1. Introduction
Constitutivists do not usually discuss disability in detail.Footnote 1 I think they should. I argue that lessons from the philosophy and history of disability expose an epistemic challenge to an important type of constitutivist view: “nature-first” constitutivism. The challenge concerns nature-first constitutivists’ claims about the nature of agency, or (for neo-Aristotelians) of specifically human agency; and their associated claims about excellence and defect, flourishing and lack thereof.
I start by explaining, in Section 2, what nature-first constitutivism is, and which familiar types of views count as instances of it. In Section 3, I examine some culturally prevalent negative assumptions about disability—that disability is bad for the person who has it, a deviation from health, and a regrettable defect—assumptions that philosophers of disability have effectively critiqued, and that some moments in the history of disability invite us to see as contingent and in principle avoidable, although long-standing and pervasive, cultural accretions. This examination forms the basis of an epistemic challenge to nature-first constitutivism, which I press in Section 4. The core idea is that the negative assumptions at issue introduce ideological distortions into our assessment of what (human) capacities for agency and the associated excellences and “defects” are like, occluding from view the worthwhile hypothesis that some disabled modes of agency might be compatible with excellence qua (human) agent. Such distortions may be avoidable, but only by seriously interrogating the assumptions about disability that underpin them, as well as specific disabled modes of agency. I develop this challenge in detail by focusing on one disability affecting agency: ADHD.
Section 5 concludes by briefly outlining related metaphysical questions for nature-first constitutivism regarding plural forms of agency. The idea of plural forms of agency is explored in different ways by, for example, Setiya (Reference Setiya, Vargas and Yaffe2014), Lavin (Reference Lavin2017), Eckert-Kuang (Reference Eckert-Kuang2026), and Jeffrey (Reference Jeffrey2026). I situate the disability-related metaphysical questions raised herein within this developing discussion, without seeking to answer them.
A terminological preliminary: It is disputed what the term “disability” means or should mean, and what disability is. These disputes are orthogonal to my concerns. I use the term “disabilities” to refer to the bodily or mental conditions that are usually classified as disabilities; and when I speak about “disability” or “disability as such,” I mean the state of having some such condition. I will not assume any view about what, if anything, unifies these conditions or constitutes their underlying nature as disabilities. My topic is culturally prevalent negative associations regarding disability or the various disabilities; and the way in which those negative associations, if unexamined, are liable to distort nature-first theorizing.Footnote 2
2. Nature-First Constitutivism
“Nature-first” constitutivism is Jeremy David Fix’s label for the following type of view (Fix, Reference Fix2021, 1):
Something is subject to standard N because it is a particular of genus G.
This view is to be contrasted with “norms-first” constitutivism (Fix, Reference Fix2021, 1):
Something is a particular of genus G because it is subject to standard N.
In each view, the “because” indicates some kind of metaphysical explanation. For nature-firsters, “what something is determines what it should be”; for norms-firsters, it is the reverse (Fix, Reference Fix2021, 1).
In some more detail, nature-firsters hold that something, x, should be a certain way because of being a particular of genus G, where G has a certain nature or essence. Because x is a cat, x should have four legs. If x were a human, x should not have four legs, but two. It is an “excellence” in cats but not in humans to have four legs, and a “defect” in cats but not in humans to have two; and this is because of the nature of cats and humans, respectively (Fix, Reference Fix2021). More importantly for those interested in norms for (human) agents and actions, nature-first constitutivism claims something of the following form: because x is a (human) agent, x should do/be Φ (e.g., take sufficient means to their ends, keep their promises, be charitable); it would be a defect in x if it did not/was not Φ, and this is because of the nature of (human) agency.Footnote 3
But how does what something is determine what it should be, for nature-firsters? Views on this vary. One idea is that Gs by their nature have a certain function, which defective members of G perform badly or not at all, and excellent members perform well. For example, Christine Korsgaard (Reference Korsgaard2009) holds that the function of agency or action is self-constitution; for Michael Smith (Reference Smith2013, Reference Smith2015), the function is desire-realization. For both, individual exercises of agency are excellent or defective (to a degree) depending on how well they perform the relevant function.Footnote 4 A similar idea is that Gs by their nature have a constitutive aim—for example, according to David Velleman (Reference Velleman2015), that of self-understanding—and particular members of G are excellent or defective depending on how well they achieve this aim.Footnote 5
Fix’s own story is slightly different. Fix argues that some essential properties of Gs—properties that are part of their nature or essence—are not necessary properties of Gs. As Fix observes, the possibility of such “contingent essential properties” seems to have been noted by Aristotle, but ignored by the recent revival of essentialist metaphysics in analytic philosophy (Fix, Reference Fix2021, 2–12).Footnote 6 In our feline case, the claim is that though cats by their nature have four legs, particular cats can lack some of them. A two-legged cat is “missing” two legs (in Aristotle’s terms, “suffers privation”), because cats by nature have four. (In contrast, a two-legged human is not “missing” two legs, because it is not in the nature of humans to have four legs.) If that is right, then there are contingent essential properties—properties that belong to the nature or essence of a genus G, but that some particulars of G can lack (Fix, Reference Fix2021, 6–12 and esp. 8).
How do we get from these claims about contingent essential properties to evaluations of particulars of G as good or bad, excellent or defective, or as doing or being as they should or should not? Related, how do we get claims about what standards Gs as such are subject to? For Fix, badness is “privation” and goodness is “perfection” (Fix, Reference Fix2021, 16). A particular cat or human agent is bad, defective or not as it should be (qua a G) because and to the extent that it lacks some contingent essential properties of its genus G; or if some of those properties come in degrees, lacks them to a degree. A particular of genus G is good, excellent or as it should be (qua a G) because and to the extent that it possesses the contingent essential properties of G (to a degree).Footnote 7 The standard for particulars of G that is defined by contingent essential properties is just to have those properties; and if they come in degrees, presumably to have them to a maximal degree (Fix, Reference Fix2021, 12–16).Footnote 8
Something like Fix’s view of the matter seems to be accepted by some other prominent neo-Aristotelians. For Michael Thompson (Reference Thompson and O’Hear2004, Reference Thompson2008), “natural” goodness and defect in an individual living organism is a matter of whether it satisfies the “natural-historical judgments” that describe the nature of the “life form” to which the individual belongs. Natural-historical judgments have something like the form “the S is/has/does F,” “Ss characteristically are/have/do F,” or “it belongs to an S to be/do/have F” (Thompson, Reference Thompson and O’Hear2004, 49): for example, “the cat has four legs,” or “humans keeps their promises.” An individual of the life form is “naturally good” or excellent as a particular of its life form to the extent that, and because, it satisfies these descriptions, and “naturally bad” or defective to the extent that, and because, it does not (Thompson, Reference Thompson and O’Hear2004, 54–55, Reference Thompson2008, 80–81). This seems to involve appeal to contingent essential properties in something like Fix’s way. Similarly, Philippa Foot (Reference Foot2001) adopts roughly Thompson’s framework; but sometimes additionally says that an individual is flourishing to the extent that it “instantiate[s] the life form of [its] species” (Foot, Reference Foot2001, 91).Footnote 9 Similar characterizations of how to ground facts about an individual’s natural goodness and defect are found in, for example, Lockhart and Lott (Reference Lockhart and Lott2024, 71) and Frey (Reference Frey and Hacker-Wright2018, 48).
These characterizations seem compatible with viewing Thompson, Foot and some other neo-Aristotelians as nature-first theorists.Footnote 10 Yet it is not always clear whether that classification is ultimately accurate. Focusing on Foot, she emphasizes that only features, F, that (if had) characteristically help the individual S secure “the good” of Ss count as excellences in Ss (and defects if lacked). For nonhumans, “the good” is self-maintenance or reproduction. The blue tit’s blue head patch, while characteristic of blue tits, is not an excellence because (let us assume) it plays no role in securing this good (Foot, Reference Foot2001, 30–33, 46). Crucially, Foot maintains that the “human good” is sui generis, consisting in being “happy” in the seemingly normatively laden sense of, roughly, being benefited through virtuous rational activity and enjoying the attainment and pursuit of right ends (Foot, Reference Foot2001, 51, 93–98). Human excellences characteristically help to secure this good, and defects detract from it. While Foot seems to think that instantiating the “human good” is in some sense part of human nature—specifically, part of the nature of human agency, practical reason or will—her normatively laden description of the human good, along with the fact that she does not attempt to derive it from a prior conception of human nature, does raise the possibility that rather than grounding evaluations of human activity in human nature, she might be doing the reverse.Footnote 11
I will not try to resolve such interpretive issues. To keep things manageable, I frame my challenge as targeting nature-firsters, and will not explore whether a related challenge impacts norms-firsters. Here, I have elucidated the general idea of nature-first constitutivism, identifying some instances of it. A final preparatory remark: We should clearly separate the question of what is first in the order of metaphysical explanation from the question of what is first in the order of discovery. Nature-first metaphysics of norms is compatible with norms-first epistemology of the nature of the kinds, G, that they govern. This will be important in Section 4.
3. Conceptions of Disability and its Place in Good Human Lives: Some Lessons from the Philosophy and History of Disability
There are pervasive negative conceptions of disability that, if unexamined, are likely to distort our assessments of whether disability is compatible with (human) “perfection” or “excellence” in the sense at issue in nature-first views. Related, they are also likely to distort our assessments of what belongs to the nature of (human) agency. In this section, I first describe the negative conceptions at issue. I then present some critiques of them and draw some further lessons, so that we can see (in Section 4) why it is plausible that these negative conceptions not merely influence but distort nature-firsters’ assessments of agential nature, excellence or defect. My main claim in the paper is that to avoid disability-based ideological distortions in our thinking about the nature and excellences of (human) agency, we should examine in detail disabled modes of agency, as well as whether and how various disability-based distortions may be influencing our theorizing. Related, we should resist the tacit assumption that disabled modes of agency are automatically incompatible with perfection qua agent.
It is hard to overstate the cultural prevalence of the idea that having a disability is bad for you and a deviation from health, so something to be cured, fixed, or avoided. This is despite explicit representations of disability being relatively absent from mainstream Western culture. When disabled people are featured in popular movies or TV shows, their representation often follows various tropes that express and reinforce the view that disability is a tragedy. These tropes include, for example, bitterness and anger as a response to being disabled; and at the extreme, the idea that one would be better off dead than living with a disability such as, say, quadriplegia.Footnote 12
The view that disability is bad for you, a deviation from health, and indeed a defect is also pervasive in many aspects of medical practice. For example, in the United States and in many European countries, prenatal testing for genetic conditions such as Down’s Syndrome or other so-called “birth defects” is common practice, with high rates of selective abortion following positive test results.Footnote 13 Another example is surgeries and amputations for children with various congenital “deformities of the upper limbs”—conditions which are often painless but detract from “normal functioning” and affect appearance.Footnote 14 The message is that “normal” functioning and appearance are so important as to justify painful surgeries and the complications that may result. The representation of disabilities as regrettable defects and unhealthy conditions to be avoided, or else fixed or managed through medical intervention, also extends to disabilities affecting agency. For example, ADHD (attention-deficit hyperactivity disorder) is commonly represented as a deficit and a disorder; it is right there in the name! Likewise for autism, often referred to as autism spectrum disorder.
Disability scholars, activists and philosophers have effectively challenged the prevalent negative associations of disability with tragedy, ill health, deficit, or defect. Disability and lack of “normal” functioning have a much more complicated relationship to health and well-being than the automatic associations suggest (see, e.g., Barnes, Reference Barnes2016, Reference Barnes2023; Campbell & Stramondo, Reference Campbell and Stramondo2017; Quill Kukla (writing as Rebecca Kukla), Reference Kukla, Arras, Fenton and Kukla2015; Silvers, Reference Silvers2003). Some cases of disability are associated with elevated risk of health conditions or lower well-being, but many are not—at least, not independently of the unaccommodatingness of contingent social and environmental arrangements—and disability as such need not be bad for you or a detriment to health. Nor need disability be automatically classified as a defect. I will now outline some reasons why these claims are plausible. Again, this is so that we can see (in Section 4) why it is plausible that the negative associations in question not merely influence but distort nature-firsters’ assessments of agential nature, excellence and defect.
Start with well-being, in the sense of what is finally prudentially good for one. Call the idea that disability as such is bad for you or detracts from your well-being—specifically, bad for you even if society were fully accommodating—the “Tragedy view” (cf. Goldiner, Reference Goldiner2022).Footnote 15 If disability as such is bad for you, it could be a final prudential bad or an instrumental one, depriving one of final prudential goods or leading to final prudential bads.
As Campbell and Stramondo (Reference Campbell and Stramondo2017, 156–157) point out, most major theories of final (or “intrinsic”) prudential goods and bads would not classify disability as a final bad.Footnote 16 Most disabilities do not intrinsically involve pain, so hedonistic views will not classify disability as such as finally bad. Nor is lack of disability usually on objective lists of final goods. On some interpretations of desire-fulfillment views, what is finally bad for one is not desire-frustration or aversion-satisfaction as such, but rather whatever, x, one is (actually or ideally) averse to or desires not to have, as an end; on such views, some disabilities or even disability as such are finally bad for some people, given their desires or aversions. Still, disabilities are not finally bad for those indifferent to or wishing to retain their disabilities (Campbell & Stramondo, Reference Campbell and Stramondo2017, 156–157 and 177–178 n.13). Many in, for example, the Deaf, autistic or ADHD communities are proud of these aspects of themselves.Footnote 17
Neo-Aristotelian theories that view well-being as human perfection may count disability as finally bad for you, if disabilities are human defects or imperfections. Many neo-Aristotelians do seem to tacitly assume that disabilities are natural defects, given their stock examples (see, e.g., Fix, Reference Fix2021, 8, 16; Foot, Reference Foot2001, 24; Thompson, Reference Thompson2008, 81). But, it seems, one could in principle be a neo-Aristotelian while holding that some disabilities are compatible with human perfection or flourishing (Barnes, Reference Barnes2016, 62; Jeffrey, Reference Jeffrey2026)—different ways of being a flourishing, perfect, excellent human. While I ultimately raise a disability-based challenge for nature-firsters, including neo-Aristotelians, it is not straightforward or obvious why neo-Aristotelians could not just change their examples and assumptions about disability while keeping the structure of their view.
Next: Is disability instrumentally prudentially bad? No doubt many disabilities bring social disadvantage and exclusion, given unaccommodating environments and prejudicial attitudes. (As do minority racial, ethnic or gender statuses.) But the Tragedy view supposes disability to be bad for one even if society were fully accommodating. Is it? A tempting argument for disability’s instrumental badness is that disabilities prevent one from being able to enjoy specific prudential goods, such as pleasant visual or auditory experiences or casual strolls with family. However, a person deprived of these goods can usually access other prudential goods of the same general type: say, pleasure, engagement with beauty, achievements, or activities with loved ones. Such general goods are multiply realizable, and indeed, specific disabilities often open new avenues for realizing them, such as fluency in ASL or taking casual rolls with family (Campbell & Stramondo, Reference Campbell and Stramondo2017, 157–160). Disabilities do often deprive one of specific goods, and so are locally instrumentally bad, but we can say the same about almost anything. My current enjoyment of a steaming cup of coffee prevents me from concurrently enjoying a brisk run. Local instrumental badness is very common, and does not support the idea that disability tends to be overall instrumentally prudentially bad—more instrumentally bad than good for one—even if society were fully accommodating.Footnote 18
The Tragedy view, then, is at least significantly problematic. Turn now to the relationship between disability and health. The assumption that disability is a deviation from health is part of the “Medical model” of disability, according to which the primary remedy for alleviating disability-related disadvantage is medical intervention aimed at “curing” impairments or minimizing their impacts on functioning (Goldiner, Reference Goldiner2022, 32–33).Footnote 19 What, if anything, is wrong with the automatic association of disability with ill health? The basic problem is that it is difficult to find accounts of health or ill health (pathology) that non-question-beggingly and clearly classify disability as pathology and are otherwise extensionally adequate.Footnote 20
For example, consider “normal function” accounts, such as Christopher Boorse’s (Reference Boorse1977, Reference Boorse, Humber and Almeder1997, Reference Boorse and Gifford2011, Reference Boorse2014). For Boorse, the “normal function” of a part or process is its statistically typical contribution to the individual’s survival and reproduction, relative to the reference class of individuals of the same species, sex, and age; and health is the “readiness” of each internal part or process to perform all of its “normal functions” on typical occasions with at least typical efficiency. Disease or pathology is an internal state that impairs health, making a part or process function below typical efficiency (Boorse, Reference Boorse1977, 555). This view has many problems. For instance, it fails to count arthritis in old age as pathology, given its frequency and (in our circumstances) lack of negative effect on individual survival and reproduction (Barnes, Reference Barnes2023, 24–25). In any case, the view does not clearly classify disability as pathology, since plenty of disabilities do not have a negative effect (in our circumstances) on survival or reproduction.Footnote 21
Normal function accounts of health seek not to make normative or evaluative assumptions in their characterizations of health. Some instead view health in partly evaluative terms (Barnes, Reference Barnes2023, 37; Cooper, Reference Cooper2002; Daniels Reference Daniels2012; Kukla, Reference Kukla, Arras, Fenton and Kukla2015; Powers & Faden, Reference Powers and Faden2006). A simple suggestion is that health just is well-being, or an aspect of well-being; and that pathologies are detriments to (the relevant aspect of) well-being.Footnote 22 But this view does not support the idea that disability as such involves deviation from health, if disability does not automatically reduce well-being, as discussed above. A different, neo-Aristotelian suggestion is that being healthy is being an “excellent or flourishing instance of one’s kind” (Barnes, Reference Barnes2023, 42). Pathology is any condition that makes one fall short of excellence or flourishing. Neo-Aristotelians do sometimes explicitly associate excellence and flourishing with health, and defect with pathology.Footnote 23 However, just as it is unobvious why neo-Aristotelians cannot in principle view (many) disabilities as contributing to different ways of being an excellent, flourishing human, it is also unobvious why (many) disabilities are not fully compatible with neo-Aristotelian health.
I have surveyed some reasons why philosophers of disability have doubted that disability as such is bad for you or a detriment to health. I have also noted that it is unclear whether and why we should view disability as a (human) defect, rather than as compatible with excellence and perfection. Now for some final lessons from the history of disability, before developing the epistemic challenge to nature-first constitutivism that all of this raises (Section 4).
First, some moments in the history of disability illustrate that the negative conceptions of disability critiqued above are contingent and in principle avoidable cultural accretions. Traditional pre-colonial indigenous worldviews of North American peoples emphasized balance and reciprocity within a community, where each person regardless of bodily and mental differences has a useful role to play and a “gift” to contribute to the life of the community, while benefiting from others. What we would today call disability was not automatically stigmatized or associated with incompetency or defect, and people with disabilities were integrated into the life of the community in various ways viewed as mutually beneficial (Nielsen, Reference Nielsen2012, chap. 1).Footnote 24 In contrast, views about disabled people and their place in society starkly shifted in North America during the colonial era and the rise of capitalism. Early capitalist colonizers viewed disability as inability to perform labor to earn an individual living. While many physical differences were initially viewed as compatible with labor, colonial communities from the 1700s to the 1930s gradually came to view anyone with mental or physical disabilities as likely to become a “public charge,” hence as to be prevented from entering the country; or if already present, forcibly institutionalized and/or sterilized, for the “health of the nation” (Nielsen, Reference Nielsen2012, chaps. 3–7). This is likely part of how the view of disability as unhealthy, bad for one, and a defect of sorts was baked into one mainstream culture over centuries. Histories elsewhere differ, but given the influence of European colonizers across the world, I speculate that some similarities are likely. In any case, my point here is precisely that the negative conceptions at issue, while prevalent, are contingent and in principle avoidable.Footnote 25
Second, and related: the various contingent exclusions of disabled individuals from society plausibly contributed to what I call disability invisibility: the relative absence of disabled people from public awareness and mainstream culture even today, despite gains in disability rights in recent decades.Footnote 26 Third, we should be wary of how capitalist ideals of individual economic efficiency, together with disability invisibility and the earlier-critiqued negative assumptions about disability, may tacitly shape our ideals about what should or should not happen in human life, and related, our ideals of excellent (human) agency. Such factors may not typically function as reflectively endorsed reasons for adopting specific ideals of (human) excellence; their influence is likely more surreptitious than that, but no less distorting. Section 4 develops these themes.
4. Ideological Distortion in the Epistemology of Agency and its Excellences
The disability-related factors discussed above, if unexamined, are likely to distort our assessments of agential excellence and defect, occluding from view the possibility that some disabled modes of agency may be compatible with agential perfection in the sense at issue in nature-first views. Related, they are likely to distort our assessments of what belongs to the nature of (human) agency. This section first articulates the idea of “disabled modes of agency,” focusing on the case of ADHD as typically involving one such mode (Section 4.1). I then explain how disability-related ideological distortions arise in our explicitly evaluative judgments about agential excellence and defect (Section 4.2). Next, I explain how these distortions infect theorizing about the nature of (human) agency as well (Section 4.3). Finally, I consider some objections (Section 4.4). The overall upshot is that without examining in detail disabled modes of agency and how disability-related distortions may be impacting their theorizing, nature-firsters should not be very confident in their views about the nature and excellences of agency. This is not an argument that any specific nature-first view is definitely mistaken; rather, that nature-firsters should pay attention to these issues when constructing and justifying their views, so as not to ignore relevant alternatives to their visions of the excellences and nature of (human) agency.
4.1. Disabled modes of agency and the case of ADHD
Some theorists have explored nonrational or proto-rational agency, as may be found in small human children (Jeffrey, Reference Jeffrey2026) or nonhuman animals (Korsgaard, Reference Korsgaard2009, chaps. 5–6). My interest is with agency consisting in exercises of practical reason—where “practical reason” is understood as the capacity to reason about what to do and to act and intend for reasons, that is, on the basis of considerations. I sometimes call this “minimally rational” agency, to underscore that exercises of such agency can (it seems) sometimes be irrational, falling short of rationality in a commending sense. This is my interest because I am interested in the potential of nature-first views to ground standards of practical reason. What might it mean, then, to say that there are different “modes” of minimally rational agency, including “disabled modes”?
Let me first describe the case of ADHD, to point toward the kind of thing I mean. I focus specifically on one feature of agency that ADHD shapes: one’s degree and kind of instrumental efficiency. While ADHD does not always manifest in the exact same ways, it is commonly agreed both in self-reports and in medical approaches that ADHD often makes it difficult to stay on task and finish projects due to, for example, distraction or impulsively switching to novel tasks; and can make it difficult to even start taking steps toward completing an intended project, until it is too late to finish it on time.Footnote 27 Specifically, ADHD makes it difficult to exercise intramental strategies of self-control to keep on task, such as attention regulation or “inhibitory control” (Koi, Reference Koi2023)—roughly, the ability, through sheer cognitive effort, to suppress impulses toward actions inconsistent with staying on a given task (Kang et al., Reference Kang, Hernandez, Rahman, Voigt and Malvaso2022). ADHD does not involve similar difficulty in using extramental strategies of self-control, such as situation selection and situation modification, to help stay on task, if such strategies are accessible (Koi, Reference Koi2023).Footnote 28 Overall, ADHD plausibly shapes what we may call one’s patterns of instrumental efficiency, by making one less generally instrumentally efficient. By this I mean that, taking into account all of one’s intended ends, ADHD makes it less likely (than for otherwise comparable non-ADHD populations) that one will take sufficient means to those ends, other things being equal.
Much research on and public framing of ADHD is “deficit-oriented” and fixates on such patterns conceptualized as negative, defective and as deviations from health. It is certainly easy to see how they can hinder meeting standard expectations at school and many jobs. However, recently popular neurodiversity movements and some subsequent research have highlighted what are often experienced as positive features or strengths of ADHD functioning. Frequently mentioned strengths include creativity and the ability to think “outside the norm,” being solution-oriented and resilient, having high curiosity, and of special interest for my argument, having the ability to “hyperfocus” on tasks that are experienced as interesting for long periods of time without getting distracted (Nordby et al., Reference Nordby, Guribuye, Nordgreen and Lundervold2023). Hyperfocus impacts ADHD-related patterns of instrumental efficiency. For it seems to involve or enable a kind of high selective instrumental efficiency, in the service of ends deemed interesting. It can help one excel at tasks one is passionate about.Footnote 29 It can also contribute to lower general instrumental efficiency, as when one’s focus on an interesting task makes one miss a more boring work meeting that one intended to attend.
Assuming that relatively low general efficiency combined with high selective efficiency (in the service of interesting ends) is indeed a common pattern in adults with ADHD, we might model this in different ways. I will not take a stance on how best to do this. On one model, different patterns of efficiency involve the agent being disposed to follow different instrumental rules. High general efficiency involves a rule that does not discriminate between intended ends based on one’s level of interest in them—or, perhaps, based on anything else. Every intended end is on a par as an “input” to the rule, with intentions for believed means as “outputs” when all goes well. Whereas the combination of high selective efficiency and low general efficiency characteristic of ADHD involves a different rule that does so discriminate, either highly prioritizing interesting ends as inputs, or at the extreme, not even accepting boring ends as inputs.
On a different way of modeling the situation, different degrees and kinds of efficiency involve the same general instrumental rule but in different ways—specifically, through having different tendencies to go or not go in for instrumental reasoning structured by this one rule on given occasions. Compare: different theoretical reasoners may differ in their tendencies to go in for the same type of deductive reasoning. Someone may be very interested in the logical implications of their existing beliefs, and another not so; this is reflected in what each ends up believing. Similarly, on this second model, ADHD’s combination of low general and high selective efficiency involves a disposition to go in for instrumental reasoning a lot when the intended ends are deemed interesting, and to interrupt or hold off from instrumental reasoning when the intended ends seem boring; and high general efficiency disposes one to go in for instrumental reasoning equally either way.
There may be further viable models. However precisely we model the situation, what I mean by the “mode” of (instrumental) agency characteristic of ADHD is (i) the combination of relatively low general efficiency with high selective efficiency, specifically in the service of ends that one finds interesting and (ii) this as a stable disposition, accompanied by an incapacity or at least great difficulty of achieving high general efficiency (at a degree comparable to that typical of non-ADHD populations). This “mode” clearly involves minimally rational agency. However, it is a “disabled mode,” in the sense that it is a mode characteristic of a condition we count as a disability.
All of the above is neutral on whether this mode of agency is compatible with excellence or perfection qua agent, or instead implies some agential defect. This is by design. For as I aim to display, it is a substantive and difficult question worthy of explicit engagement whether any particular disabled mode of agency involves agential defect.
4.2. Distortions in judgments of agential excellence and defect
With this mode of agency in mind, how to evaluate it? Is it compatible with excellence or perfection qua agent, or does it necessarily involve some agential defect or imperfection? I am hoping that this is looking like an increasingly hard question, and will look even harder by the end of this paper. I will not try to answer the question, but instead to display why it is a hard question, some of the likely reasons why questions like it have been overlooked, and why seriously engaging with it and with the topic of disabled modes of agency more generally looks worthwhile for nature-firsters.
Suppose we are considering explicitly evaluative questions about what agential excellence or perfection is like. In attempting to make judgments about this topic, a set of disability-related ideological distortions, stemming from the factors discussed in Section 3 is likely to conspire to occlude from view or from serious consideration the idea that some disabled modes of agency might be compatible with agential perfection. The most obvious route to distortion is via the idea of defect. As noted in Section 3, congenital disabilities are commonly classified as “birth defects.” This classification implies falling short of human perfection. Given the prevalent negative associations about disability, it is not surprising if, implicitly or explicitly, nature-firsters (or others) assume that all disabilities are “defects” of some sort—including disabilities that affect one’s practical reason. Certainly, the neo-Aristotelian literature is full of examples that suggest that many authors casually assume various disabilities to be human defects.Footnote 30 If one assumes that disabilities are defects, one is likely to ignore the possibility that some disabled modes of agency might be compatible with, or even contribute to, agential perfection. Even if one does not explicitly classify disabilities affecting agency as defects—perhaps because one has not even considered the topic—one’s theorizing may be distorted precisely by the failure to consider relevant alternatives to one’s picture of perfection.
Related, we can see how disability invisibility can distort judgments about agential excellence or perfection. Given disability invisibility, theorists tend to not even discuss disability or disabled agents in detail. As a result, we again overlook the idea that disabled modes of agency might be modes of perfection instead of defect.
Assumptions that disability is bad for you or unhealthy can distort judgments about agential excellence or perfection too. For these assumptions are difficult to reconcile with the idea that disabled modes of being or doing might be ways of flourishing. Indeed, such assumptions, whether explicit or implicit, make it difficult to even think of the possibility that some disabilities might involve modes of flourishing. And for neo-Aristotelian nature-firsters, lack of flourishing involves falling short of perfection. Nature-firsters focusing on minimally rational agency in general rather than specifically human minimally rational agency do not tend to put their views in terms of “flourishing.” Still, it is at least uncomfortable to view an excellent or perfect mode of agency as being, as such, bad for you or unhealthy. Thus, the assumptions that disability is bad for you and unhealthy are likely to bolster one’s sense, whether implicit or explicit, that disabled modes of agency are incompatible with agential excellence or perfection. Note that the relevant kinds of negative assumptions are easily mobilized in the case of ADHD in particular, given culturally prevalent representations of ADHD as calling for medical intervention.
Finally, I tentatively propose that capitalist ideals of individual economic efficiency may also tacitly distort our sense of what agential excellences and defects are like. Specifically, such assumptions may partly explain why general instrumental efficiency, in the service of whatever ends one has, seems like an obvious agential excellence of the sort that one’s constitutivist theory should vindicate. Effectively taking sufficient means to one’s intended ends, even when those ends are hoisted on one by the demands of socially expected schooling and labor arrangements and are experienced by the agent as boring, is obviously useful in capitalist societies for most people. And falling short of this can obviously lead to difficulties. Yet usefulness in one kind of context does not imply “natural” excellence; and one can imagine societies in which those who find it difficult to complete intended projects they find boring are not thereby seen as defective qua agents, because their efforts are exceptionally creative and sustained in areas that they find interesting, and this is seen as an asset to them and to society.Footnote 31 The general point here is that contingent social and economic arrangements and associated assumptions about how an ideal individual functions in society are a further potential source of distortion in our judgments about agential excellence and defect—including in the case of ADHD modes of agency.Footnote 32
If unexamined, all these factors—disability invisibility, negative associations regarding disability, and socioeconomic arrangements which prize certain nondisabled modes of agential functioning—seem likely to work in concert to implicitly steer one’s theorizing toward only considering nondisabled modes of agency as candidate modes of agential excellence or perfection. I suggest that, given the discussion of Section 3, we should take seriously the worry that the factors not merely steer but distort. To avoid the distortions, I claim, one should carefully examine disabled modes of agency while open-mindedly considering the hypothesis that some of them may be modes of agential excellence or perfection; and see how one’s arguments fare with this hypothesis in view.
Consider an objection. Even if the distortions are an initial concern for our explicitly evaluative judgments about agential excellence or defect, is not there a way around this? Nature-firsters say that specific agential standards, N, apply to (human) agents because of the nature of those agents: what these agents are determines how they should be. Given this, one might think that distortions at the level of evaluative judgments about agential excellence and defect are ultimately irrelevant to nature-firsters’ arguments in favor of their views. After all, cannot and should not nature-firsters simply start with arguments about the nature of (human) agency, and see what picture of agential excellence or perfection emerges? Even if the distortions impact explicitly evaluative judgments about agential excellence or perfection when we try to make such judgments without deriving them from an explanatorily prior account of the nature of agency, the distortions will not impact our theorizing about the very nature of agency. This gives us an ultimately distortion-free route to knowledge of agential excellence and defect. Or so a reasonable objection goes.
In response, I will display how disability-related distortions arise even for nature-first theorizing about the nature of agency. Methodologies differ among nature-firsters, and I cannot examine all of them. But examining a selection will help to display the general concern. While the concern arises at slightly different junctures for different views, it does arise.
4.3. Distortions in theorizing about the nature of agency
It is easiest to see how the kinds of distortions at issue arise on views according to which our knowledge of the metaphysics of agency is itself arrived at, wholly or in part, via evaluative judgments regarding agential excellence or defect. Such a “norms-first” epistemology of (human) agency is suggested by, for example, Micah Lott (Reference Lott2012) and Michael Thompson (Reference Thompson and O’Hear2004, 72–73). While Lott and Thompson do not spell out the idea in quite this way, one can see the attraction of a norms-first epistemology of agency for nature-firsters as follows. If x’s excellence or defect qua a G is a matter of the extent to which x instantiates contingent essential properties of Gs, then knowing that x is excellent or defective qua a G in virtue of having or lacking property F tells you something about what the essential properties of Gs are. It tells you that Gs are essentially F. Thus, we can move from claims about agential excellence or defect to claims about what belongs to the nature of agency, by (implicitly or explicitly) applying Fix’s general schema for how these two kinds of claims relate to each other. This kind of norms-first epistemology of agency is clearly compatible with being metaphysically nature-first. But it also makes it clear how initial distortions at the level of evaluative judgments would infect one’s views about the nature of agency as well.
To be clear, this is not a critique of a norms-first epistemology of agency. One can consistently hold that unexamined ideological distortions in evaluative judgments prevent knowledge about agential excellence and defect, and thereby prevent norms-first knowledge of the nature of agency; while also holding that knowledge of agential excellence or defect, which it is possible to gain, would allow one to gain norms-first knowledge of the nature of agency. The problem is the unexamined ideological distortions, not the norms-first epistemology. Those adhering to a norms-first methodology can and should try to get around the problem by open-mindedly examining disabled modes of agency and the credentials of one’s evaluative verdicts regarding those modes, while guarding as much as possible against the kinds of distortions I have pointed to.Footnote 33
A different proposal comes from John Hacker-Wright (Reference Hacker-Wright2009, attributing the view to Foot). He suggests we can know the human form (of agency) via a kind of “self-interpretation”: we interpret ourselves as creatures for whom it is characteristic to Φ—for example, to live in accord with charity. What rationally commits us to one rather than another interpretation? Hacker-Wright suggests that our acceptance of certain interpretations, when reinforced by others, “makes it true that ‘this is how things are with us’”—“we have a kind of species self-constitution involving reason” (Reference Hacker-Wright2009, 320). Thompson (Reference Thompson2003) appears to agree. The idea seems to be that certain judgments of the human form are true, and we should take them to be true (?), because they constitute the human form as being the way they represent it as being. The judgments are “self-validating” (Thompson, Reference Thompson2003, 7). This is an intriguing idea. But it is also subject to at least some of the disability-related distortions I have been concerned with. Specifically, disability invisibility is likely to lead nondisabled inquirers to include only nondisabled people in the “us” whose agency is being interpreted; and is likely to lead to self-interpretations that ignore or exclude disabled modes of agency without examining them. The self-interpretation proposal is in any case problematic on further grounds. For one thing, it is hard to see how it distinguishes between disagreeing judgments about human form. If disagreeing interpretations are each reinforced by some others, which ones are incorrect and why? Or do we multiply human forms according to self-interpretations, so that none are incorrect? If some are incorrect, why and how can we tell which ones? This seems to just reiterate the problem of how to know the human form (of agency). Accordingly, I set the self-interpretation proposal aside.Footnote 34
The most promising strategies for sidestepping the disability-related distortions seem to be ones that engage in what looks like straightforward metaphysical theorizing about the nature of agency, and seek to derive a view about agential excellence and defect from this ex post. I am thinking here, for example, of Korsgaard’s (Reference Korsgaard and O’Neill1996) transcendental arguments about what makes acting for a reason possible; Korsgaard’s (Reference Korsgaard2009) appeals to the function of action and unity of agency; and Smith’s (Reference Smith2013, Reference Smith2015) arguments regarding the function of agency and what is required to ideally perform it. Such arguments look to use metaphysical theorizing about agency to justify the idea that abiding by certain norms, such as instrumental or moral norms, is part of being excellent qua agent. Why should such arguments involve the disability-related distortions I have been pressing?
Here, the distortions enter in a subtler way. Although explicitly evaluative claims about agential excellence and defect do not here justify the theorist’s claims about the nature of agency (unlike on the norms-first epistemology above), evaluative claims do seem to play a role, implicitly or explicitly, in the context of discovery. That is, they play a role in identifying features of agency to focus one’s theorizing on, and in generating hypotheses about which features are essential to agency.Footnote 35 It is hard to deny that nature-firsters often focus on putative features of agency that look most relevant for justifying normative views that have some pre-existing plausibility. Indeed, the whole point of nature-first theories is not just to engage in disinterested metaphysical theorizing about agency, but to vindicate plausible norms by appeal to the metaphysics of agency. No surprise, then, that theorists tend to specifically search for putatively essential features of agency that could justify the claim that excellence qua agent involves abiding by some plausible-looking norms, such as norms of instrumental efficiency and/or moral rectitude.Footnote 36 This introduces room for precisely the kinds of distortions that I am concerned with.
To see how, consider Korsgaard’s (Reference Korsgaard2009) arguments regarding instrumental efficiency; or, in her terms, the hypothetical imperative. Korsgaard claims (plausibly enough) that it is in the nature of agents to perform actions (acting is our “plight” [Reference Korsgaard2009, 23]); and that (more controversially) actions as such have a function, which is self-constitution—unifying the agent as the “autonomous and efficacious author of his own movements” (Reference Korsgaard2009, 161). The crux of Korsgaard’s vindication of the hypothetical imperative is that agents must will the means to their willed ends, because otherwise they fail to constitute themselves as efficacious authors of their movements over and above the play of impulses and desires in them, and hence fail to constitute themselves as agents:
The reason that I must conform to the hypothetical imperative is that if I don’t conform to it, if I always allow myself to be derailed by timidity, idleness, or depression, then I never really will an end. The desire to pursue the end and the desires that draw me away from it each hold sway in their turn, but my will is never active. The distinction between my will and the operation of the desires and impulses in me does not exist, and that means that I, considered as an agent, do not exist. Conformity to the hypothetical imperative is thus constitutive of having a will. (Korsgaard, Reference Korsgaard2009, 69–70)
Korsgaard (Reference Korsgaard and O’Neill1996, 230–231) gives a very similar argument, almost word for word. And while she does not distinguish between general and selective/specific efficiency, she seems to take her argument to justify a perfectly general demand for means-ends efficiency—the imperative that one must always will the means to one’s willed (in our earlier terms, intended) ends. Abiding by this imperative is supposed to be constitutive of agency in the literal sense that, to the extent that one does not abide by it, one’s actions are less fully actions, and fail to constitute one as the efficacious author of a movement; they thus fail to perform the function of actions and are defective (Reference Korsgaard2009, chap. 8).
However, once we are aware of the idea of more selective patterns of efficiency, such as those characteristic of ADHD, it is hard to unsee that there is a gap in Korsgaard’s argument. The argument seems to only justify the conclusion that, to be an agent, one must sometimes will the means to one’s willed ends. Of course, when one does not will the means to some end, one thereby does not perform the relevant action that would have consisted in taking those means to that end. Still, this is compatible with the agent willing, and taking, means to other intended/willed ends (instead), and so performing other actions, which do serve to unify the agent as the efficacious author of the movements they do perform. There were simply some actions that they did not perform; some intentions they did not follow through on. At most, then, the argument seems to justify the idea that, in order to fully constitute oneself as the efficacious author of one’s movements, one must enact some kind of efficiency, whether selective or general. Correspondingly, selective efficiency seems sufficient for agential excellence (as far as the efficacy dimension of agency is concerned).
This argumentative gap might be missed if one never even considers the idea that selective efficiency, such as in ADHD, may be compatible with agential excellence. One often sees what one sets out to see. If one sets out to vindicate perfectly general instrumental efficiency, and does not consider alternative, more selective patterns of efficiency, then one may miss that one’s arguments have not actually ruled out a relevant alternative.Footnote 37
The lacuna in Smith’s (Reference Smith, Sobel and Wall2009, Reference Smith2013, Reference Smith2015, Reference Smith2020) discussions of instrumental efficiency is partly similar, partly different. Smith explicitly grants that different versions of an instrumental reasoning disposition may operate in different agents; although without considering possible disabled modes of efficiency. For Smith, some agents are merely “locally” instrumentally rational, disposed to “put together” some of their end-desires and means-ends beliefs to arrive at desires for the means. Others are “globally” instrumentally rational to a greater or lesser degree: disposed to put together many of their end-desires and means-ends beliefs so as to co-satisfy those end-desires, where, in the ideal case, the strength of one’s resultant means-desires is proportional to the strengths of one’s end-desires and one’s credences in the efficacy of various means. What kind of instrumental efficiency, if any, is part of excellence qua agent, and what essential features of agency are supposed to justify it?
For Smith, agential instrumental excellence corresponds to maximal global instrumental rationality. What is supposed to justify that claim is that the function of agency is desire-realization; and that performing that function optimally well requires maximal global instrumental rationality. The trouble is that the mere idea that agency’s function is desire-realization is compatible with many alternative conceptions of what it is to perform this function optimally well—specifically, optimally well simply qua agent, not according to some external standard (such as a presupposed standard of rationality). On one conception, one realizes as many desires as possible, however weak or strong. No need to prioritize stronger end-desires, or weigh them against weaker ones. Just cue the desires up and realize as many as you can! Another conception focuses on how well one realizes the desires one realizes, more than on the number or strength of the desires realized. Speculatively, transposing the idea of ADHD patterns of instrumental efficiency to the key of desires from that of intentions, this last conception might best accommodate the idea that the ADHD mode of efficiency may be compatible with agential perfection. Smith does not consider any such alternatives to his picture of perfection qua agent, and it is unclear what, if anything, in his views about the essential nature of agency is supposed to decide between them (or yield some yet further standard).
Akin to Korsgaard, Smith seems to have set out to justify the hypothesis that maximal global instrumental rationality is part of agential perfection. He ignores relevant alternatives to his hypothesis and as a result does not seem to point to anything in the nature of agency to rule those alternatives out. Granted, it seems analytic that if agency has a function, then excellent agents must perform that function excellently, and at the limit, optimally well. But the issue is precisely what the standards for performing it “optimally well” simply qua agent (not according to some external standards) are; and whether there is anything in the metaphysics of agency that vindicates Smith’s view on that issue.Footnote 38 (In Smith’s terms: while he ultimately hopes for the standards of “kind-ideal” and “rational-ideal” agency to align (Smith, Reference Smith2015, Reference Smith2020), he must still find the vindication for his view of kind-ideal standards in the metaphysics of agency.)Footnote 39
I have considered a sampling of ways, some obvious and some subtle, in which disability-related distortions may infect nature-first theorizing about agency. My aim has not been to show that any particular nature-first theory is definitely mistaken. Rather, I hope to have made it increasingly plausible that nature-firsters need to ward against such distortions; and that examining disabled modes of agency can be illuminating, whether to expose gaps in arguments for existing views or to broach the possibility of new, explicitly disability-inclusive views about agential excellence.
4.4. Objections and responses
I will briefly address three objections to the foregoing.
Objection 1: It is obvious that failing to take, or at least to intend, sufficient (or necessary) means to your intended ends, whatever they are, is always a form of practical irrationality. And one can infer from the idea that X is a form of practical irrationality that X is a defect of (human) agency, of (human) practical reason. How could a form of practical irrationality fail to be a defect of practical reason? It is thus not reasonably in question whether failures of general efficiency are agential defects. Accordingly, it is not reasonably in question whether the ADHD “mode” of efficiency is an agential defect.
Response 1: While defects of minimally rational agency are probably cases of irrationality in some sense, it does not follow that any case of practical irrationality is thereby a defect simply qua (exercise of) minimally rational agency. As I have tried to explain in the case of Smith and Korsgaard, there is a substantive gap between identifying a particular rational ideal and saying that it is part of the nature of agency to meet that ideal. In any case, “irrationality” is said in many ways. On one use, it refers to failure to respond to normative reasons within one’s ken (Kiesewetter, Reference Kiesewetter2017; Lord, Reference Lord2018). On another, it refers to failure of one’s attitudes to cohere or fit with each other (Kiesewetter & Worsnip, Reference Kiesewetter, Worsnip and Zalta2023).Footnote 40 A yet further use of “irrational” is ecological, referring to one’s reasoning strategies being ill-fitted to deliver good results in the environment one is in (Morton, Reference Morton2017; Todd & Gigerenzer, Reference Todd and Gigerenzer2012). It is highly nontrivial to claim that any of these kinds of irrationality is a failure by the lights of the constitutive standards of agency.
Objection 2: The ideological distortions I have worried about are a relatively shallow phenomenon, stemming from using the negatively valenced term ‘disability’ to refer to the various conditions labeled disabilities. If we instead use more neutral terms like ‘divergence’ or ‘functional variation’ to refer to these conditions, the ideological distortions will not arise.
Response 2: Such verbal maneuvering might help some people to some extent in avoiding the distortions at issue, but it does not remove the problem. Disabilities have been referred to by many different (derogatory) labels in different historical periods, and while changes in labeling have likely helped to reduce the stigma associated with many disabilities, the negative associations and disability invisibility that I have discussed have largely endured. Changing deep-seated negative cultural heritage associated with the conditions now labeled as disabilities takes more than switching labels. Likewise, simply switching labels will not make nature-first theorists suddenly consider a broader range of hypotheses about agential constitution and excellence. In any case, disability-positive philosophers and activists today typically use the term ‘disability’ in a way that is not negatively valenced; and ‘disability’ is nowadays the preferred term among these groups. So I have stuck with this term.
Objection 3: Nature-firsters often use the term ‘defect’ to refer to agential imperfections, but they need not do so. If we use the term ‘imperfection’ instead, and point out that (as Thompson says) nobody is perfect (Thompson, Reference Thompson2008, 72), it is not very interesting to find out that one’s exercises or modes or dispositions of agency, whether disabled or not, involve agential imperfections. Most, maybe all, instances of agency do. Perhaps this terminological choice also reduces the extent to which disability-related negative associations are mobilized when discussing how agents may fall short of agential perfection or excellence.
Response 3: I welcome the suggestion that another terminological fix might alleviate the problems I have pointed to. I would like the distortions to be addressed and avoided. Still, as before, this will not remove the problem. Insofar as disability is commonly thought of as a defect, it is thereby thought of as an imperfection of some kind. The substantive issue I have tried to raise for consideration is whether some disabled modes of agency might be compatible with agential perfection, rather than automatically being cases of agential imperfection. This question is still likely to be occluded from serious consideration by the association of disability with defect (and hence imperfection), even if nature-firsters prefer not to talk in terms of “defects.” Moreover, we can admit that nobody is perfect, even if we hold that there is such a thing as a standard of perfection. The question is what that standard is, and related, what the nature of (human) agency as such is; and whether we should be confident that disabled agency automatically falls short of meeting that standard. I have urged that we should not, until we carefully examine disabled modes of agency and any disability-related distortions that may be impacting our theorizing.
5. Concluding Reflections on the Possibility of Plural Forms of Agency
I have posed a disability-based epistemic challenge to nature-firsters on the tacit assumption that they wish to hold a monistic view about the nature of (human) agency, as is typical. On such views, the excellences of agency may be many—such as instrumental efficiency, moral rectitude, and perhaps more besides—but they are so due to different features of one kind of thing: agency. I claimed that nature-firsters should not be very confident in their views about agential excellence and defect until they carefully examine how disability-related distortions may affect their theorizing, and seriously consider whether some disabled modes of agency, usually ignored in nature-first theorizing, are compatible with agential excellence.
However, one natural response to the call to seriously consider disabled modes of agency is to view these as instantiating “forms” or “kinds” of agency so different from typical nondisabled forms that their constitutive standards of excellence also differ. Rather than attempting to figure out one shared set of standards for all agents, and how disabled agents fare by its lights, it may seem more reasonable to hold, for example, ADHD agents and typical non-ADHD agents to different standards of instrumental efficiency, each defining an agential excellence for a different kind of agent. The possibility of pluralism about forms of agency and associated standards has been explored by, for example, Setiya (Reference Setiya, Vargas and Yaffe2014), Lavin (Reference Lavin2004, Reference Lavin2017), Eckert-Kuang (Reference Eckert-Kuang2026), and Jeffrey (Reference Jeffrey2026)—although without considering the idea of disabled modes of agency.Footnote 41 A pressing issue for pluralists is to identify which differences in agents mark differences in “kinds” of agency in the relevant sense. It would be highly counterintuitive to count any old difference in agential dispositions, such as an idiosyncratic disposition to not care about what happens on future Tuesdays, as grounding its own kind of agency and agential excellence (Setiya, Reference Setiya, Vargas and Yaffe2014, 76). One might suggest that only differences in agential capacities, not mere dispositions, make for differences in kinds of agent; thus, for example, when ADHD involves an incapacity to be generally efficient, it counts as a different kind of agency from otherwise comparable non-ADHD kinds. However, should we say the same about a vicious person who, given where they are in life, finds it beyond their capacity to be anything but vicious? What about the temporarily vicious suffering from mere temporary incapacity to conform to more virtuous patterns of practical thought? Such possibilities look obviously unattractive for nature-first constitutivism, but the problem is how to ward them off in a principled and plausible way (cf. Eckert-Kuang, Reference Eckert-Kuang2026, 472). There is a “problem of proliferation” toward an increasingly fragmented and implausible normative landscape (Eckert-Kuang, Reference Eckert-Kuang2026, 471; Lavin, Reference Lavin2004, 124–125).
It seems crucial for pluralists to stop the proliferation. Eckert-Kuang promisingly suggests that kinds of agency correspond to “different ways of acting for a reason” or in pursuit of an end, not to “acting for different reasons” or for the sake of different ends as distinguished by their content (2026, 474, 478–480). This seems plausible as far as it goes: surely pursuing a quiet afternoon (or the capacity to do so) does not make for a relevantly different “kind” of agency from pursuing an evening of solitude (or the capacity to do so). However, I still worry that this casts the net too wide. Recall Smith’s (Reference Smith, Sobel and Wall2009) view that there are different patterns of and capacities for instrumental agency: some agents are only locally instrumentally efficient, others are more globally so. These distinctions do not seem to turn on which ends are pursued, in terms of their content, but how. Is the end being pursued via a disposition/capacity for merely local or for more global efficiency? If global, to what degree? If every degree of difference in this measure of “how” the agent pursues their ends makes for a difference in “kind” of agency, that would again look like too much proliferation. But if only some selected points on the continuum mark differences in kinds of agency, which ones and why? At least prima facie, the problem persists.Footnote 42
This leaves nature-first constitutivists at a fork in the road: continue down the monist path, facing a new disability-based epistemic challenge; or go down the pluralist path, with its puzzle of proliferation. (Or turn back and abandon nature-first constitutivism.) Happily, in philosophy more than most anywhere else, one can thoroughly explore both paths to see which one lands at a better place. To adapt Yogi Berra’s advice: “When you come to a fork in the road, take it.”Footnote 43
Acknowledgments
For very helpful discussion, feedback and encouragement on previous versions of this material, I am grateful to Luca Ferrero, Jeremy David Fix, David Forst, Antonio Freiles, Josh Hunt, Mikayla Kelley, Olof Leffler, Evgenia Mylonaki, Jesse O’Brien, James Pearson, and the participants in the September 2025 Constitutivism in Ethics and Epistemology workshop in Lisbon, Portugal.
Hille Paakkunainen is an Associate Professor of Philosophy at Syracuse University. Her research focuses on normativity and reasoning in ethics and in epistemology.
Open access
