Introduction

While mental health systems have come a long way since the days of 'insane asylums' and their history of cruelty and rampant abuse of people 1 under their "care", they are still plagued with controversial issues of power and control. Practices of physical or symbolic violence, degradation, infantilization (or some combination thereof) to control how people receiving psychiatric treatment and care are still rampant in mental health systems across the globe (Burstow 2015; Holmes et al. 2014; Joseph 2014; Liegghio 2013). To varying degrees, mental health service users can be subject to the administration of physical/chemical restraints while in hospital, medications that shorten life and alter reality, intensive surveillance practices, the use of involuntary incarceration in mental institutions, beatings by security officials, infantilizing punishment and reward systems, and various forms of moral judgement and stigma (Ashcraft and Anthony 2008; Johnston 2014; Johnston and Kilty 2014, 2016; Burstow 2016; Due et al. 2012; Holmes and Murray 2011; Holmes et al. 2014; Lincoln et al. 2017; Mason 2006).

Most of these instruments of power retain legislative authority in mental health acts (Federman 2012; McSherry and Weller 2010; Peay 2003; Szasz 1989), and as a result medical experts today can still forcibly incarcerate, restrain, sedate, seclude, and medicate people in hospitals as long as the doctors' decisions are not found to be punitive in nature and in the best interests of the service user (Fennell 1996). Others have questioned how the distribution of "improved" drugs to treat mental health suffering, which are supposed to reduce the side effects of older psychoactive medication, perpetuate the corporate pharmaceutical regimes that ultimately have more interest in selling drugs and controlling populations than developing (if possible) more long-term treatment plans for mental health suffering (Breggin 2008; Farber 2013; Whitaker and Cosgrove 2015).

Dynamic controversies regarding informed consent (or lack thereof) for treatment exist in psychiatry. Oftentimes, patients who seek mental health services voluntarily will report that their treatment was more effective than patients who were ordered to receive care (Alang and McAlpine 2019). Even with the recognition that individuals are autonomous and quite often can make good decisions about their care (Neilson and Chaimowitz 2015), a great deal of trust and authority is given to psychiatric experts to compulsorily treat mental health service users on the principle that their medical judgments are better than the person suffering. But this begs the question of how—if it all—can we identify an exact "threshold" after which people can no longer make decisions about their health? Additionally, what agency should service users be afforded to have their voices legitimated? By this we mean, what kind of capacities do they have to make claims about mental health and the mental health system? Tensions remain over such an ethics of care and responsibility, and continue to intersect with the person's socioeconomic status, race, gender, family support system, or broader capacities in general (Erving and Thomas 2018; Villatoro et al. 2018).

Because of the privileged knowledge of mental health professionals and suppression of other forms of knowledge on mental health suffering and distress (Cohen 2016), less is known concerning how mental health service users resist institutionalization as well as the discourses and power of the psychiatric apparatus. 2 The first author in particular has experienced involuntary psychiatric hospitalization for a psychotic disorder, and takes anti-psychotics to manage his illness. He has seen a number of psychiatrists and physicians for his mental illness. He has also taken part in psychiatric coercion, having helped restrain and chemically sedate service users during his tenure as a security officer in a psychiatric ward setting (Johnston and Kilty 2014, 2016). This experience shaped his own questions about the nature of patient-doctor relationships, and possible avenues for resistance.

This study, drawing on 5090 English reviews of 468 psychiatrists working in Canada posted on ratemds.com, examines how Canadian mental health service users enact agency and subtler forms of resistance when they are treated by doctors at the crossroads of these controversies and tensions. In doing so, we employ a post-anarchist perspective 3 in analyzing the narratives of our reviewers to discover the micropolitics of psychiatric resistance. Within this frame, anarchy is not seen as a desired Utopian end state based on the rejection of hierarchical authority and the valorization of a more fundamental level of human existence. Rather, post-anarchism takes anarchy as an ontological principle that must be adopted in any analysis of institutions of power. In this respect it is a continuation of Foucault's understanding of power as an inherently unstable relation, always susceptible to dispute, contestation, and reversal. As Newman (2016b:12) writes, post-anarchism is "a form of autonomous action, a way of acting and thinking anarchistically in the here and now". Simply, post-anarchism takes a view of politics in which essential identities are de-privileged as the point of political contestation, and one where normative categories are destabilized. Consequently, we argue that mental health service users draw on affective techniques of counter-analysis 4 to critique or appraise their doctors and the mental health system, and produce a knowledge that nuances and resists some dominant psychiatric discourses.

While much of the Mad Studies literature problematizes the functions, discourses, and practices of the psychiatric establishment (for example, see Lefrançois et al. 2013), as well as mobilizes first-person perspectives to show how service users resist and complicate its discursive power (Capponi 1992; Michener 1998), we argue our online data is unique for a number of reasons. Specifically, in micropolitical fashion, the reviews and critiques we analyze tend to focus on individual doctors rather than assert grand narratives about the psychiatric establishment, which is a unique diversion from much of the historical and current anti-psychiatry literature (see Burstow 2015; Szasz 1972, 2008, 2010). Yet interestingly, the targeted and individualist nature of our data still makes remarkable and clear links to the more macro critiques found within the realm of critical disability studies, without asserting a stable political position. We argue this online nuance is important to the future of Mad Studies, as service users in an anonymous review site can assert "distance" from the discourses and gaze of the psychiatric apparatus without necessarily committing to all of the radical political positions and divides embedded in the praxis of Mad Studies. Like other social movements in our current era (Kelly and Winlow forthcoming), Us versus Them mentalities can exclude service users' voices on the basis of the language they use or the diverging positionalities and platforms for change they invoke (Cresswell and Spandler 2016). In this sense, service users accessing online fora become included micropolitically without too much theoretical commitment, and this perspective is important to consider in the Mad Movement or any critical exegesis that is prone to becoming unproductively at odds with its own members.

Before turning to our analysis, we first flesh out the framework of this study and some of the ways in which mental health systems and psychiatric discourses have been challenged and navigated. More specifically, we overview how patient-doctor relations create tension and suffering in healthcare spaces, and how the texts upon which psychiatry draws much of its power shape experiences with pathologization and discourse resistance. We then interrogate the processes that emotions, class status, and behavioural differences play in the social construction of mentally ill people as 'deviant'. Following this discussion, we situate this research within the limited scholarship on psychiatric resistance, and then describe our online research design. We find that reviewers interrogate and disrupt discourses of pathologization, treatment, care, sexuality, and sexual function. Through these micro-political insurgencies against their doctors, reviewers re-direct clinical power dynamics in a way that demonstrates their autonomy and agency.

Mad Studies: Tensions, Problems, and Resistances in the Mental Health System

Situated under the umbrella camp of "Mad Studies," this study contributes to the body of work that places understandings of madness in the control of the authors, and often challenges the more pathologizing narratives of madness in the mental health system and their consequential impacts on service users, which sometimes work to condemn or infantilize other-minded people. There is a relatively large body of work that engages first-person narrative as a strategy to preserve and explore self-identity when illness, mental health system officials, or society in general threaten to contest or fragment it (Church 1995; McDonnell 2018; Michallat 2017; Stone 2006). Existing first-person narratives of madness confront the triggers and symptoms of madness in ways that look to rebuild, strengthen, and reinstate self-identity in the face of its disintegration (Bassman 2007; Burstow and Weitz 1988; Pembroke 2009; Russo and Sweeney 2016; Shannonhouse 2003). While what follows is a discursive map of the many theories, concepts, and empirical considerations that have shaped the critical inquiry of psychiatry, we position this work firmly within the politically diverse Mad Studies movement as a point of clarity for readers who may be unfamiliar with the porous borders of critical psychiatry scholarship.

Research suggests that mental health service users feel more dignified and safer when caregivers afford them autonomy over their treatment, making them a partner in their own care rather than a "subject" (Chambers et al. 2015; Kuivalainen et al. 2017; Lindwall et al. 2012; Quirk 2012). Others feel more powerless in settings where caregivers or security personnel misuse or abuse their power (Jacob and Holmes 2011; Johnston and Kilty 2014, 2015, 2016; Johnston 2014). Women in particular have expressed their experiences with betrayal, harm, and indifference from healthcare staff during psychiatric hospitalization (van Daalen-Smith et al. 2020). Indeed, fear of patients is a major variable influencing nurses and doctors' decisions to distance themselves from care and manage healthcare space in terms of risk and security (Jacob and Holmes 2011). Some nurses experience significant moral distress when they see other nurses and doctors refuse to talk meaningfully to in-patients, endure a hierarchal power structure of physicians, and feel like they do not have the ability to advocate for ethical change in their institution (Woktowicz et al. 2014).

There are many positions one could take in terms of how (in)effective the mental health system is at treating and helping people. However, this article is more concerned with the fact psychiatric experts have a tendency to construct service users as passive recipients of psychiatric care, perpetuating the assumption that people who suffer from serious mental illnesses are unable to comprehend their own sickness, recovery, and behaviour and therefore engage psychiatric experts about their care (Cohen 2008; Kirmayer 2000; LeFrançois et al. 2013). To people who have good experiences with doctors in the mental health system, these suppositions may be startling and troubling, but still it remains imperative to question our assumptions concerning the legitimization of mad discourses, psychiatric treatment, and who gets to advance and 'implement' such knowledge.

Doing so requires being critical of the texts upon which psychiatry draws most of its power and authority, and also because many of our reviewers problematized the processes that their doctors would engage in to diagnosis, label, and treat their condition. The third edition of the DSM in 1980 "replaced the theoretical and aetiological system of classification…with an atheoretical and purely descriptive approach to classification, which has remained the favoured methodology in subsequent editions of the DSM" (Tsou 2016:406). Psychiatry then focused on looking for indicators to classify disease rather than dig into the patient's life history to come up with a theory of why they are suffering the way they are. The DSM-III was heralded by practitioners as revolutionary for its ability to produce reliable and testable criteria for mental disorders; specifically, a cluster of various symptoms that could help explain biological, neurobiological, and molecular mechanisms occurring and interacting at multiple levels to produce unwell states of being (Clegg 2012). One limit of the DSM in a descriptive form is it has difficulty classifying and causally explaining naturally occurring disorders such as depression and schizophrenia. Though grounded in years of clinical research, another assumption is that clinicians must believe that the cumulation of descriptors and criterion are in fact a conceptually sound, uncontroversial, and valid basis for determining the existence of a disorder (Wakefield 1992).

While the DSM remains a useful text for practitioners trying to diagnose and treat people from their mental health condition or disorder, like any other discourse or narrative, the text is conceptualized and circulated through a process that is shaped by political, gendered, cultural, and socio-economic forces (Clegg 2012; Reuter 2006, 2007). For instance, the gendered disposition of psychiatric pathologization is evident by the disproportionate number of women who are prescribed psychotropic medication or diagnosed with a disorder because they express gender in ways that are deemed suitable only by men (Russell 1995). The introduction of the DSM-V, which was created in part to recognize the degrees to which culture might influence how mental health suffering is experienced and understood (Wakefield 2016), still forces patients to absorb other clinical definitions, such as depression, in order to receive attention from the mental health system. The broader shift from explanation to diagnosis and treatment results in a more volatile eruption of bottom-up forces resisting the psychiatric apparatus, especially when such definitions are too obtuse, clinical, or controlling. It is this emergence that this study intends to analyze.

In thinking more about the differences between what societies deem rationalist, sane behaviour as opposed to 'mad' and potentially sickly expressions or outbursts, Peggy Thoits (2012:201) uses the term emotional deviance to account for the "persistent, repeated, or intense violations of societal feeling or expression norms, where emotion management efforts are often ineffective". She indicates there is a perceptive (unphysical) threshold people cross when they communicate thoughts and identities that others recognize as constituting mental illness or sickness as opposed to just say, unusual forms of expression and distress. Her disconcerting yet evidentiary content analysis of the 351 Axis 1 and Axis 2 disorders located in the DSM-IV finds that 101 (30.5%) of these diagnoses are primarily established by the recognition of culturally constructed deviant feelings or emotional deviance. The labelling of mental illness begins when we lose emotional control, disturb others by expressing emotions that are unruly, disruptive, excessive, and unsightly (i.e., anger, frustration, nervousness, fear, melancholy, worry, apathy, laughter), or do not perform emotions that societies deem to be appropriate given the circumstances. She and others (Kirmayer et al. 2015) are quick to point out that the criterion that comprises definitions of mental illness as well as how it is experienced by people are heavily shaped by cultural constructions of normal behaviour and our underlying ideological orientations.

For example, patients with a lower socioeconomic status tend to get diagnosed with more severe disorders such as schizophrenia when their behaviour presents to others as unruly, confusing, irrational, or erratic (Dohrenwend 1990; Hollingshead and Redlick 1958; Kessler et al. 2005). Debate still exists as to whether socioeconomic disadvantage shapes the production and realization of psychiatric symptoms (Deleuze and Guattari 2009) or whether marginalized populations are more likely to be labeled and diagnosed with a serious mental disorder (Conrad and Bandini 2015).

Nonetheless, those whom society pits as "mad" can get taken up as a separate class of "deviants" because of how others view their behaviour. Mental illness is still not often understood well in society, but rather can be perceived and stigmatized as a dire weakness and difference, which creates a negative connotation associated with some types of behaviour (Conrad and Bandini 2015; Dobransky 2019). For instance, racial discrimination can lead to more negative mental health outcomes (Brown et al. 2019; Head and Thompson 2017; Louie 2020). Other people with more "invisible" mental illnesses, meaning sicknesses that they can conceal from others because their behavioural differences are less obvious, become obligated to perform sanity and conform to institutional behaviours so they can adequately manage the stigma, professional expectations, and vulnerabilities some may perceive as a fault (Johnston and Sanscartier 2019; Link and Phelan 2013).

While not directly related to viewpoints of psychiatrists, Lopez et al.'s (2012) study of online reviews of general physicians on ratemds.com found mostly positive results. Reviewers indicated that care extends well beyond the doctor-patient dyad, as they appraised or critiqued their doctors based on access and convenience of services, as well as their impressions of the clinic's other staff members. Negative reviews underscored the importance of good bedside manner, as doctors who were perceived to be rude were less likely to receive a positive review. In a similar study of ratemds.com, Gao et al. (2012) found that physician hierarchy influenced the results, with board certified doctors and long-term graduates tending to receive slightly higher ratings than those with malpractice claims. While both of these studies treat the reviews as their analytical object, which in turn speak to the varying functions of the medical system, we treat the psychiatrist reviews as radical tools of discourse via a post-anarchist perspective.

Psychiatric resistance in particular has manifested inside its institutions. One ethnographic study observed that male psychiatric service users problematically drew on their masculinity to emphasize their heterosexual status (Leyser 2003). They used roughhousing, touching, sexualized talk about and objectification of women as a means to resist hospital constraints that limit expressions of gender to normative enactments. Other research documents that smoking spaces found in hospital units help service users build social capital and resist institutional control, which some staff will then heavily regulate as a "protective health measure" (Wood et al. 2013). An earlier study found that ex-psychiatric service users constructed alternative forms of social organizations to oppose mainstream society and resist stigma and oppression (Herman and Musolf 1998). Some strategies involved self-labelling, selective concealment, or political activism in the forms of walkouts, sit-downs, and protest marches.

Beyond direct forms of action and activism, Montenegro (2018) recently explored how service users disengaged and distanced themselves from the psychiatric apparatus, which positioned themselves outside of the legibility, approachability, and descriptive ambitions of the mental health system. Some studies discuss the logics of professionalization (El Enany et al. 2013) and how those working in mental health are just as prone to psychiatric power and oppression as service users (Chapman et al. 2016; Tosh and Golightley 2016).

All that said, it appears that overt psychiatric resistance has only been preliminarily explored. More so, there is a gap in the literature assessing how patients engage in more subtle, anonymous techniques of psychiatric resistance, which this study seeks to explore. We are less concerned with how mental health service users engage in direct forms of resistance and action against mental health professionals, but rather are interested in how they refuse to become subjectivized by psychiatric discourse and power. We explore how digital media provide mental health service users with a community of critique to regain control over settings where there are many power imbalances. More specifically, we examine how "the online" acts as a conduit for affect, emotion, or feelings that may be considered "abnormal" under the psychiatric gaze (Thoits 2012). In other words, how does affect, which is controlled, minimized, and regulated through the normalizing forces of the medical and psychiatric apparatus, manifest online, and illustrate the discursive and emotional tactics of resistance wielded by service users? Further, (how) do these tactics serve collective functions of resistance, and create a reservoir of resistant discourse against the psy apparatus, despite the fact that posts about psychiatrists, psychotherapists, and counsellors are made by individual service users? By addressing these questions, our goal is to examine—using a post-anarchist framework—the manners by which service users establish forms of identification at a distance from psychiatric providers, destabilizing the "interpellative hail" of psychiatrists (Newman 2016b).

Methodology

To explore these issues, we analyzed 5090 English reviews of 468 psychiatrists working in Canada (313 of which are male). These reviews were posted between the periods of 2007 and 2018, and collected from ten provinces and two territories, with the Northwest Territories containing no reviews. In these data, psychiatrists were primarily responsible for diagnoses and drug prescription, although they did perform secondary services of counselling/psychotherapy that were also visible in reviews to a lesser extent. Reviews of psychologists and other psy-professionals are outside the scope of this article but should be explored in future research. Names, genders, ages, and other identifying information of the doctors are not reported in this study, though they are publicly available. Last, because the focus of this research is an exploration of the micropolitics of psychiatric resistance, we are unconcerned with the quantitative aspects of the ratings, chiefly how many stars doctors were afforded.

To fully appreciate how reviewers subvert medical subjectivization through the online circulation of affective texts, an ethnographic approach is warranted as we do not aim to simply examine qualitative themes in a more or less detached, scientific manner. Online ethnography has become a popular approach in the last twenty years, serving a variety of methodological functions ranging from the digital tracking of how certain virtual communities grow and change, participant-observation in online forums, and reading online texts left by others to capture some aspect of the human experience and virtual community-building (Garde-Hansen and Gorton 2013; Gatson 2011; Hine 2008; Horst and Miller 2012; Kozinets 2015). As Pink and colleagues (2015:5) illustrate, "virtual sociologists" frequently analyze online texts to understand how "inequality is extended, reproduced, or complicated by digital media technologies", and how asymmetrical power relations manifest online textually and otherwise.

López et al. (2012:688), taking a more scientific and systematic approach to analyzing ratemds.com, developed a set of "global themes" and "specific factors" of what keeps clients happy with their doctors, notably interpersonal manners, technical competencies and administrative efficiency. While we are certainly interested in how these themes emerge from these data, we are also interested in how clients build what digital media scholar Adi Kuntsman (2012:3) calls affective fabrics, which she defines as "lived and deeply felt everyday sociality of connections, ruptures, emotions, words, politics and sensory energies, some of which can be pinned down to words or structures; others are intense yet ephemeral." Importantly, she notes, when individuals identify and discuss their feelings in online forums, they collectively create archives and communities of feelings, where "the emotionality of the everyday starts living its own life in cyberspace" (ibid: 7). Using online ethnography, then, we treat our data as indicating more than strangers simply venting on the internet about their doctors. Rather, they are "memorial talismans that carry the affective weight of the past" contained within the "digital archive" of ratemds.com (Cvetkovich 2008:120; Kuntsman, 2012:6). In sum, we want to ensure our analysis stretches beyond a scientific, cognitive function to encompass an affective one as well.

To generate a thematic scheme, the first author first read all the reviews line by line to obtain a general impression of the epistemological and theoretical connections and digressions the reviewers made about the Canadian mental health system (Creswell 2014). After getting a feel for the data, the first author then compiled a set of codes that emerged inductively and he applied these to his field notes he took during his first read-through (Chretien et al. 2015). Qualitative data were managed in a word processor program and representative quotes were highlighted for each code. Key concepts and thoughts that appeared repeatably were tracked in the word processor, which produced three sequential critical sections such as service users feeling misunderstood by their psychiatrist, counter-analyses of their doctor's behaviour, and expressions tying psychiatric practice to broader social structures like capitalism. We then coded the more positive reviews, though they were far fewer in number (18%).

Sub-categories and themes were then produced (not mutually exclusive), which included psychiatrist aggression or apathy, feelings of pressure, conflicting understandings of mental illness between service users and psychiatrics, doctor focus on sexuality, monetary corruption, and therapeutic alliance-making. Since we do not know the demographics of participants, there may be concerns over the validity of our data in terms of it being truly representative of service user opinions about the Canadian mental health system. Although we could not always code our data in terms of the reviewer's gender, race, or social class, we could determine that power (im)balances were shaping the context of the review, and thus resistance to psychiatric power remains the brunt of our analysis. Future research should continue to analyze intersections of gender, race, age, and class in service user and doctor relations.

Last, we should note that some "netnographers" might object to this kind of passive interaction with data, as we are not directly participating in the co-construction of a community of feelings (see Bengtsson 2014; Costello et al. 2017). Nonetheless, we do follow other "netnographers" in taking this kind of approach to data collection and analysis, carefully incorporating our own reflections as needed. Examples include Ferreday's (2003) earlier exploration of the "Pro-Ana" online community and Underwood's (2017) more recent online ethnography of IPED (image and performance enhancing drug) users. Finally, while there is no clearly delineated community of which to declare ourselves "consummate and acknowledged insiders" (Gatson 2011:514), we nevertheless feel like a part of this affective fabric, having experienced the psychiatric gaze ourselves. The first author's experiences with psychosis and Schizophreniform Disorder (Johnston 2020b) especially made him a real citizen of this research, and in conjunction with the reviewers' narratives, shaped his perceptions of the mental health system.

FINDINGS

Discussion and Conclusion

Research has repeatably found that when mental health service users feel autonomous, they tend to praise their care (Chambers et al. 2015; Kuivalainen et al. 2017; Lindwall et al. 2012; Quirk 2012). Yet when they feel powerless, betrayed, or abused, they tend to criticize their care (van Daalen-Smith et al. 2020). Building on this literature, our findings indicate that not only do service users hold the capacity to point out the insufficiencies of psychiatric treatment through online critique, but in doing so, they turn the tables of psychiatric power and discourse unto their doctors. Sometimes their negative framing encounters tensions such as when two reviewers pathologized their doctors out of spite and anger, however, many of our reviews point to a subtler resistance not yet discussed in the budding literature on psychiatric resistance. Some patients have been known to engage in more direct forms of resistance such as sexualizing their healthcare space or walking out of their closed institution in protest (Herman and Musolf 1998; Leyser 2003), but our reviewers seem to enable the guise of anonymity to bluntly reject the practices, and more importantly, discourses of the psychiatric institution. Their reviews, which manifestly appear to be micro-political attacks on their doctors, also problematize the structures of power that enable psychiatrists to subordinate patients and restrict their sense of community because of the individualist nature of the clinical environment. We argue that communities of critiques are emerging in the online sphere that hold the potential to destabilize and disrupt the imbalanced relations permeating the psychiatric apparatus (see Burstow et al. 2014).

It has been documented that people with more "invisible" mental disabilities encounter difficulty in performing and conforming to all of the professional or societal expectations imposed on them, especially when they are problematically labeled or perceived by others as non-disabled or not disabled "enough" (Johnston and Sanscartier 2019; Lightman et al. 2009). We recognize the possibility that many of these reviewers may feel trapped in a clinical environment that, while often recognizing and treating their disability or condition, enforces obligations and norms of behaviour that do not coalesce with the service user's mad or non-realist ontology and way of being. Requiring compulsory obedience and ability, defined in this sense as an expectation that the service users work diligently to demonstrate they are sane, is a form of ableism we believe constricts relations with professionals. Johnston (2019) finds that direct confrontation with psychiatrists trying to regulate any kind of recalcitrant or "irrational" behaviour does not always yield positive benefits or meaningful negotiation. Caught in a kind of trap, it seems that the ratemds.com forum offers a space of hope and expression for service users to speak out against the psychiatric practices they find unjustifiable or concerning both to their doctor—by virtue of posting to their profile—and to a community of observers who may read such reviews. In doing so this way, reviewers anonymously escape the consequences of more direct resistance or relational engagement with "experts".

We also found that these critiques, lacking this kind of social re-direction, paves the way for uninhibited emotional expression, even at the risk of assuming a double standard, whereas other studies analyzing the person-to-person resistance demonstrate the sheer difficulty service users face when resisting psychiatric professionals whose control on the truths of the mind permeates much of the Western world and beyond (Johnston 2019; Wood et al. 2013). Unlike Montenegro's (2018) recent work that shows how service users may choose to disengage and distance themselves from the psychiatric apparatus when they no longer see the benefit in accessing psychiatry, our reviewers' words often represent a more private, subtler, but still, head-on attack on their caregivers. It could very well be that many of these reviewers decided to still engage with the doctors whom they critiqued, but nonetheless, their reviews are a powerful indicator that service users understand very well the relational dynamics and power (im)balances that govern psychiatric relationships. Not only that, our study confirms that they are willing and ready to resist in ways that are difficult to police and silence; in other words, their transgression is both perceptive and intelligent.

Grounded in years of clinical research, the Diagnostic and Statistical Manual of Mental Disorders (DSM)—a book Burstow (2016) criticizes for being a "boss text"—has become widely regarded as a legitimate tool to define encounters with mental illness and circumscribe appropriate treatment. Our reviewers who sought out treatment voluntarily, in some sense, implicitly acknowledge that they have may have some problems of the mind by virtue of seeking out a psychiatrist. However, it is the application of discourses found within the DSM that so many reviewers resist and find problematic. Reviewers tended to compete with their doctors' diagnoses by sometimes challenging the competency of their diagnosis and active listening skills, or in other cases highlighting that they are sicker than their doctor will give them credit for.

As some reviewers demonstrate through their experience, when activated by a clinician, the DSM carries with it a strong current of power, and thus the potential for abuse can be high (Tosh and Golightley 2016). Thus, this research demonstrates the need to think carefully about the ways in which we activate mental health labels. Medical professionals must understand that when they use these diagnostic tools, they should be open-minded to the possibility that these labels might change, and should be willing to negotiate with service users their applicability. This is a difficult suggestion to bear in mind since doctors may want to show their patients a sense of control and authority they have over their diagnoses, but in doing so, they run the risk of infantilizing the user to the extent that they are seen as unknowing of their own conditions (Cohen 2008; Kirmayer 2000; LeFrançois et al. 2013) – conditions that they, after all, are the ones experiencing. On the flip side, those critical of the psychiatric apparatus must accept the good work that some doctors are doing in giving meaning to those who experience mental illness, as some of our reviews astutely point to there being a social condition with doctors that can be welcomed and embraced by service users.

Another point we wish to stress is that reviewers' harsh critiques may give the impression that their psychiatrists are always acting with malice aforethought and awareness. However, as one study illustrates (Woktowicz et al. 2014), some healthcare professionals have to negotiate their own ethics in order to deliver care competently, and they too may feel like resistance within their profession is difficult to enact. Future research must study psychiatrists' perceptions of their work and the moral, ethical, and power (im)balances they must negotiate under the scrutiny of a great deal of medical responsibility.

Turning to a more theoretical direction, we would like to end this article with a brief consideration of the importance of service user feedback in any critical account of psychiatric practice. It has never been our purpose to examine whether or not such feedback is effective; these kinds of conclusions could be reached were the psychiatrists interviewed and asked whether they accessed and responded in meaningful ways to the feedback posted online, but this was not our goal. Neither has it been our goal to explain such feedback as a sort of capitalization upon the psychological suffering by the mental health industry. Though some users think it is important to "be a smart consumer" (in one unquoted review in our study), we are hesitant to explain sites such as "ratemds" as the surface effect of a more fundamental process of marketization or as yet another residual, individualizing process of a wider "neoliberal turn" (Sanscartier 2017). By counter-analyzing doctors to feel like they have a say over their treatment/conditions, insisting doctors are asking the wrong questions to better control their identities, and attempting to informally expose their doctors' ties to the pharmaceutical industries, service users refuse to simply accept the labels, diagnoses, and pathologies imposed on them. A post-anarchist framework, a view in which "essential identities and fixed normative categories are destabilised" (Newman 2016a:285) allows us to view these textual gestures as subverting medical subject-positions via discourse external to the psychiatric apparatus.

Drawing on such a framework, we do not privilege experiences of service users by granting them a status of "real" while the psychiatric apparatus is merely "ideological". To make these assumptions would to risk rendering these texts as simply another variant of identity politics, in which there is a struggle for representation/rights by a specific, marginalized group. But we understand these reviewers not as appealing to another subject-position, framing their marginalization through a lens representation or rights. Indeed, they are directly contesting their interpellation by the psychiatric apparatus, and, in doing so, refuse to adopt the technologies of the self that the latter tries to impose on them. Our examination of these texts was driven by a desire to find out more about this resistance, rather than appealing to discourses that would simply strengthen a diametrically oppositional subject-position in relation to the psychiatric apparatus, thereby reifying the latter as a repressive, totalizing mechanism.

In conclusion, we believe that websites like ratemds.com act as a digital agora in which people are afforded the ability to make the personal political. Through critiquing their own doctors, reasserting their own identities, and identifying inappropriate institutional coercions, service users invert the question of what is "wrong" with them to what is "wrong" with agents of the psychiatric apparatus (i.e. their psychiatrists). In this way, it is arguable that service users are in fact exercising a form of the sociological imagination, refusing to shoulder only individual blame for the problems they face in everyday living and instead drawing on and contributing to discourse that problematizes a system so concerned with regulating and "fixing" them. Finally, we hope this research will inspire other ways in which individuals destabilize forced subject-positions borne of trying to find help, both in the psychiatric apparatus and beyond.