Skip to main content

Sellars’ metaethical quasi-realism


In this article, I expound and defend an interpretation of Sellars as a metaethical quasi-realist. Sellars analyzes moral discourse in non-cognitivist terms: in particular, he analyzes “ought”-statements as expressions of collective intentions deriving from a collective commitment to provide for the general welfare. But he also endorses a functional-role theory of meaning, on which a statement’s meaning is grounded in its being governed by semantical rules concerning language entry, intra-linguistic, and language departure transitions, and a theory of truth as correct assertibility relative to such semantical rules. On these non-representationalist theories, even though moral statements are expressions of intentions and not fundamentally descriptive, they nevertheless count as assertorically meaningful, and some count as positively true. I further argue that this interpretation is capable not only of explaining Sellars’ explicitly metaethical writings, but also of unifying his scientific realism with his commitment to the ineliminable and indispensable role of the language of intentions: if this linguistic framework does not play an explanatory role, but only an expressive role, this explains both why Sellars’ commitment to it does not contravene his naturalism, as well as why, given the necessity of such language for our practical engagement with the world, the scientific image of humans in the world will only be completed once such language supplements it to enable us to relate practically to it.

This is a preview of subscription content, access via your institution.


  1. See Solomon (1977: p.149) and deVries (2005: p. 246). Other than in overviews of Sellars’ work (deVries 2005: ch. 9; O’Shea 2007: ch. 7), the only treatment of Sellars’ ethics of which I am aware written between 1980 and 2015 is Hurley (2000).

  2. See Olen and Turner (2015, 2016), Koons (2016) and Baumeister (2017). Koons is also presently writing the first monograph devoted to Sellars’ ethics.

  3. It is standard practice to cite Sellars’ writings by abbreviation. See the entries for Sellars’ works in the Bibliography.

  4. This is how Richard Joyce (2015: Supplement to §3) defines the term: “The quasi-realist is someone who endorses an anti-realist metaphysical stance but who seeks, through philosophical maneuvering, to earn the right for moral discourse to enjoy all the trappings of realist talk.” Simon Blackburn, who introduced the term “quasi-realism,” similarly introduces “the quasi-realist” (in general, without specific reference to the moral domain) as “a person who, starting from a recognizably anti-realist position, finds himself progressively able to mimic the intellectual practices supposedly definitive of realism” (Blackburn 1993: pp. 4, 15). Note also the similarities between Blackburn’s “fast-track” method of supporting quasi-realism and Sellars’ argument for it on the basis of his theories of meaning and truth: the fast-track quasi-realist “would make sufficient remarks about truth to suggest that we need a comparable notion to regulate evaluative discouse (even though this is nonrepresentational) and then say that our adherence to propositional forms needs no further explanation than that” (Blackburn 1993: p. 185).

  5. I will use the term “moral statements” to cover those of our moral utterances that have the surface grammar of assertions. I should note, though, that my use of this term isn’t a stipulation that the utterances in question should ultimately be analyzed as assertions or as truth-apt; I’m leaving it open whether our “moral statements” are genuine statements in that sense. (I will sometimes use “moral assertions” or speak of our “asserting moral statements” in a corresponding way).

  6. Of course, Sellars would grant that our moral statements are descriptive in a derivative sense, since he does hold that they are truth-apt; this distinction is analogous to that between generically factual statements and statements that are matter-of-factual in Sellars’ special sense (SM: ¶¶V.1−2). Hereafter I will sometimes leave such qualifiers as “fundamentally” implicit when talking of moral statements as non-descriptive and non-representational.

  7. In turn, Sellars holds, there is a connection, ceteris paribus, between intending to do A at some later point in time and, at that point in time, having a volition to do A (and so, if time t1 is earlier than time t2, between sincerely uttering at t1 ‘I shall do A at t2’ and sincerely uttering at t2 ‘I shall now do A’).

  8. Sellars thus risks misleading the reader when he suggests that hypothetical imperatives contain “an ‘ought’ which pertains to the coherence of our valuings” (OMP: ¶25; cf. ¶¶33, 37). Hypothetical imperatives assert relations of implication, and so they imply relations of incoherence (OMP: ¶¶19−20). But they do not constitute, but rather only presuppose, imperatives of coherence with respect to intentions.

  9. Indeed, this seems to be true even of the prudential “ought.” Sellars will attempt to secure the “univocity” of the prudential “ought” by a parallel argument to the one he develops concerning the moral “ought,” but one centered on the idea that the intention “Would that I lived a satisfying life all things considered” is categorically reasonable, since “intentions which are arrived at by taking all relevant considerations into account are, in so far forth, reasonable” (ORAV: ¶123). I do not find this argument convincing, since it fails to explain why the intention in question should be rationally required as against the equally all-encompassing “Would that I lived an unsatisfying life, all things considered.” (Of course, few are tempted to adopt this latter intention rather than the former. But this does not in itself help to meet Sellars’ burden, which is to adduce rational rather than psychological considerations that favor the former over the latter). But in any case, the parallel character of the argument lends support to my thesis that Sellars thinks he can capture the dimension of endorsement in moral “ought”-statements, and thus the univocity of the moral “ought,” without going beyond the strictures of expressivism.

  10. Thus Hurley (Hurley 2000: §§2−3) suggests that Sellars held that “ought” is a special case of “shall” only while he held that “all practical reasonableness is relative reasonableness” (p. 298); once he came to accept that there can be categorical reasonableness in the practical sphere, he recognized that “ought”-statements go beyond “shall”-statements (and so, seemingly, cannot be analyzed by them), since they do not merely express intentions but also rationally endorse them. On my reading, Hurley is correct to note that “ought”-statements, unlike “shall”-statements, involve rational endorsement. But Sellars’ constitutivist account of categorical reasonableness enables him to capture this dimension of endorsement without having to deny that “ought” is a special case of “shall”.

  11. Besides other passages throughout IILOR, Sellars states this position explicitly at TA: 106 and SE: 408; see also CDCM: ¶78.

  12. I discuss Sellars’ theory of truth in §2, but see also SM: ¶VII.92.

  13. As he states explicitly in ORAV: “I am not saying that everybody shares this shareable intention. I am simply saying that it defines the moral point of view” (¶207).

  14. Of course, not all moral statements are of the form “I ought to do A”; we might wonder whether Sellars can adequately account for statements of other forms. He can account for statements of the more general form “S ought to do A” as expressions of the speaker’s intention, qua member of the collective, that S should do A as a requirement of the community’s non-derivative collective intention. It is less obvious that he can account for past-tense moral statements, since intentions and their expressions seem essentially present- and future-directed; Sellars discusses this difficulty of “practical discourse in the historical mode” briefly in IILOR: §9, but his discussion is far too brief to resolve it satisfactorily. Finally, Sellars holds that both “‘ought’ and ‘good’ are special cases of ‘shall’” (TA: 106; emphasis added; cf. SE: 408). As far as I am aware, he does not state explicitly his view of how “good”-statements should be analyzed in terms of “shall”-statements, but presumably it would rely on the account of preference in terms of intentions he goes on to give in “Thought and Action.” Thanks to Wes Siscoe for pressing me on this point.

  15. In a certain stipulated sense of the phrase: for Sellars, matter-of-factual discourse is that which pictures—or stands in a particular complicated network of non-semantic relations to objects in—the world. I do not have space to explicate or assess Sellars’ theory of picturing here, but it is obviously an important element of the background of my argument, as it is ultimately what prevents Sellars’ quasi-realism from collapsing into full-blown moral realism. An overview of the theory of picturing may be found in deVries (2005: pp. 50−56), and more extended accounts and discussions may be found in deVries (2010), Rosenberg (2007) and Seibt (2009), to name just a few. For objections to the theory of picturing from one sympathetic to Sellars’ general philosophy of language, see Rorty (1988); for criticism of objections in this general vein, see O’Shea (2010), especially pp. 466−467, as well as Levine (2007) and, from a somewhat different point of view, Seiberth (2018: ch. 4).

  16. The distinction between representationalist and use-theoretic theories of meaning is due to Brandom (1976). (Brandom does not label the non-representationalist approach, but its proponents do seem to be unified in accounting for meaning in terms of proper use or assertion).

  17. On such a view, the statement predicates a semantic relation (meaning) between a linguistic entity (“dreieckig”) and a real, abstract entity (triangularity). But, for one thing, “dreieckig” means triangular, not triangularity. So how can the sentence relate it to triangularity? For another, this account seems to founder on statements like “Und” (in German) means and: what is the abstract entity to which this sentence relates “und”? (Conjunctivity? That seems implausible). For a fuller rehearsal of the problems in this region, together with a brief sketch of the solution Sellars defends at more length in MFC, see G&E: §XII.

  18. Actually, Sellars would want to avoid the apparent reference to a linguistic abstract entity in this sentence’s subject by rendering it instead in terms of either “the ‘dreieckig’” or simply “dreieckig”s. (See MFC: p. 94). But this complication is not relevant for my purposes here.

  19. Indeed, Sellars thinks that not merely meaning, but all “semantic relations” are pseudo-relations, and he offers a parallel treatment of reference (for a brief overview, see deVries 2005: pp. 34−36). Compare Rorty (1988: pp. 151−152).

  20. Though Sellars never makes this explicit, the paragraphs that follow make clear why the sort of proper assertibility that constitutes truth is concerned only with the rules governing language entry transitions and intralinguistic moves, not those governing language departure transitions.

  21. Michael Williams argues that this is in effect an epistemic theory of truth, “even though it is not stated in obviously epistemic terms.” This is because it defines truth in terms of assertibility in accordance with semantical rules, and the semantical rules in question are “thoroughly epistemic” (2016: p. 232): though they may license me in asserting something false, this is only because they make provision for increases in my information, with the result that truth is proper assertibility in epistemically ideal conditions (2016: p. 239). But in his manuscript “Sellars, Truth Pluralism, and Truth Relativism,” Lionel Shapiro notes Sellars’ insistence that to confuse the proper assertibility he identifies with truth with the epistemic notion of warranted assertibility is “to confuse truth with probability” (NI: 664). Shapiro argues that the additional information in terms of which Sellars defines truth is not constrained by what is accessible to the speaker, either at the time or ideally; it is simply information that an assessor would require to determine whether the assertion complies with the relevant semantical rules.

    Given that he grants that Sellars’ theory of truth is not transparently epistemic, the burden of proof thus would seem to fall to Williams to establish that Sellars constrains the correctness conditions specified by semantical rules by reference to the information that is accessible in principle to the speaker. In any case, the point that is important for my purposes—namely, that Sellars defines truth in terms of assertibility licensed by semantical rules, and so allows for the truth-aptness of statements that do not describe entities countenanced by his naturalistic ontology—is orthogonal to this dispute. (Thanks to a reviewer for this journal for prompting me to consider Williams’ position more carefully in light of Shapiro’s, as well as to Shapiro for kindly sharing his manuscript with me and permitting me to refer to it here).

  22. For, on the one hand, imperatives are subject to something like semantical rules of criticism, which rules specify their meanings. Obviously there is a conceptual tie between the meaning of an imperative and some behavioral output: one who uses imperatives without recognizing that they prescribe some particular behavioral responses and proscribe others uses them incorrectly. But, on the other hand, Sellars insists that there cannot be “reasonings which have imperatives as premise or conclusion” (IILOR: §3)—and so, by an inferentialist criterion of meaning, that they are not assertorically meaningful or truth-apt.

  23. As we saw above, Sellars explains the intersubjectivity of instrumental “ought”-statements differently, grounding it in the intersubjective bindingness of statements of causal necessity.

  24. Thus Sellars holds that “‘We would that…’ lacks the logical privacy of ‘I would that…’” (SM: ¶VII.127).

  25. Brandom (2002) argues that, given Sellars’ rejection of an ontological account of the distinction between observable and unobservable objects, he must allow that, in principle, moral statements could properly figure in language entry transitions as well.

  26. See footnote 9.

  27. I take this to be Sellars’ summary of Peirce’s argument at the end of his article “Grounds of Validity of the Laws of Logic” (1869) that an individual’s reliance on inductive inference in the short term can be justified only by her identification of herself with the whole community of inquiry and her elevation of its interests above her purely personal ones. In Peirce’s judgment: “He who would not sacrifice his own soul to save the whole world is illogical in all his inferences, collectively.” And thus the theory that the human individual can only be motivated to act by desires for personal gain or pleasure is “reduced to an absurdity” by “our principles of the objectivity of knowledge” (1869: p. 81).

  28. The textual evidence below establishes this with respect to Science and Metaphysics. One might suggest that Sellars abandoned his quasi-realism in the twelve-year period between that work and “On Reasoning about Values” (1980), since in the latter work, Sellars states that the question “Why should I be moral?” is “a perfectly meaningful one,” without offering even the start of an answer to it (¶208). Does this suggest that Sellars ceased to view the foundational collective intention of the moral community as in any way rationally defensible, and so returned to mere expressivism? I do not think so: in the latter sections of this work, Sellars is attempting to argue only that the idea of a moral point of view (as against an egoistic one) is coherent, and not that this point of view is rationally required, or that the considerations it takes to be reasons are genuinely binding. Thus Sellars’ failure to respond to moral skepticism in this article should be seen as determined by the scope of his argument, and not necessarily as due to a lack of conviction in the possibility of such a response.

    I should mention one other possible interpretation of Sellars’ late stance on this point: it may be that he returned to the position suggested in a passage in IILOR that precedes the eudaimonist argument, in which he had suggested that, though our moral discourse presupposes the existence of a community with shared collective intentions, it does not guarantee this; rather, our moral discourse “set[s] this agreement as a task. To abandon the idea that disagreement on moral matters is even capable of resolution is not to retreat to a moral solipsism; it is to abandon the moral framework itself, and to retreat to the language or ‘form of life’ of purely personal intention” (IILOR: §12). If Sellars did ultimately take this to be all that can be said in defense of the moral point of view, then, though I still think it would be fair to describe him as a quasi-realist, we would have to caveat this claim by noting that he accorded very different statuses to the two components of his quasi-realism: while his expressivism resulted from a straightforward analysis of moral discourse, his view of moral discourse as truth-apt would have to be seen as a regulative assumption of the moral point of view itself, one that does not admit of independent justification. But it should be noted that Sellars never returns to this line of thought, either, in his later metaethical writings, so that to my knowledge there is no textual basis for concluding that he eventually came once more to endorse it. Indeed, the very fact that Sellars saw the need to supplement it with the eudaimonist argument in the revised version of IILO raises the question of whether he could ultimately have thought it constituted adequate support for the moral point of view on its own. (My thanks to Michael Hicks for suggesting to me this possible Sellarsian response).

  29. Sellars is primarily concerned in this passage with denying that the aforementioned assertions entail that “the time for scientists to abandon the framework of common sense is now” (SRII: 354). Since the point at issue for my purposes is different, I have removed some of Sellars’ italics and added my own.

  30. I am unsure whether I am disagreeing here with Willem deVries’ reading of this passage. DeVries initially states that he rejects the idea that Sellars means that, in the completed scientific image, “we shall let science do all the describing and explaining, and the presence of practical discourse in our language need amount to nothing more than the presence of the intention operator ‘shall’” (2005: p. 274). I think this is exactly what Sellars means. We see additional evidence for this interpretation, for instance, in “Counterfactuals, Dispositions, and the Causal Modalities,” where Sellars contends that “neither ‘ought’ nor any other prescriptive description could be used” in a naturalistically acceptable description of the world (CDCM: ¶79). This is precisely the idea that science is to do “all the describing,” to the exclusion of our practical or prescriptive discourse.

    But as deVries’ argument against this interpretation progresses, he seems really to be claiming only that it would be much more practical to retain the shorthand term “hammer” than to force all those venturing to the hardware store to utter complex scientific descriptions, supplemented by complex statements about intentions (both individual and collective) for pounding activities. This is clearly correct, but if we take “hammer” to be precisely a shorthand for such an impure description (in which a rudimentary naturalistic description is supplemented by the ascription of a function to—that is, the expression of intentions concerning—the object), then we can accommodate this point within the interpretation of Sellars I am proposing.

  31. But recall footnote 6. Like his denial that moral statements correspond to facts, Sellars’ denial that moral statements are robustly or fundamentally descriptive is grounded in his theory of picturing. See footnote 15.

  32. That Sellars’ target is limited in this way is confirmed both by the preceding paragraph, which makes clear that the contrast position at issue is an empiricist instrumentalism, as well as by the following paragraph’s summary of the thesis just defended: “the epistemological thesis that such a direct use of theoretical language in perceptual response to the world could stand on its own feet” (SM: ¶V.91; emphasis added).

  33. I go on to argue for this claim by examining Sellars’ clarifications of his position in the expanded version of “Obligation and Motivation.” It is thus open for someone to object that Sellars might have endorsed an error theory of Mackie’s sort in the 1951 version of the paper, expanding the paper to revise his view rather than simply to state it more felicitously. But given that, as we have seen, Sellars had already signaled his support for an expressivist stance 2 years earlier, in his 1949 “Language, Rules, and Behavior,” I think it is more plausible to read the expansion of “Obligation and Motivation” as intended to clarify his position in the original version rather than to correct it.

  34. See Olen and Turner (2015: p. 8n3).

  35. This is what we should expect, since Olen and Turner commit themselves to the quasi-realist reading’s two major theses. They note both that Sellars analyzes moral statements as expressions of collective intentions (2015: p. 13, 2016: pp. 2064−2065), and also that he does hold that “there is truth and falsity with respect to obligation” (2016: p. 2059).

  36. Some may have doubts about the viability of such a distinction; I shall say something by way of addressing these in footnote 39.

  37. They single out Willem deVries and James O’Shea (Olen and Turner 2016: p. 2070n66).

  38. On the next page, they state this outright, characterizing O’Shea’s interpretation as attributing to Sellars the claim that “Normativity and practical reasoning, as sui generis categories of explanation, are needed to account for everyday human agency and action, despite their omission from the final theoretical description of the world” (2016: p. 2071). It makes sense that they would oppose this position: if normativity is really an explanatory category, then there is no basis for exempting it from competition with scientific explanatory categories to earn its keep in an ideal description of the world.

  39. This distinction between analysis and explanation is obviously the linchpin of my reading of the relation Sellars posits between the natural and the normative. It is because our “ought”-statements cannot be analyzed in naturalistic, descriptive terms that Sellars can preserve the boundary between the natural and the normative (reflected in his statement that “‘ought’ has as distinguished a role in discourse as descriptive and logical terms” [CDCM: ¶79], and consequently in his claim that the scientific image must be enriched with the language of intentions for it to be complete). And it is because our acts of making “ought”-statements can be fully causally explained without appeal to any non-natural facts or entities that Sellars can maintain that this boundary between the natural and the normative does not compromise the ontological naturalism to which he commits himself in, e.g., the scientia mensura passage. Thus it is natural for readers to wonder whether this distinction can really bear all the weight I take Sellars to place on it.

    In reply, if Sellars held that “ought”-statements could not be analyzed in naturalistic, descriptive terms for the reasons that the intuitionists held this, then there would indeed appear to be a problem with his attempting to maintain at the same time that our practice of making normative assertions could be fully causally explained in naturalistic terms. If the reason our normative statements cannot be analyzed in naturalistic terms were that they posit non-natural facts or entities, then it would be unclear how we could be warranted in making any such statements without having epistemic access to these facts or entities. And it is not obvious, to say the least, how such access could be secured by the naturalistic causal processes that give rise to our normative assertions. If such a causal explanation of our normative assertions were not incomplete, then, seemingly it would undermine the warrant we have to make such assertions. Therefore, while an error theory on which all our normative statements are false—or at least unwarranted—could easily affirm that such statements are logically irreducible but causally reducible to natural ones, it is certainly not clear that a moral realist could be warranted in affirming this. By contrast, Sellars analyzes “ought”-statements as expressions of intentions, and so as fundamentally non-descriptive. But then it seems straightforward, or even trivial, to hold that such “statements” cannot be analyzed in descriptive terms, and yet that agents’ utterances of such “statements” can be fully causally accounted for in naturalistic terms. This should seem no more mysterious than that my utterance that “I shall go to the Penguins game tomorrow”—following Sellars in reading this as an expression of an intention, rather than a self-ascription of one or a prediction of my future behavior—cannot be analyzed in naturalistic, descriptive terms (since any such would-be analysans will be truth-apt, as this utterance is not), without this having the least tendency to show that my utterance cannot be fully causally explained in naturalistic terms. (I do not see that normative statements’ exhibiting the sorts of semantic discipline that “earn us the right” to talk of them as descriptive and truth-apt conflicts with this point). Thus, on Sellars’ account of normative statements, they do seem to constitute utterances whose making can be explained in naturalistic terms, but that cannot be analyzed in such terms.

    But a deeper worry might be raised at this point: even if a prima facie case can be made for the distinction between analyzing our normative statements and explaining our acts of making them, what is the point of Sellars’ account of our normative language, if not to explain our practices or our behavior? For one thing, I don’t think Sellars’ quasi-realist account is totally orthogonal to the project of explaining our normative practices or language-use: it bears on that project at two points. First, it provides a general orientation for this project: by analyzing moral statements as expressions of collective intentions, Sellars suggests to us that an adequate causal explanation of our uses of normative language will centrally make reference to the causal processes that give rise to and disseminate collective intentions. But second, it justifies naturalistic constraints on this explanatory project: in showing that the meanings of our normative statements and the phenomenologies of agency and of normative judgment can be accounted for without violating naturalistic ontological strictures, it justifies us in maintaining those strictures in the project of causally explaining our normative utterances. In my view, then, the best answer to this deeper worry is to say that the point of Sellars’ account of normative language is to open space for this naturalistic explanatory project by showing that we can do justice to the distinctive logical and phenomenological features of normative language and its use without compromising our ontological naturalism. And since, in Sellars’ view, doing justice to these phenomena requires maintaining such endoxa as that “ought”-statements cannot be analyzed in terms of descriptive ones; that they are truth-apt; and that they express weightier states of mind than simple emotions, his quasi-realist account of them is in harmony with—because the necessary precursor to and warrant for—the naturalistic explanatory project, not in tension with it. (Thanks to a reviewer for this journal for pushing me to think more carefully about the analysis/explanation distinction and the purpose it serves).

  40. Here I take my account to agree with deVries’ claim that, for Sellars, “persons are practically real, although not fundamental ontological realities,” and that this practical reality is “a matter of the truth of prescriptive and normative claims, and that, in turn, is a matter of recognized, intersubjectively held, intersubjectively applicable, shared intentions” (2005: p. 277).


  • Baumeister, D. (2017). Social conceptions of moral agency in Hegel and Sellars. International Journal of Philosophical Studies,25(2), 249–265.

    Article  Google Scholar 

  • Blackburn, S. (1993). Essays in quasi-realism. Oxford: Oxford University Press.

    Google Scholar 

  • Brandom, R. (1976). Truth and assertibility. The Journal of Philosophy,73(6), 137–149.

    Article  Google Scholar 

  • Brandom, R. (2002/2015). The centrality of Sellars’s two-ply account of observation to the arguments of ‘Empiricism and the Philosophy of Mind.’ Reprinted in From empiricism to expressivism: Brandom reads Sellars (pp. 99–119). Cambridge, MA: Harvard University Press, 2015.

  • deVries, W. (2005). Wilfrid Sellars. Montreal & Kingston: McGill-Queens University Press.

    Book  Google Scholar 

  • deVries, W. (2010). Naturalism, the autonomy of reason, and pictures. International Journal of Philosophical Studies,18(3), 395–413.

    Article  Google Scholar 

  • Hurley, P. (2000). Sellars’s ethics: Variations on Kantian themes. Philosophical Studies,101, 291–324.

    Article  Google Scholar 

  • Joyce, R. (2015). Moral anti-realism. Rev. ed. In E. N. Zalta (gen. ed.), The Stanford encyclopedia of philosophy.

  • Koons, J. R. (2016). Toward a sellarsian ethics for the 21st century. In D. Pereplyotchik & D. Barnbaum (Eds.), Sellars and contemporary philosophy (pp. 36–52). New York: Routledge.

    Google Scholar 

  • Levine, S. M. (2007). The place of picturing in Sellars’ synoptic vision. The Philosophical Forum,38(3), 247–269.

    Article  Google Scholar 

  • Mackie, J. L. (1977). Ethics: Inventing right and wrong. Harmondsworth: Penguin.

    Google Scholar 

  • Olen, P., & Turner, S. (2015). Durkheim, Sellars, and the origins of collective intentionality. British Journal for the History of Philosophy,23(5), 1–22.

    Article  Google Scholar 

  • Olen, P., & Turner, S. (2016). Was Sellars an error theorist? Synthese,193, 2053–2075.

    Article  Google Scholar 

  • O’Shea, J. R. (2007). Wilfrid Sellars: Naturalism with a normative turn. Cambridge: Polity Press.

    Google Scholar 

  • O’Shea, J. R. (2010). Normativity and scientific naturalism in Sellars’ ‘Janus-Faced’ space of reasons. International Journal of Philosophical Studies,18(3), 459–471.

    Article  Google Scholar 

  • Peirce, C. S. (1869/1992). Grounds of validity of the laws of logic: Further consequences of four incapacities. Reprinted in N. Houser & C. Kloesel (Eds.), The essential Peirce (Vol. 1, pp. 56–82). Bloomington, IN: Indiana University Press, 1992.

  • Price, H. (2011). Naturalism without mirrors. Oxford: Oxford University Press.

    Google Scholar 

  • Rorty, R. (1988/1991). Representation, social practise, and truth. Reprinted in Objectivity, relativism, and truth (pp. 151–161). Cambridge: Cambridge University Press, 1991.

  • Rosenberg, J. F. (2007). Sellarsian picturing. In Wilfrid Sellars: Fusing the images (pp. 104–126). Oxford: Oxford University Press.

  • Seiberth, L. C. (2018). Intentionality in Sellars. Ph.D. dissertation, Universität Leipzig.

  • Seibt, J. (2009). Functions between reasons and causes: On picturing. In W. A. deVries (Ed.), Empiricism, perceptual knowledge, normativity, and realism: Essays on Wilfrid Sellars (pp. 247–282). Oxford: Oxford University Press.

  • Sellars, W. (1949). LRB. Language, rules and behavior. Reprinted in J. Sicha (Ed.) Pure pragmatics and possible worlds: The early essays of Wilfrid Sellars (pp. 129–155). Reseda, CA: Ridgeview Publishing Company, 1980.

  • Sellars, W. (1952). OMR. Obligation and motivation. Rev. ed. 1952. In W. Sellars & J. Hospers (Eds.), Readings in ethical theory (pp. 511–517). New York: Appleton-Century-Crofts. (Original version published in 1951).

  • Sellars, W. (1953) SSMB. A semantical solution of the mind-body problem. Reprinted in J. Sicha (Ed.) Pure pragmatics and possible worlds: The early essays of Wilfrid Sellars (pp. 219–256). Reseda, CA: Ridgeview Publishing Company, 1980.

  • Sellars, W. (1956). EPM. Empiricism and the philosophy of mind (p. 1997). Cambridge, MA: Harvard University Press. (Monograph reprint).

    Google Scholar 

  • Sellars, W. (1957). CDCM. Counterfactuals, dispositions, and the causal modalities. In H. Feigl, M. Scriven, & G. Maxwell (Eds.), Minnesota studies in the philosophy of science (Vol. II, pp. 225–308). Minneapolis: University of Minnesota Press.

    Google Scholar 

  • Sellars, W. (1960). G&E. Grammar and existence: A preface to ontology. Reprinted in K. Scharp & R. B. Brandom (Eds.), In the space of reasons: Selected essays of Wilfrid Sellars (pp. 126–162). Cambridge, MA: Harvard University Press, 2007.

  • Sellars, W. (1962). PSIM. Philosophy and the scientific image of man. Reprinted in K. Scharp & R. B. Brandom (Eds.), In the space of reasons: Selected essays of Wilfrid Sellars (pp. 369–408). Cambridge, MA: Harvard University Press, 2007.

  • Sellars, W. (1963). IILOR. Imperatives, intentions, and the logic of ought. Rev. ed. 1963. In H. -N. Castañeda and G. Nakhnikian (Eds.), Morality and the language of conduct (pp. 159–214). Detroit: Wayne State University Press. Reprinted online (Original version published in 1956).

  • Sellars, W. (1964). NI. Notes on intentionality. The Journal of Philosophy,61(21), 655–664.

    Article  Google Scholar 

  • Sellars, W. (1965). SRII. Scientific realism or irenic instrumentalism: A critique of Nagel and Feyerabend on theoretical explanation. Reprinted in Philosophical perspectives, (pp. 337–369). Springfield, IL: Charles C. Thomas Books, 1967.

  • Sellars, W. (1966). TA. Thought and action. In K. Lehrer (Ed.), Freedom and Determinism (pp. 105–139). New York: Random House.

    Google Scholar 

  • Sellars, W. (1967) SE. Science and ethics. In Philosophical perspectives, (pp. 389–412). Springfield, IL: Charles C. Thomas Books.

  • Sellars, W. (1968). SM. Science and metaphysics: Variations on Kantian themes. Reprint. Atascadero, CA: Ridgeview Publishing Company, 1992.

  • Sellars, W. (1974). MFC. Meaning as functional classification: A perspective on the relation of syntax to semantics. Reprinted in K. Scharp & R. B. Brandom (Eds.), In the space of reasons: Selected essays of Wilfrid Sellars (pp. 81–100). Cambridge, MA: Harvard University Press, 2007.

  • Sellars, W. (1975). AR. Autobiographical reflections (February, 1973). In H.-N. Castañeda (Ed.), Action, knowledge and reality: Studies in honor of Wilfrid Sellars (pp. 277–293). New York: Bobbs-Merrill.

    Google Scholar 

  • Sellars, W. (1980). ORAV. On reasoning about values. American Philosophical Quarterly,17(2), 81–101.

    Google Scholar 

  • Sellars, W. OMP. ‘Ought’ and Moral Principles. Unpublished manuscript. Delivered as a lecture in Pittsburgh on February 14, 1966. Available online

  • Solomon, W. D. (1977). Ethical theory. In C. F. Delaney, M. J. Loux, G. Gutting, & W. D. Solomon (Eds.), The synoptic vision: Essays on the philosophy of Wilfrid Sellars (pp. 149–188). Notre Dame, IN: University of Notre Dame Press.

    Google Scholar 

  • Williams, M. (2016). Pragmatism, Sellars, and truth. In J. R. O’Shea (Ed.), Sellars and his legacy (pp. 223–260). Oxford: Oxford University Press.

    Chapter  Google Scholar 

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Griffin Klemick.

Additional information

I am grateful to Philip Clark, Michael Hicks, Philip Kremer, Cheryl Misak, Wes Siscoe, Evan Simpson, and audiences at meetings of the Canadian Philosophical Association and the Society for the Study of the History of Analytical Philosophy, as well as two anonymous reviewers for this journal, for their helpful comments on drafts or presentations of this paper. I am also grateful to the discussion group on Sellars that met in Cambridge in the Lent term of 2015—especially Luz Christopher Seiberth, its organizer, as well as James Hutton—and to the Skype group formed out of it, for providing a cheerful context in which to puzzle over Sellars’ work. Finally, I wish to acknowledge the Balzan Foundation−University of Toronto Styles of Reasoning project, inaugurated by Ian Hacking, for its support.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Klemick, G. Sellars’ metaethical quasi-realism. Synthese 197, 2215–2243 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Wilfrid Sellars
  • Quasi-realism
  • Collective intentions
  • Error theory
  • Naturalism
  • Normativity
  • Scientific realism