Authority without privilege: How to be a Dretskean conciliatory skeptic on self-knowledge


Dretske is a “conciliatory skeptic” on self-knowledge. Take some subject S such that (1) S thinks that P and (2) S knows that she has thoughts. Dretske’s theory can be put as follows: S has a privileged way of knowing what she thinks, but she has no privileged way of knowing that she thinks it. There is much to be said on behalf of conciliatory skepticism (“CS” for short) and Dretske’s defense of it. We aim to show, however, that Dretske’s defense fails, in that (in part) if his defense of CS’s skeptical half succeeds, then his defense of CS’s conciliatory half fails. We then suggest a potential way forward. We suggest in particular that the correct way of being a Dretskean conciliatory skeptic is to deny that S has a privileged way of knowing about her thoughts, but to grant that she is nonetheless an authority on her thoughts.

This is a preview of subscription content, log in to check access.


  1. 1.

    Our focus here is on Dretske’s later writings on self-knowledge. For discussion of his earlier writings (Dretske 1994, 1995: Chapter 2, 1999), see, e.g., Aydede (2003), Kemmerling (1999), and Lycan (1999).

  2. 2.

    Dretske believes that his defense of CS can be generalized to mental states and events other than occurrent conscious thoughts (so understood), e.g., perceptual experiences (see Dretske 2003b for relevant discussion). He restricts CS to occurrent conscious thoughts because his “plate is already too full” (Dretske 2012b: p. 2, n. 1). For discussion of other theories of self-knowledge, and for references, see Gertler (2017).

  3. 3.

    How exactly (a) and (b) are to be unpacked is not entirely clear. But Dretske certainly holds that a given way of knowing can be privileged even if it is not infallible (see Dretske 2012a: p. 52). See Byrne (2018: Chapter 1) on “privileged access” and “peculiar access” for relevant discussion. See also Fernández (2013: Chapter 1) on “special access” and “strong access”.

  4. 4.

    This is perhaps most explicit in Dretske (1981b). But see also Dretske (1970, 1971, 1981a, 2005a, b, 2006a). Sawyer (2015) has a different reading. She reads Dretske’s theory of knowledge as a non-contrastivist theory on which knowledge is a binary relation—between a subject and a proposition—holding against a set of alternatives. This is okay for our purposes, however, since our main points about CS and Dretske’s defense thereof would hold even if Sawyer’s reading were assumed. For further discussion of contrastivist epistemologies, see, e.g., Blaauw (2008), Johnsen (2001), Rieber (1998), Schaffer (2005), and Sinnott-Armstrong (2004).

  5. 5.

    We defend the reading of CS just presented in Roche and Roche (2017); see it for further details. We there appeal to some of Dretske’s early work on contrastive focus (e.g., 1972, 1975, 1977) in order to make sense of his theory.

  6. 6.

    See Baumann (2006), Dretske (2005a, b, 2006a), Hawthorne (2005), Jäger (2004), Okasha (1999), Roche and Roche (2017), and Shackel (2006) for relevant discussion.

  7. 7.

    Although this puzzle is similar in some respects to the paradox introduced in Boghossian (1989), there are nonetheless important difference that in our opinion make the former more challenging than the latter. See Roche and Roche (2016) for further discussion of the former.

  8. 8.

    Even if one finds one or more of the above theses implausible, Dretske’s defense of CS is still interesting in that if it succeeds, the three theses are not inconsistent as a set.

  9. 9.

    Dretske sometimes presents himself as a reliabilist on knowledge. But he stresses that he is not a process reliabilist, and that he accepts Knowledge Requires Evidence. See Dretske (2003b: pp. 11–12, n. 15, 2012a: p. 51, n. 3).

  10. 10.

    In addition to his 2003a, see also his 2003c and 2006b. There are important respects in which Dretske’s understanding of McKinsey’s paradox is idiosyncratic. But whether or not CS solves McKinsey’s paradox, Dretske’s claim that his externalism is consistent with CS and therefore also with Privileged Self-Knowledge is of interest. See Sawyer (2015) for an alternative contrastivist answer to McKinsey’s paradox. For general discussion of McKinsey’s paradox, and for references, see Parent (2017).

  11. 11.

    CS is relevant to yet another puzzle concerning self-knowledge. Consider:

    … some writers have worried that if one can know what one thinks, then one can know that one thinks. If so, then the ability to armchair know one’s own thoughts would too easily refute eliminativism about the mental …. (Parent 2017: Sect. 4, emphasis original)

    CS says in part that S has no privileged, or “armchair”, way of knowing the specific claim that she thinks that P. It does not follow from this that S has no privileged way of knowing the general claim that, contra eliminativism, she has thoughts. But that is part of Dretske’s overall view. He would deny that if S has a privileged way of knowing what she thinks, then she thereby has a privileged way of knowing that eliminativism is false.

  12. 12.

    Dretske sometimes writes in terms of “acquaintance” instead of “o-awareness”. See Dretske (2011: p. 429, 2012b: p. 56, 2012c: p. 157).

  13. 13.

    How exactly an o-awareness of a proposition should be understood is unclear. But we are happy to bracket this issue and go along with Dretske’s claim that we are often o-aware of propositions.

  14. 14.

    Recall that for the purpose of setting aside issues of closure, Dretske concedes that S knows that she thinks that P only if she knows that she thinks that P. We suspect that for the same reason, he would concede that you know that S thinks that P only if you know that she thinks that P. This is why we are supposing that you know she is thinking.

  15. 15.

    Imagine, e.g., that S can be o-aware of at most one proposition at a time (and that S knows this). Then, if S knows she is thinking, she can deduce that she is thinking that P.

  16. 16.

    Dretske references a footnote in the passage displayed at the beginning of this section where he qualfies his claim about the evidential worth of propositions. It reads in part:

    You do not have to know (be aware that) p and q are propositions to be aware of, and thereby distinguish, p from q anymore than you have to be aware (know) that X and Y are counterfeit to be visually aware of, and thereby distinguish, X (the $100 counterfeit bill) from Y (the $50 counterfeit bill). (Dretske 2012a: p. 57, fn. 11)

    We fail to see how this supports CSCO. S’s being able to distinguish between the proposition that P and the proposition that Q is one thing. S’s being able to rule out the proposition that she thinks that Q is another thing entirely. (Recall Dretske’s point about being able to distinguish between Baptists even though nothing about them indicates that they are Baptists.) You too can distinguish between the proposition that P and the proposition that Q, but you are not thereby in a position to rule out the proposition that S thinks that Q.

  17. 17.

    The kind of modality that Dretske has in mind here is unclear. Is it logical, technological, or something in between? Notice that a similar question arises with respect to CS. Recall that, for Dretske, S has a privileged way of knowing a given fact only if she has a way of knowing it that is unavailable to others. We assume that whatever kind of modality he has in mind for the latter, he has in mind for the former.

  18. 18.

    Notice that Sarah’s mother’s affirmative answer to the question “does Sarah think that her father is home?” also fails to indicate that Sarah thinks that her father is home. Sarah’s behavior could be misleading, after all.

  19. 19.

    These points are related to a well-known passage from Evans (1982):

    I get myself in a position to answer the question whether I believe that p by putting into operation whatever procedure I have for answering the question whether p. … If a judging subject applies this procedure, then necessarily he will gain knowledge of one of his own mental states: even the most determined sceptic cannot find here a gap in which to insert his knife. (Evans 1982: p. 225)

    The idea that Evans’s procedure is knowledge-yielding arguably runs counter to Knowledge Requires Evidence. If so, and if that thesis is true, then there is a gap in which a skeptic can insert his knife! But, regardless, Evans’s procedure is reliable. Evans-style approaches to the epistemology of belief are defended by, among others, Byrne (2018: Chapter 5), Fernández (2013: Chapter 2), and Roche (2016).

  20. 20.

    Cases can be described where S*’s thinking that Pdoes indicate that S thinks that P (imagine, e.g., that S and S* have minds that are for whatever reason perfectly in sync with one another). When (6) holds—as it typically will—such cases are ruled out.

  21. 21.

    This is a “screening-off” condition to the effect that each member of partition Γ = {C, ~ C} renders D probabilistically irrelevant to A. For recent discussion of screening-off, and for references, see, e.g., Roche (2017) and Shogenji (2017).

  22. 22.

    The literature on confabulation suggests one way in which Pr(A|B) might be less than 0.5. Perhaps S will sometimes sincerely assert that P, not because she thinks that P, but rather because she confabulates. (Although confabulation is typically taken to involve a self-ascription of a mental state, asserting that P is true is closely related to asserting that one thinks that P (Evans 1982).) Suppose, for example, that S is motivated to believe that she thinks that P (and thus to assert that P). Take the “and so on” in (5) to include the condition that S is not confabulating in this way. The current point is that even if you are unsure whether (5) holds, S is still the authority on whether she thinks that P in that Pr(A|B) > Pr(A|D). For discussion of confabulation studies, and for references, see Schwitzgebel (2016: Sect. 4). We thank an anonymous reviewer for pointing out the relevance of confabulation to our project.

  23. 23.

    We claimed above that when (5) holds, S’s affirmative answer to the question of whether P expresses her underlying thought that P. This is central to the foregoing account of authority. The account has this in common with “neo-expressivist” accounts of first-person authority (Bar-On 2004; Finkelstein 2003). There is also a difference, however. Neo-expressivists focus on avowals, i.e., utterances such as “I think that P”. Our account, by contrast, focuses on utterances such as “Yes, P”, which make no mention of thought or even a thinking subject. The former semantically express propositions that ascribe mental states; the latter do not.

  24. 24.

    They can even add in a second conciliatory thesis to the effect that S has both a non-privileged way of knowing that she thinks that P, and a non-privileged way of knowing that she thinks that P.

  25. 25.

    We thank two reviewers at this journal for their helpful comments. We also thank Sarah Paul for feedback on an earlier version of this paper.


  1. Aydede, M. (2003). Is introspection inferential? In B. Gertler (Ed.), Privileged access: Philosophical accounts of self-knowledge (pp. 55–64). Burlington: Ashgate.

    Google Scholar 

  2. Bar-On, D. (2004). Speaking my mind: Expression and self-knowledge. Oxford: Oxford University Press.

    Google Scholar 

  3. Baumann, P. (2006). Information, closure, and knowledge: On Jäger’s objection to Dretske. Erkenntnis, 64, 403–408.

    Article  Google Scholar 

  4. Blaauw, M. (2008). Contesting Pyrrhonian contrastivism. Philosophical Quarterly, 58, 471–477.

    Article  Google Scholar 

  5. Boghossian, P. (1989). Content and self-knowledge. Philosophical Topics, 17, 5–26.

    Article  Google Scholar 

  6. Byrne, A. (2018). Transparency and self-knowledge. New York: Oxford University Press.

    Google Scholar 

  7. Dretske, F. (1970). Epistemic operators. Journal of Philosophy, 67, 1007–1023.

    Article  Google Scholar 

  8. Dretske, F. (1971). Conclusive reasons. Australasian Journal of Philosophy, 49, 1–22.

    Article  Google Scholar 

  9. Dretske, F. (1972). Contrastive statements. Philosophical Review, 81, 411–437.

    Article  Google Scholar 

  10. Dretske, F. (1975). The content of knowledge. In B. Freed, A. Marras, & P. Maynard (Eds.), Forms of representation (pp. 77–93). Amsterdam: North-Holland.

    Google Scholar 

  11. Dretske, F. (1977). Referring to events. Midwest Studies in Philosophy, 2, 90–99.

    Article  Google Scholar 

  12. Dretske, F. (1981a). Knowledge and the flow of information. Cambridge, MA: MIT Press.

    Google Scholar 

  13. Dretske, F. (1981b). The pragmatic dimension of knowledge. Philosophical Studies, 40, 363–378.

    Article  Google Scholar 

  14. Dretske, F. (1994). Introspection. Proceedings of the Aristotelian Society, 94, 263–278.

    Article  Google Scholar 

  15. Dretske, F. (1995). Naturalizing the mind. Cambridge, MA: MIT Press.

    Google Scholar 

  16. Dretske, F. (1999). The mind’s awareness of itself. Philosophical Studies, 95, 103–124.

    Article  Google Scholar 

  17. Dretske, F. (2003a). Externalism and self-knowledge. In S. Nuccetelli (Ed.), New essays on semantic externalism and self-knowledge (pp. 131–142). Cambridge, MA: MIT Press.

    Google Scholar 

  18. Dretske, F. (2003b). How do you know you are not a zombie? In B. Gertler (Ed.), Privileged access: Philosophical accounts of self-knowledge (pp. 1–13). Burlington: Ashgate.

    Google Scholar 

  19. Dretske, F. (2003c). Knowing what you think vs. knowing that you think. In R. Schantz (Ed.), The externalist challenge: New studies on cognition and intentionality (pp. 389–400). Berlin: Walter de Gruyter.

    Google Scholar 

  20. Dretske, F. (2005a). The case against closure. In M. Steup & E. Sosa (Eds.), Contemporary debates in epistemology (pp. 13–26). Malden: Blackwell.

    Google Scholar 

  21. Dretske, F. (2005b). Reply to Hawthorne. In M. Steup & E. Sosa (Eds.), Contemporary debates in epistemology (pp. 43–46). Malden: Blackwell.

    Google Scholar 

  22. Dretske, F. (2006a). Information and closure. Erkenntnis, 64, 409–413.

    Article  Google Scholar 

  23. Dretske, F. (2006b). Representation, teleosemantics, and the problem of self-knowledge. In G. Macdonald & D. Papineau (Eds.), Teleosemantics (pp. 69–84). New York: Oxford University Press.

    Google Scholar 

  24. Dretske, F. (2011). Skeptical doubts about self-knowledge. In S. Bernecker & D. Pritchard (Eds.), The Routledge companion to epistemology (pp. 425–432). New York: Routledge.

    Google Scholar 

  25. Dretske, F. (2012a). Awareness and authority: Skeptical doubts about self-knowledge. In D. Smithies & D. Stoljar (Eds.), Introspection and consciousness (pp. 49–64). Oxford: Oxford University Press.

    Google Scholar 

  26. Dretske, F. (2012b). Doubts about cogito. Grazer Philosophische Studien, 84, 1–17.

    Article  Google Scholar 

  27. Dretske, F. (2012c). I think I think, therefore I am—I think: Skeptical doubts about self-knowledge. In J. Liu & J. Perry (Eds.), Consciousness and the self: New essays (pp. 150–164). New York: Cambridge University Press.

    Google Scholar 

  28. Evans, G. (1982). The varieties of reference. Oxford: Oxford University Press.

    Google Scholar 

  29. Fernández, J. (2013). Transparent minds: A study of self-knowledge. Oxford: Oxford University Press.

    Google Scholar 

  30. Finkelstein, D. (2003). Expression and the inner. Cambridge, MA: Harvard University Press.

    Google Scholar 

  31. Gertler, B. (2017). Self-knowledge. In E. Zalta (Ed.), Stanford encyclopedia of philosophy (Fall 2017 edition),

  32. Hawthorne, J. (2005). The case for closure. In M. Steup & E. Sosa (Eds.), Contemporary debates in epistemology (pp. 26–43). Malden: Blackwell.

    Google Scholar 

  33. Jäger, C. (2004). Skepticism, information, and closure: Dretske’s theory of knowledge. Erkenntnis, 61, 187–201.

    Article  Google Scholar 

  34. Johnsen, B. (2001). Contextualist swords, skeptical plowshares. Philosophy and Phenomenological Research, 62, 385–406.

    Article  Google Scholar 

  35. Kemmerling, A. (1999). How self-knowledge can’t be naturalized (some remarks on a proposal by Dretske). Philosophical Studies, 95, 311–328.

    Article  Google Scholar 

  36. Lycan, W. (1999). Dretske on the mind’s awareness of itself. Philosophical Studies, 95, 125–133.

    Article  Google Scholar 

  37. McKinsey, M. (1991). Anti-individualism and privileged access. Analysis, 51, 9–16.

    Article  Google Scholar 

  38. Okasha, S. (1999). Epistemic justification and deductive closure. Crítica, 31, 37–51.

    Google Scholar 

  39. Parent, T. (2017). Externalism and self-knowledge. In E. Zalta (Ed.), Stanford encyclopedia of philosophy (Fall 2017 edition),

  40. Rieber, S. (1998). Skepticism and contrastive explanation. Nous, 32, 189–204.

    Article  Google Scholar 

  41. Roche, M. (2016). Knowing what one believes—In defense of a dispositional reliabilist extrospective account. American Philosophical Quarterly, 53, 365–379.

    Google Scholar 

  42. Roche, M., & Roche, W. (2016). Review of Declan Smithies and Daniel Stoljar’s (Eds.) Introspection and consciousness (2012, Oxford University Press). Philosophical Quarterly, 66, 203–208.

    Google Scholar 

  43. Roche, M., & Roche, W. (2017). Dretske on self-knowledge and contrastive focus: How to understand Dretske’s theory, and why it matters. Erkenntnis, 82, 975–992.

    Article  Google Scholar 

  44. Roche, W. (2017). A condition for transitivity in high probability. European Journal for Philosophy of Science, 7, 435–444.

    Article  Google Scholar 

  45. Sawyer, S. (2015). Contrastive self-knowledge and the McKinsey paradox. In S. Goldberg (Ed.), Externalism, self-knowledge, and skepticism: New essays (pp. 75–93). Cambridge: Cambridge University Press.

    Google Scholar 

  46. Schaffer, J. (2005). Contrastive knowledge. In T. Gendler & J. Hawthorne (Eds.), Oxford studies in epistemology (Vol. 1, pp. 235–271). Oxford: Oxford University Press.

    Google Scholar 

  47. Schwitzgebel, E. (2016). Introspection. In E. Zalta (Ed.), Stanford encyclopedia of philosophy (Winter 2016 edition),

  48. Shackel, N. (2006). Shutting Dretske’s door. Erkenntnis, 64, 393–401.

    Article  Google Scholar 

  49. Shogenji, T. (2017). Mediated confirmation. British Journal for the Philosophy of Science, 68, 847–874.

    Google Scholar 

  50. Sinnott-Armstrong, W. (2004). Classy Pyrrhonism. In W. Sinnott-Armstrong (Ed.), Pyrrhonian skepticism (pp. 188–207). Oxford: Oxford University Press.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Michael Roche.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.



Suppose that:

(7) 1 > Pr(A|B) = Pr(C|D).
(8) Pr(A|B) > Pr(A|C) > Pr(A).
(9) Pr(A|C & D) = Pr(A|C) and Pr(A | ~ C & D) = Pr(A | ~ C).

The aim is to show that:

(10) Pr(A|B) > Pr(A|D).

It follows from the law of total probability that:

(11) Pr(A|D) = Pr(C|D)Pr(A|C & D) + Pr(~ C|D)Pr(A | ~ C & D).

Given this, and given (7) and (9), it follows that:

(12) Pr(A|D) = Pr(A|B)Pr(A|C & D) + Pr(~ A|B)Pr(A | ~ C & D)
  = Pr(A|B)Pr(A|C) + Pr(~ A|B)Pr(A | ~ C)
  = Pr(A|B)Pr(A|C) + Pr(A | ~ C) − Pr(A|B)Pr(A | ~ C).

It follows from this and (8) that:

(13) Pr(A|B)Pr(A|B) + Pr(A | ~ C) − Pr(A|B)Pr(A | ~ C) >
  Pr(A|B)Pr(A|C) + Pr(A | ~ C) − Pr(A|B)Pr(A | ~ C) =

The following holds without exception:

(14) Pr(A|B) > Pr(A|B)Pr(A|B) + Pr(A | ~ C) − Pr(A|B)Pr(A | ~ C) iff
  Pr(A|B) − Pr(A|B)Pr(A|B) > Pr(A | ~ C) − Pr(A|B)Pr(A | ~ C) iff
  Pr(A|B)[1 − Pr(A | B)] > Pr(A | ~ C)[1 − Pr(A|B)] iff
  Pr(A|B) > Pr(A | ~ C).

Given this, and given that (8) entails that Pr(A|B) > Pr(A | ~ C), it follows that:

(15) Pr(A|B) > Pr(A|B)Pr(A|B) + Pr(A | ~ C) − Pr(A|B)Pr(A | ~ C).

By (13) and (15) it follows that (10) holds. QED

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Roche, M., Roche, W. Authority without privilege: How to be a Dretskean conciliatory skeptic on self-knowledge. Synthese (2019).

Download citation


  • Authority
  • Conciliatory skepticism
  • Contrastivist epistemologies
  • Dretske
  • McKinsey’s paradox
  • Privilege
  • Self-knowledge
  • Thought