Introduction

The term "placebo" frequently refers to forms of treatment that work in ways that defy the causative logics of medicine. Red pills that stimulate and blue pills that tranquilize, both made only of sugar (Buckalew and Coffield 1982); inactive pacemakers that resolve palpitations in cardiac patients (Moerman 2011); saline injections that improve postoperative pain as successfully as morphine (Levine and Gordon 1984): in each case, placebos prompt a cessation of symptoms but lack the properties by which drugs or other interventions cause physiological changes. For this reason, placebos are well-known and long-standing mysteries; their persistent capacities to solicit well-being (and to substantiate the marketability of new treatments) are at heart of medical practices and biomedical research itself. Less familiar, nocebos are an ominous side of the same coin: so-called "inert" substances and procedures that bring about harms and side effects, seemingly unbidden.

Placebos are essential to the "curative" ideals of biomedicine, demarcating the very boundaries of what counts as medically efficacious (Kafer 2015; Kim 2017). There are many different cases of placebos, operating in varying domains: domains such as the clinic (where placebos include the "white coat" worn by clinicians), the laboratory (where placebos may be both foil to "real" treatment and object of knowledge in their own right), and the marketplace (where placebos play a central role in the drama of the commercially driven, multi-centre clinical trial). In each of these contexts, placebos mark the other side of legitimate treatment, recognizable as deception (Groll 2011; Touwen and Engberts 2012), contaminant (Cohen 2014), or imagination (Wall 2000, 126). They, along with nocebos, tend not to be recognized as actors in their own right. This is why, as Isabelle Stengers points out, "doctors can still laugh at the placebo effect" (1997, 107).

For as much as placebos are solicited within the practices of biomedicine, however, specifically because this laughable status allows researchers to police the boundaries of what constitutes "real" treatment, so too do they undercut these boundaries. As we argue in this article, placebos model an interactivity between subjects and objects, minds and bodies, individuals and environs—issues of great interest to Feminist Science studies and Disability Studies alike. In what follows, we forge an exchange between these two fields in order to track the traffic of placebos and make the case for the relevance of placebos and Placebo Studies to wide-ranging critiques of the curative imaginary.

But a theoretical difficulty arises as soon as we begin to track the movement and activities of placebos. Just as there are different cases of placebos, there are different ways of understanding and referring to placebos, a situation that complicates the task of assessing the import of placebos and Placebo Studies for feminist and crip theorists. In fact, within Placebo Studies and within the domains of biomedicine, placebo is used to refer to such incommensurate phenomena that it has become necessary to flag the different understandings expressed by the term. Put simply, these differing terms seem to be homonyms: they sound the same but have entirely different referents.

A homonym is a word that, while looking and sounding the same as another, indicates a distinctly other referent. 1 In this article, we draw out the differences between two homonyms at play within the term "placebo." We do so in order to complicate and contest dominant definitions of placebo and nocebo. As Stengers explains, it is vital that we—as feminists—find ways to complicate methods and practices that have successfully imposed themselves "as normal and inevitable" (1997, 150), like, for example, those that subtend curative biomedicine. One way to carry out such complications, Stengers suggests, is by contesting definitions: we should "no longer be led to share the definitions of those with whom one enters into struggle" (1997, 143). By identifying two differing "placebos" as homonyms, we open up a space for such contestation. And, as will become clear, we advocate for one homonym over another, making the case for a definition of "placebo" that resonates promisingly with the insights of Disability Studies and Feminist Science Studies.

We begin our analysis by suggesting that the term "placebo", distinguishable as two homonyms, refers in one case to a noun and in another case to a verb, drawing on the work of Stengers to articulate why this distinction matters. In the sections that follow, we attend to the activities of placebo-as-verb (the second of the two homonyms) in locations where placebos are typically encountered as noun (the first of the two homonyms). More specifically, we scrutinize three sites at which placebos emerge and interact with others: the clinic, the laboratory, the commercial trial. We invite our readers to consider placebos as relational clinical phenomena, as physiological outputs reflective of biosocial dynamics, and as technique or tool for bringing drugs to market, respectively. In each case, placebos warrant the claim by disability theorists "that how we tell our stories matters" (St. Pierre and Peers 2016). It matters literally, as placebos and their effects bear witness to the embodied insights that our own bodyminds carry ("shape carries story," as Rosemarie-Garland Thomson puts it [2007; 2011]). And it matters methodologically, as the ways in which we tell stories about placebos can exacerbate or help to undo some of the harms of curative medicine.

As we track the traffic of placebos, it becomes evident that certain activities are often ignored while others are foregrounded, such that, as we explore below, the verbal placebo passes as a noun while the homonym of placebo-as-verb is hard to recognize and disappears from use. In the clinic, for example, there's a dissonance in which placebos are solicited but also rejected (the word itself tends to be an invective in clinical contexts). In the lab, there's a dissonance whereby the recognition of the effects of context conflicts with methodologies that actively and effectively seek to decontextualize (the attempt to identify the dynamic nature of placebo effects and yet reify placebo responders). And in the market, placebos play entrenched roles in clinical research trials (new treatments become legitimate because of the foil of placebo-controls). The tension between the two homonyms is dramatized in each of these settings and has important implications for both feminist science studies and disability scholarship. We conclude by turning to the counterpart of placebo, namely the harm-inducing activities of nocebo, in order to affirm Eli Clare's recent claims that healing and harm are fully entwined (2017, 5, 48, 70).

In each section, we explore the ontological dynamics of placebos-as-verbs. Instead of the unitary ontology of biomedicine, evinced by the noun placebo, we find that the verbal activities of placebos (and, as we explore in our conclusion, nocebos) point us to a kind of "promiscuous realism" in which ontologies are multiple, relational and interactionist. 2 This understanding that "matter matters" undercuts supposed binaries between "subject" and "object," "mind" and "body," "treatment" and "side effect" and even "object" and "inquiry" (Barad 2012). After all, if the traffic of "cure" includes an ever-present possibility of harm, then the curative imaginary is not restricted to the realm of biomedicine alone but saturates the very design and practices of research itself (Hamraie 2017; Dolmage 2017). It also prompts skepticism towards binaries like "impairment" (or "biological deficit," linked to a pre-social or pre-discursive domain) and "disability" (or "functional limitation," linked to social institutions, mechanisms and relations). By doing so, the verbal activities of placebos and nocebos resonate with emerging work in disability and feminist science studies that emphasize the entangled relations of bodies, identities, institutions and power (see Hughes 2002, 67; Tremain 2017, 9, 87).

But this is no neutral investigation. By examining the already-promiscuous grammatical uses of placebos, as nouns and also verbs (Eyers 2017, 136), we advocate for the verbal usage over that of the noun. (Queer is "a kind of verbing," Melanie Yergeau writes, and this phrasing by Yergeau, which echoes now-canonical work by Judith Butler on the "verbal" activities of gender, suggests that there are ethical and political ramifications to "verbing" dynamics like those of placebos [2018, 18]). And by including nocebos in this discussion, we acknowledge that the vexed line between healing and harm extends to our own relationships, practices and institutional affiliations. We are not apart from placebos and nocebos, in other words, and the encounters in the following sections raise the stakes for how we live out our own investments in meaning and meaning-making.

I. Placebos as Homonyms

In this article, we focus on two ways of deploying and understanding the term "placebo." Since the word itself is identical, we suggest that these two usages can be distinguished if we identify them as homonyms: two placebos, whose differing definitions we track by way of the practices that they enact and make possible. In the first case, "placebo" is defined as a noun that marks a crucial bifurcation at the heart of biomedicine: either it is real (as in treatment), or it is imagined (as in placebo). This placebo performs functions central to the service of the curative imaginary—the constellation of values, symbols, institutions, and practices that organize the orientation of biomedicine toward an imagined disease- and disability-free future (Clare 2017, 26, 70, 76). It legitimates the production of new treatments, a task that is vital for the curative imaginary's orientation towards futurity (Kafer 2013, 27-28; see Dumit 2012, 89). And it tends to be unitary; its usage points, almost always, to the placebo. This ontological premise underscores the curative promise that new cures are always forthcoming, such that "a cough is not a cough but a potential sign of sickness and health trajectory" (Reynolds, 2018). A unitary ontology secures a connection between ailments in the present and treatments in the future. After all, if something is "real," then it will be the same "across time and place without regard for personal history" (Staiano-Ross 2012, 40) precisely because "diagnostic categories are universal" (Adams 2002, 670). Above all, this ontology adheres to a logic that has precision, internal consistency and teleology (Bennett 2017, 95), of which biomedicine in particular holds the tools for surveying and capitalizing upon.

Put otherwise, this placebo enjoys the status of an "object." According to Isabelle Stengers' account of scientific research, an object occupies a special role within the leading science. For those who work within the leading science, an object is a phenomenon that has already testified, as Stengers puts it, which means that its properties or effects have been so well-documented through experimental testimonies that they have yielded a theory that takes the place of its testimonial. Rather than relying upon the witness of an actual object, then, scientists can anticipate in what way the now-theoretical object will continue to testify (1997, 145). Similarly, Bruno Latour describes an object of science in this way: an entity acquires ontological status through various trials, experiments, transformations—in short, performances—that enable the transformation of a non-entity into an entity that is then irrefutably demonstrated to exist. At the same time, the experiments and tools of the scientific apparatus as well as the experimenter(s) (those that made the testimony possible) fade into the background (1999, 116-129).

An object is what authorizes the leading science to judge other sciences as ideological or nonobjective (Stengers 1997, 147). In the case of placebos, defined as noun, the placebo enables the scrutiny and management of the very boundaries of biomedicine, such that whatever exceeds its boundaries is rendered as noise. In this way, the placebo, as object, quickly authorizes biomedical researchers to judge noisy phenomena as "mere placebo," an invective whose pejorative force is the foil to a dependable, evidence-based biomedical treatment. The placebo-as-noun, in other words, underscores the ontological presumption that "either you built it, or else it's real" (Latour 2010, 7). In this bifurcation, to be "built" means to be subjective, imagined, artificial or otherwise constructed; to be "real" means to be objective, verifiable, authentic and effective (Stengers 2011, 18; 38; 275).

We are struck by how accurately Stengers's account of "the object" aligns with the activities of the placebo-as-noun. Consider the objects that Stengers describes as exemplary of the leading science and the consequent judgements that arise from such objects: "If a germ is the truth of a contagious illness, the fact that, exposed to the same germs, some people fall sick and others do not, becomes secondary, and preventive medicine clearly becomes a useful, but not fundamental, research field. If the genetic program is the truth of the living, the essential is the resemblance between a bacteria, an elephant and a man, all genetically programmed. What distinguishes them may certainly be interesting, but it will have to be redefined on the basis of the notion of the genetic program" (1997, 145, italics ours). Let's continue along Stengers's lines, adding "placebo" to her list of exemplary objects.

If a placebo is the truth of efficacious treatment, the fact that placebo effects occur outside and as well as within the bounds of recognized treatment points to a problem of noise, rather than to a competing definition of placebo. The problem of noise requires ongoing and resolute deployment of placebo-controls, so that the efficacy of new treatments can be ascertained. This homonym, then, underscores the theoretical value of the placebo-as-object because it authorizes the very distinction between efficacy and "mere" simulation.

There is a counter strand in Placebo Studies, one that we draw out as a competing homonym because of its affinities with emerging interests in Feminist Science Studies and Disability Studies. This counter strand understands the phenomena of placebos in grammatically divergent terms from its homonym the noun, although this grammatical difference has yet to catch on within Placebo Studies. In Stengers's terms, this homonym marks the emergence of a new scientific proposition, one that asks scientists to accept a phenomenon as a reliable witness, capable of intervening in discussions (1997, 144). The hypothesis of a new proposition can be scandalous to the leading science. It will likely be denounced as ideological or nonobjective "because, if the questions that they pose and the testimonies that they seek were taken seriously, they would bring into question the theoretical object, implying that the resemblance affirmed by the theory is not essential, that some of the phenomena belonging to the field of the theory testify to another type of truth" (1997, 145, italics added). On our assessment of Placebo Studies, this is precisely the challenge that this homonym poses to the placebo-as-noun: it testifies to another type of truth or ontology, one that holds feminist and crip promise, and one that jeopardizes the authority of the theoretical object.

In this second usage of the term, "placebo" is not a noun but a verb. Patrick Wall is representative of this much less prevalent usage, in which, as Wall explains, placebo can be defined as "appropriate action" (2000, 152). Rather than an object that subtends a theory that divides reality from illusion, this placebo undercuts the possibility of bifurcations between objects and subjects, between "placebo" and "effect." Similarly, Laurence Kirmayer makes reference to placebo as a "social situation," as opposed to an object of medical science (2016, 137). Rather than demarcating placebos from their effects, the verb homonym affirms the fact that placebos are always engaged—adulterated, to use Elizabeth Wilson's word (2015, 136)—with the responses they provoke. From this perspective, placebos are verbs, and to claim otherwise is to miss something crucial about their ontological nature: namely that they are a kind of traffic (to use another of Wilson's descriptors [2015, 156]). There is something lively about verbs (and, conversely, something inert about nouns [Roosth 2014, 58]. And there is something relational with respect to this liveliness. As Stengers puts it, glossing Whitehead, "To feel is 'to be affected by,' that is, at the same time, to confer upon what is felt 'there' the power to have an effect 'here'" (2011, 294). To experience a cessation of symptoms, for example, is to confer meaning-making powers upon the placebo, as a verbal activity that draws out such effects. Sound Studies scholars make a similar argument when they argue that "voice" and "sound" should be understood as verbs, rather than as nouns, as a way to foreground the relational and interactionist nature of ontology: "That is, if we reduce and limit the world we inhabit, we reduce and limit ourselves" (Eindsheim 2015, 3; see also Voegelin 2014, 2 and Feld 2015, 14).

This verbal, active placebo has perhaps become more recognizable within Placebo Studies because of a recent shift in the scholarship away from polemics about whether placebos "exist" at all and towards wide-ranging reflections on the activities of placebos. Its recognition by scientists as a phenomenon that bears witness to its own activity, however, requires other conceptual resources than those that prevail within curative biomedicine. This homonym asks researchers to pay attention to singular and local interactions, to investigate rather than to judge (Stengers 1997, 148). It shifts the locus of causation itself, since the status of "cause" moves from efficacious treatments to dynamic and relational activities. And it prompts a more relational approach to research practices, one in which knowing is "always experiential, contextual, fallible, changeable, contingent, emergent, opportune, subjective, constructed, selective" (Feld 2015, 14). The verbal placebo does not fit into the unitary logic of biomedicine, since it reflects tendencies rather than fixed logic. It therefore requires other methods than those honed by biomedicine: as Jane Bennett explains, "A logic can be surveyed from the outside; a tendency must be inhabited to be understood" (2017, 95).

Etymologically, the earliest usage of the term "placebo" is that of a verb, not a noun: it will please (Wall 1996, 162). 3 In later years, this shifted to the more common usage, to please. By listening to the echo of its long-ago etymology in contemporary biomedical practices, we discover the striking fact that, in many cases, placebos are verbs that pass as nouns (Haraway 2003, 18). Moving our attention away from the noun (the it that brings about pleasure) and toward the verb (the acts of pleasure themselves), there are contradictions and dissimulations that become open for recognition when we attend to this other homonym, placebo as verb.

To define placebo as a verb is more than trifling semantics. As Rey Chow explains, naming is a performative gesture, one that establishes a contact zone between methods and matter (2014, 3-4). Placebo, as a noun that preoccupies much biomedical research, is a contact zone that serves the purposes of delineating causal mechanisms and determining efficaciousness of treatments, reinforcing a "real" versus "imaginary" distinction that echoes an objective/subjective binary. 4 In contrast, the placebo as verb is a different contact zone, we argue, one that resonates with feminist and crip critical work that affirms and solicits the interactive and relational aspects of ontological life. This verbal placebo is, constitutively, an active phenomenon; as such, it poses challenges, as well as proffering helpful resources, for addressing the harms and curative ambitions of biomedicine. The contact zone that opens up when verbal placebos are acknowledged is one, we argue, that disrupts biomedical presuppositions: the verb no longer passes so comfortably as a noun.

II. Placebo and Cure: Traffic in the Clinic

"Use the new treatment before it stops working." This statement, shared jokingly and knowingly amongst physicians, discloses a challenge that placebos pose to biomedical treatments. On the one hand, the phrase expresses a positive expectancy—a hopefulness—for the curative promises afforded by biomedical knowledge. On the other hand, it undermines that very belief in the march of progress and signals the presence of more complexity within the evidence base of medicine. As such, it is a phrase that gets to the heart of the curative imaginary, while locating placebos squarely within it. By "curative imaginary," we refer to the interpretative framework, practices, and disciplinary matrix that make up biomedicine and see intervention in the name of cure as the only alternative. This phrase, a locution proposed by Alison Kafer (2013, 27), points us to a fundamental premise of the clinic: if a "cure" does not already exist for a given set of symptoms, then one will surely exist in the future, as disability or illness is not a desirable temporal location (2013, 27; see Clare 2017, 57 and Dumit 2012, 89).

Though the ethics of knowingly employing placebos as a curative strategy have changed in the latter half of the twentieth century, placebos continue to saturate clinical encounters. 5 As one key placebo study's title declares, "diagnosis is treatment" (Brody and Waters 1980, emphasis added): the very act of receiving a prescription from a clinician in a white coat yields, at least sometimes in some patients, marked improvement (Blumhagen 1979). In clinical settings, having therapeutic assurances, a firm diagnosis, or promises of pain relief as part of the verbal context of clinical care have been shown to activate endogenous opioids (endorphins) and produce marked differences in rates of symptomatic recovery (Thomas 1987; Benedetti 2002; Levine et al. 1978). An especially striking early study of placebos within clinical encounters, for example, points to the clinician's very confidence in a proposed treatment alone as a major determinant of beneficial outcome. (To believe that something is "real," in other words, improves its efficacy). Similarly, a subsequent double-"blinded" study of placebo versus fentanyl (a potent opioid) versus naloxone (an opioid antagonist) in the context of post-operative dental pain demonstrated that it was only under conditions where the clinicians believed an opioid could potentially be delivered that a strong placebo analgesic effect was observed. 6 Even with strict "blinding" conditions for randomization, the clinician's belief that no active analgesic would be available resulted in an average increase in pain, whereas the belief that a potent analgesic was a mere possibility significantly improved pain ratings. 7 The very mode by which a clinician inhabits the rituals of treatment, it seems, solicits well-being in patients: "enthusiastic doctors could heal twice as many people as skeptical ones could with the same drugs" (Moerman 2002, 39).

But consider that in the most common understanding of placebos, they are non-potent, strategically-colored sugar pills that contrast with the potency of treatments. Colloquially, the term "placebo" tends to refer to deception, sham, and manipulation. This is why the term can serve as an invective, a way for medical practitioners themselves to clarify the distinction between real treatments and their decoys. 8 To ascertain that a treatment is a placebo, Anne Harrington explains, is to claim that it does not "really" work (2001, 36). The phrase, mere placebo, emphasizes this pejorative usage, where "mere" marks a category divide between inert and active substances. Placebo, in such cases, stands for "mere psychology" (beliefs upheld by a subject), and "biology" stands for measurable objective effects (effects that are recognizable through scientific methods). For as much as the "cure" depends upon the very expectancies and rituals that confer healing, the frame of the clinic simultaneously rejects placebos for their sham qualities and lack of "real" effect.

When one clinician advises another to choose a new treatment over an older one, when a clinician can carry a belief that subtly shifts the clinical interaction so as to produce pain relief in a patient from an "inert" substance, when the act of being handed a highly stylized piece of paper (a prescription) brings about clinical improvement well before any medication is ingested, the very traffic of placebos, their "verbal" activity, is being hailed. Yet at the same time, biomedical rationality rejects this verbal activity, denouncing the reality of placebos as "mere psychology." In Stengers's terms, this is an instance of "judging and silencing" (1997, 150) through the use of the invective. When we recognize these "verbal" activities, however, we open up the possibility of representing placebos as intersubjective and context-dependent activity. And when this alternate placebo is allowed to testify, we can follow this traffic to a fascinating conclusion: the very activity that a biomedical lens would consider to be end-organ or neurophysiological changes located within an individual as the result of an "inert" object (a noun) is, in fact, a complex biosocial phenomenon that is fundamentally relational. 9

Following the traffic in placebos, in turn, signals that there is a temporality to such relationships: placebo effects index the embodied significance of past experiences and meanings. That my symptoms cease to cause me distress, after my visit to the lab-coat-clad physician, testifies to my own traffic through biosocial networks: my body and mind (my bodymind, to use Margaret Price's phrase [2015]) bear witness to what I have learned about the efficacious powers of white coats and MD designations. 10 (Conversely, that the white coat induces panic, heightens pain or increases blood pressure testifies to similar lessons, experienced by my bodymind, about the harmful powers of biomedicine; such effects are known as "nocebo" effects). Such witnessing is at odds with the much more common understanding of placebos as nouns. Nouns are discrete and ontologically bounded; nouns are specific (we refer to the placebo pill) and not relationally porous. Such an interpretative framework renders the traffic of placebos irrelevant at best and nonsensical at worst. Following the traffic in placebos, then, allows us to emphasize the intersubjective materiality and ontological relationality of treatment-encounters.

Moreover, what becomes especially salient in placebo-analysis is the culturally specific aspects of "cures". Red pills, for example, lead to excitement, and blue pills depress (Blackwell et al 1972; deCraen et al 1996; Da et al 2017). Except, as medical anthropologist Daniel Moerman points out, in the case of Italy, where blue pills elicit enthusiasm—likely, Moerman muses, because of the colour of the national soccer team's jerseys (2006, 234). This example might seem superfluous, yet the culturally contingent flavours of placebos tend to evade notice. (As we explore below, clinical research trials follow research protocols that ascribe universality to the very nature of a cure). This point is an important one in the context of Disability Studies in particular because it foregrounds the epistemic overreach of much biomedicine. If our very presumptions about cures hinge upon our own sociocultural travels, then the normative claims of biomedicine can be called out as such: normalizing ideals, rather than treatments that are effective regardless of context. Crip theorists and Disability Studies scholars have long argued for the importance of research that draws attention to the oft-ignored, culturally and historically specific mechanisms and practices that constitute biomedical endeavours as well as designations like "disability" or "disabled." Placebos and their activities offer an especially vibrant site for such reflections.

III. Traffic in the Laboratory

While the contrast between placebos and so-called verum or evidence-based treatments in the clinical setting continue to secure the rhetorical power of "mere placebo" as an insult, there is increasing consensus amongst those conducting research in the field of Placebo Studies that the effects of placebos are (to varying degrees, with differing properties and expressions) real. Following more than half a century of research into the mechanisms of placebo effects in a range of biomedical situations (including analgesia, movement disorders and immunology), there is widespread recognition that placebos generate broad changes in behaviour and physiology, and that these changes are measurable and traceable. As we discuss below, this research largely engages with placebos as nouns, an ontological position that is tied to the stakes of this research: namely, a certain empirical legitimation of placebos as such. Despite this, the studies themselves contain material that demonstrates the verbal activities of placebos.

Laboratory investigations of how placebos work attempt to secure the ontological status of placebos-as-nouns by demonstrating definitively that placebos are "real" insofar as they reliably produce end-organ changes that are observable through an experimental apparatus. As the field has expanded, the scale of study has shifted to ever-smaller terms of reference—moving from psychological theories such as classical conditioning, to pharmacological mechanisms at the end-organ level, and more recently to single neuron outputs. One of the first studies to point to a molecular mechanism underlying the placebo effect in humans demonstrated a reversal of placebo analgesia with the opioid antagonist naloxone; this suggested that endogenous opioids were implicated in at least some aspect of placebo responses to pain (Levine et al. 1987). It also suggested that study subjects "created the biochemical conditions that allowed his or her expectation to come true (Harrington 2016, viii). Neuroimaging studies have now become an indispensable technology for the field. Functional MRI (fMRI) and positron emission tomography (PET), which promise spatial and temporal data at the level of neural networks, are methods of choice for a wide range of placebo studies. Imaging technologies are being utilized in a wide range of studies, from placebo relief of pain, to relief of motor symptoms in Parkinson's disease, to the overlap of placebo and antidepressant responsiveness, each drawing greater attention to signaling pathways and neurotransmitter systems as mechanisms of placebogenic action. 11 Donald Price and colleagues describe a shift in thinking that has accompanied the rise in neuroimaging: from an early focus on the "inert" content of a physical placebo agent to the more contemporary concept of a simulated therapeutic intervention that has neurobiological underpinnings and "actual" effects on the brain and body (Price et al. 2008). Commenting on the role of imaging research (such as functional magnetic resonance imaging or fMRI and positron emission tomography or PET) within Placebo Studies, the authors suggest that brain imaging modalities enable testing of the explicit mechanistic hypotheses surrounding placebo effects at a refined neurobiological level (ibid, emphasis added).

A preoccupation with what is real versus not real echoes throughout the literature: neuroimaging modalities in Placebo Studies, in particular, are meant to answer the question of whether placebos actually influence various physiological processes, and such studies are described as reflecting a central tenet of neurobiology—namely, that "subjective" experiences such as values and expectations have physiological bases. Implicit in this statement is the notion that having a physiological basis thereby renders placebo-related phenomena "objective" (see Benedetti et al. 2005). It is this body of research that attempts to confirm the "reality" of placebos by providing a credible foundation within the disciplinary matrix of biomedicine.

What these diverse methodologies have in common, expressed by their intent to demonstrate the realness of placebo effects, is a tendency to individualize and decontextualize the processes that generate these effects. It is within these transformations that we find a denial of the placebo-as-verb. Rather than taking place in the spaces of the lab, in the handling of the mice, or in the conversations between technician and participant on the way to the scanner, the actions of placebos are localized in the plasma or brains of individual actors (human and non-human), signified by particular end-organ changes. Within Placebo Studies that engage in decontextualizing practices, animal models and human patients are, arguably, "merely parts" (Birke 2012, 164). This is evidenced by the removal of subjects whose placebo responses are too little or too great, in order to standardize a given data set; it is also the case when brain images are taken apart, re-worked, and put back together in an amalgamation, regardless of the differences between research subjects. 12 As Science Studies scholar Lynda Birke writes, research subjects, both human and non-human, are a specific accomplishment of the networks of people working to sustain carefully designed, controlled, and standardized bodies for research (2012). This work is frequently left out of the methodologies, discussions, and the "thicket" (Latour 1987) of references, figures, tables, and legends that are part of data production; the networks of cage-mates, handlers, lab personnel, and other research assistants that situate human as well as animal research subjects do not enter the written record, especially because the recalcitrant actors—those bodies whose experimental parameters are too variable to be reliably included—are cleaned from the data sets. 13 It is through this decontextualization that placebos continue to be identified as nouns in most placebo studies, which treat the placebo as an object that brings about end-organ change, rather than the dynamic process of change itself.

But just as in the scenario of the clinic, where the meanings of the white coat, the medical paraphernalia, the diplomas on the wall, or the touch of a hand are metabolized by the bodymind, when we follow the traffic in placebos within the laboratory, we can likewise appreciate the ways in which placebos are relational activities that leave their traces. Placebo analgesia, for instance, not only generate changes in brain activity that can be mapped on to changes induced by opioid medications themselves—this effectiveness of placebo analgesia depends on how strongly the individual administering the substance believes pain relief is a possibility. As with the clinical setting, the disciplinary matrix of biomedicine makes it difficult to appreciate the verbal (or ontological relationality) placebos in the lab, even as the studies themselves are demonstrative of the verbal situation.

What is at stake in the purifying efforts of placebo research, which seek to decontextualize placebo-action as much as possible within the methodologies and reporting of study results, while at the same time articulating the "realness" of expectancies, conditioning, and context as integral to placebo effects? The reductiveness ultimately undermines what should be understood as relational ontologies in which natureculture, bodymind and biosociality entangle and interact.

IV. Placebo Traffic in the Market

As Wilson demonstrates in Gut Feminism, there is, in fact, no treatment that is not in some way intertwined with placebo. Wilson points out, for example, that "an antidepressant drug is most clearly itself—indeed, can only fully be itself—when it has been adulterated by placebo" (2015, 136). We explore some of the implications of this claim below, but for now, we want to foreground the relational significance of this description: "Any pill is part placebo." 14 This account is an implicit invitation to approach placebos as verbs, not nouns. As a corrective to prevailing methods in biomedicine, Wilson explains that "a relation is not simply the smallest possible unit of analysis, it is the only possible unit of analysis" (2015, 133). 15 By identifying "relation" as the only possible unit of analysis, Wilson is making a formal claim about the very nature of conceptual labour; methods will only get traction if they focus in on the "smallest unit" possible—namely, the most salient relationship at issue for the researcher. In this case, Wilson's analysis focuses on the relational dynamics of drugs with placebos. And this relation is one that both motivates and perplexes placebo researchers.

This relationship—the adulterating dynamic that, in part, enables the very efficacy of drugs as treatments—is one that is, like all relationships, ecologically and contextually specific. Germans, for example, find ulcer treatments much more persuasive than other Europeans (Moerman 2003). Antidepressant pills work, as another example, through what Wilson describes as "biochemical relationality" (2015, 113). The precise nature of a given biochemical relationship is particular and therefore resists certain kinds of generalizations.

In fact, placebo effects are currently on the rise—at least in America (Tuttle et al, 2015). Statistically speaking, ever more patients are benefiting from the ease of symptoms that placebos elicit. Increasing rates of placebo responses, however, undermine the gold-standard methods of research trials in ways that dramatize the verbal activities of placebos. Anti-depressant research, for example, shows that high placebo responders are also those who respond best to treatment: they are asked to be purifying agents (by participating in placebo-controlled trials) but enact the adulterating activities of placebos (that undercut the unitary ontology of biomedicine). At the outset, we identified three areas of interest concerning placebo effects. Having discussed the intersubjective effects of the clinical context and having raised ire with some aspects of laboratory-based studies of placebo effects, one domain remains to be considered—the role of placebos in bringing new pharmaceuticals to market. As "the placebo effect" increasingly drew attention in biomedicine, the very consistency of placebo responses generated an impetus for their inclusion in pharmaceutical research practices. If placebos were employed systematically through the design of research protocols, the information that researchers investigated would be all the more robust—protected from a range of contaminants that might pollute data—and medical research itself all the more reliable.

And so, with the rise of the randomized controlled trial (RCT) in the second half of the twentieth century, placebos have been incorporated in studies of drug efficacy. 16 The placebo effect, in such instances, acts as a purifying control: through its contrast with placebos that mimic their precise appearance and even replicate their actual side effects, new medical treatments emerge out of placebo-controlled research trials, "triumphant," in the words of Stengers, because of their superiority to the placebo (2003).

If placebo effects are "contaminants" that "deceive" with "irrationality," they represent precisely what is most in need of purifying, from the vantage point of medical science. But strikingly, the very method that has become enshrined as gold-standard in biomedical research for achieving such purification solicits the work of placebos. In the words of Haraway, they pass as nouns but evince verbal attributes, activities and relations (2003, 18). Indeed, the import of placebos for clinical research trials involves the creation and solicitation of other supposed nouns as well: the "high placebo responder," for example, is an increasingly common entity in the context of research trials and placebo studies. This locution, high placebo responder, presumes that there is some natural kind of participant in research trials, one who reliably assents to the curative promises of treatments and thereby serves as a template for effective research (Garland-Thomson 2011. Recent excitement about the possibility of a "placebome" evinces similar epistemic confidence in the noun-like quality of placebo effects (Hall et al, 2015).

Despite such hopes, however, researchers cannot predict placebo reactors. Indeed, this marks one of the ongoing dilemmas of pharmaceutical companies and their need for new treatments that can be brought to the market. The very rise of placebo effects in America, for example, necessitates an increasingly global reach of clinical trials (Petryna 2011). Such global trials follow the gold standard design model. They make use of the purifying powers of placebos and placebo controls because of the need to produce evidentiary justification for new treatments (Dumit 2012). The very design of clinical research trials tends to presume a kind of baseline of health, what Mel Y. Chen describes as a "pov of mythic health" (2011, 273). Since there is no such thing as a disembodied or "dis-embedded" body, 17 one that would stand in as the universal template for every human body, the use of such standards or baselines risk reinstantiating the unmarked status of bodies that are able, white, and otherwise compliant with "mythic" ideals.

Conclusion: Nocebos and the Harms of Cure

"Holding it all—sickness and human vulnerability, health and disability, the need for and the rejection of cure—is much harder work than writing anti-cure diatribes. And much more necessary." (Clare 2017, 62)

We would like to conclude this investigation into the "traffic" of placebos by turning to a key insight from Disability Studies: namely, that harm, injury, exclusions and and other forms of violence result from curative biomedicine. This is a complicated insight, of course, given that Disability Studies scholars also affirm and acknowledge the import of biomedical research, treatment and access to healthcare (Kafer 2013, 4; Clare 2017, 45). This seeming contradiction resonates with a striking finding of feminist research into placebos: "There is no absolute distinction between the pill and the world, and between the remedies and the injuries they each enact" (Wilson, 2015 154). Wilson points out that "remedies are always already breached by their capacity to injure" (2015, 146), and this porous line between "healing" and "harm" is perhaps most dramatically enacted by the "verbal" activities of nocebos.

Just as placebos index the relational ontology of specific contexts, so too do nocebos mark the meaning-making activities by which effects come to pass. In the case of nocebos, however, these effects prompt symptoms that are undesirable. Learning that a treatment might have side effects, for example, might well lead to the onset of those side effects—even if that treatment was, in fact, a placebo (Benedetti 2014; Justman 2015). Like placebos, nocebos and their activities dramatize the entanglement of belief and physiology: in the case of medicine's healing and its harming properties, "people need explanations that match their own cultural and cosmological systems" (Adams 2010, 9; see Crum et al 2017). Nocebos bear witness to the relational and interactionist ontologies of harm. When we know that something will hurt us, it very well might hurt us, despite the assessment by biomedicine that this "something" is inert and lacks any recognizably potent ingredients. Nocebos, like placebos, index the powerful articulations of embodied experiences, in which bodyminds act "in defiance of what we know about them" (Greco 2008, 29).

In this way, verbs might pass as nouns, but their activities belie that designation. But while placebos testify to defiance and other significant expressions of embodied life, our analysis has demonstrated how they are persistently called into obedience as explanatory nouns. Likewise, in the first published use of the term "nocebo," the term refers to a quality inherent in the patient themselves (Kennedy 1961). In more recent biomedical publications, "nocebo" tends to refers to "the induction or worsening of symptoms induced by sham or active therapies" (Planès et al 2016). In both cases, the term "nocebo" is used to index the bifurcation between imagined beliefs and objects that are delimitable, ontologically, and that in principle do or will in the future yield to the explanatory prowess of biomedicine. A more relational and interactionist account of nocebos, however, substantiates the wide-ranging critiques by Disability Studies scholars that elucidate the harmful workings of curative biomedicine. The very protocol of distributing a consent form, for example, draws participants into an expectancy of harmful side effects: only one of a tremendous range of harmful effects of clinical research trials (see Jain 2010).

We want to "celebrate" placebos, to adopt a verb used by Stengers, while also, as she puts it, celebrating them "against the bifurcation of nature with which they are associated" (2011, 247). This kind of celebrative learning, Stengers explains, requires us to pay close attention to the constitutive links between phenomena like placebos and the scientists who solicit them "into integral and reliable protagonists in their practices." There are contact zones, in other words, between placebos and those who experience them, study them and employ them in the service of other activities (like the production of new pharmaceuticals). Stengers writes, "Whereas we usually have a choice between looking at the moon while forgetting the finger pointing at it, or looking at the finger and disqualifying what is designated as relevant merely to this designating gesture, the point is to succeed in affirming at the same time both the finger and what this gesture requires and presupposes, and what this gesture gives us to perceive" (2011, 248). Stengers's formulation aligns with what we've discovered through our inquiry into the traffic of placebos: contact zones that promise "cures" are always, at the same time, beset with the promise of harms. The very term "nocebo" was coined, in fact, by biomedical researchers who began to realize that curative interactions prompted harmful side effects, most notable in scenarios when physicians were simply trying "to please" their patients by prescribing them placebos. Our own hope is that the counter strand of Placebo Studies, in which researchers draw out the activities of placebos (and nocebos) as relational, biosocial and dynamic, might become an ally for the work of Disability Studies scholars. The insights we have drawn out here, through this counter strand, dispute the notion of cure proffered by the medical-industrial complex while making space to ease burdens and celebrate well-being.