Introduction
Recent years have seen an ever-growing stream of concerning reports and personal disclosures of bullying in academia (Kinman, 2001; Morrish, 2019; Morrish & Priaulx, 2020; Jayman, Glazzard & Rose, 2022) as well as an escalation in mental illness and high-profile cases of work-related suicide among academic workers (Oswald, 2018; Waters & Palmer, 2021). These tendencies, and the psychic pain they evidence, exacerbate the negative features of a sector that has for a while been confronted with growing levels of precarity, intensification of work, growing managerialism, emergent cultures of aggressive interpersonal behaviours (Hodgins & McNamara, 2017), stalled progress towards gender equality and continuing legacies of – often tacit – racism and elitism.
The phenomenon of academic bullying, understood here as the intent to cause distress through an abuse of power, often systemic power, through acts that denigrate, humiliate and disempower for personal gratification (Cilliers, 2012), consequently sits at the intersection between several very different bodies of research. Changes in the organisational structures and management of universities in the neoliberal era, including the introduction of ‘new public management’, have, inter alia, been viewed through the lenses of political economy and organisational theory (e.g. Lorenz, 2012; Rustin, 2016; Brandist, 2017). While cognitive labour has been the object of critical analysis in the works of French and Italian Marxists (Vercellone, 2005; Moulier-Boutang, 2011). In the field of communications studies, academic work began to be analysed seriously as a form of labour in the 2010s, for example, by Mark Fisher (2011) Toby Miller (2011) and Rosalind Gill (2014). This literature, to some extent, overlaps with studies of academic work as labour in the fields of labour sociology (e.g. Hey, 2001; Maitles, 2007; Norkus, Besio & Baur, 2016; McCarthy, Song & Jayasuriya, 2017; Ivancheva & Garvey, 2022). More recently, these have been augmented by a growing literature on the algorithmisation, automation and digitalisation of academic labour processes (e.g. Wolff & Almeida, 2013; Ovetz, 2017) as well as analyses of academic labour in the fields of industrial relations (e.g. Law & Work, 2007; Hall, 2017; Allmer, 2019) and gender studies (van den Brink & Benschop, 2012; Lund, 2018). These studies have tended to focus on structural factors or on collective forms of agency.
There is currently a large gulf between this – essentially sociological – body of literature and the literature which analyses the causes and dynamics of bullying at the individual level, including its bodily and psychic impacts (Cilliers, 2012; White, 2013). Yet the empirical evidence suggests that these impacts are immense, though poorly understood. To gain insights into this, we must turn away from the studies that view workers as interchangeable parts of a larger workforce and focus, instead, on them as individual suffering subjects. In other words, we must investigate the literature in the field which we loosely term here as ‘psychological’ – a category that spans a spectrum between psychiatric literature at one extreme, and, at the other, studies in the field of occupational health and safety (OHS) that focus on psycho-social impacts of work practices.
In this article we attempt to make a start on bridging this large disciplinary divide by developing a conceptual framework that combines elements from these diverse fields in such a way that new research can be designed and operationalised to make it possible to understand the dynamics of bullying in academic settings as a multifaceted phenomenon.
This framework, we hope, will make it possible to explore the intersections between social and personal factors, including the impacts of processes of algorithmisation, automation and digitalisation on organisational procedures, work cultures and behaviours in academia and their connections to the continuing presence of bullying. We also hope that drawing together insights from bodies of theory that currently occupy very disparate spaces in the literature will make it possible to sketch out an agenda for future research.
One ambitious aim of this exercise is to make visible the bodily experience of violations that bullying entails in a context in which labour is increasingly depersonalised and rendered abstract both in management discourses and in the processes of digitalisation: to bring into view the material aspects of the dematerialisation produced by algorithmisation. In order to do so, we develop the concept of ‘algorithmic violence’ in higher education. Here, we expand on the concept of bullying as ‘the intent to cause distress through an abuse of power, often systemic power, through acts that denigrate, humiliate and disempower for personal gratification’ presented earlier, understanding it to refer to practices of targeting a worker or a group of workers through undermining, exclusion, unreasonable demands and, crucially, through the weaponisation of bureaucratic and algorithmic measures of control and evaluation. This last part of our definition aims to shed light on the increased entanglements of algorithmic violence and interpersonal bullying in academia. While the weaponisation of metrics is one important element of this connection, another key aspect of it is the alignment between bullying and algorithmic compliance. Through this alignment, the bullying behaviour is exercised in such a way that the position of the bully is strengthened and secured through their contribution to university rankings, while the targets of bullying are systematically excluded or prevented from participating in the kinds of activities that are considered more productive and valued in the contemporary academy, such as obtaining grants, publishing research and networking. We argue that the exercise of algorithmic governance through performance metrics leads to a fragmentation of the academic workforce in ways that privilege a particular type of academic subject and that this hierarchisation of labour creates the conditions for abuse of power and privilege in the form of bullying, while simultaneously masking it beneath an apparently neutral veil of standardised protocols.
Methodology
The processes by which this article was produced can best be described as deriving from an experimental, multi-method approach. The four co-authors have between them considerable experience, both empirical and conceptual, of research in the overlapping fields of organisational restructuring, de-professionalisation of knowledge work, academic labour, algorithmic management, new public management, comparative cultural studies, gender studies, trade union studies, occupational health and safety, and the precarisation of labour, including academic labour. They also have experience of psychotherapeutic and counselling practice, as well as scholarly analysis of the psychiatric literature. They were thus well placed to evaluate the literature in these fields critically and pool their knowledge in an attempt to combine them to build an overarching conceptual framework. However, they are also all current or past employees of universities with direct first- or second-hand experience not only of recent changes in management culture but also of workplace bullying. This creates an unavoidable dimension of self-referentiality. But it also makes them vulnerable and exposes them to career risks. After long discussion, all the co-authors, apart from one who is retired, felt that their future employment prospects might be damaged if they were named, and that they would prefer to remain anonymous, even though this would lead to the loss of recognition for their contribution to the development of knowledge in the field, a decision which, in itself, makes a telling point about academic culture in the 2020s.
Even before the decision to remain anonymous was reached, rather than attempt to distance the object of study from their own experiences of it, it was decided collectively to embrace the self-referential perspective this opened up, following a tradition perhaps best developed in the field of feminist research (e.g. Harding, 2008), though also followed in other branches of ethnography, especially autoethnography (Adams, Jones, Holman & Ellis, 2015). Accordingly, a series of online group discussions was organised in which the participants both reflected on their personal experiences, and those they had witnessed, and sought to map these onto the bodies of literature with which they were familiar.
In this process, certain key concepts emerged which seemed to provide fruitful objects of study at the intersection of the various disciplines under review. The next stage involved the refinement of these concepts, with the aim of exploring the extent to which they could be analysed through different disciplinary lenses to produce multifaceted definitions which might potentially enhance understanding of the bullying phenomenon, in particular concepts that could meaningfully be viewed both through sociological and organisational lenses and psychological ones. Building on these discussions, a series of literature reviews was carried out by the participants to gain an overview of the relevant state of the art. These were then brought together to construct a draft conceptual framework.
In the next and final stage of what is still very much a work in progress, we attempted to test this framework by applying it to some empirical data in the form of records of interviews carried out between January–March 2021 with academic workers in which their experiences of workplace bullying were discussed, along with other issues.
It is important to emphasise that these data were not originally collected with this purpose in mind and should be regarded merely as illustrative examples, not as a definitive piece of research designed to test specific hypotheses about bullying in the academy. The research on which they were based was originally carried out for a different purpose: to explore tensions in the culture of Business Schools in the UK in the very specific context of the end of the Research Excellence Framework (REF) in March 2021 in the UK (the UK system for assessing the quality of research in higher education institutions, used to inform the allocation of around £2 billion per year of public funding for university research).
It focused in particular on a wave of redundancies at around the same time, many involving academics in the field of labour studies whose approach could broadly be regarded as ‘critical’. Since the 1970s, when sociologists and political economists began to be employed in significant numbers in UK business schools, critical scholarship has occupied an important, if contradictory, position in their development. On the one hand, there was a growing need to equip future HR managers (whose training was an important business school function) with the knowledge to negotiate with trade unions and set in place policies to ensure that their employers were compliant with equalities legislation and address racism, sexism, homophobia and discrimination against people with disabilities. Many of the people who were recruited to teach such courses had campaigning backgrounds which positioned them broadly on the left politically and, as neoliberalism advanced in the academy, this put them increasingly in conflict with university managers. However, because of their active engagement with politicians, trade unions and third sector organisations, they were also much more likely to be able to generate the kinds of ‘impact’ that could be substantiated according to REF rules and hence monetised by these business schools.
This research asked, first, whether the concept of ‘critical management studies’ (Alvesson & Willmott, 1992) had reached the end of its useful life, and, second, whether the REF had, somewhat paradoxically, served to prolong the employment of critical business studies scholars until after their ‘impact’ had been evaluated.
In the course of this research (Huws, 2021), interviews were carried out with research-active academics currently or recently employed in 11 UK Business Schools, ranging from top schools in Russell Group Universities 1 to lower-ranked ones in former polytechnics, to investigate perceptions of change in attitudes to critical business studies, and explore the extent to which changing REF requirements might have shaped the retention, or dismissal, of these academics. Although not a strictly random sample, politically mainstream scholars were targeted, as well as self-declared leftists. The sample also included senior and junior, male and female staff and included respondents who were happy and unhappy in their current jobs, in schools with a reputation for critical scholarship and in ones where this was absent. With the exception of one respondent on a fixed-term contract, it was, however, made up entirely of respondents on permanent employment contracts, making it untypical of the academic workforce as a whole, in which, according to a Higher Education Policy Institute study, only 67% of academics were in permanent employment in 2020–21 (Ogden, 2023:25). Given the greater vulnerability of non-tenured staff, this might mean that the negative experiences of precarious academic workers were under-represented. Perhaps reflecting their higher likelihood of obtaining tenure, and being awarded professorships (and hence being included in REF submissions), the sample was also disproportionately white and male. It therefore lacked strong representation from people of colour, migrants, people with disabilities, people on precarious contracts and women, especially younger women, and cannot be regarded as representative either of the academic workforce in general or of those likely to be victims of bullying in particular. Again, this suggests a probable under-representation of victims of bullying.
In carrying out the research, one surprise was a marked reluctance by respondents to speak on the record, despite the relatively secure position of the interviewees. When presented with the option of an email interview, few respondents took this up, with the great majority preferring to be interviewed by telephone. As the research progressed, although there were a number of narratives that provided useful evidence on the declining status of critical management studies in UK business schools and some of the contradictory impacts of the REF on research agendas, what emerged, to the further surprise of the researcher, was a series of very strong and emotionally charged narratives about bullying. Some of these narratives were autobiographical, while others described experiences respondents had witnessed, relating to colleagues, sometimes in other university departments. A few of these were followed up, using snowball methods that led to the identification of further respondents, not all of whom were in business schools. This produced a sample of 27 testimonies, not all of which directly address bullying and not all of which emanated from business schools, but which provided us with a data bank of quotations against which our conceptual framework could to some extent be tested and which took the co-authors beyond our own subjective experiences. This exercise should, nevertheless, be regarded as a preliminary and tentative exercise which could usefully be tested more rigorously in further research.
Understanding bullying in the academy
Although research linking mental health problems and precarious work is growing (e.g. Irvine & Rose, 2022) the literature connecting this to the specific impact of algorithmic management within higher education is still emergent, in part related to the ongoing stigma experienced by those who speak up about bullying within the profession (Korika, 2022). This has led to a disconnect between the experiences of bullying by academics and what is formally known from the research literature, which, in turn, leads to a marked lack of critical thinking about the systemic and dynamic nature of how the widespread phenomenon of bullying might be interrelated with other forms of workplace violence (Fevre, Lewis, Robinson & Jones, 2012) in the sector. In the rest of this section, we review some of the relevant bodies of literature with the aim of extracting from each the key ingredients it contains for the development of a more integrated conceptual framework.
Algorithmic management and new academic cultures
Algorithmic management refers to a broad range of practices of control across multiple domains of work. Aneesh (2009), in his study on call centres in India, notably compares algorithmic management or ‘algocracy’ as he calls it, to classical bureaucratic control and deems the imposition of algorithmic control to entail a materially imposed constraint on what can and cannot be implemented as action. As algorithms prescribe a sequence of steps to be performed they, at the same time, negate the possibility of any other approach to a task by obscuring the mechanics of performance behind the codes and firmware and limiting the possibility of intervention and deviation. Unlike other forms of control, where the desired behaviour is taught and where deviation is either weeded out or corrected, algorithmisation simply eliminates the possibility of deviation. This represents an, often invisible, coercion hidden behind the technological configuration of interfaces, standards and algorithms. This may be as simple as the preset and limited options to choose from in a software program, which constrict your possibility for action to a predefined horizon of possibilities. You can click yes or no, you can choose a specific font but there is no way to intervene in the array of choices and the sequence of actions without technical skills and access to the source code of the programs.
The violence that results from such coercive restriction might seem rather benign, but the inclination of algorithmisation to weed out deviation and difference, to eliminate and invisibilise them, represents part of a much broader process, which ranges from labour management to policing, extraction of intellectual labour and algorithmic bias. Media scholar Wendy Chun (2016) argues that the logic of algorithms creates a propensity to homophily – a process through which differences and deviances are erased. Chun sees this process in social media that tend to create ‘bubbles’ of like-minded people, in machine-learning algorithms where statistical probability tends to solidify dominant narratives, worldviews and representations, and in the overall culture of erasing diversity that is coalescing around new digital technologies. This culture of violence, which is rarely recognised as such, perpetuates the exclusion, policing and penalisation of specific groups, such as the poor (Eubanks, 2018) or Black people (Benjamin, 2019) outside of a purely workplace monitoring. There is rather little literature linking algorithmic management to psychological suffering, but one partial exception is Dejours (2006), from the French ‘psychodynamics of work’ school, who draws on Marx’s theory of alienation to explain the role of new forms of management (such as ‘total quality management’) in workplace suicides. An explicit link between neoliberal management practices and workplace suicides has also been drawn by Waters (2020).
Algorithmic management is not, of course, exclusive to academia, but universities represent one of the many contexts into which it has been introduced in recent years, as an ingredient in a range of different functions including managing student selection and admissions, marking, staff recruitment, performance appraisal and a range of different aspects of employee management and quality control.
The literature on algorithmic management is scattered across different disciplinary fields, and much of it focuses on technological and economic aspects, rather than the impact on workers. An important exception to this is the work of Kellogg, Valentine and Christen (2020) who conducted a systematic review of the literature and concluded that there were six distinct ways in which employers use algorithms to control workers which they call ‘the “6 Rs”: to help direct workers by restricting and recommending, evaluate workers by recording and rating, and discipline workers by replacing and rewarding’ (ibid.: 368). All of these are evident to some extent in higher education.
Academic exceptionalism and the link to algorithmic management
The role of algorithmic management in universities and the violence it enables there is, however, less straightforward. Part of the reason this relationship is complicated in academia is the very nature of intellectual labour. Information and knowledge are both the products of academic labour and the tools for algorithmic management that confront university workers (Woodcock, 2018), a paradox that serves as a model of self-perpetuation (Nicklich & Pfieffer, 2023) of the increasing striving for metrics of excellence. Information or, more accurately, the knowledge produced in academia, is subsumed and valorised through metrics that measure the quality of research and that are generated through the participation of the whole academic community through practices of citation, through preference for specific outlets of publication and through the illusion of meritocracy that underpins the professional notion of good research. This dynamic interaction between professional standards for quality in academic research and their use in metrics for the measurement of performance foregrounds a contradiction that makes the relationship between algorithmisation and bullying particularly insidious because it presents itself as safeguarding academic integrity and a high standard of research (although similar practices in other fields might be regarded as deskilling).
These dynamics lead to the normalisation of divisions of labour, where metrics of research quality are increasingly cherished by universities as ways to secure leading positions in global rankings and embraced by researchers themselves in order to create and sustain distinctions between elite researchers and staff engaged predominantly in teaching. Mariya Ivancheva and Brian Garvey (2022) argue that these processes reveal the role of algorithmic management and metrics in bringing about the real subsumption of labour – subsumption through which the very nature of academic labour is reconstituted and through which, they claim, new hierarchies of academic labour are imposed. These hierarchies stem from the high value placed on research excellence in global university rankings. As they note:
This progressive enclosure of academic knowledge production and the competitive growth model that now underpins each of the UK institutions has a marked signature on internal, stratified divisions of labour that polarises research and teaching work. While institutions boast rankings based on research output to attract students, these students are less and less likely to be taught by the highest ‘valued’ academics who boost these ratings. Those who regularly win prestigious grants and have high international visibility are afforded (or rewarded) additional research time free of teaching commitments. (Ibid.:389)
Ivancheva and Garvey argue that this division of labour tends to oppose ‘productive’ research to what they term the ‘reproductive’ labour of teaching, mimicking the already existing divisions of labour in larger society that have been the target of feminist critique. Moreover, the workers who end up taking over the reproductive tasks of teaching are increasingly the subjects who have traditionally been marginalised within an elitist Eurocentric academy: women, working-class academics, people of colour and migrants (see, for example, Burlyuk & Rahbari, 2023)
As Michał Zawadzki and Tommy Jensen (2020) show in their co-authored autoethnography on the subject, the hierarchies imposed through research ranking and the valorisation of successful external funding are masterfully exploited in the case of academic bullying. Academic bullies embrace and weaponise the privileged position in which the culture of metrics puts them if they have won a grant or maintained a successful record of publications. The persistence of such hierarchies replicating traditional class, gender and race hierarchies converging with the perceived ultramodernity of algorithmic management in academia appears to reproduce a retrograde historical reality of exclusions and exploitations. It also creates what Gigi Roggero (2011) refers to as ‘blockages’ to the possibility of developing organised resistance against the neoliberalisation and automation of academia. Roggero sees these blockages as examples of how the technological constitution of class (i.e. how class is divided and organised by capital) subsumes and ‘crushes’ the political composition of workers’ organisations and the solidarities built from resistance against these divisions and exploitations.
At the same time, as many have noted (Swinnerton et al., 2020), universities are ramping up the experimentation and implementation of automated modes of teaching, where digital technology and online platforms are mobilised in an effort to make tutoring and lecturing work standardised and replicable (Ovetz, 2017) and, where possible, even record teaching sessions to make them available for future use, for which their authors are not remunerated. The use of digital technology and algorithms deepens the gap between winners and losers in the contemporary academy in several ways, including the mechanisms of transforming the ‘living knowledge’ (Roggero, 2011) of academic knowledge production into the ‘dead knowledge’ of metrics, platforms and algorithms that confront academic labour, solidify workers’ hierarchies and enable labour intensification and exploitation.
The algorithmisation of academic labour further deepens inequalities between workers by maintaining a conservative and black-boxed system of evaluation of research ‘excellence’ on the one hand, while increasing the precarity of teaching labour, on the other. While journal impact factors and citation indexes are presented as objective measures of the quality of professional work, the reality behind these metrics is that they often reflect the incorporation of scholars within elite academic networks to a much greater extent than anyone cares to admit. Thus, the entanglement of research excellence with markers of socio-economic belonging and identity that reaffirm the dominance of white, middle- and upper-class male academics becomes invisibilised behind the illusion of objective measures of knowledge production. Malniak, Powers and Walter (2013) provide an example of this in their study of citations in the field of international relations, which showed that women are systematically cited less than men. A study by van den Brink and Benschop (2012) on the measurement of ‘excellence’ in HE found that it exacerbates existing gender inequalities by privileging a gender-biased measurement of excellence, particularly within recruitment and promotion processes. This was exacerbated during the COVID-19 pandemic lockdown, reflected in the post-COVID-19 academic publishing statistics which show a distinct and growing gender gap between academic contribution and publication (Squazzoni et al., 2021).
The disability studies literature further reveals that there is a tension between models of excellence and the logics of ableism and disablism that exist in HE. Disablism can be understood as the differential and unequal treatment of people with disabilities based on conscious or unconscious assumptions about and devaluations of disability. Connected to this value system is the logic of ableism in HE, where ‘able-bodiedness’ and ‘able-mindedness’ (Dolmage, 2023) are positively valued to the detriment of disability. Further, within HE cultures of exceptionalism this logic of ‘hyperability’ (Dolmage, 2023) plays into the ‘fetishisation’ of excellence (Moore et al., 2017) in the academy.
In stark contrast to these deep and systemic biases within HE are organisational attempts to address them. The ‘thin’ standard of wellbeing that currently dominates workplace wellbeing programmes (Purser, 2019) that leaves the structural and material causes of low wellbeing intact is founded on the paradigm of positive psychology in the work of Martin Seligman (2011), which focuses on individual cognitions and behaviours, representing a highly decontextualised and individualistic mental model. Consequently, positive psychology allows for a ‘rationalisation and technocratisation’ (Dehler, 2009:34) of wellbeing responses, such that complex organisational work is reduced to applying individual psychological interventions where wellbeing is mobilised and thus weaponised as a ‘politically convincing technology’ (Freidli & Stern, 2019:85), projecting organisational failure into the individual.
Summary
From this brief overview of the literature on algorithmic management we can extract several promising ingredients for a conceptual framework for analysing academic bullying. First, by standardising and anonymising bureaucratic procedures and creating an illusion of impartiality it renders invisible the psycho-socio harms to individuals. Second, by intensifying the pressures on workers it can itself contribute to or exacerbate psycho-social harms. Third, it must be understood in the broader context of socio-economic inequalities in which some socio-economic groups (such as women, precarious workers, ethnic minorities and people with disabilities) are rendered vulnerable, while others (notably white men from privileged backgrounds) are empowered.
Occupational health and safety framing
Despite a large literature on work-related stress (EU-OSHA, 2014), the specific links between psycho-social wellbeing and precarious work remain under-theorised, notwithstanding the useful overview carried out by Irvine and Rose (2022) which points to a link between precariousness and stress. By far the biggest shift in academic labour has been the increase in casual labour and contracts, with an impact on a range of key precarity factors including financial insecurity, lack of career progress, burnout and mental health problems and loss of a sense of belonging (Socio-Legal Studies Association, 2023). Other research clearly links mental health problems to insecure forms of employment such as zero-hours contracts and self-employment (Marmot, 2020) with an emerging literature that links wellbeing to the specific impact of performance management on conditions of work (Boxall & Macky, 2014). Reports on mental health in universities link unsustainable workloads, work insecurity for staff and performance management systems to the evidenced increase in mental health problems within the academic workforce (Wray & Kinsman, 2021; Kotouza, Callard, Garnett & Rocha, 2022). The crisis has been compounded by the experience of working through the pandemic – linked to work intensification, increased student distress, isolation and lack of social support at work (Morrish & Priaulx, 2020).
Although the wellbeing of students in higher education (HE) is named as a priority in many policy documents, little is reported about support offered to staff. Multiple surveys indicate the clear link between workloads and work intensification on academics’ wellbeing – although stopping short of linking the model of performance management and metrics used within HE management explicitly with wellbeing and mental health problems, including the link to suicide (Waters & Palmer, 2022). While universities routinely offer wellbeing programmes for staff, this is dominated by a focus on positive psychology and short-term and individual interventions (Kotouza, Callard, Garnett & Rocha, 2022), supplied by contractors providing Employee Assistance Programmes (EAPs) including those provided by large online therapy and health platforms. Experience of EAPs is mixed but, in the UK, they often only offer short-term (six sessions) limited Cognitive Behavioural Therapy (CBT) placing pressure on both client and therapy practitioner to show high recovery rates, as is required across the wellbeing and short-term therapy sector (Cotton, 2021). In many cases it could be argued that trade union structures and HSE management systems, such as the HSE Stress Management system, offer a more solid basis for organisational change. However, because enforcement is weak and there is a lack of effective sanctions against non-compliance, they offer only a minimal solution to a widespread problem. More concerning is the growing ‘weaponisation’ of wellbeing and mental health issues in the workplace (Kenny, Fotaki & Scrivener, 2019) as a way of driving a normative model of academic performance and individualising the link between mental health and working conditions, in particular through the development of performative systems and the use of command-and-control management in HE, thus rendering invisible any systemic link.
Other factors that have been explored in relation to stress that could relate to the conditions of digitally managed workers include the impact of job insecurity (Ferrie, Shipley, Stansfeld & Marmot, 2002), multiple job-holding (Brun & Milczarek, 2009) and the role played by emotional labour (Pugliesi, 1999). The link between algorithmic management and states of mind is not well researched, in part due to the bias within the psychological research towards focusing on individual psychology rather than the impact of systemic and employment relations environments (Holmqvist, 2009). An emerging literature links mental health problems to insecure forms of employment (Chandola & Zhang, 2017; Gallie, Felstead, Green & Inanc, 2017), such as zero-hours contracts and self-employment (Marmot, 2020) and the impact of performance management on conditions of work (Boxall & Macky, 2014). This suggests a link between algorithmic management and occupational health and safety.
Psychological framing and the dynamic nature of workplace bullying
Workplace bullying has multiple formulations as ‘trouble at work’ (Fevre, Lewis, Robinson & Jones, 2012) but encapsulates the presence of an abuse of power. Within the context of algorithmisation and managerialism in HE, this abuse of power becomes institutionalised and de-personified but nevertheless operates on a profoundly psychological and personal level, both through the drives of the bully and through the effects these have on the bullied. The interplay of technological, institutional and psychological mechanisms in workplace bullying makes it especially challenging to conceptualise, but it is also not a novel phenomenon in capitalist management. Frederick Taylor (1911), in his early twentieth-century experiments in scientific management, posited exactly this kind of connection – that establishing a technologically standardised and optimised process for the purposes of capitalist profit necessitates the exercise of coercion and psychological manipulation of the labouring subject. Behind the coercion, however, there is a destructive drive for annihilation, which Marx (1976 [1867]) saw as the ‘vampiric nature’ of capital that sucks vitality and power out of the labouring subject. Power and vitality feed the flows of capital but they are also, crucially, consumed in the process of technological innovation and standardisation: living labour is continuously absorbed and encased in the technological means of production. We can see this tendency quite clearly in the recent moves by universities to record lectures, standardise and multiplicate the courses developed by tutors and transform intellectual collaborative practices like networking, citations and interdisciplinary projects into metrics used to confront, divide and penalise the academic workforce. At the same time, these technological practices also exacerbate the internal divide within academia that juxtaposes precarious undervalued university workers to workers in a position of relative privilege whose ability to produce high-end research depends on the exploitation and marginalisation of their precarious colleagues. As the labouring academic subject is embedded in a system defined by the cannibalising death-drive of capital, the mechanisms of bullying as individual behaviour and institutional practice are inescapably characterised by pathologies of power and resistance. Similar interpretations of labour-capital relations as metabolic relations of depletion and mutual digestion have been used by feminist scholars like Shiloh Whitney (2018) who says that it is not just physical labour that partakes in this dynamic but also emotional labour. Part of the logic of this metabolic process is that capital consumes and exploits all valuable or positive emotions and forces the worker to digest and metabolise all negative ones (sadness, shame, anger, etc.).
One of the distinct models of bullying that emphasises its systemic and dynamic nature comes from a psychoanalytic tradition. This paradigm derives from Elliott Jaques (1955) and Isabel Menzies Lyth (1960) who looked at unconscious defences against anxiety with a psychoanalytic orientation towards changing social practices and institutions. This model understands unconscious defences as socially constructed and institutionalised to minimise contact, for example, in ritualistic rule-following, hierarchies and restricted discretion, with limited opportunity to reflect. These means of coping with anxieties that become embedded and enforced within structures with rules and cultures where defences are learned tend to marginalise those who resist.
Psychoanalytic ideas allow us to think deeply about the interpersonal and unconscious aspects of bullying, including the exploration of anxieties in work, projections and social defences against anxiety within groups and organisations (Armstrong & Rustin, 2015). Psychological research includes work on the increase of bullying as a result of public sector management – mainly based on studies of performance management systems and algorithmic management in the health sector (Rizq, 2014; Cotton, 2021)
This psychoanalytic model recognises the intersubjective nature of bullying where the victim is engaged in a dynamic relationship with the bully. This is where internal and often unconscious aspects of the self, such as critical voices and trauma can become activated as if an ‘internal bully’ (Cotton, 2017) has become enlisted in the attack. This allows us to understand the disorientation and distress caused by bullying which can, within this model, involve a dynamic relationship between external and internal bullying. The dynamic and systemic aspects of bullying rely on core psychoanalytic ideas including projective identification (PI), which is a process that occurs when we cannot tolerate aspects of ourselves and project them into another, for example, internal vulnerability is projected onto the victim of bullying. This dynamic model looks at how power and submission come to be displaced within the relationship between bully and victim, in a dynamic and complex way. Bullying at work also involves group dynamics and the roles and positions adopted by the range of people involved in bullying whether active or passive perpetrators or bystanders. Again, this draws on the idea of projections and adopting ‘gang-like’ states of mind (Canham, 2002): defences against anxiety – commonly involving scapegoating and denial – as a way of attempting to maintain group cohesion and processes within groups.
Envy, spoiling and coercive control
In a system that is increasingly defined by deficiencies and scarce resources, bullying becomes a tactic of accumulation, encroachment and boundary-making. The PI mechanism that is enlisted in the dynamics of bullying is envious by nature: the dynamic is about ridding the self of what is not wanted but also about appropriating what is lacking and therefore desired. This mechanism is enlisted in cases of bullying both through the projection of feelings of vulnerability onto the victim and also in the projection of envy of the desired qualities of the other. This dynamic of envy in bullying can result from an unconscious desire to acquire what is desired where projection is used to take control of and possess the other.
Although this gender pattern is by no means unique (with women represented among bullies and men among their victims), this envy seems to take a particularly virulent form in the bullying of women by men. The research literature on bullying reveals a high prevalence of women, often exceptional women, among those who are bullied (Baumgartner, Zarestky & Lechuga, 2022). As the 1904 cases analysed by Sherry Moss and Morteza Mahmoudi (2021) demonstrate, a specific targeting of women academics is common, in part because of their precarious positions, combined with rigid enforcement of metricised performance management and associated surveillance (Gill, 2014). There is a further intersection with class, as Johansson and Jones (2019) outline, with women being regarded as double interlopers both in class and gender. A contradictory dimension of this, is that many women and men who experience bullying in the academy are excelling in their work and have a high standing in their relative academic fields but are nevertheless perceived as intruders because of their successful navigation of frameworks that seem designed to exclude them.
Narcissism framing: bullying subjectivity as authoritarian personality
The vicious circle of the core complex evokes powerful psychotic and pre-oedipal anxieties experienced as a threat of complete disintegration of the ego. These anxieties themselves are characteristically infantile, concentrated around narcissistic, omnipotent and psychotic states. Two common defences include narcissistic withdrawal, where aggression is directed at the self (masochism), and sexualisation of the aggression that forms the basis for sadism (Ruszczynski, 2007). Importantly, in these cases the object is not destroyed, but the anxiety is removed, albeit at the expense of care and intimacy, and replaced by feelings of excitement and triumph over the other.
Bureaucratic violence and shame
The revitalisation of sociological perspectives on the impact of bureaucracy (Weber, 2009), particularly workplace bureaucratisation through the advancement of performance management and data systems, has opened up a useful frame through which to understand our, often traumatic (Graeber, 2015), experiences of bullying within HE. Ideas of bureaucratic violence (Arendt, 1969; Bauman, 1989) and how bureaucracy can be weaponised to attack populations (in this case workers), allow us to look more deeply at the ‘technocratic veneer’ (Norberg, 2022) or ‘slow violence’ (Mayblin, Wake & Kazemi, 2020) of the performance management of academics including its datafied and algorithmic formulations. This then allows us to link this context of standardised and weaponised metrics and connect them to their violent consequences on the lives of academics both structurally and bodily, including mentally.
One of the key consequences of these tendencies is the internalisation of power hierarchies and the violence of their enactment through institutional penalisation and interpersonal bullying. Those who are bullied are often the ones who do not fit into the standardised academic subjectivity that the culture of academic algorithmic management promotes. As already noted, they are often women, people of colour, working-class academics, migrants, (single) mothers, people with disabilities and/or people who have worked outside of academia.
This internalisation can be understood as a dynamic relationship between feelings of shame and how they become enlisted by the industrial and digital ‘shame machines’ (O’Neil, 2022) that can be observed in higher education. In the emancipatory writings within the Freirian tradition (Freire, 1972), bullying involves the external bully forming an alliance with the internal bully, the part of ourselves that blames and shames us into agreeing with our own and our institutions’ demands for exceptionalism. Additionally, we are aware that the measurement of any poor performance as knowledge workers in highly metricised professions will be held as digital evidence to be enlisted in future professional attacks. Furthermore, any visible hit on our states of mind, mental health and nervous systems that results can then be used to feed the projection of deficiency that is used in bullying.
In yet another dimension, the attack involves onlookers, more successful colleagues, who become complicit in the violence of algorithmisation, disguised behind a culture of passivity and ‘sarcastic compliance’, a trend that the collective behind The Analogue University (2017) names as a second wave of critique of the neoliberal university. This second wave is marked by attitudes that regard ‘education as a public good being forced to mimic the market where academic values could be maintained by strategically (yet often sneeringly) “playing the game”’. The maintenance of this dual position of ironic compliance enables a bullying subjectivity that can continue to purport to be critical of the algorithmic subsumption of academic labour without challenging or confronting it. Instead, these subjects negotiate the gains of both the symbolic capital of critical research and the material capital of job security and prestige through their compliance with university metrics. They are instrumentalised by and instrumental for capital.
One further line of analysis relates to the obsessional-punitive nature of algorithmic management. Within a psychoanalytic model, performance can be understood as an obsessive defence combining repeated checking, lack of decision making and a severe superego which is signed up to the demand for exceptionalism. The drive of the superego within professions emphasises surveillance, punishment for missing targets and inspections, and subsequent compliance, for fear of reprimand and collusion, involving an identification with management systems. Performance management can be understood as an organisational anxiety about aggression projected into staff and then defended against through an obsessional-punitive social defence in order to control work and the relationships within it.
Summary
For the purposes of our study, perhaps the most important lesson to be drawn from this highly abbreviated survey of the psychological literature is not so much the identification of certain psychological ‘types’ per se, but rather the kinds of behaviours (and associated personality traits) that are encouraged and privileged in particular organisational contexts. On the flip side, this also points to a need to identify the behaviours and personality traits that are liable to be regarded as inappropriate and to be penalised, demonised or render their possessors vulnerable to gaslighting. Our attention is also drawn to the need to focus on the subjectivity of academic workers and how this is impacted both by algorithmic violence and by bullying behaviours.
Towards a conceptual synthesis
The lessons learned from this brief literature review provide us with the ingredients for a multifaceted framework that draws on scholarship in several different fields with the potential for developing a rounded understanding of bullying in academia.
From the literature on algorithmic management we can derive the insight that by standardising and anonymising bureaucratic procedures and creating an illusion of impartiality, new forms of management render invisible the psycho-socio harms to individuals.
Furthermore, by intensifying pressures on workers to meet predetermined targets they can contribute to or exacerbate such harms. The literature on occupational health and safety, tells us that there is substantial evidence of a causal link between the precarisation and intensification of work and poor mental health among academic workers. Taken together, these point to a need to take into account the managerial context of the neoliberal university. Beyond the immediate organisational situation, it is also necessary to take into account the broader context of socio-economic inequalities in which some groups (such as women, precarious workers and ethnic minorities) are rendered vulnerable, while others (notably white men of European origin) are empowered, even while these differences may be rendered invisible by the apparent neutrality of digitally captured metrics.
However, we still need insights from the psychological literature to enable us to observe individual pain and suffering (including actual bodily injury), and how this is related to the aggressive behaviours associated with bullying, which are shaped not only by these social factors but also by deeper psychological ones which are rarely addressed in the sociological literature. This draws our attention to the kinds of authoritarian behaviours that are legitimised and enabled by bureaucratic management. Dialectically related to these aggressive and authoritarian behaviours are those behaviours and attributes which are regarded as inappropriate and punished in these same cultures and which might, indeed, be regarded as posing threats to the managerial authority which lurks behind the technologies which enable it. This in turn leads to a focus on exceptionalism.
In a contradictory twist which seems to be particularly acute in academia, it appears that many of the attributes which contribute to ‘excellence’ in research and teaching (such as intellectual creativity, questioning of existing scientific assumptions, altruism and empathy with students) may be precisely those that are perceived as challenging the hegemony of algorithmic control, exposing those who hold them to the most savage forms of bullying. These negative impacts are, however, both legitimised and normalised through a process of internalisation – a process which, in itself, may exacerbate the emotional pain.
The short-hand term we developed for this interconnecting cluster of concepts is ‘algorithmic violence’. In the next section of this article we test its usefulness as an analytical tool by applying it to the empirical results of interviews with academic workers, focusing in particular on four aspects: their perceptions of the ways in which their experiences have been shaped by technologically-enabled bureaucratic management practices; their personal experiences of bullying, including physical and psychological harms; the behaviours which were rewarded and punished in the academic contexts in which these took place; and their reflections on the ways in which these were shaped by social and personal attributes including gender, disability, ethnicity and citizenship status, class background and political views.
Testing the concept
We now proceed to apply this framework to our empirical evidence.
Techno-bureaucratic (algorithmic) management and cultural change
Increasing bureaucratisation and the requirement to meet ever-more stringent performance targets formed part of the backdrop to all the interviews. As one male mid-career academic put it:
There is consistent expectation/pressure to publish in what the school ranks as high-profile journals; performance is judged by 4-star publications across the school and university.
This was perceived to have become progressively more intense in recent years. A senior male professor in a top-ranked business school described the ways in which this was affecting management culture:
Increasingly driven from the top, increasingly centralised; replacement of awkward academics in senior positions with those more aligned with vision and strategy as devised by those at the top; very awards-driven, competitive, income-oriented, corporate business model of organisation. Staff receive ‘corporate communications’ one or twice a week.
He added ‘Now that we have a corporate management, many of the more opportunistic creative impulses have gone to ground’.
A respondent based in a lower-ranked business school in a former polytechnic put it like this:
What is clear is there is now an agenda that privileges the utilitarian and the income-generating above the academic and research-based.
A male professor at another university described the cultural change:
This has definitely got worse in recent years; 10 or 15 years ago whenever I had anything critical to say about the institution the Dean was happy to engage with me and others to explore whether we could address the issue. This certainly wouldn’t happen now.
There was a widespread view that this culture was associated with growing intolerance and authoritarianism. In the words of this male senior lecturer at a different institution:
The changes have been a gradual inching towards what is now almost complete suppression of any critical voices. It has become implicit that to criticise any aspect of management policies or behaviour is committing career suicide.
This view was echoed by a male mid-career lecturer at a Scottish university, who commented:
The level to which critical scholars have adapted to a ‘normal’ that was not enforced, and without the obvious risk of penalty is rather astounding … no one has been told what they can or cannot teach or research and yet still, the conservatism of erstwhile critical researchers in their bids and teaching is astounding.
In a similar vein, a third respondent, a professor at an English university, said:
I think that the system has been designed so that anything that is critical of the system needs to be dissolved. It’s why sociology, philosophy, psychoanalysis have been under attack for so long. Anything that stands in contrast to how universities are now managed in England needs to be picked off. It’s also convenient that most people won’t risk their jobs protecting something as esoteric as ‘critical management studies’ – it’s code for ‘difficult people’ – it’s not well known enough to defend it.
While the new management systems may appear to be universal and neutral, it is clear that in the hands of individual managers they can be used to bring pressure to bear individually. An important instrument here is the institutional process, usually carried out annually or six-monthly, known variously as a ‘development review’, ‘performance review’ or ‘appraisal’ typically carried out by line managers under the supervision of HR departments. One male senior lecturer reported:
In our annual development reviews that are very target-driven, the university has tried to implement performance improvement plans for those not publishing … Voluntary redundancies are consistently invited, and staff deemed not to be excelling in their annual reviews/REF outputs have encountered suggestions for exit packages… I have even heard of cases where people are forced onto performance plans or threatened with disciplinary procedures but told that these will be withdrawn if they consider an exit package.
Another aspect of new university management is an increasing standardisation and specialisation of roles, eroding the traditional model of an academic as someone who combines research with writing and teaching. The head of research in one leading business school put it like this:
Now the REF is over they are looking ahead and thinking it will be another five years or so before we need to think about research again. These guys are a dead weight. Let’s get them into the classroom or get rid of them.
Despite the experience of increased vulnerability, few respondents felt supported by their representatives in finding means to resist it. This was attributed to the intensification of teaching work, considerably exacerbated during the COVID-19 pandemic. One respondent, who was a trade union rep, explained that this had directed attention away from issues which are seen as more individual and personal (such as bullying, which he was reluctant to highlight as an issue):
We have a problem recruiting young academics into the union at the moment. For most of them the overwhelmingly important issue since lockdown has been workload. There has been a massive increase. This is the issue we need to address if we want to recruit them. For them, bullying and the freedom to speak out are not seen as relevant.
Bullying behaviours below the bureaucratic radar
Below the radar of standardised techno-bureaucratic systems, and rendered invisible by them, there is considerable scope for individual or collective behaviours to develop which can be construed as bullying. One professor put it like this:
If you are a senior academic you don’t tend to think of yourself as bullied. It was more of a creeping process of marginalisation. I felt ignored and left out of key decisions about things like recruitment and course development. Nobody was hostile to my face, but it was reported back to me that I was being badmouthed behind my back. And it seemed that my research, and that of my PhD students, was being sabotaged in various ways.
Often, coercive pressures were described as part of a broader culture in which tightening management regimes were used to keep academics feeling insecure. One female lecturer informant described a department where:
The people on short-term contracts are disproportionately women. The underlying story being that if you’re nice and pleasing to the largely male professoriate you get your contract renewed.
Strategies adopted by the bullies described in these testimonies sometimes bear striking resemblance to those of people described in the literature on coercive control in personal relationships, in which an initial period of ‘love bombing’ is followed by gaslighting and abuse. The same informant went on to describe how two colleagues, one a woman and the other a gay man, ‘found confidences about their mental health used effectively against them’.
The weaponisation of HR procedures by bullies seemed common. Another woman, a professor, described a situation whereby a junior team member was encouraged to take out a grievance against another female professor, claiming harassment. ‘The union officer representing the claimant was a close friend of the senior manager who wanted to get rid of her’. A female senior lecturer in another university recounted how her tormentor ‘coached a vulnerable young woman researcher into raising a grievance against me based on allegedly stealing research that actually never existed’.
Accusations of ‘unprofessional behaviour’ were particularly common in the testimonies of women. In one case, the behaviours that were regarded as ‘unprofessional’ included supposedly being ‘emotional’, referring to a work situation on Twitter, failing to attend a departmental meeting while suffering a migraine attack, identifying with a political position viewed as left-wing and expressing sympathy for a student protest.
In some cases, interviewees referred explicitly to a gap between the official procedures and the actual behaviour of particular managers (the invisible space in which bullying occurred), as in this statement from a female senior lecturer.
From my first week I was subject to a sustained campaign of bullying and harassment by the head of research which carried on for the entire year I worked there, including six months of the pandemic. The point though is that this person was supported by the school’s senior management. His sadism and cruelty, particularly to women and BAME researchers was well known in the sector and we assumed he’s been employed to do exactly the same – strip out the senior and expensive researchers once their REF contributions had been secured.
The painful experience of algorithmic violence
The psychological impacts of bullying are heavy, even for the most apparently privileged academics. The male head of research in a prestigious school reported that:
Previously we had a Dean who was very much focused on generating income and he was pushing me to leave because I was not bringing in large grants. It was a hard time. I really hated going into the office and lost interest in my work. But thankfully he left.
Less fortunate was the male senior professor who confided that:
It was made perfectly clear that I was only there because of what I could contribute to the REF submission and once that was completed I immediately became expendable. This period was one of the blackest that I have encountered and I have really struggled to get through it, to the extent that I was seriously considering leaving academia.
A female interviewee described how it affected her:
It was devastating. There were months when I could hardly walk, hardly function at work but I knew that, if I didn’t, I would lose my job. As a single parent I couldn’t afford it. The most painful aspect was that I couldn’t protect my toddler from what was happening to me.
Another woman linked such impacts to retraumatisation, saying:
I had really bad, traumatic stress symptoms: PTSD symptoms, which were such a scary thing to go through, with flashbacks and nightmares and I have a history of trauma from my childhood, so it kind of just brought … in a really scary way, it brought all sorts of ghosts and very scary experiences, just back to the surface, to basically, re-live every single nightmare.
I started doubting myself, I started doubting my perception of what had happened, I started thinking, ‘Oh, maybe it’s just in my head; maybe I just made everything up’.
The internalisation suggested in that last sentence was echoed by another respondent who reflected ‘Where these things happen, you do tend to internalise and start to question very deeply what it is you seem to provoke in people and their responses’. This was echoed by another woman who spoke of a
mindset in which you can become so absorbed by these toxic systems that tell you that if you don’t make it work, it is a systemic failure on your part not to make it work and that that takes away something from you.
Given the importance of empathy in the emotional labour of academic work, it was clear that many interviewees expected it to be present in relationships with colleagues. Several commented with surprise and shock about its absence in the bullying context. As one put it, ‘there was just no warmth or understanding that you might want to have support in your way through; I was just treated as a problem’. Another (a migrant) said:
And then I just couldn’t believe the callousness of my supervisor, who actually works on migration … yes, who just told me the first time I had a meeting with him afterwards that I should just look for a job somewhere where I don’t have to apply for a visa.
The lack of empathy from managers and perpetrators was echoed by a lack of solidarity from colleagues, summed up by one woman in the words. ‘No one wants to go in there with you … because it’s a very dark place, isn’t it?’. Another recounted how:
… nobody asked how I was doing afterwards; they just, every person who knew about it just tried to pretend it never happened and was acting just super-uncomfortable around me, as if they were just like walking on eggshells, being scared that I might bring up this again, you know. So it made me very angry, it made me very angry.
A caveat
It should be pointed out that the interviews in this section were not originally designed to capture information about bullying and do not necessarily represent extreme cases. Indeed, by over-representing staff on permanent contracts and those involved in research, they may well represent a relatively rosy picture.
They were carried out at a time when over half (53%) of respondents in a 2021 UK survey reported ‘probable depression’ (Wray & Kinman, 2021) and 47% described their mental health as ‘poor’ in another online study (Dougall, Weick & Vasiljevic, 2021) and it seems reasonable to assume that they represent the tip of a very large iceberg.
Conclusion
We can conclude from this brief analysis that a direct line can be drawn between the apparently abstract and impersonal algorithmic management practices gaining force in the neoliberal university and the deeply personal pain and suffering of individual academic workers. Algorithmic management plays a double role here: on the one hand, it serves as an instrument of intensification of work, standardisation, de-professionalisation and disempowerment, while on the other it renders invisible the personal cost of these developments to academic workers. In the invisible arena below the umbrella of algorithmic management, a space is created in which abusive behaviours can flourish, a space in which some personality traits are privileged while others may be demonised. This space is enabled, and takes its characteristics from a context, particularly pronounced in academia, in which workers are expected to perform exceptionally. This raises particularly acute contradictions, creating a situation where exceptionalism is simultaneously both required and punished. In a pattern that is, furthermore, shaped by broader social variables, the vulnerable academic worker, striving to excel, is placed in a ‘no-win’ situation: failure to excel brings the risk of redundancy; while high achievement attracts envy and the risk of running foul of the rules, which demand the impossible combination of conformity and excellence.
This article has attempted to demonstrate that this phenomenon can best be investigated by drawing together insights from diverse bodies of literature. We hope that it will provide a basis for future research that can inform a fuller understanding of academic bullying.