Abstract
Pragmatist responses to skepticism about empirical justification have mostly been underwhelming, either presupposing implausible theses like relativism or anti-realism, or else showing our basic empirical beliefs to be merely psychologically inevitable rather than rationally warranted. In this paper I defend a better one: a modified version of an argument by Wilfrid Sellars that we are pragmatically warranted in accepting that our perceptual beliefs are likely to be true, since their likely truth is necessary for the satisfaction of our goal of effective agency. On the version of the argument I defend, the great good for human life of control over our empirical circumstances renders our goal of effective agency reasonable. But only if our perceptual beliefs are likely to be true—and only if we accept that this is so, assuming it as a premise for inference and a guide for action—will the success of our actions be due to our effective agency, not mere luck. Since we’re warranted in taking the necessary means to our reasonable ends, we’re warranted in accepting that our perceptual beliefs are generally justified, and so that skepticism about empirical justification is false.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The reader may notice that I have considered only pragmatist attempts at resolution of the skeptical problem here, not attempts at dissolution. But the latter may be at least as prominent in the history of pragmatism: Hookway suggests that Peirce’s “fundamental” response to skepticism is a form of doxastic conservatism that rejects the skeptic’s demand for justification for our current beliefs as simply misguided (2012: p. 29n13; cf. §1.3 throughout), and the guiding impulse of Dewey’s pragmatism was its “reject[ion of] the dualistic epistemology and metaphysics of modern philosophy” (Field undated), including its felt need (so to speak) for an answer to the skeptic. On this point (if few others!), I instead side with Rorty, who remarks that: “The pragmatist philosopher’s context of inquiry is … the context in which one wonders whether and why justification should be thought to lead to truth” (2000: p. 266), and in which one does not accept facts about when we in fact do take claims for true as adequate answers to such wonderings. Due to my continuing to accept the early modern skeptical problematic and standards for knowledge, a reviewer suggests that my position might fairly be described as realist neo-pragmatism. I find the label congenial.
Such a warrant might seem straightforwardly impossible: after all, doesn’t our very concept of perception arise experientially? Yes, but this does not prevent us from deploying it in a priori knowledge, any more than the fact that our concept < dog > arises from experience renders the judgment Dogs are animals an a posteriori one. What is relevant for the a priori/a posteriori distinction is the means of justification of the judgment, not the means of acquiring the concepts of which it’s composed.
In case these stipulations need motivation: suppose I’m an archer, and at one afternoon’s archery session, I hit the target every time. This doesn’t manifest effective agency if my arrows only found their mark due to fluke winds, or if every time I drew back, you nudged me into the right position as I released the bowstring. In the former case, the means and end aren’t sufficiently closely connected; in the latter, I lack control.
I’ll talk of perceptual beliefs rather than judgments. Nothing turns on this.
Mightn’t this end be idle, however? Isn’t it true that aiming at goals presupposes being an effective agent in the first place, obviating the need to aim at effective agency itself? No: aiming at goals presupposes being an agent, but not that one’s agency is effective. (What is further true, however, is that aiming at particular goals presupposes aiming—at least implicitly—at being an effective agent. Indeed, this will be crucial to my argument in Sect. 7.)
He tells us that he “use[s] ‘accept’, in the first instance, as roughly equivalent to ‘come to believe’” (1974: p. 438, n2). (Moreover, he notes, claims can be reasonable to accept in that sense without being reasonable to base one’s actions on [ibid.: pp. 410ff.].)
Wright similarly frames the non-evidential, pragmatically grounded species of warrant (viz., entitlement) he develops as a warrant for “a mode […] of acceptance of a proposition which can be rational but which [is] not tantamount to believing,” one that involves “acting on the assumption that P or taking it for granted that P or trusting that P for reasons that do not bear on the likely truth of P” (2004: pp. 176–77).
I think Sellars actually offers two independent responses to skepticism: a semantic argument meant to directly refute it [one similar to Davidson’s (2001: p. 213) argument that “our view of the world (must be), in its plainest features, largely correct,” since “the stimuli that cause our most basic verbal responses also determine (…) the content of the beliefs that accompany them”], and the pragmatic skeptical solution under consideration here. I cannot defend here, though, either this interpretation of Sellars’s epistemology or my judgment that the latter anti-skeptical argument’s prospects are better than the former’s. (I do so in other work in progress.)
Despite my disagreement with Wright here, my argument owes much to his epistemological writings. I differ from him, though, in centering my anti-skeptical proposal on (a non-psychologistic version of) entitlement of rational deliberation (2004: §VII), not entitlement of strategy or entitlement of cognitive project, the two species of entitlement most central to Wright (2004, 2014), and the latter of which seems key to his epistemology of perception (2004: §VI).
Someone might object that the considerations cited here still represent a merely psychological basis for the end of effective agency: that we desire control over our circumstances, after all, is just another psychological state, not in itself a reason for anything. In reply, this is the reason for noting that we do not merely desire control but, on reflection, judge it to be good: I am according these judgments an authority that enables them to confer warrant on aims that are instrumental for achieving that which they recognize as good. (Why grant evaluative seemings and judgments this authority when I do not grant it equally to perceptual seemings and judgments? Because the former seem to be immune from at least some of the skeptical worries to which I take the latter to be vulnerable: for instance, I don’t think I even understand the suggestion that a brain-in-a-vat might be deceived in judging that a pain that feels intrinsically and intensely bad really is bad.)
It is possible (and, indeed, commonplace) to be an effective agent with respect to some goal even as one’s efficacy depends on one’s use of some sort of means or implement: I can be effective at hitting the bullseye from 50 m even as my efficacy depends on my use of my bow. Could it be objected, then, that the supplicants are like this: their efficacy depends on the higher power, and yet they are no less effective agents for that? I think this assessment of the case would be implausible. For an agent’s dependence on an implement not to diminish the efficacy of her agency with respect to her goal, it seems to me that she must meet two conditions: first, she must have control over the implement, and second, she must have at least a rudimentary sort of knowledge how her use of the implement will enable her to realize her goal. (If I am reduced to flailing around spastically with the bow, or if I lack even rudimentary knowledge how a bow might propel an arrow toward my target, then I don’t seem to be an effective archer even if, fortunately, I should happen to strike the bullseye every time.) But in this case, the supplicants don’t control the higher power at all: they are dependent on its pity. And lacking any awareness of their environment (let alone of the higher power’s nature), they can’t be said to have even the most rudimentary sort of knowledge of how the higher power might achieve their goals in that environment. Accordingly, it seems quite strained to frame them as effective at realizing their ends, or in control of whether their ends are realized, through the means of the higher power.
Actually, the claim needn’t be conditional: any instance of effective human agency presupposes reliable perceptual faculties. An apparent counterexample is merely epistemic agency. But I can know introspectively (bracketing the external world and the veridicality of perception) that my epistemic agency’s success is vulnerable to my phenomenal circumstances: I won’t attain my epistemic goals if distracted by sharp pains or intense feelings as of cold or hunger. Thus even effective epistemic agency requires, not just reliable powers of reasoning, but also faculties that imbue one with sufficient control over one’s empirical circumstances to reliably ensure that one’s phenomenal state conduces to epistemic success.
Thus Lupyan’s (2017: p. 82) view, on which “the goal of perception is not truth, but rather […] guiding adaptive behavior,” seems consistent with (2) and with our being effective agents, since on his view, perception nevertheless does generally yield “information that is true enough for normal human goals (and sometimes generalizes beyond them)”—and further, to the degree that perception is not veridical, this is “oftentimes” not because it distorts the truth, but rather because “there is simply no truth of the matter for perception to provide.” His view still allows our agency to be guided by a reliable baseline of perceptual information. Similarly, I needn’t disagree with Churchland’s (1987: pp. 548–49) claim that “the principal function of nervous systems is to enable the organism to move appropriately,” and that representations are valuable only inasmuch as they serve that function, with truth “definitely tak[ing] the hindmost.” That our perceptual system is fundamentally oriented toward enabling adaptation, not true representation—and that this sometimes leads it not to yield true representations—is consistent with my thesis that perception couldn’t enable adaptive, effective agency if it didn’t generally yield true representations.
Compare Martin’s (2014: pp. 21–22) argument that reasonable hope doesn’t imply reasonable action-as-if.
It’s sometimes reasonable to adopt an end that is unlikely to be realized, I’ve noted, if its realization would be sufficiently valuable. But however great some good might be, it will still be unreasonable to aim to realize if it’s sufficiently unlikely to eventuate. (That violence between humans should entirely cease forever, starting tomorrow, would be an almost incomparably great good. But it’s still unreasonable for me to aim to realize this state of affairs, since it’s just not going to happen.)
At least, they do if it’s reasonable for me to accept that memory and induction are reliable, too. I can’t treat these topics here. But note, first, that Sellars takes his pragmatic argument to establish our warrant to accept memory’s reliability, too; and second, that Wright (2004: §VII) offers a similar argument to establish our warrant to accept induction’s reliability. If these extensions succeed, a Sellarsian pragmatic approach can fill this lacuna.
This differentiates my pragmatic response to skepticism from Rinard’s (2017: §9; incidentally, the flying example is hers). I want mine to accommodate Hieronymi’s (2005) judgment—which Rinard rejects—that, generally, our wanting to hold a belief or feeling enjoyment from doing so aren’t merely insufficient reasons to hold it, but positively inappropriate ones. By following Wright’s suggestion, I think we can hold these thoughts together: we can hold that, in the skeptical context, where we have no evidence for any empirical belief as against any other, goods crucial to human flourishing generate pragmatic warrants for accepting that certain basic empirical beliefs of ours are likely to be true, but that these basic beliefs found standards of evidence that constrain our warrant to believe—or even accept—subsequent empirical claims.
I owe the point that follows to Imogen Dickie.
Let me address one point I don’t raise below. Someone might worry that, even if I’m right that the Sellarsian argument yields warrant to accept that some of our cognitive states are likely to be true, this can only be true for acceptances, not beliefs, and so the argument doesn’t establish any conclusion about the attitude it’s reasonable for us to take toward our empirical beliefs. This worry is misplaced. What is true is that, as I conceded in Sect. 3, the argument can only yield pragmatic warrant to accept, not pragmatic warrant to believe. But there’s no reason it cannot yield pragmatic warrant to accept a claim about our beliefs: namely, that they’re likely to be true, and indeed justified. (The objector might retort by asking, if the basis of empirical warrant is not epistemic but merely pragmatic in the way I suggest, why we should ever have any empirical beliefs in the first place, and not only acceptances. But the doxastic involuntarism that motivates the belief/acceptance distinction undermines this retort: we just do have empirical beliefs. As Hume discovered playing backgammon, we can’t get rid of them if we try. What the skeptic threatens, as Wright (2004: p. 210) perceptively argues, is our reflective stance toward those beliefs: in particular, “our right to claim” that they’re justified. It’s this right that the Sellarsian argument aims to safeguard.)
This argumentative strategy—insisting that what non-evidential considerations entitle us to accept regarding our perceptual beliefs is that we’re justified, not merely entitled, to hold them—is indebted to Wright (2004: pp. 207–8).
Compare Wright (2004: p. 206): skeptical arguments show that we cannot claim to know “certain cornerstones” of our procedures of inquiry, and suggest that we are therefore irrational or capricious in “proceeding in the ways we do.” A skeptical solution concedes the former but can resist the latter by identifying a different sort of warrant for accepting these cornerstones, skeptical arguments notwithstanding.
The language of angst is Pritchard’s. I share his opinion that, in light of that angst, “what reasons we [can ultimately offer] for our everyday beliefs will be of a pragmatic, rather than an epistemic nature” (2005: p. 204). But my own account of these reasons is fundamentally pragmatic, not constitutivist (i.e. rooted in the necessary conditions of “the practice of offering grounds in the first place”: ibid.) While my Sellarsian proposal constitutes a hinge epistemology, then—it views empirical justification as beginning from unjustifiable general propositions that “are necessary in order for us legitimately to take perceptual experiences to bear [on] beliefs about […] physical objects” (Coliva, 2016: p. 92)—it’s based on an epistemic construal of hinges, close to one approach suggested by Wright (see my footnotes 9 & 11), not on a framework construal of hinges, like Pritchard’s and Coliva’s accounts (ibid.: pp. 84–86, 94). I think this is necessary for it to avoid psychologism.
References
Churchland, P. (1987). Epistemology in the age of neuroscience. The Journal of Philosophy, 84(10), 544–553.
Cohen, L. J. (1992). An essay on belief and acceptance. Oxford University Press.
Cohen, S. (2002). Basic knowledge and the problem of easy knowledge. Philosophy and Phenomenological Research, 65(2), 309–329.
Coliva, A. (2016). Which hinge epistemology? International Journal for the Study of Skepticism, 6, 79–96.
Davidson, D. (2001). Subjective, intersubjective, objective. Oxford University Press.
Elgin, C. (2010). Persistent disagreement. In R. Feldman & T. Warfield (Eds.), Disagreement (pp. 53–68). Oxford University Press.
Field, R. Undated. “John Dewey (1859–1952).” The Internet Encyclopedia of Philosophy. https://iep.utm.edu/john-dewey/. Accessed 21 Sept 2023.
Harman, G. (1973). Thought. Princeton University Press.
Hieronymi, P. (2005). The wrong kind of reason. The Journal of Philosophy, 102(9), 437–457.
Hookway, C. (2012). The pragmatic maxim. Oxford University Press.
Kripke, S. (1982). Wittgenstein on rules and private language. Harvard University Press.
Lupyan, G. (2017). How reliable is perception? Philosophical Topics, 45(1), 81–106.
Martin, A. (2014). How we hope. Princeton University Press.
McDowell, J. (1994). Mind and world. Harvard University Press.
Peirce, C. S. (1931–58). Collected Papers. 8 vols. Edited by C. Hartshorne and P. Weiss (vols. i–vi), A.W. Burks (vii–viii). Belknap Press.
Pritchard, D. (2005). Scepticism, epistemic luck, and epistemic angst. Australasian Journal of Philosophy, 83(2), 185–205.
Proffitt, D. (2006). Embodied perception and the economy of action. Perspectives on Psychological Science, 1(2), 110–122.
Proffitt, D., & Baer, D. (2020). Perception. St. Martin’s.
Pryor, J. (2000). The skeptic and the dogmatist. Noûs, 34(4), 517–549.
Rinard, S. (2017). No exception for belief. Philosophy and Phenomenological Research, 94(1), 121–143.
Rorty, R. (2000). Response to Akeel Bilgrami. In R. B. Brandom (Ed.), Rorty and His Critics (pp. 262–267). Blackwell.
Sellars, W. (1974). Essays in philosophy and its history. Reidel.
Sellars, W. (1979). More on givenness and explanatory coherence. In G. Pappas (Ed.), Justification and knowledge (pp. 169–182). Reidel.
Sellars, W. (1980). Pure pragmatics and possible worlds. Ridgeview.
Sellars, W. (2007). [1963] Phenomenalism. In K. Scharp & R. Brandom (Eds.), In the space of reasons (pp. 303–349). Harvard University Press.
Shah, N., & Velleman, J. D. (2005). Doxastic deliberation. The Philosophical Review, 114(4), 497–534.
Wright, C. (2004). Warrant for nothing (and foundations for free)? Aristotelian Society, 78(1), 167–212.
Wright, C. (2014). On epistemic entitlement (II): Welfare state epistemology. In D. Dodd & E. Zardini (Eds.), Scepticism and perceptual justification (pp. 213–247). Oxford University Press.
Acknowledgment
Thanks to CJ Guth, Cheryl Misak, Steven Levine, Gurpreet Rattan, Andrew Sepielli, and several anonymous referees; to audiences at Hope College (especially Kate Finley), the fourth European Pragmatism Conference (especially Brandon Beasley), and three times at the University of Toronto (especially Imogen Dickie); and, most of all, to David James Barnett for very helpful discussion of versions of this paper. I also gratefully acknowledge support from the National Endowment of the Humanities in the form of a stipend to attend its 2019 summer school “Philosophical Responses to Empiricism in Kant, Hegel, and Sellars,” which stimulated my development of the Sellarsian argument I defend here. (And thanks to all the summer school participants for wonderful conversations!)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Klemick, G. Perceptual justification and the demands of effective agency. Synthese 203, 34 (2024). https://doi.org/10.1007/s11229-023-04453-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11229-023-04453-3