application of his work in current research, Foucault left no extended methodological commentary. Accordingly, this study uses his work as a ‘toolkit’ (Foucault 1980) to construct a broad analytic grid (See Fig. 1). His ideas, given their sensitivity to political dynamics, are appropriate for this study given that the concept of the WCU itself is always contestable and fluid. The analysis would be focused on how the concept of ‘top universities’ is articulated in the ATU and the power techniques used in the policy to make the process of subjectification and objectification possible. Moreover, the analysis would also pay attention to counter discourses and resistance in the ATU power network. In this context, the value of applying Foucault’s conceptual ideas in the current study is to produce a critical reflexivity of the ATU rationale, legitimacy and its impact on idea of a ‘top university’. In this way, we might better understand how the power dynamic within the ATU policies actually worked; in Dean’s (2010) term, to ‘denaturalise’ the ATU (p. 6).

Fig. 1
figure 1

A mind map of the possible implication of Foucault’s conceptual ideas in my research

Full size image

Research design

Since this study is itself a discourse that provides particular ways of understanding the issues at hand, varied sources of data have been used to present a balanced picture and to explore the complexity of the ATU discourse (as summarised in Table 1). The two interviewees, Mary and James, are both managers in the Department of Higher Education (DoHE). They are directly in charge of the ATU, and in most cases they represent the MOE in negotiations with leaders of ATU-recipient universities, interactions with the public media and relations with other government officials. It could, therefore, be argued that they are the officials in the MOE that know the ATU best and are potentially among the most appropriate advocates for the government.

Table 1 A summary of the main data sources
Full size table

Data analysis

In his analysis of power, Foucault emphasises the analysis of strategies and tactics as a basis for power struggles (Bent Flyvbjerg 1998). In this sense, the ATU can be interpreted as a strategic move by the MOE to compel its top universities to compete with other, internationally outstanding HEIs. In this section, this study examines the two phases of the ATU to further compare the MOE’s strategies and tactics in the following terms: how the MOE interprets the idea of the world-class university and the political strategies and tactics adopted by the MOE through which recipient universities become the objects of particular kinds of knowledge that construct them as ‘top universities’. Moreover, the case of Chengchi University’s resistance to the ATU is also discussed in the first phase section to provide a background for the transformation from the first phase to the next.

The first phase of ATU (2005–2010)

The MOE’s imagination of the idea of a world-class University

In the context of ATU, the most important concept is the ‘World-Class University’. In the first phase of the ATU, the main objective was outlined as follows: “In 10 years, at least one university will become one of the world’s top 100 universities. In 15–20 years, that university will become one of the world’s top 50 universities, with several research centres in that university having the potential of becoming some of the world’s Top-level research centres” (MOE 2006, p. 1). In brief, from the very beginning, the term world-class university was simply the same as ‘the university ranked as top 50 in any one of the world-class university rankings which has several top research centres’ (MOE 2006, p. 1).

This first conceptualisation was seriously criticised by various institutional leaders and individual academics. The Control Yuan (2010), an investigatory agency that monitors the other branches of government in Taiwan, raised a critical concern about the ATU’s strategies and its purposes. In its correction reports, the Control Yuan argued that ‘global university rankings are many but it is not clearly indicated which ranking would be the reference for ATU...Too much emphasis is placed on the rankings which are questioned about their validity in explaining what a WCU is’. Moreover, it is argued by the Control Yuan that since this funding is from taxpayers across this country, the funding recipients should also respond to local or regional economic, social and political issues and needs rather than those that feed into global markets or rankings.

Having Taiwan University better ranked in world-class university rankings means nothing but only satisfies the government officials excessive prid...ATU only has turned universities into expensive factories producing cheap research papers which contribute nothing to this country...The MOE should be ashamed of it...They waste tax payers’ money on inefficient public universities (Song and Tai 2007, p. 335).

Amongst all the other governmental and non-governmental institutions, the Taiwan Higher Education Union (THEU) is the most critical of the ATU. It is a non-government organisation aiming to protect academics’ employment rights. Its members and leaders are academics from local HEIs, including ATU-recipient institutions. The above quote indicates that the limited focus of the ATU could intensify inequalities among HEIs, especially between public and private institutions. The inequality has long exist between the public and private institutions given that The public HEIs received about 56% of their annual revenue from the MOE while 20% from private sources. Moreover, public institutions received funding unconditionally from the MOE while whether the private institutions are funded based on their evaluation performance held every 5 years by the MOE. If the MOE fails to pay due attention to the ‘overall development’ of higher education in Taiwan it could impact on the rights of students in non-recipient institutions to an equitable education. The Control Yuan (2010) also warns that the concentration of funding in the 11 recipient institutions could undermine the overall development of Taiwan’s 153 institutions. This is especially true for the private HEIs, which, as the quote above suggests, have become more and more marginalised. It is argued that what the MOE is doing amounts to “robbing the poor (private universities) to feed the rich (public universities, especially ATU recipients)” (Control Yuan 2010, p. 4). If continued, this could lead to a vicious circle in which non-ATU intuitions are more and more marginalised, especially the private ones.

Embodying the WCU imagination through disciplinary techniques: surveillance, normalisation and division

The success of disciplinary power lies in ‘technologies of power - simple instrument, hierarchical observation, normalizing judgment and … the examination’. (Foucault 1995, p. 170)

Being included as ATU recipients does not mean that these top universities can keep the title ever after. The continual competition and funding insecurity that universities have to deal with is clearly promoted by the MOE (2006, 2011a) with recipients expected to submit an annual self-review report to determine whether they and their departments are ‘on the right track’. In the third year of each phase, there is a formal review conducted by the MOE review committee. In practice, when conducting self-reviews, each institution would ask its departments to submit their own self-review reports too. As a result, a hierarchical, three-level, observational mechanism is produced.

To effectively govern this structure, it is also necessary to adopt a simple instrument to render visible the space over which government is to be exercised (Rose 1996). The evaluation criteria for the first phase include three measures: academic performance, internationalisation and financial gain form the private sector (MOE 2006). These quantitative instruments provide a base for developing performance measures, normalising judgment regimes and instituting a hierarchical observation structure through negotiations between departments, institutional managers and the MOE.

The performance indicators are provided as a public reminder of what aspects recipients should pay attention to … and yes, academic performance, internationalisation, national scientific, and educational collaboration are the basic … the most important things we want them to work on specifically for the ATU … the last group (other aspect group) provides a chance for recipient institutions to set goals based on their institutions’ advantages or their specific development objectives (Mary, a manager in the DoHE).

The above quote seems to echo Foucault’s (1995, p. 171) argument that ‘Discipline operates by a calculated gaze’. Although the MOE has explained that university rankings are not the only indicators it uses to determine funding decisions, there is always criticism of the MOE’s increased emphasis on the use of rankings and other quantitative information for evaluating funding recipients. In addition to the widespread use of the performance indicators, the way the MOE communicates ATU outcomes and achievement also demonstrates this inclination towards the use of data. According to James, these outcomes are usually summarised in several newspaper articles that emphasise the numbers, growth rates and the results of the submission reviews. On TV, the reporters usually focus on two things: the rankings and review results. ‘There is no other alternative; it’s the easiest way for them to do their work, but as you can imagine, people had the impression that the MOE only cares about rankings and numbers’, James added.

Interviewer: Was it difficult to determine the review criteria?

Mary, a manager in the DoHE: Yes, the main problem is that it’s impossible to please everyone, even within the government. The Control Yuan criticised the use of rankings as performance indicators, but in fact this was done due to the suggestion made by the Legislative Yuan (the parliament in Taiwan). Their points all make sense to me; however … WCU is a vague concept…sometimes quantitative indicators might provide an easier way for universities to follow. …The ranking says itself: ‘Behold, these are WCUs!!!’ The rankings might not be perfect but I think it is still worth our attention…The point is the rankings are not the only indicator to inform the MOE and the review committee’s judge but the public’s unfair criticism reveals this kind prejudice (that the MOE only cares about rankings).

As noted, to govern, it is necessary to render visible the space over which government is to be exercised (Rose 1996). Flow charts, maps, pie charts, graphs and tables are examples of tools that make visibility possible. All of these technologies spare the central state the trouble of imposing direct surveillance on individuals at all times to induce the ‘permanent visibility’ which enables the automatic functioning of power (Foucault 1995, p. 200). While the MOE recognises the limitations of rankings and other quantitative indicators used in ATU reviews, the MOE argues they might still be useful for enhancing the visibility of problems that need further interventions. In other words, they simply tell the MOE and funding recipients which aspects they can improve upon which at the same time makes the funding recipients productive in the way defined by these criteria set by the MOE.

Another dilemma that makes quantitative indicators a sensible choice is the problem of policy communication. According to James, the ATU captures everyone’s attention, and its outcome always needs to be examined. While the review committee would love for every funding recipient’s submission report to find a wide audience, they understand that other government officials and the general public will only ever gain a partial picture. The political reality is that the MOE has to represent taxpayers and other sectors of the government (like the Control Yuan and the Legislative Yuan). The MOE needs to explain the outcomes to them simply and concretely, which again reinforces the perception that these quantitative indicators, especially the WCU rankings, are the MOE’s only concern.

The distribution according to acts and grades has a double role. It creates gaps and arranges qualities into hierarchies, but also punishes and rewards. Discipline rewards and punishes by awarding ranks (Foucault 1995, p. 171).

The ATU might well be seen as another form of classification and ranking activity for national universities. The funding recipients are officially granted the status of ‘first-class universities’, while the others are not. This division establishes an inclusive us and an exclusive them. Based on this division, the ATU automatically rewards or punishes those it classifies in terms of how much extra funding they received from the MOE. Moreover, in return, it is common practice for the public media to create a table listing the recipients in the order of how much funding they receive from ATU. This generates an informal national ranking of Taiwanese HEIs that reinforces the hierarchy produced by the ATU results. Because of the fear of being excluded (punishment) and the desire to be included (reward), institutions have to play by and identify with the MOE’s rules to maintain their competitive and comparative advantage over other universities. Through these disciplinary techniques and ‘the constraint of a conformity that must be achieved’ (Foucault 1995, p. 183), ATU constitutes productive and docile subjects at the same time. Therefore, the power of ATU is not primarily repressive, but productive.

In next section, this study explores how the play of inclusion and exclusion promotes and imposes an alternative (desired by the MO) subjectivity (WCU) that results in the objectification of ATU recipients.

Subjectification and objectification in the ATU: Chengchi University’s resistance

In the first round of reviews in 2005, the MOE selected 12 universities as recipients of the ATU. In 2008, 2 weeks before the publication of these results, information was leaked out by one of the review committee members. The release revealed that Chengchi University, a school that is reputed for its social science and humanity schools, and Yuan Ze University (a private institution) were to be excluded from the project because their performance was comparatively unsatisfactory to the review committee (they both ranked as ‘average’ in 2007 and 2008).

This news prompted a dramatic response from Chengchi University and other social science scholars. It took the form of articles in newspapers and commentary in other public media that expressed their disappointment about this decision. The director of research and development at Chengchi University argued that more than 80% of the members of the review committee had a background in natural science and that review indicators like the number of research papers published in the SCI, SSCI and EI also suggested that the MOE did not show adequate respect for the social sciences (Chen 2008). Other scholars argued that this augured the death of the humanities and social sciences in the ATU and demonstrated the MOE’s ignorance of the importance of this field (e.g. Han 2008). According to a newspaper article at the time, the MOE responded quickly to the controversy:

The president of Chengchi University called on the Minister of MOE right after the leakage of the bad news. After their private, 40 min meeting, the Minister of MOE said the MOE would respect the professional decision made by the review committee, he will call for another meeting with the committee before the final decision is made. One week later, the MOE announced the final list. All 12 recipients stayed except Yuan Ze University. .. Ironically, Chengchi University was well prepared for the press conference to condemn and denounce the MOE, but turned out to celebrate its success when they know the result. .. the president said ‘Thanks for the MOE for the decision from an academic-field balanced thinking which makes the young in this country believe there is still hope to major in humanity and social science subjects’ (Wang 2008).

These debates about Chengchi University’s ‘survival’ reveal two divergent attitudes. The campaign in support of the humanities and the social sciences celebrated what they deemed to be a good decision. It was interpreted as a victory of humanity and social science and was understood to have advanced the overall development of higher education in Taiwan. In short, Chengchi University offered an interesting example of how to challenge the dominant, official ATU discourse. Instead of presenting its exclusion from the ATU as a failure, the university described the original decision as a general threat to the humanities and social sciences. Competition for ATU recognition was conceived of as a war between the science and social science campaigns. The controversy in the public media drew the attention of other social scientists from across the nation and united them to resist the MOE. This was the first—and, so far, last—time that the MOE’s central role in the ATU was challenged by non-governmental actors.

In the aftermath, some university presidents questioned the legitimacy of Chengchi University’s claim to top university or WCU status. They argued that this process had been unfair, especially for Yuan Ze University, which was excluded from the ATU simply because it “respected and trusted the professional decision of the review committee” (Han et al. 2008, p. 1) and stayed silent. Many HEI leaders believed that political concerns had interfered with professional judgment. The concern was that if the president of Chengchi University was able to influence the ATU result, other presidents might do the same. In response to a reporter’s question, the Minister of the MOE (2011a, p. 2) explained:

The exclusion of the Chengchi University from the ATU was just a rumour … Chengchi University’s review results were average but better than Yuan Ze University, the committee’s decision was always to keep it in the ATU … the decision was absolutely objective and fair without being influenced by anyone or particular groups. … The president of Chengchi University did pay a visit to me, all I said to him was: ‘Relax! You are safe’.

It is impossible to determine whether the claim quoted above is true. The curious thing is that since the Chengchi University case in 2008, the MOE refused to publish its annual ATU review results (in which the performance of recipient universities would be assigned one of three grades: outstanding, good or unsatisfactory). Instead, the MOE has only published the list of ATU recipients and the funding allocation results. Funding received by schools from 2005 to 2017 is presented in Table 2.

Table 2 Funding received by schools from 2005 to 2016 (NT$500 million) (30 NT = 1 US dollar)
Full size table

The change in funds received by the institutions indicates a trend towards an increased concentration of funding in fewer institutions. The first half of the list (institutions which received more funding) shows an average growth in the funding received in the time frame from 2008 to 2016, while the other half shows a declined. It seems possible that one or more institutions in the bottom half of the list could be excluded from the project if this concentration continues. Moreover, in 2011, the National Taiwan Normal University, another social science-based school, was included in the elite group. The president of the university has stood up against the MOE with the president of Chengchi University in 2008. The MOE (2011a) explained that this decision was based on its leading role in Chinese literary research and language education. The inclusion of the National Taiwan Normal University seems to support the president of Chengchi University’s claim that the committee has adopted ‘academic-field balanced thinking’.

Although funding allocations and the selection of universities for inclusion in the ATU are determined by the MoE ‘based on’ suggestions from the ATU review committee, the case of Chengchi University suggests that the MOE still holds ultimate power for the final decision. In other words, the claim that decision-making within the ATU is absolutely fair and objective is questionable. Debates surrounding the process also suggest that the ATU does not simply influence how the included and excluded institutions identify themselves. The ATU is also involved in the normalising judgment and examination process through which institutions are objectified so that they can be checked against the rules and standards set by the MOE. It thus contributes to the consolidation of the ATU discourse.

The resistance and ‘success’ of Chengchi University reflects Foucault’s argument that power comes from below and circulates. There is “no binary and all-encompassing opposition between rulers and ruled at the root of power relations” (Foucault 1972, p. 94). The discourse/power of the ATU must be acknowledged by the dominated recipients or its power disappears once it is not recognised by either the ruler (the MOE) or the ruled (funding recipients). This was a rather important case, for in the past the MOE was more like an absolute power, the boss who was rarely challenged in this way.

After the Chengchi University case, the MOE modified the review criteria for the second phase of the ATU. One important change has allowed the funding recipients to define their own criteria for evaluating whether their research areas/centres are world-class. This also enables them to add/define other relevant criteria for evaluating their overall outcomes and performance as top universities (as shown previously in Table 2).

If knowledge is a power over others, the power to define others and self (Geeiene 2002), then the funding recipients are empowered to co-construct the rules and discourse. As Rose (1996) observed, “to govern humans is not to crush their capacity to act but to acknowledge it and to utilize it for one’s own objectives” (p. 4). Through this facilitative mechanism, universities govern themselves and others towards what they see as the ‘truth’ (utilising the right indicators to define a world-class research area, field and university). In this sense, the ATU can be argued to be engaged in reflexive activities; technologies of the self permit individuals to “effect by their own means or with the help of others a certain number of operations on their own bodies and souls, thoughts, conduct, and ways of being, so as to transform themselves in order to attain a certain state of happiness, purity, wisdom, perfection or immortality (in this case, to be recognised as WCUs)” (Foucault 1988, p. 18).

The second phase of ATU (2011–2017)

The MOE’s imagination of the idea of a world-class university: a remake?

In response to the criticism from the Control Yuan, public media and academics from the universities, especially the case of Chengchi University, for the second phase of the ATU (from 2011 to 2017) the MOE changed the project objectives as follows:

Internationalising top universities and broadening students’ global perspectives, promoting universities’ research/innovation quality and international academic reputation/visibility, building academic and student capacity to provide excellent human resources for the country, strengthening collaboration between universities and industry to enhance national competitiveness and responding to societal needs and market demands. (MOE 2011a).

The way the MOE (2011b) imagines the WCU is clearly represented in the second-phase plan of the WCU strategy blueprint (Fig. 2). The university must first identify its top research centres/fields before establishing them as the institution’s developmental core. It is then provided with abundant resources from the private and public sectors for investing in its development. This includes elevating its research, teaching, internationalisation and university-enterprise cooperation in order to improve its overall academic quality and possibly become a WCU. Meanwhile, the top universities are expected to facilitate national higher education quality goals and enhance Taiwan’s international competitiveness.

Fig. 2
figure 2

Source: MOE (2011b)

WCU strategy blueprint.

Full size image

The main changes in the second phase were the deletion of the term ‘world university ranking’ and the specification of WCU characteristics or elements. In addition to research, functions emerged as priorities that would be subject to evaluation; namely, teaching, internationalisation and connection/contribution to the local enterprise. Moreover, it was stressed that the funding recipients should also play a role in elevating the overall quality of Taiwan’s higher education by means of sharing teaching and research equipment and resources with other non-recipient intuitions. These changes can be interpreted as not only as a compromise between the MOE and other stakeholders, but also a ‘localization’ of the concept of the WCU in Taiwan and an indication of the contingent and fluidity nature of power.

Embodying the WCU imagination through disciplinary techniques: the renewed performance indicators. In accordance with the change of the objectives of ATU, Table 3 is composed of the performance indicators provided by the MOE (2011b). The evaluation criteria can be divided into two groups: the ‘specified aspect group’, which has existed since the first phase of ATU and is fixed and applies to all recipients, and the ‘other aspect group’. The specified aspect group includes three measures: academic performance, internationalisation and national scientific and educational collaboration. The specific indicators in the ‘other aspect group’ are determined by the university through an institutional academic affairs committee and are usually negotiated with and approved by the MOE’s review committee. These quantitative instruments provide a base for developing performance measures, normalising judgment regimes and instituting a hierarchical observation structure through negotiations between departments, institutional managers and the MOE.

Table 3 Performance indicators
Full size table

In terms of the transformation of the power network from phase 1 to phase 2, the researcher summarised it as shown in Fig. 3. In contrast to the first phase, the performance indicators of the second phase denote a more flexible and more bottom-up thinking of what a ‘top university’ is and how it should be evaluated in the sense that the MOE didn’t set up all the indictors but left some space for the recipients to creating their imagination of a top university of their own. This change, based on the analysis of previous section, partly resulted from the mediation of some stakeholders in the ATU power network, namely the Control Yuan, the public media and some university managers as well as academics. In phase 1, the MOE set up ATU mostly to pursuit universities’ WCU ranking performance and thus the ATU seemed to be fully dominated by the WCU ranking indicators while in phase 2, the influence of WCU indicators was still obvious but its power was reduced as a result of the MOE modified the aim and performance indicators based on other stakeholders’ suggestions. On the other hand, the phase 2 of ATU might be as well interpreted as a result of compromises in the sense that the MOE included more elements into the idea of the top university without the increase of ATU funding.

Fig. 3
figure 3

A comparison of the power network of the two ATU phases

Full size image
Productive and docile subjects in the more ‘balanced’ power network?

It seems to be the case that after 5 years’ implementation and some modification made as a response to the publics’ criticism, the phase 2 of ATU worked more smoothly. Echoing the power network as shown in Fig. 2, James and Mary both consider the phase 2 of ATU is more like a ‘compromise’ among different stakeholders which makes the power between them more ‘balanced’ and the project more acceptable. According to James, institutions, ATU funding recipients are ‘now more used to identifying themselves with the ATU performance indicators which provide clear criteria for institutions to move on, to improve themselves.

Regardless of the changes made in terms of the performance indicators, the MOE reports the performance of ATU recipients to the public much the same way. The following table is a sample from an MOE press conference to announce review results for the second-phase submission (This was done every year since 2006). In the outcomes section of the press release, there were 11 tables listing the rank, number, per cent and growth rate of each funding recipient under different indicators. For the rest, the MOE explained how the ATU successfully promoted overall university qualities based on these tables. In this way, it is demonstrated that the ATU and funding recipients are ‘productive’ and worthy the highly concentrated funding (Table 4)

Table 4 A table from the MOE’s press release explaining the outcomes of ATU
Full size table

Conclusion

The ATU is a Taiwanese, state-initiated, higher education policy, backed with unprecedented funding and designed to foster world-class universities. It could be interpreted as yet another example of direct government control, leading funding recipients in a direction set or approved by the MOE. This echoes Sassen’s (2006) argument that globalisation is, in fact, constructed from within the nation-state. The state still plays a key role in opening up the nation to new forces, which in turn transform the nation (Robertson et al. 2012). In Taiwan’s case, the ATU has the potential to accelerate the globalisation of selected top universities as they pursue world-class status. While it is claimed that HEIs are increasingly autonomous, the government has the power to allocate funding and supervise and evaluate universities. To a great extent it is able to shape the discourse determining the roles and functions of higher education in Taiwan.

In his study of European countries, Erkkilä (2014, 2013) argues that global university rankings have led to ongoing patterns of convergence at some levels of global higher education that have come at the cost of diversity. This has resulted in the stratification of university systems and an increase in inequality between HEIs. In Taiwan’s case, the strong role of the MOE seems to have mediated the effects of WCU rankings on Taiwan’s higher education sector. Specifically, the MOE’s assumptions about the elements of a WCU have evolved during the last decade. In the first phase of ATU-implementation, the MOE’s imagining of the WCU was criticised for its dependence on WCU rankings. During the second phase, the emphasis shifted to maximising the contributions that top universities could make to the development of non-recipient institutions and Taiwanese society more generally. This change echoes U-Multirank’s emphasis on regional engagement and is reflective of higher education’s increasing connectedness to localised contexts, especially in societal and economic terms (Van Vught and Ziegele 2011). However, due to the funding scheme selected by the MOE and its effort to promote university classification, other non-recipient institutions have been prevented from joining the global competition game. This has kept universities focused on specific missions, such as functioning as research-intensive or teaching-oriented institutions. To some extent, the ATU can be described as a national intervention to selectively boost the global presence of some institutions and simultaneously block the potential exposure of other institutions.

The ATU has been described as an example of top-down state power. It has been employed to influence the funding of universities and shaped academic practice in the Taiwanese higher education sector (Chou 2014; Chou et al. 2013; Guo 2009). In a similar vein, the analysis indicates that, through the incentive of selective funding and varied disciplinary techniques, the MOE tried to discipline ATU recipients and bend them to embody the subjectivity of so-called ‘top universities’. However, some evidence presented in this study indicates that management of these funding recipients has been moving away, in Foucault’s term, from top-down disciplinary power in favour of networked governance. The case of Chengchi University reflects Foucault’s argument that power comes from below and circulates. With respect to the ATU, the MOE no longer acts as a sovereign power; its legitimacy has to be acknowledged by the dominated recipients and can also be challenged. Moreover, it is also evident that by introducing block funding and empowering funding recipients to decide their own development strategies and evaluation indicators (subject to approval by the MOE), the MOE has chosen to extend governance beyond prohibitory mechanisms to include facilitative mechanisms. This acknowledges the recipients’ capacities and thus empowers them to govern themselves. In sum, the MOE now governs ATU recipients with a combination of top-down and bottom-up mechanisms, granting them greater autonomy, even while more NPM principles and management mechanisms are imposed. This also echoes Foucault’s argument that power comes from below and circulates.

Higher education funding is a zero-sum game. This study shows that even if the ATU is necessary, the MOE fails to pay due attention to the ‘overall development’ of Taiwan’s higher education sector and the education rights of students in non-recipient institutions. The ATU grant itself indicates the positive reputation and research performance of an HEI, which in turn could lead to a crowding in effect that makes it easier for ATU recipients to attract private funding. This could create a vicious circle in which non-ATU intuitions and their students are more and more marginalised. Private institutions are at particular risk of isolation in the sector. Accordingly, a greater emphasis on local and regional perspectives in thinking about the role of top universities in Taiwan might be more relevant. The MOE must rethink the ATU in the context of higher education as a whole. It is necessary for the ministry to account for the relative advantages and consequences of the programme launch, so that the ATU can benefit from further improvements in the future.

Current study explores the world-class university discourse in Taiwan at the national level; however, ATU imposes hierarchies among not only institutions, but also subject fields and academics (Chang et al. 2009). Our analysis also indicates that even within the same project (ATU), different institutions have different perspectives and interpretation of the ATU and this might also be the case for departments within the same institution. Future research might consider investigating the impact of the ATU on management and academic practices at the institutional and departmental levels which might provide a complimentary view of the world-class university discourse.