Theorising pedagogical dimensions of higher education leadership—a non-affirmative approach

Today, multilevel analytics on educational governance, management, and leadership are common in educational leadership research, drawing on a variety of approaches and academic disciplines. This article develops a threefold critique of the state of the art. First, this article argues that research on higher education leadership often represents an unreflected position regarding the societal role of higher education—decontextualising leadership from external practices and disregarding its internal object. Second, the approaches applied in leadership as a multilevel phenomenon can be problematic. On one end of the spectrum, we find particularist approaches focusing on individual levels representing disparate and often incompatible theoretical perspectives. As these positions rarely communicate, they have difficulties producing a coherent representation of higher education leadership. Universalist approaches, in turn, study leadership on several levels but offer identical conceptual tools for any societal practice, thereby losing the sensibility of the societal, cultural, and economic tasks of higher-education institutions, and for their specific character as institutions for research and teaching. Thirdly, research on educational leadership mostly fail to provide adequate theory of pedagogical interaction and influence. This is a twofold challenge. On the one hand, research lacks a theory of the object of educational leadership, namely teaching and studying. On the other, although research often defines leadership as a process of influence aimed at supporting learning, it lacks a theory able to explain what constitutes this influence. This article elaborates the possibilities for non-affirmative theory of education to provide a theory and language to overcome these challenges.


Introduction
As a growing field, higher-education leadership (HEL) research applies an increasing variation of perspectives and theoretical positions. Approaches range from individualistic heroic conceptions, viewing leadership competence as individual characteristics, to positions emphasising the social, structural, or processual nature of leadership, in addition to transformational, transformative, and transactional approaches (Crevani, 2018;Shields, 2014;Wang, 2018). One might assume that this variety of perspectives demonstrate vitality in the field. However, studies on leadership and educational leadership (EL) have not been ideal regarding theoretical foundations (Alvesson, 2019;Niesche & Gowlett, 2019;Wang, 2018). The pace of theoretical and conceptual development appears more modest than the increase in the volume of empirical research (Alvesson, 2019;Wang, 2018). Niesche and Gowlett (2019), however, observe a sort of theory turn in recent critical EL research. In this article, we contribute to this increased focus on theory by addressing three critiques pointing at some of the conceptual challenges in contemporary research on EL and HEL. With HE we are referring mainly to universities. In the following, we shortly outline the three critiques.
First, EL research demonstrates an unclear position in conceptually and theoretically relating HEL practices to other societal fields of practice such as economy or politics. The lack of an articulated position risks resulting in isolated or partial interpretations of the relations between HE and other societal fields and interests framing HE institutions. Second, many approaches to EL as a multilevel phenomenon draw on universal terminology or generic theories, thereby losing a necessary conceptual sensitivity for leadership of educational institutions. Universal approaches conceptually treat multilevel leadership the same, regardless of which institution is in question. On other occasions, when the object of research is specifically education institutions, particularistic research approaches tend to focus on separate levels of leadership in isolation, losing the 'big picture' of HEL. Third, research on leadership and EL concludes that one aspect of leadership is to create direction, create conditions for change, and intervene in or influence others' learning. Thus, research has widely recognised that leadership includes a pedagogical element. While understanding pedagogical change processes theoretically is crucial, the field is severely undertheorized. Despite various initiatives (e.g. Kasworm & Bowles, 2012), HEL research lacks a language both for the pedagogical dimensions of leadership, as well as for the ultimate object of HEL, namely research, teaching, and supervision.
This article argues that a way to overcome the limitations sketched above is to recognise that (i) HEL requires an idea of how education relates to other societal practices, (ii) EL and pedagogical leadership (PL) are phenomena occurring at different leadership levels simultaneously, and (iii) HEL theory requires an idea of the pedagogical process because pedagogical processes constitute its object, and because HEL itself features a pedagogical dimension. Theory acknowledging these dimensions could better explain the pedagogical dimensions of leadership at and between different levels, while understanding the object of HEL: HE teaching, studying, and learning. Based on these assumptions, our aim is to take educational theory as a starting point for approaching HEL by studying whether nonaffirmative theory of education and Bildung (NAT) (Benner, 2015) may provide a theoretical language for elucidating the pedagogical character of relational leadership interaction in HE, at and between all levels of governance and leadership. NAT draws on the relational and processual theory of Bildung, positioned in a Western tradition identified as the Humboldtian model of HE.

Difficulties related to the societal role of HE
HEL always occurs in relation to, and in complicated dialogues with, different stakeholders. In a historical and contemporary perspective, most research agrees that it is not possible to understand HEL in a decontextualized fashion (McLendon, 2003). The question is not if, but how, conceptual positions on HEL, policy, and teaching understand the relation between HE and other societal practices.
Given the complexity of the field, we illustrate our critique by identifying two broad positions on how research relates education and EL to societal fields and interests. First, traditional leadership research has been grounded in structural functionalism, partly through organisational systems theories (see Uljens & Ylimaki, 2017, 48-54). These ideologically naïve positions see education functionally, often subordinating education and EL to contemporary economic or political interests. Second, critical-transformative research often superordinates education and EL to contemporary society (e.g. Giroux, 1980;McLaren, 2014;Shields, 2014). We view these positions as end positions on a spectrum, as much of contemporary research falls somewhere in between (e.g. Mezirow, 1991).
EL research in the first position either remains unarticulated to how HE and HEL relates to other societal fields or focuses on uncritically reaching some predefined present-day need. Examples of the many positions could be distributed leadership (e.g. Leithwood & Riehl, 2003), instructional leadership (e.g. Apkarian & Rasmussen, 2021), or team leadership (e.g. Koeslag-Kreunen et al., 2021). If HEL is treated as a closed system, not addressing the critical question of the role of HE in society, leadership, and leadership research can principally be put into the service of virtually any ideology or agenda. Positions focused on meeting present-day conservative needs could be focused on topics such as student employability, labour market expectations (e.g. Varga, 2006), or HE rankings. These positions risk instrumentalizing HE and HEL to serve interests external to the core tasks of HE that are independent academic research and education of students towards self-determined, ethically reflecting, and politically aware subjects.
In contrast to the ideologically naïve functionalist positions, critical positions approach HE and HEL as normatively driven societal transforming forces. In traditional critical education research, it is typical to view education as having the potential to 'shape those who will go on to become future educators, lawmakers, and politicians' (Tolman, 2019). In their radical form, transformative HE and HEL not only defend education's emancipatory task as aiming at negative liberty (Berlin, 1969), i.e. liberating students from represented prejudices, but this leadership also positions itself as superordinate to existing society by aiming at leading students to predefined future ideals representing own interests. Examples of EL research positioned in this category include specifically targeting normatively closed understandings of sustainability, gender perspectives, or equality and social justice, a topic on the rise in EL research (McArthur, 2010;Wang et al., 2017), rather than identifying various positions to these issues as objects for elaborations.
What unites both leadership research positions above is that they risk subordinating HE and HEL either to existing societal interests or to normative ideals. In Kantian terminology, normatively closed external interests do not treat universities, research, or university students as representing ends in themselves but rather see education as a strategic instrument for reaching something decided upon in advance. Broadly speaking, economic, political, cultural, and religious interests and ideologies all represent societal practices regulating HE 1 3 research and teaching, which can operate both as conservative and transformative forces. We see it as crucial that a theory of HEL represents a reflected position in this matter, that seeks to avoid instrumentalism. The non-affirmative position elaborated on later, positions itself beyond those described above.

Difficulties with approaching multilevel HEL
HE has in many countries witnessed a shift in leadership policies and practices from a collegial and bureaucratic model towards a neo-liberally inspired model characterised by accountability, managerialism, and leaderism (Bolden et al., 2014;Crevani et al., 2015;Croucher & Lacy, 2022;O´Reilly & Reed, 2010;Välimaa et al., 2016). The shift is a move from government in Old Public Administration to governance in New Public Management. Parallel to, and partly because of this, situational, contextual, and multilevel approaches to studying educational governance and leadership have strengthened, drawing on a variety of academic disciplines (Alvesson, 2019;Wang, 2018;Wang et al., 2017). We agree that understanding leadership, governance, and management of HE requires a multilevel perspective (e.g. Bolden et al., 2008;Frost et al., 2016;Uljens & Elo, 2020). However, we find it troublesome how this multilevel character of HEL is conceptualised, which we explain below.
Today, there are several multilevel approaches to EL, such as actor-network theory (Czarniawska & Sevón, 2005), discursive institutionalism (Schmidt, 2008), and refraction (Goodson & Rudd, 2012). Despite a variation in terminology, leadership comes across as a mediating activity between different levels and actors. Mediation in terms of negotiations, re-contextualising, and discourses always includes an element of interpretation, pointing towards a hermeneutic dimension of translation. Despite their obvious strengths, we argue that the weakness of many previous multilevel approaches to EL is that they offer universal vocabularies that are neutral towards the practice/praxis in question, which, for HE, obviously, is education and the creation of new knowledge through the research that HE is based on. The dilemma with universal approaches is their insensitivity towards what is lead and where leadership occurs. Universal approaches offer identical conceptualisations for understanding the policy and practice of any societal field, such as healthcare, education, and private businesses. The strength of these frameworks-their intended universal validity-is simultaneously their weakness, as they lack a specific vocabulary for grasping the studied object. We argue that understanding the leadership of HE at different levels requires insight both in research-based teaching and studying (the object) and in the societal role of education (the context). To theorise leadership as separate from its object and its context fails to capture the unique character of leadership of education. However, we are not suggesting disparaging, for example, the rich tradition of organisational or policy implementation research. These and other fields of leadership expertise provide important additional perspectives, but their limitation operating as foundational points of departure needs to be acknowledged.
Viewing context-and content-neutral universalist approaches to multilevel EL as one end of a continuum, the other end is represented by particularist positions, focusing on specific levels and aspects of EL in isolation. Thus, research on policy and policy borrowing has generally focused on the macro level (e.g. Capano & Pritoni, 2020;Rhoades & Sporn, 2002). Research on academic leadership generally focuses on the meso-or organisational level (e.g. van Ameijde et al., 2009;Floyd & Preston, 2018), while curriculum, instructional theory, teaching theory, and research on learning and instruction (Didaktik) typically focus the micro level (e.g. Bovill & Woolmer, 2019;Stes et al., 2010). Particularistic approaches have the advantage of providing vocabularies that capture nuances and details of the studied objects. However, the lack of a uniting theoretical frame hinders research-informed communication and dialogue between research on different levels, leaving findings from different fields and levels unconnected, resulting in 'siloed' research (Wang et al., 2017). Thus, the 'big picture' of HEL is lost. We propose NAT as a position with the capacity to conceptualise the multilevel character of HEL without losing the how, what, and where of leadership-pedagogical dimensions, the content and context.

Difficulties related to understanding pedagogical processes
In the current HEL research, pedagogical interaction is under-theorised. This is a challenge in two respects. Firstly, this means that the pedagogical qualities of leadership interaction that occur within and between different levels are under-theorised. Secondly, it means that HEL research often lacks a developed idea of the core object of HEL, namely researchbased teaching and studying.

Pedagogical qualities of EL
EL at any level covers a wide variety of tasks and assignments related to, for example, economy, organisation, jurisprudence/law, technology, facilities, and communication. One of these areas involves creating favourable conditions, either directly or indirectly, for professional learning, growth, and the development of all staff (academic and professional), as well as initiating and participating in change processes concerning the organisational culture. In these respects, leadership in any institution includes a pedagogical dimension: to intentionally support the learning of others by direct or indirect measures. Even though supporting professional learning is a decidedly central dimension of leading in any expert organisation, leadership literature on this topic is insufficient. Statements such as leadership being about 'influencing' (Alvesson, 2019) or 'influencing learning' are frequent, but what these statements mean theoretically or conceptually is surprisingly vague. In international literature, there is a disturbing gap between the central role of pedagogical activities and the core notions of EL theories and models explaining how we can understand pedagogical influence (Alvesson, 2019;Niesche & Gowlett, 2019;Wang, 2018).

The object led-study programmes, teaching, and research
As previously observed, theories in HEL seldom pay attention to curricular issues (aims, contents, and methods of study programmes) or to teachers' educational professionalism. In HEL research, instructional leadership (Shaked, 2020) and curriculum leadership (Stark et al., 2002) are rare, despite being very common in school leadership research (Hallinger, 2005). Curriculum leadership is sometimes defined as 'a facilitating process in which the leader works with others to find common purpose' (Wiles, 2009, p. 21). Curriculum leadership also covers the university's freedom to formulate the aims, contents, and structure of research and teaching, connecting directly to the autonomy of research and teaching. Freedom to research, to teach and to learn are inseparable dimensions of academic freedom with a long history (Robertson, 1969). Given that the study programmes are typically in the hands of the university, leading collaborative work developing these programmes is a main issue at any leadership level. Developing study programmes in HE is naturally related to and intertwined with the creation of new knowledge through research. Creating conditions for research (to study something) has a pedagogical element at its core, as it aims at supporting a form of academic professional development characteristic for HE. Leading and developing the university as a haven for academic learning includes focusing on teachers' professional development as teachers as well. If empirical studies on instructional leaders' practices indicate a lack of knowledge of teaching and learning, this also appears as a lack of pedagogical interaction theory.

Bildung centred non-affirmative education theory as an analytical framework
To overcome the indicated limitations productively, we will draw on non-affirmative theory of education and Bildung (NAT) (Benner, 2015, forthcoming;Uljens & Ylimaki, 2017;Uljens, 2015). NAT includes an interpretation of the modern tradition of Bildung as developed by Rousseau, Fichte, and Herbart (e.g. Benner, 2015, forthcoming;English, 2013;Horlacher, 2004;Uljens, 2015), but it must be related to the concept of education to be pedagogically meaningful (Siljander et al., 2012). Figure 1 lays out the fundamental regulative and constitutive principles of NAT. The two regulative principles to the right focus on how education relates to society, while the two constitutive principles to the left focus on pedagogical interaction.

Regulative and constitutive principles
The first regulative principle in the bottom right-hand corner concerns how we define the relation between education and other societal practices including politics, culture, religion, and economics. This principle relates to our first critique and argues that modern societies are characterised by a non-hierarchical relation between societal practices; they all influence each other while simultaneously being influenced by each other. For example, while politics decide on laws, laws regulate political practice itself. Education operates under the influence of many societal practices while simultaneously preparing the individual for participation in and transformation of all of them. A political system influences how HE is organised, yet it is dependent on how education prepares new generations for the system in question. Benner (forthcoming) reminds us that although the principle of the nonhierarchical relation of societal practices is non-pedagogical, it is necessary as a point of departure for a general theory of education. It regulates educational activities, as we cannot meaningfully outline any human education without having an idea of the world and the future. This Bildung theoretical position accepts a non-teleological view of history and the future, meaning that the future is radically open and depends on what we and future generations make of it. The second regulative principle (top right-hand corner) relates to our second critique and asks how policy, financing, administration, and other forms of governance and leadership, occurring on levels ranging from supranational down to teacher levels, contribute to transforming societal interests to pedagogical work. Given the many levels of decisionmaking in the education system, this principle asks to what extent autonomous action to determine the meaning and value of aims and contents of educational influences exists on and between levels of EL.
Both constitutive principles to the left of Fig. 1 relate to our third critique on the absence of a theory of pedagogical interaction in EL models. Pedagogical interaction is dependent on, what in the German language is called the learner's Bildsamkeit, the first constitutive principle (bottom left-hand corner). In the literature, the notions of Bildsamkeit (originally developed by J. F. Herbart) and Bildung are interpreted very differently (Benner & Brüggen, 2004;Lenzen, 1997). In this context, Bildsamkeit refers not to the human ability to learn, to human plasticity, but to the subject's self-active, spontaneous, and never-ending dynamic relation to the world, in which we relate to and may transcend our current way of understanding and being in the world (Benner, forthcoming). If the first regulative principle describes an assumption regarding the interdependent dynamics on the societal level operating in a non-hierarchical way, Bildsamkeit describes a similar relational assumption regarding dynamics on the individual level.
The second constitutive principle (top left-hand corner) defines a pedagogic intervention as a summons of self-activity. Pedagogical intervention is understood as an invitation or provocation to an already self-active other, to direct her attention in a certain direction and engage in self-transcending activity that likely will result in intended changes through a process of learning. A pedagogical intervention is thus an interruption in the relation between the other and the world. Due to constitutive subjective freedom, the summoner does not possess coercive power over the Other's way of perceiving themselves and the world, and is unable to directly transfer ideas, knowledge, values, and competencies, to the other. PL, understood as a pedagogic summons, entails directing an Other's self-activity to transcend their current state through a process of self-directed transformation. PL is not tied to any formal leadership positions, as all actors are potential objects as well as initiators of pedagogical summoning.
NAT has made use of the concept of recognition, originally developed by Fichte and Hegel (Williams, 1992), and more recently by Charles Taylor (1989) and Axel Honneth (1995). In the present context, the notion does not itself refer to pedagogical activity. Instead, it means, first, that each recognises the other as an anthropologically indetermined subject (free, actively relating to the world) and second, paying due respect to the experienced life realities, possibilities, limitations, and potentials of the other. By recognition, summoning takes the other's life reality and orientation as its starting point. Recognition is a dual relation and includes the other recognising the summons directed at them.
Learning, emanating from study activity, which in turn is influenced by a PL summons, is not a direct linear consequence of the intentions of the summoner. Rather, learning may occur as a consequence of the learners' own actions, resulting from the interplay between the summons, the context, and the interpretations of the summoned. In this way, summoning and Bildsamkeit are relational concepts-summoning assumes Bildsamkeit, which always points to experiencing influences. In this respect, how the subject develops depends on, but is not determined by, pedagogical intervention. As actors in HEL stand in an open relation to the world, they are in constant transformation and do not identify as stable entities. A pedagogical intervention, defined as the summoning of self-activity through PL or teaching, can thus be understood as an act of directing the others' self-activity in a specific direction with the ambition of inducing activities that may result in learning. In a teaching context, Uljens (1997) describes this as the teaching-studying-learning process, denoting that learning is not something we do; it is something that may happen to us resulting from the activity of studying. Teaching, in turn, does not lead to learning, but facilitates studying as the teacher and learner intersubjectively construct a situational shared experience within which the teacher may direct the learner's attention to new forms of self-activity (Uljens & Kullenberg, 2021).

Answering the critiques: non-affirmative HEL
In this section, we answer the three critiques presented at the outset.

Critique 1-the dynamic relation between HE and other societal practices
The fundamental freedom in research, teaching, and studying characterising universities following a Western tradition, identified as the Humboldtian model of HE, states that new knowledge and its teaching must not be determined by economic, political, or religious powers and authority. Such a view relates education constitutively to the societal role of an institution. However, soon after the ideal of independent university research and teaching was, once again, confirmed by the Bologna Declaration of 1988-the 'Magna Charta of European universities'-the division between East and West Europe ended by the collapse of the Soviet Union and fall of the Berlin wall. The 'market state' became the dominant model, resulting in obvious steps to bringing together liberal and vocational forms of higher education (Anderson, 2006). The rhetoric around the emerging knowledge-based society, requiring research and teaching to be economically relevant, has influenced governance, leadership, and financing of HE. Thus, beyond any doubt, HEL theory needs to provide a relevant point of departure for dealing with these issues.
To clarify how non-affirmative theory deals with education's relation to other societal practices, an analytical-descriptive lens is required. From this perspective, NAT asks to what extent education is sub-ordinate and superordinate to other societal practices. Universities can seldom, if ever, act independently of the rest of society, standing above other interests in absolute autonomy. It is, however, equally rare that universities are completely sub-ordinate to or determined by political, economic, or religious interests. This means that different societal practices, of which HE is one, must recognise the interests of each other. Typically, educational institutions and educational actors have relative autonomy, as they are not operating totally without either outside influence or boundaries, nor are they in total subordination to these influences (Uljens & Ylimaki, 2017).
The contribution of non-affirmative theory of education is a non-linear and non-hierarchical view, offering a theoretical construct for empirical analysis to what extent instances owning a superior role in relation to other practices recognise the relative autonomy of these action levels. If external interests strictly govern universities, or if there is a top-down hierarchy within them, leadership forces actors to affirm external or internal interests. If universities not only recognise but also affirm such interests, education institutions subordinate themselves in relation to these.
While the above argumentation applies for empirical studies regarding how and to what extent various interests and actors require affirmative action from others, non-affirmative education theory itself takes a critical stand in this question. It argues in favour of recognising the interests external or superior to each operational level, reminding of the importance of maintaining a space and capacity for autonomous action, recommending actors not to affirm these interests. The motive for supporting non-affirmative EL and teaching relate to the societal task of the universities described above. Affirmative pedagogy and leadership run the risk of not reaching the aims of the universities of promoting the development of self-determining, reflective, and critical citizens able to contribute to existing practices and develop new ones.
To conclude, the non-affirmative position offers an analytical construct for how to relate HE to other societal practices beyond functionalist and transformation-oriented positions, providing an analytical gaze, emphasising that HE institutions are indeed expected to recognise legitimate interests and aspirations of different stakeholders, yet maintaining that these aspirations cannot in any case be expected to be affirmed without serious consideration. At different levels, recognition without affirmation creates a space for collaborative, joint reflection and repositioning activities of individuals and organisations. Nonaffirmative analytics asks to what extent EL manages to take into account interests such as those of the labour market, science, and politics, but avoiding instrumentalization, which would violate HE's relative autonomy in education and research. Autonomy is crucial for educational institutions, as education in liberal economies and political democracies has an emancipatory task aiming at developing students' professional, personal, and societal self-determination. This entails supporting students to develop their ability of analytical and critical reflection as individuals, problematizing existing theoretical answers to various dilemmas. In this respect, existing knowledge offers itself as a necessary medium through which the learners' reflexive ability is developed.

Critique 2-approaching the multilevel character non-affirmatively
Our second critique is that many current approaches either appear to run the risk of providing universal languages covering several levels of any societal practice, thereby making the nature of education invisible, or generating a particular vocabulary limited to specific levels or aspects in isolation, thereby missing the big picture. We requested a third position, providing a vocabulary that is (a) relevant for the leadership of educational institutions and (b) addressing the pedagogic character of leadership and governance initiatives across the different levels of HEL. We illustrate this dialogue in Fig. 2.
The formal HEL hierarchy, as illustrated in Fig. 2, operates in the symbiotic tension between the scientific community and various external stakeholders, but at the same time, universities contribute to the self-construction of these stakeholders. HEL is a hierarchy and a rhizomatic network simultaneously (Välimaa et. al., 2016), with individual actors located at specific levels, interacting with actors at other levels in dynamic and changing networks. Any actor can participate in a leadership process that emerges in these networks. The circular arrows between actors on different levels represent the reciprocal influences between these actors. NAT brings a system perspective to the forefront, arguing that understanding the pedagogic dimensions of HEL cannot be limited to focusing on individuals in isolation, nor the activities of any particular group in an entitizing sense. Grasping HEL requires seeing it as a part of a larger dynamic process of creating direction collaboratively, spanning several leadership levels, including a multitude of actors. This means that when a single study focuses on PL between two individuals or within a group, for example, this interaction cannot be meaningfully understood in isolation from the pre-existing larger context, while this larger context simultaneously is discursively co-constructed by micro-processes (Crevani et al., 2010).
EL covers a wide range of tasks including economy, law, organisation, distribution of work, or communication. One of these tasks is to create conditions for the learning of others, both directly and indirectly, and educational theory helps us conceptualise this pedagogical dimension. From a PL perspective, a formal leader or actor, or more broadly, a  Uljens & Elo, 2020) level of leadership, must recognise summons from many actors, levels, and directions. These influences and initiatives may be contradictory or point in different directions and be driven by different interests. Affirming all of them is not an option. Instead, the actor must make a judgement of an appropriate course of action given the cultural and historical context at hand. Different levels of EL are thus not entirely sub-nor superordinate to each other but maintain a certain space and capacity for autonomous action, as they reciprocally influence each other in complex, rhizomatic webs of summons.
As leadership in general often includes mediation, this certainly holds true for the pedagogical dimensions of leadership. When others are summoned to self-activity, the actor must recognise the life-realities, values, and aims of the other while maintaining a capacity to challenge the Other to transcend the current state by not affirming them. Affirming the other 'all out' would entail uncritically accepting every aspect of the other's understanding of the world and render pedagogic influence impossible. PL is a process of interpretative mediation involving the recognition of external influences without affirming them, as well as the recognition but not affirmation of the other. In PL, actors control certain degrees of freedom to deliberately engage others when co-constructing a mediational space whereby Others are invited to self-transcending activity.
As an analytical concept, affirmation is not a binary question of yes or no, but rather a continuum of different degrees and forms of affirmation. Different actors, institutions, or nations in different contexts presumably have different prerequisites and capabilities for, as well as practices of, recognising and responding to summons in more or less non-affirmative ways. As an analytical concept, non-affirmative summoning provides a tool to analyse in what ways and to what extent pedagogic actors, leaders, or institutions affirm either horizontal or vertical interests in their collaboratively mediating leadership activity in a multilevel-networked system. It also provides a tool for analysing to what extent pedagogical summons are affirmative in character, i.e. to what extent they require affirmative response.

Critique 3-a language of non-affirmative education for the what and how of EL
Our third critique states that although research on leadership, EL, and PL often claim that leadership is about influencing people or influencing learning, this pedagogical influence is undertheorized. Typically, EL research does not elaborate further on how we could conceptualise this influence or the pedagogical relation. Previously, we pointed at the two constitutive principles of NAT, providing us with a language to address PL influence regardless of context.
First, this position draws on philosophical anthropology as developed within the German and Nordic tradition of Bildung. 1 Philosophical anthropology elaborates on assumptions of what a human being is. Bildung, a complex concept as such, was originally developed by emphasising that the human being is not determined by anything innate or by external conditions, thus moving from a premodern, teleological view of the subject and the world. One of the core concepts in this tradition, Bildsamkeit, points out that being in the world features a subject-world relation referring not to the external object as such, or the isolated experiential dimension within the subject, but to the world as actively experienced, constituted by meaning making activity. The process operates in both directions: while the subject embraces culture in this process, coming to share it with others, the individual simultaneously develops her unique identity or personality in this process.
Thus, personalisation (to develop a unique identity) and socialisation (to come to share the culture with others) are interrelated phenomena. When the individual makes culture her own, she makes it her own, interpreting and relating to it in unique ways. While a Bildung tradition analyses human growth and learning in terms of the experiential cultural contents rather than through psychological processes such as cognitivist learning theory, it still accepts that education may develop general capabilities such as critical thinking or ethical responsibility.
Bildung's idea of education goes back to J. G. Fichte's (1796Fichte's ( /1992) critique on Kantian transcendental idealism. Kant had argued that a human being's awareness of herself as free and indetermined is made possible by her a priori awareness of the categorical imperative in the form of moral principles, as well as innate structuring categories such as time and space. In contrast, Fichte argued that while a human being is indeed born indetermined, her awareness of herself as culturally free results from being treated as such by the empirical other. Fichte's idea was that our awareness of ourselves as culturally free depends on that we are recognized and treated as such by others. In other words, we come to realise ourselves as free by being summoned to self-active engagement with the world. In this light, the modern paradox of education is to treat the Other as if the other already would be somebody or something that she may become by her own activity, in the relational pedagogical process.
This relational tradition of thought, which emphasises the importance of the empirical other, began with Fichte but was further developed by Hegel, Vygotsky, Mead, Dewey, Habermas, Honneth, and Taylor and manifests itself today in non-entitative, processual approaches to leadership research and communication. In this interpretation, being and becoming human builds upon a relational social philosophy that is processual in nature.
Being is thus constantly about becoming, where the individual continuously establishes and reforms her relation to others, the world, and herself. There are no immanent sources determining the direction or end of human activity. Rather, 'determining direction', often considered central in leadership theory, is an inherent dimension of the process of Bildung; namely, to live with the question of which direction to choose as an open one. Assuming that the future is not predefined, but something that follows from indetermined human action, the question of direction requires permanent engagement and position taking.
Pedagogical activity builds upon the recognition of the Other, within a cultural space that is already shared. Within this space, an already self-active Other is summoned to redirect her attention by engaging in activities that may lead to the Other transcending her current way of understanding some aspect of the world, herself, or her relation to the world. In other words, in this shared process, PL initiates conditions for learning. NAT emphasises that pedagogical influence is not linear; it is not the summons that leads to learning, but rather the activities of the summoned that may result in learning and a change of perspective.
NAT accepts that no leadership theory is devoid of values. The normative dimension of the theory manifests itself in educational ideals to support the development of a mature, critically reflecting, self-determinate citizen and subject in an autonomous nation-state able to actively collaborate and contribute to a non-determined development of society in a globalized world. The distinction between negative and positive liberty (Berlin, 1969) clarifies this educational ideal. While negative liberty refers to freedom from external restraints or limitations, positive liberty refers to the capacity or possibilities to self-determination and practicing one's intentions in relation to other's interests. Formally recognising civil rights (negative liberty) does not guarantee that an individual has the true capacity to execute these rights productively. For this to occur, 1 3 positive liberty is necessary-the citizen must be recognized as having the right to be offered the cultural tools for acting in one's own interests as related to others' interests. Education is central for reaching this capacity for self-determination in practice.
The processual character of PL is visible in three ways. First, from a Bildung theoretical perspective, each subject is in a lifelong open processual relation to the world. Being constantly 'in the making', PL intervenes in this ongoing process, providing necessary input without determining the outcomes. Second, the PL relation itself is an open processual relation being both symmetrical and asymmetrical with respect to roles and experiences. The result of a pedagogic summons is not knowable beforehand, as it evolves in the interplay between the summoner, the summoned, and the context. Thirdly, the processual character of PL is visible in the processual nature of the development of organisational culture. Directing others' self-activity in a way that results in learning, regardless of whether this occurs on an interpersonal or organisational level, is an act of 'shaping movement and courses of action', which is at the core of leadership work (Crevani, 2018, p.89). Non-affirmative PL, understood as a multilevel socially shared phenomenon, is therefore in line with a process ontology of leadership.
Approaching PL in HE in this manner portrays it as a processually evolving multilevel and multi-actor phenomenon that is not strictly tied to traditional positions of leader and follower or academic and professional staff. Bildung-based PL reminds us of what Crevani (2018) describes as a processual production of direction in various forms of relations and interactions evolving over time, as well as over organisational space. The direction of the development, shaped in the interplay of summons between actors, is not predetermined.
Our argument is that being able to conceptualise pedagogic influence has three advantages. First, it enables us to conceptualise one of the core activities in HE, namely teaching, and thus conceptualise one of the core objects of HEL. We argue that academics and professional staff engaged in leading in HE are more capable and successful if they can theoretically conceptualise the work carried out within the organisation-in this case, teaching. To lead pedagogical praxis pedagogically is a very specific type of leadership.
Secondly, being able to conceptualise PL influence provides us with a language for understanding the leadership influence that occurs within and between actors at all levels in HE, NAT gives us a language to talk about PL influence regardless of level or context. Using NAT to approach PL in HE provides a language and framework for studying PL interaction as an ongoing process of production of direction, and thus a theoretical foundation to talk about PL influence, which is rarely theorised in research hitherto. The non-affirmative concept directs our attention to degrees of freedom created in a pedagogic summons, be them on a local, national, or global level.
Finally, leadership and management are increasingly related to the development of organisations' operative culture. Leadership of such change processes is pedagogical, as it aims at influencing and developing culture through the learning processes of groups of individuals. Developmental leadership is thus an object for educational theory to study. We believe the above argumentation provides strong reasons to approach EL, especially the pedagogical dimensions of it, utilising a theory of education. We argue that a non-affirmative approach can overcome some of the difficulties identified, avoiding the typical dilemmas of both functional-reproductive and ideological-transformative approaches to education.

Discussion and conclusions
Based on three critiques of perceived shortcomings in HEL research, this article elaborated how non-affirmative theory of education and Bildung apply to overcome them. NAT originally developed within German and Nordic theories of education, drawing on a longstanding tradition of Bildung and a related concept of education, where human and societal development are intertwined and interdependent. Political democracy and a liberal economy thus influence education, but are, at the same time, dependent on citizenship and professional education. They are thereby related to EL emphasising individual self-and co-determination, visible in the principles of freedom in research, teaching and studying.
Consequently, education of university students and related leadership supporting faculty and staff require a deliberational and processual notion of leadership activity. NAT makes pedagogical qualities of HEL explicit by asking to what extent such practices are nonaffirmative in character. In its normative stance, NAT argues that while external legitimate interests need to be recognized, they should not be affirmed one-sidedly, as affirmative leadership instrumentally subordinate higher education to external interests, which violates the idea of Western universities. Thus, NAT suggests that EL research should incorporate the idea of a non-hierarchical relation between education and other societal practices. Accepting a non-hierarchical view as a point of departure for understanding HEL, NAT offers a leadership language that incorporates dynamics across levels (i.e. bottom-up and top-down) and within distinct levels of administration. HEL mediates and transforms external interests, providing various degrees of freedom for enactment processes on other levels. In short, non-affirmative leadership allows for non-affirmative curriculum work, teaching, and research.
The approach maintains a distinction between EL and PL. From a structuralist systems perspective, EL refers to a multilevel networked phenomenon concerning the governance (management, leadership, and development) of institutionalised education. EL involves leading educational institutions at a teacher-, programme-, faculty-, university-, national, or supranational level. Working at any of these levels, EL can include all aspects of what it means to manage, lead, and develop an educational institution (e.g. legally, organisationally, economically, relationally, pedagogically, or technologically). One of the aspects of EL is PL, leadership interventions with pedagogical interaction at the core. Recent managerial leadership policy reforms (Gunter et al., 2016) obviously favour affirmative leadership, thus jeopardising those non-affirmative education principles historically of central concern and principal aims of universities.
Originally, the terminology developed to understanding teaching as related to Bildung (summoning to self-activity and Bildsamkeit) described dyadic teaching, studying, and learning processes. In this context, we extend the usage of summoning the other. Both the summoning agent and the 'other' can be understood both as a 'generalised other', such as an individual, organisation, a board, a policy, or a nation. One implication is that PL occurs in many places, beyond traditional pedagogical situations. Defining PL in this manner is not in conflict with but goes beyond the mainstream interpretation of PL concerning leading practices of teaching and learning in a school context (Male & Palaiologou, 2015). The direct or indirect engagement of others' self-directed activities to reach beyond a present state through a process of learning can occur in any societal field or organisation where human resources are crucial for the organisation's activity. Defined in this manner, neither EL nor PL are thus necessarily 'types' of leadership, such as transformative, transactional, authentic, or distributed leadership.
Adopting a Bildung theoretical point of departure, where the subject's relation to the world, others, and herself is constitutively open, provides a processual view of being human. In Bildung, being is an unending process of becoming but always in relation to something other than the subject herself (English, 2013). For the individual, this processual view includes reflection of the direction of one's own development. In essence, Bildung is connected to a non-teleological view of societal development. There are no inherent ideals or determinations directing individual or cultural development, and the future is radically open. Thus, 'being as becoming' is not reducible to either sociality or subjectivity, but a relational concept. Following such a relational and processual social ontology (Uljens & Kullenberg, 2021), PL is an activity where roles take turns, but it is always an intervention in the other's relation to herself, to other human beings, and to the world. By invitational summoning of the other, PL creates a temporally limited reflective, intersubjective, or shared space aimed to enable the other to engage in switching her gaze, perhaps resulting from a comparative reflection developing a renewed understanding.
Our discussion highlights the importance of having a theoretical framework enabling approaching and elaborating pedagogic influence and the extent of its coercive character, namely, the degrees of non-affirmative action it renders possible, in a theoretically coherent way. An important point of departure is seeing the relation between education and other societal domains as non-hierarchical. This allows asking to what degree and in what ways the relations between societal domains such as politics, economy, or religion allow HE an autonomous space to recognise the interests directed at HE, but not affirm them, thereby allowing HE to fulfil its task in a democracy.
Funding Open access funding provided by Abo Akademi University (ABO).

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.