The puzzle of transparency and how to solve it

ABSTRACT

According to the transparency approach, achievement of self-knowledge is a two-stage process: first, the subject arrives at the judgment ‘p’; second, the subject proceeds to the judgment ‘I believe that p.’ The puzzle of transparency is to understand why the transition from the first to the second judgment is rationally permissible. After revisiting the debate between Byrne and Boyle on this matter, I present a novel solution according to which the transition is rationally permissible in virtue of a justifying argument that begins from a premise referring to the mental utterance that is emitted in the course of judging ‘p.’

KEYWORDS:

• Transparency approach to self-knowledge
• justification
• belief
• Alex Byrne
• Matthew Boyle

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1. I deal with self-knowledge of non-doxastic attitudes such as wishes, desires, and intentions in Barz (2015). For more on self-knowledge of one’s visual experiences, see Barz (2014).

2. Cf. Gertler (2011), Cassam (2014), Finkelstein (2012).

3. I owe this label to Andi Müller.

4. See Peacocke (1998, 90) for an example that is often assumed to cast in an unfavorable light the thesis that judgment is sufficient for belief.

5. Cf. Byrne (2005, 95, 2018, 74–98). See also Dretske (2003, 2), Evans (1982, 231), Gallois (1996, 47), Martin (1998, 110) and Moran (2003, 413).

6. My formulation slightly deviates from Byrne’s as he uses ‘believe’ instead of ‘judge.’ However, in his (2005), footnote 22, Byrne himself admits that using ‘judge’ would be better.

7. Cf. BonJour (1980, 63).

8. Maybe this is not quite fair to Boyle’s view – for Boyle might claim that knowledge is not a form of awareness but a kind of ability. See Marcus (2016) and Campbell (2018) for defenses of Boyle’s reflectivism along these lines. However, since my chief objection to Boyle’s account does not depend on whether knowledge is a form of awareness or not, I do not pursue the matter any further here. Thanks to David Hunter who drew my attention to this point.

9. Note, again, that the assumption that there are mental words is neutral concerning the question of how mental content is physically encoded. Thus, thinking a mental sentence is not the same as ‘tokening a string of symbols of the language of thought’ in Fodor’s sense. Thanks to Peter Kuhn for encouraging me to be clearer on this point.

10. I owe this objection to Henning Lütje.

11. To forestall possible misunderstandings, let me emphasize that the justification in question licenses the subject merely to believe that the author of the utterance – whoever that may be – is sincere. It does not license the subject to believe that the author of the utterance is identical to herself. Thus, the argument cannot directly proceed to the conclusion ‘I believe that Toronto is in Ontario,’ but needs premise (2) for this purpose.

12. One might object that it is conceptually impossible to sincerely assert p without knowing that it is oneself who asserts p. Thus, nobody can sincerely assert that p but, for example, wonder whether it is him or her who asserts p. However, it seems that my description of the case presupposes that it is possible to sincerely assert that Toronto is located in Ontario without knowing that it is oneself who asserts it. Hence, there is a problem. In my opinion, this objection is based on a misunderstanding. Note that I do not describe a point in time of a real subject’s mental life here. Instead, I describe the logical stage of the justifying argument I am about to formulate. Compare: from a psychological point of view, it might be impossible to believe ‘A is a bachelor’ without believing that A is male – for believing the first proposition without the second would show that one does not master the concept ‘bachelor.’ However, from a logical point of view, the premise ‘A is a bachelor’ does not license one to conclude that A is male; the further premise ‘All bachelors are male’ is needed. So, the fact that one is not licensed to conclude (without further ado) from ‘A is a bachelor’ that A is male does not imply that it is possible to believe that A is a bachelor without believing that A is male. Similarly, in the case at hand, the fact that premise (1) does not license the subject to conclude that she is the one who asserts that Toronto is located in Ontario does not imply that it is possible to sincerely assert that Toronto is located in Ontario without knowing that it is oneself who asserts it. Thanks to Eric Marcus for prompting me to reconsider and improve my thoughts on this matter.

13. I owe this and the following objection to Sarah Paul. I thank her for pressing me on these points.

14. Cf. Enoch and Schechter (2008, 568).

15. This reply might seem perplexing at first sight. One may say: ‘Given that the Enoch and Schechter strategy is a form of externalism, why do you adopt it when justifying the first conjunct of premise (2)? Isn’t that detrimental to the purpose of explaining why transitions from first- to second-order judgments are rational from the subject’s point of view?’ Answer: no, it isn’t. Assume that the first conjunct of premise (2) is justified in the Enoch and Schechter way and that the subject does not believe that believing the first conjunct of premise (2) is indispensable for being rational. This does not imply that the subject does not believe the first conjunct of premise (2). On the contrary, it is perfectly possible that the subject does believe the first conjunct of premise (2), though she has no grasp of its justifier. Note that, to be justified in believing B on the basis of E, one needs to be justified in believing E, but one does not need to have cognitive access to E’s justifier. Hence, the fact that the justifier of the first conjunct of premise (2) might be inaccessible from the subject’s point of view does not mean that the justification of the conclusion ‘I believe that Toronto is located in Ontario’ is likewise inaccessible. In short: even if the subject does not know why she is justified in believing the premise, she is in a position to know why she is entitled to draw the conclusion. Thus, use of the Enoch and Schechter strategy in connection with the first conjunct of premise (2) is not detrimental to the purpose of explaining why transitions from first- to second-order judgments are rational from the subject’s point of view.

16. Byrne (2018, 14). See also Byrne (2005, 92).

17. I consider this to be the received opinion about inferential justification. If the subject were required to actually entertain and accept all propositions that constitute the evidence justifying her belief, then virtually no belief would ever be inferentially justified. Cf. Fumerton (1976, 566).

18. An earlier version of this paper was presented at the workshop ‘Transparency and Apperception’ organized by Boris Hennig, David Hunter, and Thomas Land at Ryerson University, Toronto, in Mai 2018. Thanks to the audience on that occasion for helpful discussion, especially David Barnett, Boris Hennig, Ulf Hlobil, David Hunter, Thomas Khurana, Thomas Land, Eric Marcus, Sarah Paul, Gurpreet Rattan, Houston Smit, and Jonathan Way. For extremely helpful comments on a previous draft many thanks to Philipp Hey, David Hunter, Peter Kuhn, Andi Müller, Henning Lütje, and Sarah Paul. Finally, special thanks to Mark Davies (who knows why).