Shared misunderstandings? Competing and conflicting meaning structures in quality assurance

ABSTRACT This article shows how the professional discourse on quality assurance in higher education is building on latent meaning structures that can be competing with each other and even subvert the messages on the manifest level. Taking the case of the Austrian higher education system as an example and employing a reconstructive-interpretative approach rooted in social science hermeneutics, five different meaning patterns are presented: a consumer protection pattern, an educative pattern, an entrepreneurial pattern, a managerial pattern and a quality engineering pattern. By analysing and comparing these patterns, the study argues for paying more attention to latencies and implicit meanings that might be overlooked by focusing on the manifest level of the discourse, in order to not lose track of important contradictions and sources of potential conflict.

Introduction and research aims: the meaning(s) of quality assurance In many ways, quality has the rather dubious honour of being one of the most intangible key concepts in higher education discourse. Despite the abundance of publications dedicated to quality-related questions (for an overview on recent topics see Alzafari, 2017), the issue remains elusive and difficult to define. In the past, quality in higher education has been framed as 'relative' (Harvey & Green, 1993), 'subjective' (Doherty, 2008) 'dynamic' and 'contextual' (Vettori & Lueger, 2008), 'contested' (Barnett 1992;Newton, 2002) or 'value-laden' (Kemenade et al., 2008); all characteristics that may be polemically summarised as follows: 'it depends'. Leaving it at that has its own dangers, however, as Laske et al. (2000) have found, the less quality notions are defined the more they run the risk of becoming a tool for safeguarding and enforcing (political) interests. Correspondingly, Newton (2002Newton ( , 2007 has shown, more than a decade ago, how academics resent quality assurance as a purely bureaucratic practice that impedes their activities, yet even more recent research shows that the idea of academics' ownership of quality assurance is still not working out (Cardoso et al., 2018).
The situation is neatly summarised by a 25-year-old quote from Harvey and Green (1993, p. 10): '(Quality) means different things to different people, indeed the same person may adopt different conceptualisations at different moments'. The implications have been taxing quality assurance professionals for decades. On the other hand, there is still only little research to be found on such conceptualisations, in particular if they are not understood as mere definitions but as complex forms of sense-making (Weick, 1995) or interpretive patterns (Vettori, 2012).
This article, therefore, aims at contributing to the scholarly debate on quality assurance in higher education and its effects, intended or unintended, by making visible how the discourse on quality assurance in higher education can be shaped and structured by competing interpretive patterns and how the resulting momentum causes actions and effects far beyond the actual discourse, providing an outlet for questions that tackle the very future of higher education and higher education institutions. Being firmly rooted in the tradition of social science hermeneutics and employing a reconstructive-interpretative approach, this study takes the respective analysis to a level of latency that has hardly found attention in previous research on the issue (Ramirez, 2013;Lueger & Vettori, 2014). By drawing attention to shared meanings instead of 'shared vocabularies' (Morley & Aynsley, 2007) the article shows how important contradictions and sources of potential conflicts are overlooked when the focus is set on the purely manifest level of the discourse, arguing that the underlying interpretive patterns have considerable impact on the way evaluations, accreditations and rankings are framed and used.
Contextually, the frame for this analysis is set by the Austrian higher education system. For this reason, the article will start with a quick introduction into how quality assurance was and is organised in Austria, followed by an explanation of the theoretical framework and the applied methodology. The final part will describe the five main interpretive patterns that were (re)constructed in the wake of this research and discuss their implications for practice.

Quality assurance in the Austrian context
Structurally, the higher education landscape in Austria is rather fractured: the four types of higher education institutions, the public universities, private universities, universities of applied sciences and university colleges of teacher education (Pädagogische Hochschulen), each have a different legal basis. The public universities are by far the biggest sector, as measured by number of students as well as government expenditures. They still dominate the public perception of higher education as well as the public discourse. Fachhochschulen, which are often regarded as 'hybrid institutions' (Pechar & Klepp, 2004) were introduced in 1993 and private universities only came into existence in 1999 on the basis of the University Accreditation Act 1999. Historically, the landscape of Austrian universities had long been characterised by a pronounced centralism (Konrad & Fiorioli, 2007). Until the 1990s, Austrian universities were basically state agencies. Autonomy was defined as a constitutional right of the individual academic, not of the university as an institution (Pechar & Klepp, 2004). Quality assurance was, though barely mentioned as a concept, largely achieved through centralised governmental control (at least if the concept is understood as compliance to common standards). Even though some basic elements of performance monitoring and reporting were already included in the Universities Act 1975, it was only in the University Organisation Act 1993 that evaluations had even become an integral part of the Austrian university system, though mostly in the form of student evaluations of teaching.
The emergence of the two new sectors mentioned above marked the beginning of a new approach to quality assurance in Austria. Now, for the first time, higher education institutions (and in some cases programmes) needed to be formally accredited, rendering the development of more rigid internal quality assurance processes a necessity.
In the public university sector, the respective changes were largely induced by Austria's signing of the Bologna Declaration in 1999 and the Universities Act 2002, which concluded a major reform process of the public university sector and came into full effect on 1 January 2004. The new act not only introduced a different governmental model but also required universities to develop a comprehensive institutional quality management (not quality assurance) system. Interestingly enough, neither the law nor the Ministry for Research and Higher Education defined what such a system should look like and formal external quality assurance was still confined to the other sectors.
This was changed in 2012 by the new Quality Assurance Act (QS-HRG), which formalised the quality assurance processes across the sectors by introducing mandatory external quality assurance cycles also for public universities and established the new Austrian Agency for Quality Assurance and Accreditation. Mitterauer (2013) argued that this marks the final shift from an internal assessment of institutional performance to an external examination. By 2018, most public universities are already preparing their second cycle of external quality assurance, marking a considerable shift of internal priorities towards compliance orientation.

Theoretical foundation
Deeply rooted in the sociology of knowledge-oriented perspective of social science hermeneutics, the theoretical focus of this article lies on those lifeworld structures that people's subjective interpretations of the social world draw from, 'i.e. the socially approved typifications available in a concrete historical socio-cultural Lebenswelt' (Meyer, 2008;p. 522;see also, 2006;Soeffner, 2003). This builds on the phenomenological theory of Schütz, that perceptions are already existent within coherent meaning structures (Schütz, 1972(Schütz, , 1982. From this perspective, present experiences can never be separated from previous ones. To a large degree, such experiences are gained and developed through interactions and communication with others and build on already existing socially shared stocks of knowledge and experience. From a social hermeneutics point of view, the basic aim lies in comprehending the construction of realities within specific socio-historical contexts and the ways in which they are constituted. Transferred to the field of higher education institutions, hermeneutic approaches can help to understand the dynamics of (higher) education by examining the contexts, life-worlds and meaning horizons in which educational arrangements are embedded in and upon which actors' perceptions of realities build (Lueger & Vettori, 2014).
The study makes use of a specific concept within social science hermeneutics; the concept of interpretive patterns or Deutungsmuster (Oevermann, 2001), which provides a theoretical model to explain social actions on the basis of shared reservoirs of meaning. Initially defined as an ensemble of socially communicable interpretations of the physical and social environment, interpretive patterns can be regarded as collectively shaped routines of sense-making for overcoming critical problems of action, yet not as mere 'interpretive options' that refer to single situations but rather as 'interpretive necessities' that are anchored in the practical key problems of the actors' life world (Kassner, 2003, p. 54).
From a sociology of knowledge perspective, interpretive patterns are not so much seen as compulsive structures but as providing directions for interpretations and actions, thus helping actors to make sense of their environment (similar to the works of Weick (1995), Weick (2000)) and adapt their actions to specific situations (Höffling et al., 2002). As collective reservoirs of knowledge and meaning they help to organise and structure our daily perception, interpretation and action processes. In order to become a legitimate orientation frame of their own, interpretive patterns must bear a functional relation to (objective) problems of action. Consequently, the question of how actors and action structures are associated with each other lies at the heart of the concept (Meuser & Sackmann, 1991). Similar to Schein's underlying assumptions (Schein, 2004), interpretive patterns are mostly latent and pre-reflexive (Meuser & Sackmann, 1991), hinting at a knowledge level that lies below or beyond the consciously available intentions, opinions and attitudes of single actors (Lüders & Meuser, 1997).
To become collectively shared, interpretive patterns need to be imparted and exchanged with others, either as part of a socialisation process or 'situatively' (Höffling et al., 2002, p. 2). Through this process, the patterns are reaffirmed but also changed, following the premises of symbolic interactionism (Blumer, 1969). As a result, the definition and framing of interpretive patterns is also a matter of space and time; and the normative power, which they can unfold, is limited to a specific social frame.
In this regard, the scope of an interpretive pattern can vary between different social levels, for example, societal level, field level, organisational level, group level. It is therefore necessary to identify and explain which level a certain pattern refers to or is limited to and in which regard it is viable to other patterns and levels (Kassner, 2003, p. 43). In the case of this study, a fieldlevel perspective was chosen, yet not in the original concept introduced by DiMaggio and Powell (1983) but rather following Scott's idea that fields are organised around shared cognitive or normative frameworks (Scott, 1995). Here, a field is not just understood as an ensemble of influential and closely connected organisations but as a '. . .community of organisations that partakes of a common meaning system and whose participants interact more frequently and fatefully with one another than with actors outside the field' (Scott, 1995, p. 56). This approach to the field concept shows new institutionalism's closeness to phenomenological positions (Meyer, 2008) and also redirects organisational field research towards 'understanding the processes that guided the behaviour of field members in unconscious ways' (Wooten & Hoffman, 2008, p. 132), providing an effective conceptual link to research in hermeneutics. Reconstructing the actors' and institutions' relations from the discourse they are taking part in instead as from their formal dealings, the field in this research is basically understood as an issue field (Hoffman, 1999) in which 'competing interests negotiate over issue interpretation' (Hoffman, 1999, p. 351). This concept provides a well-suited focal point for analysing the struggle over the meaning and purpose of quality assurance that is currently pervading Austrian higher education and corresponds well to the main data corpus of this research as described in the next section.

Data and method
As any related latent concepts, for example, action frames, or topoi (Höllerer et al., 2013;Jancsary, 2013), interpretive patterns cannot be directly observed but require a methodology that enables the researcher to carefully reconstruct the meaning structures from those texts and artefacts in which they manifest themselves, without succumbing to speculation or subsuming over-interpretations (Lueger & Vettori, 2014).
Corresponding to the principle that hermeneutic studies should focus on 'natural' data material that is structured by the field itself and not by the researchers (Hitzler & Honer, 1997;Froschauer & Lueger, 2009), this article's analysis is based on data that was generated within the research field as part of the documentation of the Austrian professional debate on quality assurance. The study's primary data corpus consists of the pro verbatim documentations of three conferences organised by the Austrian Quality Assurance Agency (AQA) between 2005 and 2010. Even though the data itself may thus appear a bit dated, crosschecks with later documents show that the underlying patterns are still highly relevant and a longitudinal analysis (which is hardly ever attempted in social science hermeneutics) is currently being prepared.
The programmes of the three conferences included various formats, speeches, presentations, panels and plenary debates, all of which were printed and, in the last case, literally transcribed in the conference proceedings. By bringing together the most important institutionalised key actor groups with ties to the issue field (representatives from the Ministry, the quality assurance agencies, the national student union, the bodies that represent the institutions from the public, private and Fachhochschulen sector as well as quality assurance professionals from most Austrian higher education institutions and international experts), these conferences not only provide an overview of the status quo of the Austrian discussion at that time but also gives a detailed insight into the field's internal dynamics. All in all, the proceedings contained about 60 different texts (or cases as they are referred to in this article).
This primary corpus was complemented by about three dozen position papers on quality assurance made public by the field's main institutional actors, including their publicly recorded official statements on a new act on external quality assurance in Austrian higher education that came into effect in 2012. This secondary data corpus was also analysed, yet mostly for the manifest content contained within, not for the underlying latent patterns as described below. This decision was not only taken for reasons of research economy but also because conceptual and theoretical saturation was well reached before the entire data corpus had been scrutinised. In addition, the entire logic of a hermeneutic analysis is not geared towards the redundant coding of every single piece of text (Knassmüller & Vettori, 2009;Lueger, 2010).
With regard to the analysis and interpretation of the data material, the basic interpretative model for this study follows a multi-step reconstruction process of the relevant meaning structures, deconstructing the original text from a level of everyday meaning, where the text's main issues and topics are identified and sorted, down to a level of latent yet inter-subjectively shared (objective) meaning, where the structurally most important interpretive patterns are reconstructed (see below; full description of the model in Lueger & Vettori, 2014).
Starting point is always a specific case, which is first analysed and then contrasted to other cases with regard to the typifiable patterns that transgress any individual case as an indication of the underlying and socially shared structures of meaning. In this study, every contribution to the conferences (a presentation, speech or debate) was treated as an individual case and subjected to the following analytical steps.
• The identification of manifest issues and topics (level of every day meaning) by means of paraphrasing the text's denotative (literal) meaning.
• The reconstruction of typical subjective interpretative patterns (level of subjective meaning) by taking the perspective of the communicator. • The reconstruction of latent structures and meanings (level of objective meaning) by looking for the structural conditions that have to be presumed in order for a statement or text fragment to make sense (usually indicated by the actor's choice of expression including the wording, grammar or the overall sequentiality). • The construction of hypothetical consequences and follow up options (level of pragmatic meaning) by asking about the hypothetical structural effects that would arise from the previous interpretations (for example logical consequences if the respective conditions were in effect; influence on actions, perceptions, communications and interpretations of the field actors).
These steps were cyclically repeated for every text sequence of every case, being periodically interrupted by phases of critically appraising the preliminary results, providing a kind of falsification function through defining criteria against which the plausibility of the preliminary thesis fragments are checked in subsequent interpretation cycles. This first interpretative cycle on the level of the individual case was complemented by three additional steps on the aggregated level: • Condensation phase: after the end of each cycle, the interpretative results were gradually summarised and compared with previous results. During these phases, the first tentative patterns were constructed and evaluated for their structural stability and coherence. • Final reconstructive phase: carried by the principles of constant comparative analysis (Glaser & Strauss, 1967), the structural elements identified in the previous steps either led to the (re)construction of a new type (abduction) or were subsumed under an already-existing type (qualitative induction). • Structural comparison phase: after the structurally dominant patterns had been reconstructed, they were broken down into their main components and systematically compared with regard to their main sponsors in the discourse, the actor-constellations and instruments they seemed to be mainly associated with, as well as the main problems of action and sensemaking the patterns had emerged to solve.

Findings and discussion
Quality assurance is of the utmost importance because everyone in Europe seems to think so and there are no good arguments against it. This short summary of the quality assurance related discourse in Austrian higher education may be ironically exaggerated, yet also demonstrates one of the discourse's most striking characteristics: as all field actors agree on the issue's importance, the question what the issue is actually about gets almost completely overlooked or is drowned in the technicalities of purely methodical discussions on how to improve a certain instrument or approach. The use of an institutionalised and standardised professional language and terminology further conceal that the actors in the field are talking about different, sometimes even opposite ideas and perceptions of reality. It is almost like the Tower of Babel effect in reverse: although almost every actor refers to the same sources and uses similar expressions, the latent meanings these impressions and references are imbued with, differ considerably. It is by such shared misunderstandings that an impression of harmony and mutual appreciation is created where most potential conflicts are framed as mere questions of methodological disagreement or unavoidable role-dependencies that can be eventually discarded, as the common purpose of improvement renders such differences unimportant. The fact that 'improvement', like quality itself, is very much a relative, context-bound and observer-dependent construct (Vettori & Lueger, 2008) that has different meanings for different people seems to get completely ignored. Terms such as 'quality', 'improvement', 'learning' and 'stakeholder satisfaction' are almost always positively connoted, adding to Morley's observation, that 'quality has become a universalizing metanarrative' (Morley, 2003, p. vii). As one consequence, arguments containing positively imbued references and terms and transporting the notion of improvement are rarely put to test or even contradicted.
The lack of conflicts on the manifest level of the discourse, however, should not be read as a general absence of competing and contradicting logics: taking a look at the interpretive patterns that emerged as structurally dominant frames of sense-making through the analysis, it becomes apparent that the field is not prevailed by one or two patterns, yet by five of them, which, though also partly complimentary, are strongly competing with each other: A consumer protection pattern that is organised around the notion that higher education institutions are basically service providers (or, as indicated by the language and use of words, even goods providers!) with a specific group of clients or stakeholders whose interests have to be safeguarded against the kind of 'bad quality' that can result from a not fully functional market. Quality assurance here is mainly intended to ensure that the consumers' investment is adequately rewarded. The demand for an appropriate regulative is attributed to the emergence of an international higher education market and the competition that results from it. The frequent market references in the entire discourse remain very opaque, however, indicating the concept's pre-reflexivity; either because it is already deeply ingrained in the field or because it functions as one of the signalling terms mentioned above that seem to cover any differences in meaning for the sake of an ongoing communication. The main quality assurance approaches and instruments that correspond to this pattern are minimum standards that are checked via certifications and accreditations, mostly in the form of national accreditations or evaluations conducted by the various European quality assurance agencies. The consumer protection pattern becomes highly visible on the manifest discourse level through the frequent use of product metaphors. During the open debates at the conferences this research was mainly focused on; higher education was compared to various consumer goods including pizzas or car tyres, with quality assurance being held responsible for measuring their 'tread depth'. Yet this is also the part, where the analogies of the consumer protection pattern start subverting their own logics: taking into consideration that processes of teaching, learning and research are not tangible and cannot be standardised in the same way as production processes, the corresponding quality assurance approaches have to focus on those aspects that are either easy to define and examine (for example the existence of policy documents, research output measures, learning outcome sections in syllabi) or revert to a meta-level where even the sole existence of certain instruments and processes is already regarded as a reliable indicator for an institution's quality awareness (such as the isomorphic demand for student satisfaction surveys, graduate surveys or staff development programmes). Yet the field appears also to be diffused with a certain awareness of the shortcomings of such proxies, resulting in continuous search for better quality indicators, on the one hand, and in an escalating spiral of attempting to establish trust, on the other. The benefit for the constructed consumer at the heart of the pattern, however, becomes rather invisible during the process.
An educative pattern that builds on the premise that universities, though in principle autonomous, have to be carefully developed in order to make them fit for their purpose. The label 'educative pattern' refers to the instructive logic that dominates the relationship of the relevant actors in the field. The assumptions that underlie the respective discourse have a rather patronising flair: from the perception of the pattern's main sponsors, higher education institutions are in need of some gentle 'guidance', because they are either unable to change in a constructive manner or are even resistant to change altogether. This situation calls for an overarching governing or regulatory body that, using a mixture of rules, regulations, incentives, sanctions and 'learning opportunities', such as pilot projects and cooperative initiatives, facilitates the institutional 'learning process'. Complementarily, the same body then has to check whether the learning goals were met. This is strongly reminiscent of Barnett's observation of the United Kingdom experience in the early 1990s, where 'Fitness for purpose' turns out to be a coded form of educational instrumentalism (Barnett, 1992, p. 87). On the practical level, the educative pattern favours approaches such as benchmarking and good practice exchanges but also performance contracts and external evaluations or audits as a means of assessing the state of development or the institutional performance. Last but not least, the pattern's subversive power clearly unfolds in the choice of words and expressions that are used to convey it: concepts of learning (from each other), of institutional development and (albeit directed) improvement resonate well with the educational mission and institutionalised vocabulary of most higher education institutions, making it difficult to escape the pattern's logic without reframing it as a pure struggle for power and thus lifting it to the level of political power struggles.
An entrepreneurial pattern that is increasingly emerging from the idea that higher education institutions are competing for students, reputation and funding in one big (international) market, forcing them to develop business strategies and mechanisms to gain an adequate share of the respective resources. Within this pattern, quality becomes an important competitive factor, either as a matter of cost-effectiveness and efficiency or as an image factor ('value for money') that could, in an interesting take on the 'return of investment' concept, lead to more resources or at least a higher degree of international recognition and reputation. Hence, the main difference to the consumer protection pattern is the way in which quality is framed as a business opportunity, either for individual higher education institutions or for quality assurance agencies and related consultants. Consequently, the differences between instruments of quality assurance, management and marketing become increasingly blurred: here accreditations are not regarded as a means of ensuring minimum standards but of signalling that an institution belongs to an elitist network. One of the most visible examples in which this logic manifests itself is the way in which business schools or faculties of economics and business administration within the field show their prestigious seals from institutions such as the European Foundation of Management Development (EQUIS accreditation) or the American Association of Collegiate Schools of Business (AACSB accreditation). The way, higher education institutions deal with university rankings shows similarly ambiguous traits (Wedlin, 2011).
A managerial pattern that equates quality with corporate values such as 'effectiveness', 'efficiency' and 'productiveness' and is in search for the best cost-benefit-ratio. The performance of a higher education institution can be improved by helping the organisation (or rather its management) to define the relevant performance goals more clearly and to achieve them by ways of strategy-formulation and performance measurement. This rather instrumentalist view is carried by the underlying assumption that the unregulated decentralism that has supposedly been a characteristic of most higher education institutions for centuries is limiting the institutional capacities for constructive change or is simply too expensive. There are even passages in the discourse where this assumption comes to the surface in a bluntly poetic way: 'It has not been too long, since universities were behaving like governmentally protected playing fields on which the flowers of knowledge were blooming in uncontrolled growth' (Case A-10, translation OV). Correspondingly, higher education institutions are framed as a type of organisation with certain specifics that would benefit if they were treated as any other kind of (commercial) organisation. The pattern is of particular structural dominance in the field, mirroring the diagnosis that higher education is in general taking a turn towards commercialism and managerialism (Parker, 2007(Parker, , 2011Ryan & Guthrie, 2009). Directing its 'meaning focus' decidedly on the functionality of individual organisations, the pattern makes a rather simple distinction between the only two roles it deems relevant: the 'managers' and the 'managed', with the former clearly acting as the pattern's main sponsors in the issue field. Partial resistance from the other parties is hardly surprising, considering the pattern's premise that the prospering of the organisation as such is dependent on the subordination of individual or group interests to the greater institutional good, that is, the financial or reputational prosperity of the organisation (or, as a substitute, the top management's strategic goals). This mirrors Cret's (2011, p. 428) findings how accreditations allow decision-makers in business schools to push reforms and to 'progressively create a decision monopoly in their favour'. Instrumentally, the pattern favours institutional management information systems that 'feed' the managers and decision-makers with information in accordance with Deming's (1982) 'Plan-Do-Check-Act' cycle but also formalistic rules and guidelines (in the form of policy or strategy documents) that provide clear definitions of the quality goals, the roles and responsibilities of the various actors, or the resources allotted to the goals. External stakeholders and external quality assurance procedures play a minor, mostly contextual role. It can be easily seen that within this logic, the boundaries between quality management and the university's management mode in general become blurred.
A quality engineering pattern that builds on similar premises as the managerial pattern, yet develops them in a more Tayloristic and mechanistic direction. This pattern is deeply infused with the ambition to create a 'better' organisation by re-engineering its internal processes and structures. Carried by strong beliefs in the rationality of organisational life and the causality of actions, the pattern seeks to establish and maintain order by means of an all-encompassing quality management system based on a classic scientific paradigm. The underlying image of organisations shares a lot of characteristics with the machine metaphor as described by Morgan (2006), for example, the routinisation of processes or instrumentalisation of people and ideas. Consequently, within this pattern, quality is no longer regarded as a goal, a dimension of social action or as a value per se, but as a phenomenon that has to be identified, operationalised into various dimensions, criteria and indicators, measured and, eventually, when it is sufficiently understood, increased and improved. In other words, quality can be found if the functional chains between the different stages of the 'production process' work according to a pre-defined plan (indirect quote taken from case C2). The focus lies on an organisation's inner processes, which need to be adjusted independent of the actors involved (which are rather reduced to the roles and functions that they fulfil). The depersonalisation principle of the quality engineering pattern is even taken so far as a fully developed quality management system is not only expected to monitor if the previously defined objectives have been achieved but also to generate new objectives based on environmental and internal analyses, eventually making decision-makers rather obsolete. Quality assurance and quality management are two functionally equivalent ways of achieving the desired purpose, yet require suitable methodologies and trained professionals. In the end, solving a certain quality problem, even in the realm of communication or social interactions in general, is just a matter of identifying and using the right approach or tool. Hence, the quality engineering pattern cannot be easily associated with a specific line of approaches, as in principle all instruments hold similar value if properly used. Nevertheless, the reliance on quantitative data and indicators is clearly visible.
Taking a closer look from a comparative point of view, the five patterns are not only different understandings of quality assurance or quality management but means of rearranging instruments, actors, resources in different, yet meaningful ways (Table 1). As can be seen from the comparison, the patterns, in their function as latent frames of orientation and legitimacy, are also playing an important part in empowering certain actors while re-defining or devaluing the roles of others. Yet even though the main sponsors and models of role allocation may differ, there are also important similarities: practically all five interpretive patterns strengthen the role of the senior management, whereas the role of an academic (be it as a teacher or researcher) is hardly ever touched upon. Leading to the concluding section, this observation shows the patterns' relevancy, not only with regard to quality assurance as an issue field but to shaping the very notions of what higher education is and should be about.

Conclusion
One of the key findings of this study is the observation that, even though the actors tied to the issue field of quality assurance in Austrian higher education are speaking the same professional language and are seemingly working towards the shared goal of 'improving higher education', the latent interpretive patterns at which the different actors and actor groups are oriented can differ substantially; and with far-reaching effects. The patterns themselves hardly ever appear as manifest themes in the discourse; rather they function as latent drivers of the discourse's internal dynamics and as demarcations of lines of conflict that are hardly visible but still effective. In other words, actors in the field may have rather different ideas of what they are talking about but these differences never come to light. On the downside, this kind of superficial consensus (for example on the 'importance of quality assurance' or the 'need for improvement') is not able to create a common goal. Hence, the real conflict, does not lie with the question of how to solve a certain problem (even though the question of who is authorised or expected to do something is an important part of it) but with the definition of the problems that need to be solved. Making an organisation more 'manageable' is a completely different mission than providing stakeholders with trustworthy information and both aims also require different approaches. In a way, even though quality assurance is arguably one of the most powerful and influential issues in the current higher education discourse (Westerheijden et al., 2007), it is almost completely unclear, what the issue is actually about, at least on the manifest level. Summing up, the results of the research suggest that it might be necessary to take the discussion on differing interpretive frames, that are clearly connected to the development of the field, to a more manifest level. The article is but one step towards a more differentiated discourse on quality assurance in higher education, arguing that the implicit purposes that are served by a particular approach need to be further investigated and discussed more explicitly. There is something truly unsettling to be found (and to be further researched) in the observation that even critics of current tendencies might be latently oriented at the same kind of logics they are opposing on the manifest level. By taking an in-depth look into the latent orientation frames that assist individual and collective sense-making, might help in our understanding of some of the factors and influences that take part in changing the social, political, economic and cultural contexts in which universities currently exist. In this regard, it can be argued, that, even though the patterns in the study hold a strong 'local flavour' when it comes to the details of the discourse they stem from, their structural core is not limited to the Austrian context but rather mirrors general key trends in European higher education, such as commercialisation, managerialism, consumerism or internationalisation. As was shown by examining the notions of higher education that are linked to different meanings of quality assurance, the relevancy of tackling the latent contradictions and incompatibilities in the field goes far beyond the question of the purpose of quality assurance but touches upon the future of higher education itself.

Disclosure statement
No potential conflict of interest was reported by the author.