I

Let us begin with some terminology. The conduct of science takes place, according to social epistemologists, primarily within a system of institutions (universities, research hospitals, commercial research institutions, journals, professional societies, etc.) that has a complex internal organization. I shall call scientific activity performed in one’s capacity as a member of one of these institutions institutional science. Not all scientific activity is so performed. Sometimes science is performed by people who are not members of these institutions or by members of these institutions but not in their capacity as such: This is what I shall call lay science.Footnote ⁴ I take the distinction between lay and institutional science to be purely sociological: What differentiates the two forms of scientific activity is that the latter, but not the former, is performed by professional scientists in their capacity as such. Using a purely sociological demarcation criterion, as opposed to an epistemic one, allows me not to settle in advance the question whether there can be “lay experts” (Epstein Reference Epstein1995; Arskey Reference Arskey1998) who may potentially have an epistemic advantage over institutional scientists in specific domains.Footnote ⁵

The forms of lay science that have been identified in the literature may be divided into two kinds, based on how they relate to the conduct of institutional science. In what I shall call community science (after Kovaka Reference Kovaka2021), laypeople come together to research a particular phenomenon that affects their community.Footnote ⁶ Although institutional scientists are sometimes asked for their input, this form of lay science remains largely separate from the conduct of institutional science. Indeed, in cases of community science as I define it, community scientists aim to produce their own findings, either for direct use by their community or to pressure the authorities into policy changes. For instance, they may seek to evaluate food contamination as a guide for their own consumption (Kimura Reference Kimura2016) or to evaluate the impact of pollution on their health to extract protective policies from regulators (Allen Reference Allen2017). Thus community science takes place in relative disconnection to the complex social mechanism of institutional scientific research, and any links between community science and the conduct of institutional science are tenuous.

By contrast, in what I shall call participatory science,Footnote ⁷ laypeople are invited by institutional scientists to take part in the scientific research they conduct.Footnote ⁸ It takes place on a spectrum of lay involvement: On one end, volunteers perform mere data collection for a professionally designed and run project,Footnote ⁹ and on the other end, lay scientists are “full research collaborators” who “help … to shape research questions, study methods, conclusions drawn, and all the other decisions made along the way” (Potochnik Reference Potochnik2024, 51). But even in the case of “full collaborat[ion],” the emphasis is always on how “members of the participating public” (Dunlap et al. Reference Dunlap, Corris, Jacquart, Biener and Potochnik2021, 287) can “help” (Potochnik Reference Potochnik2024, 51) or be “included” (Potochnik Reference Potochnik2024, 50) or “participate” (Evans and Potochnik Reference Evans, Potochnik and Anderson2023, 13) in the research that “professionally trained researchers” (Dunlap et al. Reference Dunlap, Corris, Jacquart, Biener and Potochnik2021, 287) or “scientists conduct” (Evans and Potochnik Reference Evans, Potochnik and Anderson2023, 12). Participatory science may involve lay researchers to various degrees, but it takes place under the purview of institutional science: The initiative and decision-making roles belong to institutional scientists.

These two forms of nonprofessional science are not jointly exhaustive: People living with HIV/AIDS and IACCs have been involved in scientific activities that do not fall under either of the two kinds surveyed so far. They have independently established the existence and characteristics of their diseases when institutional science remained oblivious to them.Footnote ¹⁰ They have collated, summarized, and spread the subsequent findings of institutional science on their diseases; but much more than that, they have collectively analyzed, criticized, and disputed almost all of its aspects, from conceptual questions (diagnostic criteria, symptom definitions) to clinical trial methodology (eligibility criteria, control type, outcome measures, posology, statistical analysis) to research into disease pathophysiology (theory building, hypothesis generation, research methodology). In response to a scarcity of correctly run trials, they have sourced medications outside formal channels, run underground trials, performed experiments on themselves, and used the knowledge acquired in this way to demand changes to the way institutional science was producing findings on their diseases. In other words, they have left no stone unturned in the collective criticism of what they have considered to be the inadequate research pursued by institutional science as a whole.Footnote ¹¹

These cases cannot be understood as forms of participatory science. As we have seen, participatory scientists are invited to perform some specific set of tasks such that their activities are decided on by institutional scientists. By contrast, the groups of sick people at hand have not been sought out or preemptively invited to participate in institutional-run research but have initiated and run their own projects without institutional oversight. Thus, unlike participatory scientists, their scientific activities cannot be said to fall under the purview of institutional science. But these cases cannot be understood as forms of community science either. For unlike community scientists, whose scientific activities are aimed at producing their own findings, the groups of sick people I am discussing here seek to exert modifications on the way in which institutional scientists produce their findings: They direct their efforts toward altering the course of institutional research. The scientific activities under discussion are thus neither subsumed under institutional science, as participatory science is, nor separate from it, as community science is. A new theoretical model for understanding these activities is therefore needed.

Lucas Dunlap et al. (Reference Dunlap, Corris, Jacquart, Biener and Potochnik2021) argue that understanding lay science requires a sense of why people engage in it. They discuss, for instance, the fun and educational aspects of participatory science for participatory scientists and conclude that one of its benefits might be to increase public trust in science (see also Potochnik Reference Potochnik2024; Potochnik and Jacquart Reference Potochnik and Jacquart2025). The sick people I have been writing about, however, do not engage in their activities primarily because they find them interesting (though many do) but because they desperately want a safe and effective treatment that will allow them to survive and/or regain some quality of life. In a context where they see institutional science as systematically producing errors, such that no safe and effective treatment exists on the horizon, they feel compelled to make institutional science change course: “It is by reaction that organisations of sick people were created” (ACT UP Paris 1994, 95, emphasis added). The kind of science I have been describing, therefore, has two core features. First, it emerges as a response to the failings of institutional science such that these failings constitute the reason for its genesis.Footnote ¹² And second, it is neither subordinate to (like participatory science) nor separate from (like community science) institutional science but characterized by sustained critical engagement with it. By virtue of these features, it can be viewed as a kind of lay counterpart to institutional research and called extitutional science. Footnote ¹³

II

—ACT UP New York, FDA Action Handbook

The epistemic success of extitutional science must be accounted for, and there are immediate difficulties: Extitutional scientists tend to have very little scientific training and no access to the kinds of resources necessary for the production of meaningful scientific findings. The source of their competence must instead reside in their experience of living with the condition at hand, but the mere fact of living with a particular condition cannot possibly give one direct insight into its biomedical aspects. What needs to be developed, then, is an explanation of how the lived experience of extitutional scientists indirectly provides the kind of competence that can propel scientific understanding of the condition forward. We saw that extitutional science has two characteristics: It emerges as a response to the failings of institutional science, and it consists in a form of sustained critical engagement with it. This suggests that the contribution of extitutional scientists is not primarily additive but critical: Their most successful activities consist not in the production of new findings but in the criticism of existing ones. In other words, extitutional science is successful because it is able to detect and correct errors committed by institutional science—or so I argue in this section.

Until the early nineteenth century, it was widely believed that scientific outputs were always true and therefore that science progressed through the mere accumulation of new findings. But developments in the sciences began to reveal that science was fallible: It could produce errors. A new philosophical tradition emerged, most closely associated with Charles Sanders Peirce, that accounted for this fallibility while maintaining that science was epistemically successful: The trick was to argue that, although science produces errors, it can also correct these errors. In other words, scientific progress came to be seen as requiring not only the production of new findings, new theory, and so on but also the correction of any errors that had been introduced. In that sense, science could be seen as self-corrective and therefore as progressing toward the truth: “Inquiry … has the vital power of self-correction and of growth” (Peirce [Reference Peirce1898] 1998, 47).Footnote ¹⁴

One way to parse this claim draws on the work of Helen Longino (Reference Longino1990, Reference Longino2002). Longino (Reference Longino1990, chap. 3) begins from the observation that the interpretation of data and the evaluation of hypotheses based on them require making a number of what she calls background assumptions. Indeed, data can never settle the truth or falsity of a more general hypothesis on its own, and background assumptions are required to bridge this inductive gap. These background assumptions can be empirical, conceptual, metaphysical, or even political: They are “substantive and methodological hypotheses that … form the framework or proximate intellectual context within which inquiry is pursued” (Longino Reference Longino2002, 127). Longino then argues that different scientists are liable to make different background assumptions, therefore conducting scientific inquiry in different ways, and even drawing different conclusions based on the same data. Progress in science therefore requires not only the collection of new data but also a “collective give-and-take of critical discussion” (Longino Reference Longino1990, 79) that enables the identification and correction of errors in background assumptions. What she calls the “transformative criticism” (Longino Reference Longino1990, 76) of background assumptions therefore has a central role to play in scientific progress, for the “conflict and integration of a variety of points of view” (69) comes to increase the epistemic quality of the outputs of science.Footnote ¹⁵

Scientific communities in which background assumptions are routinely subjected to thorough criticism, and which are responsive to this criticism, are thereby more self-corrective and yield epistemically better outputs. This, I submit, is how we should understand the primary contribution of extitutional scientists: It is by providing extensive and profound transformative criticism on the background assumptions made by institutional scientists that they have improved the scientific understanding of the pathophysiology of their disease and enabled the discovery of effective treatments. This, to be sure, is a bona fide scientific activity. Evaluating the conceptual, methodological, and other background assumptions against which data are rendered meaningful is a core part of what enables the production and correction of scientific results. Extitutional scientists are indeed doing science. But if they propel science forward by providing this kind of transformative criticism, it must be explained how they are able to notice errors committed by institutional scientists and make proposals for correcting them. It must be explained, that is, where their scientific competence springs from.

Standpoint theory (Hartsock Reference Hartsock, Harding and Hintikka1983; Harding Reference Harding1986, Reference Harding1991; Collins Reference Collins1986; Haraway Reference Haraway1988) can be read as providing this kind of explanation. According to feminist standpoint theorists, the sexist ideology that pervades our culture impacts the background assumptions that institutional scientists adopt. For instance, misogynistic assumptions about the supposed passivity and weakness of women led scientists to assume that the egg plays no active role in the process of fertilization such that they focused their attention solely on the role of the sperm and produced erroneous theories of fertilization (Martin Reference Martin1991). Because it is so pervasive, people of all genders tend to absorb some measure of sexist ideology and are thus susceptible to this kind of error. However, those targeted by sexist ideology may have an inchoate sense that this ideology does not accurately describe their own lives and, for this reason, may be motivated to form social movements through which it becomes possible to collectively pierce through sexist ideology and produce what is known as a feminist standpoint: an integrated and comprehensive understanding of gender and gendered oppression. The production and adoption of a feminist standpoint allowed feminist biologists to recognize sexist ideology where it appeared and therefore to spot the errors that it had engendered in the practice of science. They were able to see that extant theories of fertilization were erroneous by recognizing the sexist background assumptions that led to them for what they were and to produce nonsexist, epistemically superior accounts of fertilization that identified and described the active role that the ovum plays in the process.

This kind of mechanism is at play in our illustrative cases. For instance, homophobic assumptions led some scientists to deny that HIV is the cause of AIDS, defending instead the so-called immune overload hypothesis, according to which AIDS is caused by a “gay lifestyle”: People with AIDS have supposedly subjected their bodies to so many stressors, in the form of non-HIV sexually transmitted infections and use of recreational drugs, that their immune systems have simply given up (Epstein Reference Epstein1996, chap. 1). Similarly, ableist and misogynistic assumptions continue to cause some institutional scientists to argue that IACCs are essentially forms of hysteria: According to them, IACCs are perpetuated, not by organic disease, but by “maladaptive cognitions and behaviours” caused by “psychosocial factors” and the unconscious desire for the “secondary gains” of illness (Geraghty et al. Reference Geraghty, Jason, Sunnquist, Tuller, Blease and Adeniji2019; Hsu Reference Hsu2024). In these cases, it is the lived experience, not of the disease itself, but of the social meanings and forms of oppression and marginalization associated with the disease that provides the grounds for people to see through these erroneous political background assumptions. Because people with HIV and people with IACCs have formed close-knit communities in which their political situation is collectively analyzed, it is clear how they might have communally produced a standpoint from which these ideological posits can be seen for what they are.

However, the transformative criticism that extitutional scientists have leveled at institutional science has not been confined to these fringe, if dangerously influential, ideas baked in active bigotry. Instead, much of their criticism has been targeted at institutional scientists, who, on the face of it, are not particularly prone to ideological assumptions and are actively working to solve the health crises at hand. The background assumptions they have criticized have, on the whole, not been of an obviously political nature: They have been methodological and conceptual assumptions that one would tend to classify as epistemic. Because, on standpoint theory as it was developed by feminists in the 1980s and 1990s, it is ideological distortions that introduce error, and the shared experience of the associated form of oppression that allows piercing through it, this classic form of standpoint theory cannot explain the epistemic advantage of extitutional scientists. But a looser form of standpoint theory can perhaps be articulated that retains the idea that those theorized about can collectively detect error in their theorization, on the basis of their lived experience.

III

Extitutional science, I have argued, propels scientific understanding forward by correcting the (nonideological) errors of institutional science. This raises the question of how extitutional scientists are able to detect these errors. To answer this question, let us examine two examples of extitutional science:Footnote ¹⁶

1. 1. HIV/AIDS. After HIV was discovered to be the cause of AIDS, institutional scientists focused heavily on slowing and reversing HIV-induced immunodeficiency, which they saw as the upstream and therefore most important aspect of the disease. It was widely understood among institutional scientists that progress on antivirals would be met with widespread acclaim.Footnote ¹⁷ In their early trials for HIV antiretrovirals, institutional scientists borrowed methodological background assumptions from other medical fields, according to which trial data could not be “tainted” by concurrent medication use: Any data obtained in this way would count as “impure” and not be taken to reliably inform the trial hypothesis. In both the intervention and the control arms, participants were therefore required to stop any and all medication. This included all medications for the control of opportunistic infections.

   Extitutional scientists argued against these methodological assumptions. They were acutely aware that a narrow focus on immunodeficiency at the expense of its downstream effects in the form of opportunistic infections and cancers was inadequate: “To die of AIDS is always to die of an [opportunistic disease]…. It is opportunistic diseases which kill people with AIDS” (ACT UP Paris 1994, 156). Not only did trials that forbade the concurrent use of medication risk hastening the deaths of those enrolled in the trial (an ethical consideration) but they would also fail to inform the hypotheses that needed evaluating (an epistemic consideration): These trials would produce results that would not be applicable to a population that made extensive use of these medications and would thus be unable to inform AIDS health care. Although institutional scientists initially resisted this criticism, they eventually changed their ways. (For more details on this example, see Epstein Reference Epstein1996, 253–56.)

2. 2. IACCs. All people with ME and many people with long COVID have a markedly reduced capacity to perform a range of activities.Footnote ¹⁸ On the milder end of the spectrum, people are unable to exercise; on its most severe end, they are fully bedbound, unable to tolerate light and sound, and need to be tube fed. Many institutional scientists designing clinical trials for these conditions have parsed this functional impairment in the familiar-to-them terms of “fatigue” and assumed that interventions would be effective to the extent that they reduce fatigue. Accordingly, the outcome measures (ways of evaluating the impact of an intervention) of some of the most prominent trials for IACCs have been assessments of fatigue intensity.Footnote ¹⁹

   The sick people at hand have argued against these conceptual-methodological assumptions. They have pointed out that the symptoms underpinning their functional impairment go far beyond fatigue and also include pain, orthostatic intolerance, cognitive dysfunction, and more (see, e.g., the #NotJustFatigue campaign). More importantly, they have insisted that the primary factor in their functional impairment is not the intensity of their symptoms but the looming threat of what is known as postexertional malaise or PEM: an exertion-induced, often-delayed, dangerous, and debilitating pathophysiological state during which their functional capacity is markedly reduced and their symptom load is significantly increased. Because PEM carries the risk of temporary or permanent disease progression, it is essential that people refrain from whatever level of activity induces it (see, e.g., the #StopRestPace campaign).

   Accordingly, extitutional scientists working on IACCs have argued that clinical trials targeting functional impairment in people with PEM that use fatigue scales as outcome measures do not accurately capture disease-induced functional impairment. As a result, these outcome measures may fail to accurately identify intervention effects and produce false negatives (Goxhaj et al. Reference Goxhaj, McCorkell, Femke van Rhijn-Brouwer, Vogel and de Canson2025). Instead, extitutional scientists have been promoting the use of functional capacity questionnaires, such as the FUNCAP (Vogel et al. Reference Vogel, Pollack, Spier, McCorkell, Jaudon, Fitzgerald, Davis and Cohen2024; Goxhaj et al. Reference Goxhaj, McCorkell, Femke van Rhijn-Brouwer, Vogel and de Canson2025). The FUNCAP, which was co-produced by nonsick and sick researchers with feedback from thousands of sick people, evaluates not symptom intensity but the postexertional consequences of performing a range of activities (Sommerfelt et al. Reference Sommerfelt, Schei, Seton and Carding2024). To date, numerous trials continue to use fatigue scales as outcome measures,Footnote ²⁰ and the only trial for an IACC that uses the FUNCAP is sponsored by the Open Medicine Foundation, a nonprofit funded largely by sick people and their kith and kin (NCT06366724; Meadows et al. Reference Meadows, Squires, Dibble, Palwayi, Felsenstein, Tannenbaum, Li, Xiao, Bergquist and Systrom2025).

These two examples share a common structure: Institutional scientists made erroneous conceptual and/or methodological background assumptions, extitutional scientists detected these errors and argued for their correction, and institutional scientists resisted uptake of this criticism.Footnote ²¹ In this section, we examine what enabled extitutional scientists to detect errors; in the next, we turn to institutional scientists’ reaction to this criticism.

In both examples, institutional scientists new to so-far immature fields adopted erroneous conceptual and/or methodological background assumptions. In the HIV/AIDS example, this is the methodological assumption that trial results are reliable only if they disallow the concurrent use of other medications. In the IACC example, it is the conceptual assumption that functional impairment must be parsed in terms of fatigue—and the attendant methodological assumption that fatigue scales are an adequate outcome measure. In both our examples, extitutional scientists were able to notice and correct these errors, despite their lack of training and resources. This requires an explanation, and ours shall be two-stepped. Identifying an error in the conduct of institutional science requires two things: It requires knowing precisely how such science is conducted, and it requires independent knowledge of the phenomenon against which to spot error therein.

Let us begin with the first requirement: that extitutional scientists know how institutional scientists conduct their research. We have seen that extitutional scientists are invested in their scientific activities because they seek an effective treatment for the condition that brought their community together. When an effective treatment fails to materialize, they have an incentive to inquire into why, and thus to investigate in detail how institutional scientists are conducting their activities. Moreover, institutional science is governed, inter alia, by the communist norm, according to which scientists must widely share the details of their work (Merton Reference Merton1942; Heesen Reference Heesen2017). Now, understanding the relevant details of scientific work requires training. Communities of sick people therefore often make up what McKenna (Reference McKenna2025) calls “research collectives” whose activities include educating one another and the community at large about scientific issues, with the aim to provide everyone with the capacity to understand the crucial aspects of scientific research being done on their condition.Footnote ²² To summarize, the failure of institutional science to produce the desired outcomes provides extitutional scientists with an incentive to inspect the details of its conduct, its communist norm ensures that these details are available to extitutional scientists, and the internal organization of extitutional science ensures that the details are widely understood therein. It is easy therefore to see how extitutional scientists would have detailed knowledge of the conduct of science.

This brings us to the second requirement: that extitutional scientists have independent knowledge of the phenomenon against which to spot errors. Extitutional scientists are part of communities that have produced a collective understanding of the disease’s clinical picture (what it presents as, how it evolves, what treatments it’s responsive to and how) that is the result of the structured amalgamation of ample and detailed evidence. This understanding is constructed from the ground up: The “research collectives” (McKenna Reference McKenna2025) that sick people make up have a more or less formally structured internal organization, whereby individuals share aspects of their disease experience and which enable the “clashing and meshing,” in Longino’s (Reference Longino1990, 69) terms, of these various experiences. Through the activities of research collectives, a picture of the clinical features of the disease emerges. Because there is significant variation in individual experiences of disease, this kind of collective triangulation is essential. On one hand, it allows noticing particular aspects of the disease that might otherwise have eluded perceptual attention: It is common for the gathering of people with similar inchoate experiences to facilitate or even enable the articulation of these experiences.Footnote ²³ On the other hand, it enables the generation of a general picture of the condition—of what exists and what doesn’t, of what is systematic, what is common, what is rare, and so on—which requires many data points.

Thus, in virtue of their incentives, experience, and collective organization, extitutional scientists have two characteristics: They have a detailed grasp of the conduct of science, and they have robust knowledge of the condition’s clinical features against which to spot error therein. I contend that these two characteristics endow extitutional scientists with the ability to correct the errors of institutional science and thereby propel science forward. Note that this ability does not require extensive formal biomedical training or resources.Footnote ²⁴ Its first step, understanding the way in which science is being conducted, does require a sufficient grasp of research methodology and some disease-specific scientific knowledge, but this can be achieved by any research collective with an incentive to make out why institutional science is not making sufficient progress. And its second step, detecting the errors of institutional science, requires a robust understanding of the condition as it impacts the lives of those who have it. This is made possible, not by formal study, but by the collective organization of sick people.

Thus, in one sense, the capacity of extitutional scientists to detect error resembles the ability of political collectives to pierce through ideological formations, as theorized by standpoint epistemologists. For in both cases, it is the collective articulation of a group’s experience that reveals errors in the way this experience has been theorized by others. But in another sense, extitutional scientists’ activities differ from ideology-busting activities. For as we have seen, the errors of institutional science that extitutional scientists are able to detect are not particularly ideological: They are conceptual or methodological errors that one could only class as epistemic. We thus have a mechanism, related to but distinct from that theorized by standpoint epistemologists, that explains the capacity of extitutional scientists to detect the errors of institutional science.

IV

If institutional science is liable to commit critical errors, and if lay scientists have the capacity to detect them and emit proposals for their correction, the epistemic success of science will be enhanced to the extent that the criticisms of lay scientists are taken up by institutional scientists. An obvious reaction would be to argue for an increase in participatory science: If lay scientists were invited to participate, as research collaborators, in the design of studies including clinical trials, the errors committed by institutional scientists would be overcome, and the findings from these studies would be more reliable. In this way, participatory science would provide an epistemic benefit to institutional science: It would allow institutional science to produce more accurate and useful results. The fact that participatory science can procure epistemic benefits to institutional scientists is not widely recognized in the philosophy of science literature. For example, Angela Potochnik and Melissa Jacquart (Reference Potochnik and Jacquart2025, 10) list five rationales for engaging in participatory science (increasing scientific understanding among laypeople, improving the way science relates to personal and social identities, increasing lay trust in science, improving access to science, and demonstrating science’s value to laypeople), none of which amounts to an epistemic benefit for institutional scientists. Those who do discuss the epistemic benefits of participatory science for institutional science tend to focus on the benefits to institutional scientists of lay-performed data collection (Elliott and Rosenberg Reference Elliott and Rosenberg2019; Bedessem and Ruphy Reference Bedessem and Ruphy2020).Footnote ²⁵ Alison Wylie (Reference Wylie, Padovani, Richardson and Tsou2015) is an exception: She argues that scientific collaborations between institutional and Indigenous archaeologists have yielded many epistemic benefits to the former, “not only adding useful detail but generating new questions and forms of knowledge” (192; see, for discussion, Koskinen Reference Koskinen2023).

It is much more common in the medical field to recognize that participatory science can yield major improvements in the outputs of institutional science (see, e.g., Sacristán et al. Reference Sacristán, Aguarón, Cristina Avendaño-Solá, Carrión, Gutiérrez, Kroes and Flores2016). Yet participatory science has either not existed or not sufficed to ensure the uptake of lay criticism in early HIV/AIDS research or in ongoing IACC research. This can be gleaned from the very existence of extitutional science: Had institutional science organized the correction of its own errors by inviting lay scientists at scale and taking up their insights, sick people would not have been spurred to attempt to correct them from the outside. In fact, both examples presented herein reveal that early HIV/AIDS researchers and IACC researchers have also been reluctant to incorporate the criticisms leveled at their conduct by extitutional scientists. So, not only has institutional science failed to seek out the insights of lay scientists but it has actively resisted taking them up when they were presented to it. Although isolated instances of spontaneous course correction in response to requests by lay scientists with interactional expertise have occurred,Footnote ²⁶ these have been exceptions. This reveals a reluctance, on the part of institutional scientists, to take up lay criticism.

Because the success of science depends on the correction of its errors, this reluctance to take up lay criticism impedes scientific progress and must be overcome. Increasing participatory science will not suffice to secure the uptake of criticism, because participatory scientists are by definition subordinated to institutional scientists: It is characteristic of participatory science that participatory scientists can offer their opinions but have no decision-making power. If institutional scientists want to ignore the criticism of lay scientists, as many manifestly do, a mere increase in participatory science will not succeed in ensuring responsiveness to it. This reveals that lay scientists must be able to interfere with the conduct of science: They must be able to exert at least some degree of control on its activities or, in other words, to have an impact thereon that is not mediated by institutional scientists. In accordance with this analysis, in response to the manifest lack of interest in their criticism, HIV/AIDS activists proclaimed they had a right to “ingérence”: interference, intrusion, or immixture in the conduct of science (ACT UP Paris 1994, 66); as we shall see, their success in propelling science forward can be attributed to their exercising this right.

Steven Epstein (Reference Epstein1995, Reference Epstein1996) explains the initial reluctance of institutional scientists to take up extitutional criticism in the case of HIV/AIDS by appealing to the fact that institutional scientists initially did not recognize extitutional scientists as experts. It is true that the question of who had expertise loomed large in early struggles over HIV/AIDS science; Robert Gallo, an institutional scientist who played a major role in the discovery that AIDS is caused by HIV infection, initially viewed AIDS activists as incompetent: “I don’t care if you call yourself ACT UP, ACT OUT, or ACT DOWN, you definitely don’t have a scientific understanding of things” (cited in Epstein Reference Epstein1996, 116). In reaction to this kind of attitude, Mark Harrington, a prominent treatment activist, argued, “We have to break down the cult of the expert in every area of this society. People with AIDS are the experts in this disease” (as cited in Hubbard Reference Hubbard2012, 49:50). The failure on the part of institutional scientists to recognize the expertise of extitutional scientists stemmed, according to Epstein (Reference Epstein1995, Reference Epstein1996), from the fact that extitutional scientists did not speak the “language” of science. As a result, Epstein (Reference Epstein1995, 417) claims, “the most crucial avenue pursued by treatment activists in the construction of their scientific credibility has been … learning the language and culture of medical science.” In the same lineage, Harry Collins and Robert Evans (Reference Collins and Evans2002, Reference Collins and Evans2009) argue that the construction of “interactional expertise” (roughly, their capacity to interact with institutional scientists) is what eventually enabled the recognition of the extitutional scientists’ expertise.

If this explanation were complete, the success of the HIV/AIDS movement would be attributable in its entirety to the fact that some extitutional scientists did become fluent in the language of science, and the general problem of how to ensure the responsiveness of institutional science to extitutional criticism would be solved by enhancing extitutional scientists’ interactional expertise. But as Collins and Evans (Reference Collins and Evans2002, 262) themselves note, there is an air of “naïveté” to the suggestion that the development of interactional expertise is sufficient to provoke a change in the behavior of institutional scientists. In line with this, accounts of HIV/AIDS activism emerging from the collective analysis of AIDS activists (Schulman Reference Schulman2021) have attributed its success to what is known as the inside/outside strategy: one arm of the movement, made up of extitutional scientists, was collaborating with institutional scientists, “working with them, cajoling them, proposing solutions, having intimate interactions, and creating identification” (Schulman Reference Schulman2021, 86), and another arm was exerting pressure through antagonistic protest and direct action. The aim of this outside pressure was to force institutional scientists (and other targets of the movement, such as government officials) to absorb and respond to the criticism that extitutional scientists were leveling at them. As Schulman explains, “It was the movement’s analysis that the reason change didn’t happen was that there was no pressure on people in power to make change. So the strategy was to bring things out into the open so that people would be horrified, and so that the people in power would be embarrassed and would make change out of that embarrassment” (122).Footnote ²⁷ In other words, the inside/outside strategy was successful because the inside arm could emit criticism, and the outside arm could enforce its uptake.

Let us summarize. Extitutional scientists have the capacity to detect errors in the conduct of institutional science and to make proposals for their correction. In this way, they have the capacity to propel scientific understanding forward. Yet their criticism is not spontaneously taken up. The progress of science thus depends on the possibility of interfering with the conduct of institutional science to secure its uptake of extitutional criticism. The HIV/AIDS movement achieved this by exerting significant and sustained pressure on institutional scientists, making the continuation of error more costly than its correction. This suggests that, if we want to ensure the responsiveness of institutional science to extitutional criticism, and thereby enable scientific progress, we would do well to find a way to make refusing to engage with extitutional scientists more costly than the opposite.

V

—Gregg Bordowitz, ACT UP New York

Since Merton (Reference Merton1973), it has been well understood that institutional science is a credit economy. Institutional scientists are motivated by the prospect of receiving grants, being published in prestigious journals, and other markers of recognition, and their activities are oriented toward this quest for credit. This suggests that, if we want to make refusing to engage with extitutional scientists more costly than its opposite, we could intervene on the credit economy in such a way as to reduce the amount of credit that scientific work that fails to take on extitutional criticism receives. In this section, I make a proposal for one way in which this could be achieved. I do not purport to provide any more than a sketch, and further work is obviously required to work out the proposal in detail. Nor do I claim that my proposal is the only possible solution to institutional science’s failure to take up extitutional criticism, though it does have the benefit of chiming with the demands of extitutional scientists themselves. My aim here is simply to show that there exists a possible way to secure the interference of extitutional scientists within the current institutions of science and that it does not immediately fall victim to major objections.

I propose to call on extitutional scientists to serve as reviewers for scientific publication and grant/funding awards.Footnote ²⁸ This proposal has precedents: The use of lay scientists as reviewers has been suggested by extitutional scientists (Fitzgerald et al. Reference Fitzgerald, Cohen and Jaudon2024) and philosophers (Santana Reference Santana2022)—in the latter case, only for the review of grant applications. It is also an existing practice, for instance, at the British Medical Journal, where it is known as “patient and public review.” My proposal would have this nascent practice become systematic where extitutional science exists.Footnote ²⁹ It would give extitutional scientists significant power to prevent a particular finding from being published, project from being funded, or clinical trial from taking place. In this way, it would create a strong incentive for institutional scientists to ensure that their research does not contain the major errors that extitutional scientists have highlighted, which in turn would promote the engagement of institutional scientists with the work of extitutional scientists, including by inviting them to collaborate as participatory scientists on their research projects.

The proposal to grant extitutional scientists a right of interference may be thought to pose certain epistemic risks (Biddle and Kukla Reference Biddle, Kukla, Elliott and Richards2017). Participatory science, despite, unlike extitutional science, proceeding “under the supervision of professional scientists” (Bedessem and Ruphy Reference Bedessem and Ruphy2020, 633) such that institutional scientists have complete control on what participatory contributions influence the research project, has already been argued to carry risks of introducing errors or weakening the error-correcting mechanisms of institutional science, thereby making science less objective (Elliott and Rosenberg Reference Elliott and Rosenberg2019; Bedessem and Ruphy Reference Bedessem and Ruphy2020; Koskinen Reference Koskinen2023, drawing on Koskinen’s Reference Koskinen2020 account of objectivity). The intensity of these risks would soar under my proposal, because the power dynamics are inverted: On my proposal, it is lay scientists who exert control over the outputs of institutional scientists, rather than the other way around. In the remainder of this article, I survey a number of potential risks and argue that they are minimal:

1. 1. Risk of granting interference rights to science deniers. Lay groups that do not share the standards of institutional science, such as some antivaccination activists, would likely be interested in gaining the right to interference that I propose to grant extitutional scientists. These groups may in some cases have lived experience of the condition for which they are advocating (e.g., AIDS denialism in ACT UP San Francisco). But extitutional scientists are defined by their sustained and precise critical engagement with institutional science. This critical engagement presupposes shared standards: Extitutional scientists aim to show that institutional science falls short of its own aims (e.g., producing an effective treatment) because of errors that itself can and should recognize. This is profoundly different from the activities of antivaccination activists, and from those of AIDS denialist HIV/AIDS activists, who by and large fail to engage comprehensively with institutional science and do not share its standards. I propose to grant interference rights only to extitutional scientists.

2. 2. Risk of granting interference rights to unrepresentative laypeople. Lay scientists may fail to represent the interests of the community for which they advocate. Those who have garnered scientific credibility from institutional scientists tend to be less marginalized, and may fail to represent the interests of the more marginalized within the group—a phenomenon known as elite capture (Táíwò Reference Táíwò2022). For instance, there was criticism from the floor of ACT UP New York that those who communicated with institutional scientists did not take seriously enough the issues specific to people of color, drug users, and women (Schulman Reference Schulman2021). But if the competence of extitutional scientists stems from their being embedded in communities where a collective process takes place to distill and collate myriad personal experiences into a comprehensive account of the condition’s presentation, laypeople who fail to live up to this standard will not count as extitutional scientists and should therefore not be granted interference rights.

These two risks highlight the importance of accurately identifying extitutional scientists. To detect and correct the errors of institutional science, extitutional scientists must closely and comprehensively engage with institutional science and be responsive to the concerns of the community as a whole. If an individual fails to meet either of these criteria, they do not count as an extitutional scientist under my definition and therefore should not be granted interference rights under my proposal. Applying my proposal thus comes with a practical challenge: It requires editors, funders, and state agencies, as well as institutional scientists, to find bona fide extitutional scientists to serve as reviewers. Note first that this does not require tracking down extitutional scientists among the undifferentiated mass of the public: Because, by definition, extitutional scientists seek to correct the conduct of institutional science, they must make themselves known to institutional scientists (either individually or, more commonly, through organizations like ACT UP or the Patient-Led Research Collaborative) to achieve their goals. Finding extitutional reviewers thus only requires determining, of any candidate lay reviewer, whether they are an extitutional scientist. Initially, this can proceed by trial and error. If a review betrays a lack of competent engagement with institutional science, editors/funders should dismiss the review and commission a new one. If a review promotes the interests of a greater number of groups concerned by the piece of research at hand, editors/funders should prefer this review over others and may want to commission this reviewer again. Eventually, some individuals and collectives of extitutional scientists become well established as such, and editors/funders can rely on their recommendations to find further reviewer candidates. Thus the reviewer-vetting process becomes more reliable as it unfolds.

But, even if extitutional scientists are accurately identified, some epistemic risks may remain with using them as reviewers:

1. 3. Risk that extitutional scientists trespass beyond their domain of expertise. Extitutional scientists, I have argued, are able to detect errors based on collective experience-based knowledge. This knowledge has limits: Some specific technoscientific points, for instance, may lie beyond its scope. Accordingly, it is likely that there are aspects of research papers and proposals that extitutional scientists are not competent to review.Footnote ³⁰ It is unlikely that extitutional scientists would seek to have these aspects of the research amended, for it is unclear what reason they could have to do that. But if they did, editors could easily detect it by looking at the criticism provided in the review: If the criticism is not ultimately grounded in collective experience, there is no reason to take it seriously enough to compel uptake. This is not to say that such comments are to be automatically dismissed, however, because many extitutional scientists develop serious competence about nonexperiential aspects of the topic at hand. But because extitutional scientists do not have an epistemic privilege over institutional scientists when it comes to these nonexperiential aspects, this criticism cannot have the same weight, and its uptake need not be enforced.

2. 4. Risk that the activist values of extitutional scientists unduly influence science. Extitutional scientists are driven by the value of ending the crisis that brought their communities together. Because this activist value is clearly aligned with the goals of science, it is not an inherent risk. Individual extitutional scientists may also have personal values, in the same way that institutional scientists, who may also be activists (Hauswald Reference Hauswald2021), do (Longino Reference Longino1990; Douglas Reference Douglas2000; Solomon Reference Solomon2001). These can be prevented from leading science astray through the same mechanism that applies to institutional peer reviewers: Scientists hold a diversity or plurality of values and, through a “collective give-and-take,” edit out inadequate values (Longino Reference Longino1990; Solomon Reference Solomon2001).

This concludes my survey of the epistemic risks associated with granting extitutional scientists the power to review for publication and funding allocation.

VI

We are now in a place to conclude. It is widely thought that laypeople may interfere only in nonepistemic aspects of science: in agenda setting, in decisions regarding applications of scientific findings, and, if one believes that nonepistemic values are involved in the conduct of scientific research itself, in decisions regarding these values. Yet, I have argued, the interference of some lay scientists in the specifically epistemic aspects of the conduct of science can be crucial to scientific progress. I came to this conclusion by establishing the following claims. (1) In response to epistemic errors on the part of institutional science, laypeople with an interest in their correction sometimes come together to attempt to have institutional science correct these errors: This is called extitutional science. (2) Extitutional scientists are able to detect errors and make proposals for their correction in virtue of their collective experience-based knowledge. (3) This criticism is not spontaneously taken up, which shows that the uptake of extitutional criticism is currently not secured by the norms and reward system of science. Finally, (4) changing the reward system of science by using extitutional scientists as reviewers is one possible way to incentivize the uptake of extitutional criticism and increase the epistemic success of science.

This argument was made in reference to two specific cases, both medical. This raises a question about the scope of extitutional science: Is extitutional science restricted to the medical field, or can it also occur in other sciences? Extitutional science requires those who engage in it to have a direct interest in the correction of erroneous background assumptions and to be embedded in communities of people or “research collectives” (McKenna 2025) whose collective experience stands at odds with these background assumptions. These two requirements are easily met in the biomedical sciences, hence our two case studies. And although they presumably cannot be met in nonapplied sciences like theoretical physics, they can be met in nonmedical fields with a direct application to human life, such as environmental science. Consider, for instance, the struggle waged by the Black inhabitants of Diamond, an unincorporated community located next to a Shell plant in Louisiana (Ottinger Reference Ottinger2010). Regulators had posited that only ongoing, low-level, long-term exposures to toxic chemicals could affect human health. Accordingly, only the long-term concentration of these chemicals was monitored. But residents also connected short-term spikes in air pollution levels to health events, such as shortness of breath and eye irritation. They began taking their own air quality measurements and argued that the air was unsafe based on these. Gwen Ottinger explains that, in so doing, they challenged the background assumption made by institutional scientists to the effect that short-term spikes in pollution are irrelevant to human health. In other words, their collective experience allowed them to detect and emit proposals for the correction of errors made by institutional scientists: They were doing extitutional science.Footnote ³¹ It follows that extitutional science is possible outside of the biomedical sciences.

When extitutional science does arise, what can we hope for? Remember that the existence of extitutional science points to the presence of error in institutional science, for it is in response to error that extitutional science arises. As such, the very existence of extitutional science can be used to gauge the success of institutional science. This suggests a horizon for extitutional science. We have seen that the uptake of extitutional criticism corrects institutional errors and furthers the progress of science. If this uptake could be secured, through my proposal to use extitutional scientists as reviewers or otherwise, the raison d’être of extitutional science would disappear, for institutional science would cease to commit errors, and the need for extitutional science to correct them would be dispelled. Thus the goal that extitutional scientists pursue is the abolition of their own activity: not its disregard or repression (clearly) but a change in underlying circumstances that would render it obsolete. In fact, now that HIV/AIDS research proceeds with broadly correct background assumptions, extitutional scientific activity in the field has decreased significantly.Footnote ³² The pursuit of epistemic success in science should lead us to compel institutional science to take extitutional criticism so seriously that extitutional science disappears.